research on dog training methods

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Publications
Account settings
My Bibliography
Collections
Citation manager

Save citation to file

Email citation, add to collections.

Create a new collection
Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

Search in PubMed
Search in NLM Catalog
Add to Search

Improving dog training methods: Efficacy and efficiency of reward and mixed training methods

Affiliations.

1 Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal.
2 i3S -Instituto de Investigação e Inovação em Saúde, Universidade do Porto, Porto, Portugal.
3 Polícia de Segurança Pública, Lisbon, Portugal.
4 CINAMIL, The Military Academy Research Center of the Portuguese Army, Lisbon, Portugal.
PMID: 33606822
PMCID: PMC7895348
DOI: 10.1371/journal.pone.0247321

Dogs play an important role in our society as companions and work partners, and proper training of these dogs is pivotal. For companion dogs, training helps preventing or managing dog behavioral problems-the most frequently cited reason for relinquishing and euthanasia, and it promotes successful dog-human relationships and thus maximizes benefits humans derive from bonding with dogs. For working dogs, training is crucial for them to successfully accomplish their jobs. Dog training methods range widely from those using predominantly aversive stimuli (aversive methods), to those combining aversive and rewarding stimuli (mixed methods) and those focusing on the use of rewards (reward methods). The use of aversive stimuli in training is highly controversial and several veterinary and animal protection organizations have recommended a ban on pinch collars, e-collars and other techniques that induce fear or pain in dogs, on the grounds that such methods compromise dog welfare. At the same time, training methods based on the use of rewards are claimed to be more humane and equally or more effective than aversive or mixed methods. This important discussion, however, has not always been based in solid scientific evidence. Although there is growing scientific evidence that training with aversive stimuli has a negative impact on dog welfare, the scientific literature on the efficacy and efficiency of the different methodologies is scarce and inconsistent. Hence, the goal of the current study is to investigate the efficacy and efficiency of different dog training methods. To that end, we will apply different dog training methods in a population of working dogs and evaluate the outcome after a period of training. The use of working dogs will allow for a rigorous experimental design and control, with randomization of treatments. Military (n = 10) and police (n = 20) dogs will be pseudo-randomly allocated to two groups. One group will be trained to perform a set of tasks (food refusal, interrupted recall, dumbbell retrieval and placing items in a basket) using reward methods and the other group will be trained for the same tasks using mixed methods. Later, the dogs will perform a standardized test where they will be required to perform the trained behaviors. The reliability of the behaviors and the time taken to learn them will be assessed in order to evaluate the efficacy and efficiency, respectively, of the different training methods. This study will be performed in collaboration with the Portuguese Army and with the Portuguese Public Security Police (PSP) and integrated with their dog training programs.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Grants and funding

Linkout - more resources, full text sources.

Europe PubMed Central
PubMed Central
Public Library of Science

Other Literature Sources

scite Smart Citations

Miscellaneous

NCI CPTAC Assay Portal

Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare

Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliations Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal, i3S –Instituto de Investigação e Inovação em Saúde, Universidade do Porto, Porto, Portugal, Instituto de Ciências Biomédicas Abel Salazar, Universidade do Porto, Porto, Portugal

Roles Formal analysis, Investigation, Writing – review & editing

Affiliations Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal, i3S –Instituto de Investigação e Inovação em Saúde, Universidade do Porto, Porto, Portugal, Royal (Dick) School of Veterinary Studies, The University of Edinburgh, Edinburgh, United Kingdom

Roles Formal analysis, Writing – review & editing

Affiliations Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal, i3S –Instituto de Investigação e Inovação em Saúde, Universidade do Porto, Porto, Portugal

Roles Formal analysis, Investigation

Roles Conceptualization, Resources, Supervision

Affiliation Instituto de Ciências Biomédicas Abel Salazar, Universidade do Porto, Porto, Portugal

Roles Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Validation, Writing – review & editing

Ana Catarina Vieira de Castro,
Danielle Fuchs,
Gabriela Munhoz Morello,
Stefania Pastur,
Liliana de Sousa,
I. Anna S. Olsson

Published: December 16, 2020
https://doi.org/10.1371/journal.pone.0225023
See the preprint
Peer Review
Reader Comments

Dog training methods range broadly from those using mostly positive punishment and negative reinforcement (aversive-based) to those using primarily positive reinforcement (reward-based). Although aversive-based training has been strongly criticized for negatively affecting dog welfare, there is no comprehensive research focusing on companion dogs and mainstream techniques, and most studies rely on owner-reported assessment of training methods and dog behavior. The aim of the present study was to evaluate the effects of aversive- and reward-based training methods on companion dog welfare within and outside the training context. Ninety-two companion dogs were recruited from three reward-based schools (Group Reward, n = 42), and from four aversive-based schools, two using low proportions of aversive-based methods (Group Mixed, n = 22) and two using high proportions of aversive-based methods (Group Aversive, n = 28). For evaluating welfare during training, dogs were video recorded for three sessions and six saliva samples were collected, three at home (baseline levels) and three after training (post-training levels). Video recordings were used to examine the frequency of stress-related behaviors (e.g., lip lick, yawn) and the overall behavioral state of the dog (e.g., tense, relaxed), and saliva samples were analyzed for cortisol concentration. For evaluating welfare outside the training context, dogs participated in a cognitive bias task. Results showed that dogs from Group Aversive displayed more stress-related behaviors, were more frequently in tense and low behavioral states and panted more during training, and exhibited higher post-training increases in cortisol levels than dogs from Group Reward. Additionally, dogs from Group Aversive were more ‘pessimistic’ in the cognitive bias task than dogs from Group Reward. Dogs from Group Mixed displayed more stress-related behaviors, were more frequently in tense states and panted more during training than dogs from Group Reward. Finally, although Groups Mixed and Aversive did not differ in their performance in the cognitive bias task nor in cortisol levels, the former displayed more stress-related behaviors and was more frequently in tense and low behavioral states. These findings indicate that aversive-based training methods, especially if used in high proportions, compromise the welfare of companion dogs both within and outside the training context.

Citation: Vieira de Castro AC, Fuchs D, Morello GM, Pastur S, de Sousa L, Olsson IAS (2020) Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare. PLoS ONE 15(12): e0225023. https://doi.org/10.1371/journal.pone.0225023

Editor: Carolyn J. Walsh, Memorial University of Newfoundland, CANADA

Received: October 24, 2019; Accepted: October 30, 2020; Published: December 16, 2020

Copyright: © 2020 Vieira de Castro et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information files.

Funding: The current research study was supported by FCT - Fundação Portuguesa para a Ciência e Tecnologia (Fellowship SFRH/BPD/111509/2015) and UFAW – Universities Federation for Animal Welfare (Grant 14-16/17), with grants awarded to ACVC. SP was supported by PIPOL - Regione Friuli Venezia Giulia. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. FCT - Fundação Portuguesa para a Ciência e Tecnologia: https://www.fct.pt/index.phtml.pt UFAW – Universities Federation for Animal Welfare: https://www.ufaw.org.uk/ .

Competing interests: The authors have declared that no competing interests exist.

1. Introduction

To fulfil their increasingly important role as companion animals, dogs need to be trained to behave in a manner appropriate for human households. This includes, for example, learning to eliminate outdoors or walk calmly on a leash [ 1 , 2 ]. Dog behavioral problems are the most frequently cited reason for rehoming or relinquishment of dogs to shelters and for euthanasia [ 2 ], which suggests that such training is often missing or unsuccessful.

Dog training most often involves the use of operant conditioning principles, and dog training methods can be classified according to the principles they implement: aversive-based methods use mainly positive punishment and negative reinforcement and reward-based methods rely on positive reinforcement and negative punishment [ 3 ]. Within a given training method, several factors may influence how dogs react, such as the characteristics of the behavior under training and the timing of reinforcement/punishment [ 4 ]. However, the use of aversive-based training methods per se is surrounded by a heated debate, as studies have linked them to compromised dog welfare [ 5 – 10 ]. Some aversive-based tools, such as shock collars, have indeed been legally banned in some countries [ 11 ]. However, a recent literature review by [ 3 ] concluded that, because of important limitations, existing studies on the topic do not provide adequate data for drawing firm conclusions. Specifically, the authors reported that a considerable proportion of the studies relied upon surveys rather than on objective measures of both training methods and welfare; that they focused on sub-populations of police and laboratory dogs which only represent a small portion of dogs undergoing training; and, finally, that the empirical studies have concentrated mainly on the effects of shock-collar training, which is only one of several tools used in aversive-based training. In summary, limited scientific evidence exists on the effects of the entire range of dog training techniques on companion dog welfare.

Furthermore, previous empirical studies have focused on the effects of training methods on dog welfare within the training context. Behavioral and physiological indicators of welfare, such as the frequency of stress-related behaviors and the concentration of salivary cortisol, have been collected in and around the training situation (e.g., [ 9 , 12 ]; see also [ 3 ]). However, the welfare impact of training methods beyond the training scenario has not yet been examined. To our knowledge, only one study evaluated the effects of training on welfare outside the training context. Christiansen et al (2001) [ 13 ] found no effect of shock collar training on dog fear or anxiety, but this was based on dog owner reports of behavior and temperament tests rather than on objective and animal-based welfare indicators. Importantly, a suitable assessment of the effects of training methods on dog welfare should comprise an evaluation of their effects both during and beyond the training scenario.

The affective states of animals are influenced by both immediate rewarding or punishing experiences (effects on shorter-term states), and by the cumulative experience of rewarding or punishing experiences (effects on longer-term states) [ 14 ]. Hence, due to the repeated exposure to aversive stimuli, training with aversive-based methods is expected to also affect dogs’ affective states in a longer-term, transitioning to outside the training context. One way to assess affective states is through the cognitive bias paradigm (e.g., [ 15 ]). The cognitive bias task has been validated as an effective tool to evaluate the affective states of non-human animals and has been extensively used with several species, including dogs [ 16 – 18 ]. The rationale behind the paradigm is based on theoretical and empirical findings that an individual’s underlying affective state biases its decision-making and, specifically, that individuals experiencing negative affective states make more ‘pessimistic’ judgements about ambiguous stimuli than individuals experiencing more positive affective states [ 14 , 15 , 17 ].

Therefore, the aim of the present study was to perform a comprehensive evaluation of the effects of different training methods on the welfare of companion dogs both within and outside the training context. By performing an objective assessment of training methods (through the direct observation of training sessions) and by using objective measures of welfare (behavioral and physiological data to assess effects during training, and a cognitive bias task to assess effects outside training), we assessed the effects of reward-based and aversive-based methods on companion dog welfare. We hypothesized that dogs trained using aversive-based methods would display higher levels of stress during training, as determined by behavioral and physiological indicators of stress during training sessions, and more ‘pessimistic’ judgments of ambiguous stimuli during a cognitive bias task performed outside the training context, as compared to dogs trained using reward-based methods. We used a quasi-experimental approach in which dog-owner dyads were recruited to participate through the training school at which they were enrolled. As treatment could not be randomized, data on potential confounders was collected to be included in the analysis of treatment effects.

Understanding the effects of training methods on companion dog welfare has important consequences for both dogs and humans. Both determining and applying those training methods that are less stressful for dogs is a key factor to ensure adequate dog welfare and to capitalize on the human benefits derived from interactions with dogs [ 19 , 20 ].

2. Materials and methods

2.1. ethical statement.

All procedures were approved by ICBAS (Abel Salazar Biomedical Sciences Institute) ORBEA (Animal Welfare Body). All head trainers of dog training schools and owners completed a consent form authorizing the collection and use of the data.

2.2. Training schools

2.2.1. recruitment..

Dog training schools within the metropolitan area of Porto, Portugal were searched on the internet. Eight schools were selected based on both their geographical proximity and on the listed training methods. The head trainers were invited by telephone to participate in the study. They were informed that the aim was to evaluate dog stress and welfare in the context of training and the methodological approach was thoroughly explained. To avoid bias during recorded training sessions, the trainers were not made aware that study results were going to be further compared among different training methods. Of the eight contacted schools, seven agreed to participate. After study conclusion, a debriefing with the participating training schools was performed in order to communicate the results.

The training schools had different training sites and class structures. Depending on the school, training sites were located either in rural or urban environments, and classes were conducted either indoors or outdoors. Classes were either individual or in group sessions of 15 to 60 minutes, and varied in frequency and time of day depending on the school: the frequency of classes per week among schools ranged from one to three sessions and classes were taught in the mornings, afternoons or evenings. The type of behaviors trained, on the other hand, was fairly standard across training schools and included teaching the dog to sit, lie down, stay, come when called, not to jump on people and to heel or walk on a loose leash; tricks were also taught in some of the participating schools.

2.2.2. Classification of training methods.

We performed an objective assessment of the training methods used by each school. To this end, we randomly selected six video recordings of training sessions per training school (see section 2.4.1) and analyzed the videos for the frequency of the intended operant conditioning procedures used for training, namely positive punishment, negative reinforcement, positive reinforcement and negative punishment (see Table 1 for the detailed definitions). In order to be coherent with the standard for classification of operant conditioning procedures as reinforcement or punishment (which is based not on the procedure itself but on its effect on behavior) [ 21 ], throughout the paper we refer to the procedures as intended positive punishment, intended negative punishment, etc. The analysis was performed using The Observer XT software, version 10.1 (Noldus Information Technology Inc, Wageningen, The Netherlands) and afterwards the proportion of intended aversive-based techniques [(number of intended positive punishments + number of intended negative reinforcements)/total number of intended operant conditioning procedures)] was calculated for each training session (see S1 Appendix for the results). Although Schools A, C, D and F all used some form of intended positive punishment and/or negative reinforcement and, as such, their training methods can be classified as aversive-based, the fact that two highly different levels of the use of such techniques were observed lead us to divide these schools in two groups. Schools A and D which used, on average, a proportion of 0.76 and 0.84 of intended aversive-based techniques, respectively, were categorized as Group Aversive, and Schools C and F which used, on average, a proportion of 0.22 and 0.37 of intended aversive-based techniques, respectively, were categorized as Group Mixed. Schools B, E and G, which did not use any intended aversive-based techniques, were classified as Group Reward.

PPT PowerPoint slide
PNG larger image
TIFF original image

https://doi.org/10.1371/journal.pone.0225023.t001

2.3. Subjects

The head trainer of each training school was asked to indicate at least fourteen dogs fitting our inclusion criteria (described below), and we then approached the owners to ask if they were willing to participate. The information about the study given to the owners was the same that was given to the head trainers of the schools. The inclusion criteria for the dogs were: 1) to have attended the training school for less than two months, in order to mitigate familiarization to training methods, and 2) to be free of certain behavioral problems (e.g., aggression, fearfulness and separation anxiety, as determined by the owner and the first author), in order to prevent any confounding stress.

Over the course of the study, which was conducted between October 2016 and March 2019, the owners of 122 companion dogs agreed to participate. However, 30 dog owners dropped out of the training schools before any meaningful data could be collected. Specifically, these subjects dropped out before meeting our requirement that at least two training sessions were video recorded and that the owner completed a written questionnaire. Consequently, our final sample comprised 92 subjects, 28 from Group Aversive (Schools A and D: n = 14), 22 from Group Mixed (School C: n = 8, School F: n = 14), and 42 from Group Reward (School B and G: 15 dogs, School E: 12 dogs).

As for subjects’ demographics, the average age was 11.9 (SEM = ±1.0) months, 54% were male and 35% were neutered/spayed. Thirty-four percent were mixed-breed dogs and the remaining 66% belonged to a FCI-recognized breed group: 18% belonged to Group 1: Sheepdogs and Cattledogs (except Swiss Cattledogs), 13% to Group 2: Pinscher and Schnauzer–Molossoid and Swiss Mountain and Cattledogs, 5% to Group 3: Terriers, 4% to Group 6: Scent hounds and related breeds, 2% to Group 7: Pointing dogs, 20% to Group 8: Retrievers–Flushing Dogs–Water Dogs, and 3% to Group 9: Companion and Toy Dogs.

2.4. Data collection

The study was conducted in two phases. The goal of Phase 1 was to evaluate the welfare of dogs within the training context and the goal of Phase 2 was to evaluate the welfare of these same dogs outside the training context.

2.4.1. Phase 1 –Evaluating welfare within the training context.

In order to evaluate behavioral indicators of welfare during training, each dog was videotaped for the first 15 minutes of three training sessions using a video camera on a tripod (one Sony Handycam HDR-CX405 and two Sony Handycam DCR-HC23). Five experimenters were responsible for data collection. The cameras were positioned to get an optimal view of the specific participant without interfering with training. The day and time of the training sessions were determined by the training schools and by the participants’ availability.

To obtain physiological data on stress during training, six saliva samples were collected per dog to allow assay of salivary cortisol [ 9 , 22 ]. Three samples were collected 20 min after each training session (PT–post-training samples) and three were collected at home on days when no training took place, approximately at the same time as PT samples (BL–baseline samples). Owners were asked not to give their dog water in the 20 minutes preceding each sample collection, nor a full meal in the hour preceding each sample collection, respectively. The timing for sample collections, as well as other recommendations regarding saliva collection for cortisol analysis, were drawn from previous relevant research on dogs’ cortisol responses to potentially stressful stimuli [ 9 , 22 – 24 ]. Owners were instructed on how to properly collect samples of their dog’s saliva during the first training session, when the first sample (PT1) was collected by the first author of the study. The following samples were always collected by the owners. A synthetic swab (Salivette®) was rubbed in the dogs’ mouth for about 2 minutes to collect saliva. For samples collected at the training schools (PT), the swab was placed back into the provided plastic tube and immediately stored on ice. It was then transferred to a -20°C freezer as soon as possible. For samples collected at home (BL), owners were instructed to place the swab back into the plastic tube and immediately store it in their home freezer. Owners were provided with ice-cube plastic makers to transport the BL samples to the training school during the next scheduled training session without them unfreezing, and they were stored at -20°C as soon as possible. Owners were also provided with detailed written instructions for saliva collection and a phone contact in case any owners had questions related to sample collection. For standardization purposes, we ensured that Phase 1 did not last more than three months for each dog.

2.4.2. Phase 2—Evaluating welfare outside the training context.

After finishing data collection for Phase 1, dogs participated in Phase 2, which consisted of a spatial cognitive bias task. The end of Phase 1 did not correspond to the conclusion of the training programs for the dogs, as this would result in different dogs being exposed to substantially different amounts of training before being assessed for cognitive bias. Instead, for standardization purposes, we ensured that 1) dogs had attended the training school for at least one month prior to Phase 2 and that 2) the cognitive bias task was conducted within one month of completing Phase 1. Due to limited owner availability, 13 subjects either dropped out or did not meet the criteria for Phase 2, resulting in 79 (24 from Group Aversive, 20 from Group Mixed and 35 from Group Reward) of the original 92 dogs participating in Phase 2. The cognitive bias tasks were scheduled according to owners’ availability, both on weekdays and Saturdays.

The test was conducted in an indoor room (7.7 x 3 meters) within a research building at the Abel Salazar Biomedical Sciences Institute (ICBAS), University of Porto in Portugal. All dogs were unfamiliar with the room prior to testing. Two experimenters conducted the test while the dog’s owner(s) sat in a chair in a corner area of the room (see Fig 1 ). Dog owners were asked not to look into the dog’s eyes or to speak to the dog during the test, unless the experimenters instructed otherwise. The entire test took place over one meeting for each dog. The room was cleaned with water and liquid detergent at the end of each test.

https://doi.org/10.1371/journal.pone.0225023.g001

2.4.2.1. Familiarization period. Prior to the start of the cognitive bias task, the dogs were given the opportunity to familiarize with the test room and the researchers. This consisted of a 10-min period during which the dog was allowed to freely explore the room and engage with the researchers and the owner(s).

2.4.2.2. Training phase. The methodology of Phase 2 was based on [ 18 ]. During the training phase, dogs were trained to discriminate between a ‘positive’ (P) location of a food bowl, which always contained a food reward, and a ‘negative’ (N) location, which never contained a food reward. At the start of each trial, the dog was held by one trained experimenter—hereafter the ‘handler’, behind a barrier (2 x 2 m, see Fig 1 ), while a second trained experimenter—hereafter the ‘timer’, baited (or did not bait, depending on the type of trial) the bowl with a piece of sausage (approximately 1.25 g for smaller dogs and 2.5 g for larger dogs). To ensure that the dog, the owner and the ‘handler’ were blind to whether or not the bowl contained food during each trial, the bowl was baited out of their sight, on the opposite side of the barrier. Additionally, the food reward was rubbed onto the food bowl before every trial to prevent the influence of olfactory cues. The height of the food bowl was such that the dog could not visually judge the presence or absence of food from the start position.

After baiting (or not baiting) the bowl, the ‘timer’ placed it at one of the two training locations. The ‘timer’ then determined the start of the trial, by verbally signaling to the ‘handler’, upon which the ‘handler’ led the dog to the start position and released him. The ‘handler’ always led the dog to the start position on her left side. Because we found that dogs had some difficulty noticing the bowl at the end of the room during pilot tests, the ‘handler’ walked towards the bowl and pointed it out to the dog during the first four trials. For the remaining trials, the ‘handler’ simply walked the dog to the start position and released him. After the dog reached the food bowl and (when applicable) ate the reward, the ‘handler’ collected him and led him behind the barrier to start the next trial. The latency to reach the bowl, defined as the time elapsed between release at the start position and the dog putting its head in line with the edge of the bowl, was recorded for each trial by the ‘timer’ using a stopwatch.

The position of the ‘positive’ and ‘negative’ locations was counterbalanced across subjects and training schools, such that for half of the dogs from each training school, the ‘positive’ location was on the right hand side as they faced the test area, and for the other half it was on the left. Initially, each dog received two consecutive ‘positive’ trials (bowl placed in the ‘positive’ location) followed by two ‘negative’ trials (bowl placed in the ‘negative’ location). Subsequently, ‘positive’ and ‘negative’ trials were presented in a pseudorandom order, with no more than two trials of the same type being presented consecutively.

All dogs received a minimum of 15 training trials to learn the discrimination between bowl locations. Dogs were considered to have learned an association between bowl location and food (the learning criterion) when, after a minimum of 15 trials, the longest latency to reach the ‘positive’ location was shorter than any of the latencies to reach the ‘negative’ location for the preceding three ‘positive’ trials and the preceding three ‘negative’ trials. Each trial lasted a maximum of 20 seconds. If the dog did not reach the bowl by that time, the trial automatically ended and a latency of 20 seconds was recorded.

All but two dogs were able to complete the training phase. For the two dogs that failed to complete training, one did not show any interest in the food reward and the other was food-motivated but could not focus on the task. These two dogs belonged to Group Mixed. Therefore, the total number of subjects completing Phase 2 in Group Mixed was 18.

2.4.2.3. Test phase. Testing began once the learning criterion was achieved. Test trials were identical to training trials except that the bowl (empty) was placed at one of three ambiguous locations equally spaced along an arc 4 m from the dog’s start position, between the ‘positive’ and ‘negative’ locations. The three test locations were: ‘near‐positive’ (NP: one third of the way along the arc from the ‘positive’ location), ‘middle’ (M: half way along the arc), ‘near‐negative’ (NN: one third of the way along the arc from the ‘negative’ location). Three test trials were presented at each test location (nine test trials in total) in the following order for all dogs: M, NP, NN, NP, NN, M, NN, M, NP (each location was presented first, second or third in each block of three test trials). Each test trial was separated from the next one by two training trials identical to those conducted in the training phase (one ‘positive’ and one ‘negative’ trials presented in a random order), in order to maintain the associations between the ‘positive’ and ‘negative’ locations and the presence or absence of food, respectively. Thus, the test phase included a further sixteen training trials interspersed in blocks of two between the nine test trials.

To end the test phase, a final trial was conducted by placing an empty bowl in the ‘positive’ location to determine whether dogs ran to the empty bowl as quickly as they did to the baited bowl. This was meant to establish that the dogs were not relying on olfactory or visual cues during the test. During the entire test, each trial was kept as similar as possible in terms of preparation time and activity, and dogs were handled in the same way throughout the test.

Due to circumstances beyond our control, namely people speaking loudly and other dogs barking in the building during some of the tests, some subjects were clearly distracted and disengaged from the task during some trials. Whenever this happened, no latency was recorded for that trial. The experimenters waited for the dog to resettle and moved to the following trial.

2.5. Questionnaire

All owners were asked to complete a brief written questionnaire regarding dog demographics and background, and owner demographics and experience with dogs and dog training. The questionnaire was based on [ 10 ].

2.6. Data analysis

2.6.1. phase 1 –evaluating welfare within the training context..

2.6.1.1. Behavior coding. We developed two ethograms based on previous literature to record the frequency of different stress-related behaviors and the time spent in different behavioral states and panting during the training sessions [ 8 , 9 , 23 ]. The behaviors and their definitions are described in Tables 2 and 3 .

https://doi.org/10.1371/journal.pone.0225023.t002

https://doi.org/10.1371/journal.pone.0225023.t003

Behavior coding was conducted by three observers, which, with the exception of the first author, were blind to the schools’ classification based on their training methods. Each video was coded twice, once with the ethogram for stress-related behaviors, using a continuous sampling technique (by the first and second authors, see Table 2 ), and a second time with the ethogram for overall behavioral state and panting, by scan-sampling at 1 minute intervals (by the first and fourth authors, see Table 3 ). The Observer XT software, version 10.1 (Noldus Information Technology Inc, Wageningen, The Netherlands) was used to code for stress-related behaviors and Windows Movie Player and Microsoft Excel to code for overall behavioral state and panting.

The second and fourth authors were trained to become familiar with the ethograms and inter-observer reliability was assessed for each ethogram by having the corresponding pair of observers watch and code sets of four videos at an early stage of analysis [ 9 ]. Cohen’s Kappa coefficient was calculated for each pair of videos using The Observer XT. After analyzing each set of four videos, if there was poor agreement for any video (r<0.80), the observers checked and discussed all the inconsistencies and, if needed, decided on how to refine the description of the ethogram behaviors. After this, they re-analyzed the same videos and the process was repeated until r>0.8 was achieved for the entire set of videos. Once this level was attained, the observers analyzed a new set of four videos. The whole process was repeated until a value of r>0.8 was achieved for the four videos of a new set in the first attempt (i.e., without the need for a re-analysis). At this point, the observers were assumed to be in strong agreement and began coding videos independently [ 9 ]. A total of 265 videos were coded. For the ethogram for continuous sampling, the analysis of 16 videos was needed before agreement was achieved, whereas for the ethogram for scan sampling, agreement was achieved only after the analysis of four videos. Afterwards, for each ethogram, the remaining videos were distributed randomly between observers, while ensuring that each observer coded a similar percentage of videos from each experimental group. The first author coded 76% of the videos with the ethogram for stress-related behaviors and 64% with the ethogram for overall behavioral state and panting.

During some training sessions, we were not able to videotape the intended full 15 minutes of training. For these sessions, those that were at least 10-min long were included in the analysis and those that were less than 10-min were excluded.

2.6.1.2. Cortisol analysis. Two dogs (one from School B and one from School E, both from Group Reward) did not cooperate with the saliva collection procedure and, as such, no saliva samples were extracted from them. For the remaining 90 dogs, only 23 dog owners (seven from Group Aversive, five from Group Mixed and 11 from Group Reward) were able to appropriately collect six saliva samples. The samples from these subjects were selected for analysis. An additional 40 dog owners (14 from Group Aversive, 11 from Group Mixed and 15 from Group Reward) were able to properly collect at least four saliva samples. From these 40 dogs, eight were randomly selected to have their samples analyzed (one from Group Aversive, three from Group Mixed and four from Group Reward). In total, 8 dogs from Group Aversive, 8 dogs from Group Mixed and 15 dogs from Group Reward had their samples selected for analysis (Schools A, C, D, E and F: n = 4; School B: n = 5; School G: n = 6). These samples were sent to the Faculty of Sport Sciences and Physical Education of the University of Coimbra, Coimbra, Portugal, where they were assayed for cortisol concentration using standard ELISA kits (Salimetrics®).

In order to investigate potential changes in salivary cortisol concentration as a result of training methods, for each dog the baseline sample values (BL) and the post-training sample values (PT) were averaged, and the difference between the post-training and the baseline averages computed (hereafter the post-training increase in cortisol concentration).

2.6.2. Phase 2 –Evaluating welfare outside the training context.

For each dog, we calculated the average latency to reach the food bowl during each of the three types of test trials (NP, M, NN) as well as the average latency to reach the ‘positive’ and ‘negative’ training locations during the test phase.

Seventy-three dogs completed the cognitive bias task. From these 73 subjects, 14 disengaged from the task during some trials due to noise outside the test room. Thirteen disengaged for one test trial (Group Aversive: five dogs at location NP, and three dogs at location NN; Group Reward: one dog at location M, three dogs at location NP, and one dog at location NN), and one (Group Reward) for three test trials (one at each test location). For these dogs, the average latencies to the test locations were calculated from the remaining test trials. Of the remaining four dogs, one (from School G) completed the first seven test trials (at locations M, NP, NN, NP, NN, M, NN), two (one from School A and one from School E) completed the first five test trials (at locations M, NP, NN, NP, NN), and one (from School G) completed the first three test trials (at locations M, NP, NN); then they stopped cooperating with the task. Their average latencies to the test locations were calculated from these trials.

2.7. Statistical analyses

Statistical tests were conducted using SPSS ® Statistics 25.0 and SAS University Edition ® . All data were tested for normality by performing the Shapiro-Wilk test prior to data analysis. Except for the number of trials required to reach the learning criterion in the cognitive bias task, the data were not normally distributed. Thus, Procedure GENMOD was used on SAS University Edition ® to perform negative binomial regressions with repeated measures (dog) of all the stress-related behaviors, behavioral states, and panting scans as a function of group (Aversive, Mixed, Reward), training session (Session 1, 2, and 3) and the interaction between these two categorical factors. Procedure GLIMMIX was used on SAS University Edition ® to analyze the latency to reach the bowls in the cognitive bias task as a function of bowl location, group, and the interaction between these two categorical variables, considering dog as a random effect. To correct for the unbalanced distribution of potential confounders in the dataset, all known confounders for which sufficient data were available were considered in the analysis as follows. First, each confounder was tested, one at a time, in addition to the variables of interest, to verify if there was a significant relation between confounder and response variable, and if the confounder substantially changed the model estimate of the independent variables (i.e. group and training session estimates). This way of testing confounders, in which they are tested one at a time, allows to maintain enough statistical power to verify their significance and influence in the models. If more than one confounder was found to be significant, then all the significant confounders were tested in the whole model. Non-significant confounders, variables of interest and interactions were removed from the final models. A numeric (dog age) and two categorical factors (children in the household and owner gender) were tested as possible confounders for each stress-related behavior, each behavioral state and panting, as well as for the latencies to reach the bowl in the cognitive bias task. A numeric (dog weight) factor was also tested as a possible confounder in the cognitive bias test. Multicollinearity was checked among confounders and variables of interest prior to testing of statistical models. Least-squared means were compared among groups and training sessions with Bonferroni corrections for multiple pairwise comparisons. Non-parametric tests were used to compare cortisol concentrations among groups, as well as to perform a preliminary assessment of the data, as follows:

Fisher’s exact tests were used to compare the three groups (Aversive, Mixed, Reward) regarding dog demographics and background, and owner demographics and experience with dogs and dog training (variables collected with the questionnaire).
Kruskal-Wallis tests were used to evaluate the effects of group (Aversive, Mixed and Reward) on the post-training increase in cortisol concentration, on the baseline and the post-training levels of cortisol, as well as on the number of training classes attended by the dogs before Phase 2.
A Wilcoxon signed-rank test was used to compare, in the cognitive bias task, the latency to reach the P location during test trials and during the final trial, when the bowl contained no food, to verify whether dogs were relying on olfactory or visual cues to discriminate between bowl locations.

Finally, a one-way ANOVA for independent samples was used to compare the number of trials needed to reach the learning criterion in the cognitive bias task among groups (Aversive, Mixed, Reward). All the statistical tests were two-tailed and the level of significance was set at α = 0.05. When multiple comparisons were performed, a Bonferroni correction was applied. Specifically, corrected p-values were used for the post-hoc pairwise comparisons performed for the post-training increase in cortisol concentration and for the number of trials to criterion in the cognitive bias task. The effect sizes for all the reported results were calculated as Cohen’s d. The entire dataset is available in S2 Appendix .

3.1. Questionnaire

3.1.1 dog demographics and background..

Concerning dog demographics, the three groups did not differ in sex and neuter status ratios, but they differed with regards to age (F = 13.9, p = 0.013) and FCI breed group (F = 25.3, p = 0.008). As for dog background, the groups differed only in the age of separation from the mother (F = 20.8, p = 0.001, see Table 4 ).

Fisher’s exact tests were used to compare the three Groups (Aversive, Mixed, Reward).

https://doi.org/10.1371/journal.pone.0225023.t004

3.1.2. Owner demographics, experience with dogs and dog training.

Regarding owner demographics, the three groups did not differ in owner age and family household size; however, they differed in owner gender (F = 8.4, p = 0.013) and in whether the household included children (F = 6.2, p = 0.044). Regarding owner experience with dogs and dog training, the groups did not differ in whether owners had attended training classes with a previous dog nor in whether they had had other dog(s) in the past, but they differed in the information owners used to choose the dog training school (F = 19.9, p = 0.005, see Table 4 ).

3.2. Phase 1 –Evaluating welfare within the training context

3.2.1. behavioral data..

3.2.1.1. Stress-related behaviors. Dogs from Group Aversive performed significantly more body turn (Group Aversive: (M±SEM) 3.14±0.64 vs. Group Reward: 0.39±0.08; Z = 7.4, p<0.001, d = 1.77), crouch (Group Aversive: 3.56±0.71 vs. Group Reward: 0.59±0.15; Z = 4.8, p<0.001, d = 1.10), body shake (Group Aversive: 1.29±0.19 vs. Group Reward: 0.63±0.10; Z = 2.8, p = 0.014, d = 0.66), yawn (Group Aversive: 2.30±0.28 vs. Group Reward: 0.28±0.07; Z = 6.6, p<0.001, d = 1.37) and lip lick (Group Aversive: 55.90±4.36 vs. Group Reward: 4.11±0.37; Z = 16.6, p<0.001, d = 3.91) than those from Group Reward. Dogs from Group Aversive also performed more yawn (Group Aversive: 2.30±0.28 vs. Group Mixed: 0.80±0.20; Z = 3.4, p = 0.002, d = 1.03), lip lick (Group Aversive: 55.90±4.36 vs. Group Mixed: 17.84±2.15; Z = 5.7, p<0.001, d = 1.67) and tended to perform more body shake (Group Aversive: 1.29±0.19 vs. Group Mixed: 0.64±0.14; Z = 2.2, p = 0.090) than dogs from Group Mixed. Dogs from Group Mixed performed more body turn (Group Mixed: 2.13±0.39 vs. Group Reward: 0.39±0.08; Z = 6.4, p<0.001, d = 1.46), crouch (Group Mixed: 1.83±0.44 vs. Group Reward: 0.59±0.15; Z = 3.1, p = 0.006, d = 0.70), yawn (Group Mixed: 0.80±0.20 vs. Group Reward: 0.28±0.07; Z = 2.4, p = 0.050, d = 0.57) and lip lick (Group Mixed: 17.84±2.15 vs. Group Reward: 4.11±0.37; Z = 7.3, p<0.001, d = 1.99) compared to Group Reward. There was also a tendency for move away to be affected by group (X 2 2,265 = 5.3, p = 0.073). Overall, when group affected stress-related behaviors, these were performed less frequently in Group Reward than in both Group Aversive and Group Mixed. No effect of group was found for scratch, yelp, whine and paw lift. None of the stress-related behaviors was affected by training session. The average frequencies of stress-related behaviors by group are depicted in Fig 2 .

Number of occurrences of each stress-related behavior averaged across the three training sessions for Group Aversive (white bars), Group Mixed (grey bars) and Group Reward (black bars). Vertical bars show the SEM. *stands for statistically significant differences for least square means at α = 0.05.

https://doi.org/10.1371/journal.pone.0225023.g002

Presence of children in the household was a significant confounder which increased the frequency of body turn (Z = -2.4, p = 0.018), but decreased the frequency of body shake (Z = 2.4, p = 0.016). Additionally, as dog age increased, the frequency of body turn (Z = -2.6, p = 0.011) and yawn (Z = -2.8, p = 0.006) decreased.

There were not enough occurrences of salivating and lying on side/back to perform negative binomial regressions. Salivation frequency was (M±SEM) 0.29±0.14, 0.03±0.03, and 0.02±0.02 in Groups Aversive, Mixed, and Reward, respectively. In a similar pattern, the frequency of lying on side/back was 0.99±0.09 and 0.02±0.02 in Groups Aversive and Mixed, respectively, and no occurrences were observed in Group Reward. Fear-related elimination was never displayed during this study. Model details are available in S3 Appendix .

3.2.1.2. Overall behavioral state. The frequency of scans in which dogs were observed in an excited state was lower in Group Aversive compared to both Group Mixed (Z = 6.2, p<0.001, d = 1.62) and Group Reward (Z = 9.0, p<0.001, d = 2.63). Dogs from Group Mixed were also observed less frequently (Z = -4.2, p<0.001, d = 1.31) in an excited state than those of the Group Reward ( Fig 3 ). Similarly, dogs were observed in a relaxed state less frequently in Group Aversive than in Group Mixed (Z = -2.5, p = 0.033, d = 0.66) and Group Reward (Z = -2.8, p = 0.017, d = 0.78), but no differences were observed between Group Mixed and Group Reward (Z = -0.1, p = 0.999). On the other hand, dogs from Group Aversive were observed in tense and low states more frequently than those from Group Mixed (Z = 5.9, p<0.001, d = 1.85 for tense; Z = 3.7, p<0.001, d = 1.07 for low) and Group Reward (Z = 14.6, p<0.001, d = 2.96 for tense; Z = 3.9, p<0.001, d = 0.81 for low). Dogs were also tense more frequently in Group Mixed compared to Group Reward (Z = 7.6, p<0.001, d = 1.72 see Fig 3 ). Dogs were observed more often in a low state when children were present in the household (Z = -2.6, p = 0.011).

Average number of scans in the different behavioral states in training sessions 1 (S1), 2 (S2) and 3 (S3) for Group Aversive (left), Group Mixed (middle) and Group Reward (right).

https://doi.org/10.1371/journal.pone.0225023.g003

Training session tended to affect the occurrence of the behavioral state relaxed (X 2 2,265 = 5.1, p = 0.077) and significantly affected the behavioral state excited (X 2 2,265 = 10.3, p = 0.006), in that dogs were observed more frequently in an excited state in the first, S1, than in the second, S2, (Z = 2.7, p = 0.019) and in the last, S3, (Z = 3.3, p = 0.003) training sessions.

3.2.1.3. Panting. The frequency of scans in which dogs were observed panting in Group Aversive was higher than in Group Reward (Z = 4.6, p<0.001, d = 1.02). Panting frequency was also observed to be higher in Group Mixed compared to Group Reward (Z = 2.5, p = 0.042, d = 0.59, Fig 4 ). Training session did not affect the frequency of panting.

Number of scans panting averaged across training sessions for Group Aversive (left), Group Mixed (middle) and Group Reward (right). Vertical bars show the SEM. *stands for statistically significant differences for least square means at α = 0.05.

https://doi.org/10.1371/journal.pone.0225023.g004

3.2.2. Physiological data.

Baseline cortisol concentrations did not differ among groups [Group Aversive: 0.15±0.02 vs. Group Mixed: 0.14±0.02 vs. Group Reward: 0.13±0.02 μg/dL; H(2) = 1.689, p = 0.430], but differences among groups were found for post-training levels [Group Aversive: 0.26±0.05 vs. Group Mixed: 0.23±0.05 vs. Group Reward: 0.13±0.02 μg/dL; H(2) = 8.634, p = 0.013]. As a result, there was an effect of group in the average post-training increase in cortisol concentration [H(2) = 9.852, p = 0.007]. Specifically, as depicted in Fig 5 , the average post-training increase in cortisol was higher in Group Aversive than in Group Reward (U = 107.0, p = 0.003, d = 0.13), but no differences were found between Group Mixed and Group Reward (U = 90.0, p = 0.183) nor between Group Mixed and Group Aversive (U = 39.5, p = 0.826).

Average post-training increase in cortisol concentration (PT-BL) for Group Aversive (left), Group Mixed (middle) and Group Reward (right). Vertical bars show the SEM. *stands for statistically significant differences for the averages at α = 0.05.

https://doi.org/10.1371/journal.pone.0225023.g005

3.3. Phase 2 –Evaluating welfare outside the training context

Prior to the cognitive bias task, dogs from Group Aversive, Mixed, and Reward attended (M±SEM) 6.29±0.47, 7.14±0.65 and 6.07±0.36 training classes, respectively, with no significant differences observed among groups [H(2) = 2.7, p = 0.258].

3.3.1. Training phase.

Dogs took (M±SEM) 27.14±0.85 trials to reach the learning criterion. Dogs from Group Aversive, Group Mixed and Group Reward took, respectively, 28.71±1.35, 29.61±1.79 and 24.80±1.26 trials. The differences among groups were statistically significant [F(2,74) = 3.5, p = 0.037], with dogs from Group Mixed showing a tendency to require more trials to reach criterion than Group Reward [t(51) = -2.2, p = 0.090], but no differences being found between Group Aversive and Group Reward [t(57) = -2.1, p = 0.124] nor between Group Aversive and Group Mixed [t(40) = 0.4, p = 0.968].

3.3.2. Test phase.

Fig 6 depicts the average latencies to reach the two training locations (P, N) and the three test locations (NP, M, NN) for Group Aversive, Group Mixed and Group Reward. Group and bowl location affected the latency for the dogs to reach the bowl, but there was no significant group*bowl location interaction. Dogs of Group Aversive took longer to reach all bowl locations (t = 2.6, p = 0.032, d = 3.99) compared to dogs from Group Reward, but no differences were found between Group Mixed and Group Reward (t = 2.0, p = 0.153), as well as between Group Aversive and Group Mixed (t = 0.4, p = 0.999).

Average latency to reach the food bowl as a function of bowl location: P—‘positive’, NP–‘near positive’, M–‘middle’, NN–‘near negative’, N–‘negative’, for Group Aversive (white circles), Group Mixed (grey circles) and Group Reward (black circles). Vertical bars show the SEM.

https://doi.org/10.1371/journal.pone.0225023.g006

Lastly, an analysis comparing the average latency to reach the P location during test trials and the latency to reach this same location during the final trial (when the bowl contained no food) revealed no significant differences (T = 1295.5, p = 0.328), confirming that the dogs were not relying on olfactory or visual cues to discriminate between bowl locations.

4. Discussion

This is the first empirical study to systematically investigate the effects of different training methods on the welfare of companion dogs within and outside the training context. We objectively classified training methods, extended the study of aversive-based methods to other techniques and tools besides shock collars, and used objective and validated measures for the assessment of dog welfare within the training context (behavioral and physiological stress responses during training) and outside the training context (cognitive bias task). Since it became evident during data collection that the recruited dog training schools that employed aversive-based methods did so to a substantially different extent, for analysis the participating schools were divided into three groups: Group Aversive, composed by two schools that used over 75% of intended aversive-based methods, Group Mixed, composed by two schools that used less than 40% of intended aversive-based methods, and Group Reward, composed by three schools that used no intended aversive stimuli. Overall, our results show that Group Aversive and Group Mixed were in poorer welfare during training than Group Reward, and that Group Aversive was also in poorer welfare than Group Reward outside the training context. Additionally, although no differences between Groups Aversive and Mixed were found outside the training context, Group Aversive displayed poorer welfare during training.

During the welfare assessment in the training sessions, dogs from Group Aversive were observed more frequently in low behavioral states than dogs from Group Reward, and dogs from both Group Aversive and Group Mixed were observed more frequently in tense behavioral states and more frequently panting than dogs from Group Reward. Dogs from Group Aversive were also observed more frequently in tense and low behavioral states than dogs from Group Mixed. Tense and low body postures reflect states of distress and fear in dogs (e.g., [ 25 ]), while panting has been associated with acute stress in dogs (e.g., [ 9 , 26 ]). Additionally, overall, dogs from Group Aversive displayed stress-related behaviors more frequently than dogs from both Group Mixed and Group Reward, and dogs from Group Mixed displayed stress-related behaviors more frequently than dogs from Group Reward. In previous studies, high levels of lip licking and yawning behaviors have been consistently associated with acute stress in dogs (e.g., [ 10 , 27 ]). Importantly, lip licking has been associated with stressful social situations [ 27 ]. This most likely explains the large magnitude of this behavior observed in Group Mixed and the even larger magnitude in Group Aversive, as aversive-based training methods comprise social and physical confrontation with the dog. The display of avoidance behaviors such as body turn, move away, crouch and lying on side/back, specifically in response to training techniques, highlights the aversive nature of the training sessions at the schools employing aversive-based methods. Notably, lying on side/back was only displayed in Groups Aversive and Mixed (and mostly in School A, which employed the highest proportion of intended aversive-based training methods). Finally, no differences were found among groups for scratch, paw lift, whine and yelp. Previous studies on dog training methods have also failed to identify significant differences in scratch and paw lift [ 9 , 10 ], suggesting that these may not be reliable indicators of stress, at least in the context of training. It is possible that scratching and paw lift behaviors are also associated with excitement and arousal rather than just distress. In turn, whining has also been associated with attention seeking and/or food begging behavior in dogs [ 28 ], and as such, is most likely also not a reliable indicator of distress. Finally, yelping may be interpreted as a response to pain [ 27 ]. However, besides the fact that no differences were found between groups, this behavior occurred very rarely in the present study. Combined with the observed differences in other stress-related behaviors, this seems to suggest that the aversive-based methods used in the present study caused emotional rather than physical distress. Hence, the present study shows a strong association between the use of aversive-based training methods and an increased frequency of stress behaviors during companion dog training. However, our results also show that the proportion of aversive-based methods used also matters, with lower proportions of aversive stimuli resulting in lower frequencies of stress behaviors exhibited by the dogs. These results strengthen and extend the findings of previous studies on companion dogs, which suggested a positive correlation between the use of both shock collars [ 9 ] and other aversive techniques [ 10 ] with stress behaviors in the context of dog training.

An effect of training session was found for the behavioral state excited, with dogs being more frequently excited in Session 1 than in Sessions 2 and 3. This result is most likely a consequence of dogs’ familiarization with the training context. The tendency of the relaxed behavioral state to increase with training session possibly reflects the reduction in excitement.

With regards to physiological measures of stress, the average post-training increase in cortisol concentration (PT-BL) was higher in Group Aversive than in Group Reward, whereas no differences were found between Group Mixed and Group Reward nor between Group Aversive and Group Mixed. Previous studies investigating cortisol levels in dogs in relation to training have yielded contradictory results. Schalke et al (2007) [ 29 ] found significant differences in the cortisol levels of three groups of laboratory dogs trained using shock collars with different degrees of shock predictability (the lower the predictability, the higher the cortisol levels). However, studies comparing aversive- and reward-based training methods have found either no significant differences or the opposite pattern: the effect on cortisol from shock collar and lemon-spray bark collars did not differ from the control treatment [ 9 , 30 ], and a negative punishment training method (a quitting signal) resulted in higher levels of cortisol than the use of a pinch collar (aversive-based technique) [ 31 ]. Hence, the present study is the first to report a significant increase in cortisol levels in dogs trained with aversive-based methods as compared to dogs trained with reward-based methods.

The average post-training increase in cortisol levels observed in the present study (M = 0.11 μg/dL for Group Aversive and M = 0.08 μg/dL for Group Mixed) was lower than those reported in other studies that found significant increases after dogs were exposed to aversive stimuli (0.20–0.30 μg/dL in [ 29 ] and 0.30–0.40 μg/dL in [ 23 ]). One possible explanation for this difference in magnitude may be related to the nature of the stimuli used in the different studies. Whereas the reported elevations in cortisol in [ 29 ] and [ 23 ] appeared after the presentation of non-social stimuli (shocks in [ 29 ], and shocks, sound blasts and a falling bag in [ 23 ]), the stimuli used during training in the present study were mainly of a social nature (i.e.: leash jerks, physical manipulation or yelling at the dog). Stimuli administered in a social context may be more predictable or better anticipated and, therefore, generate less acute stress responses [ 23 ]. In support of this view, [ 23 ] did not find elevations in cortisol after the presentation of social stimuli (physical restraint and opening an umbrella).

When considering welfare outside the training context, we found that, in the cognitive bias task, dogs from Group Aversive displayed higher latencies for all the stimuli than dogs from Group Reward, with no differences being found between Groups Aversive and Mixed nor between Groups Reward and Mixed. Although affect is hypothesized to exert a greater influence on decision-making under ambiguity (i.e., for the test stimuli: NN, M, NP) than under certainty (i.e., for training stimuli: N, P), other studies in cognitive bias have also found differences for both test and training stimuli [e.g., 32 – 35 , see 35 for a review]. This type of result, with differences found for (at least one of) the training stimuli has also been interpreted as evidence for differences in the valence of the affective states. The fact that differences can emerge for both training and test stimuli has been proposed to result from the fact that choice in the cognitive bias task depends on two different components of the decision-making process: perceived probability and perceived valuation of rewards (and punishments). An individual may be less likely to make a less ‘risky’ or more ‘pessimistic’ response if they consider the reward to be less probable (or punisher more probable) and/or if they consider the reward to be less valuable (or the punisher more aversive) [ 35 , 36 ]. In summary, affective states may influence the responses to both the training and the test stimuli in the cognitive bias task, although different components of the decision-making process may be playing a role. Therefore, the most likely explanation for the present findings is that dogs from Group Aversive considered the food reward less probable (as indicated by the higher latencies to the test stimuli) and also showed a higher valuation of reward loss relative to win (as indicated by the higher latencies to the training stimuli) [ 36 ]. Overall, these results indicate that dogs from Group Aversive were in a less positive affective state than dogs from Group Reward. To our knowledge, the only other study in dogs that addressed the welfare effects of training methods outside the training context was performed by Christiansen et al (2001) [ 13 ], who studied the use of shock collars to train hunting dogs not to attack sheep. No general effects of the use of shock collars on dog fear and anxiety were found one year after training took place. However, unlike the test used by Christiansen et al (2001) [ 13 ], which was a modified version of a temperament test used by the Norwegian Kennel Club, the cognitive bias approach used in the current study is a widely established and well-validated method for evaluating animal welfare (e.g., [ 14 – 18 ]). Hence, to our knowledge, this is the first study to reliably assess and report the effects of aversive- and reward-based training methods in the affective states of dogs outside the training context.

Dogs from Group Reward showed a tendency to learn the cognitive bias task faster than dogs from Group Mixed. Similar findings were observed previously by Rooney et al. (2011) [ 37 ], who found a positive correlation between the reported use of reward-based training methods and a dog’s ability to learn a novel task (touching a spoon with its nose). In another study, Marshall-Pescini et al (2008) [ 38 ] found that dogs with high-level training experience were more successful at opening a box to obtain food than dogs which had received either none or only basic training. Although Marshall-Pescini et al (2008) [ 38 ] reported that all subjects’ training included positive reinforcement methods, they did not specify whether positive punishment and/or negative reinforcement were used in combination. Altogether, previous research suggests that training using positive reinforcement may improve the learning ability of dogs. It remains unclear why a difference was not observed between Group Aversive and Group Reward in the present study. Still, it is important to mention that in all previous studies cited above, animals were required to perform a given behavior in order to obtain a positive reinforcer. Thus, it is unclear whether the same effect would stand if the dogs had to learn a task whose goal was, for example, to perform a behavior to escape from an unpleasant situation. It may be the case that dogs trained with positive reinforcement develop a specific ‘learning set’ [ 39 ] for tasks involving positive reinforcement, but that dogs trained with aversive-based methods perform better in tasks involving some sort of aversive stimuli. Further research is needed to clarify the relationship between training methods and learning ability in dogs.

Notably, we found that the higher the proportion of aversive stimuli used in training, the greater the impact on the welfare of dogs (both within and outside the training context). This result is in line with the findings of a previous survey study, which showed that a higher frequency of punishment was correlated with higher anxiety and fear scores [ 8 ]. Still, in the present study, welfare differences were found even when comparing Groups Reward and Mixed, which used a lower proportion of intended aversive-based techniques as compared to Group Aversive. Dogs from Group Mixed showed higher frequencies of stress-related behaviors, were found more frequently in tense states and panted more frequently during training than dogs from Group Reward. When comparing Group Mixed and Group Aversive, the latter showed a higher frequency of stress-related behaviors and was more frequently found in tense and low behavioral states during training. This seems to suggest that, although dogs trained in ‘low aversive’ schools do not show as many indicators of poor welfare as those trained in ‘highly aversive’ schools, their welfare may still be at stake.

Moreover, our results suggest that the proportion of aversive stimuli used in training plays a greater role on dogs’ stress levels than the specific training tools used. As an example, one school from Group Mixed used pinch and e-collars, whereas another school from Group Aversive only used choke collars during training. Although the tools used by the former school may be perceived as more aversive, the frequency of stress behaviors was higher in dogs being trained at the latter school. The type of (intended) positive reinforcers also appears to be relevant. All schools except the aforementioned school from Group Aversive used primarily food treats as rewards, whereas the latter only used petting. Although this was not the school using the highest proportion of aversive stimuli, it was the school whose dogs showed the highest frequency of stress behaviors (data not shown). Previous research has shown that petting is a less effective reward than food in training [ 40 ]. Having a highly valuable reward might thus be important in reducing stress when aversive stimuli are used in training. The goal of the present study was to test the overall effect on dog welfare of aversive- and reward-based methods as they are used in the real world, but it may be interesting for future studies to focus on disentangling the effects of the different types of stimuli used in training (as has been done with e-collars) [e.g., 9 , 25 ].

Finally, some limitations of the present study must be considered. Firstly, because this was a quasi-experimental rather than an experimental study, we cannot infer a true causal relationship between training methods and dog welfare. To do so would require a randomized control trial. However, conducting an experimental study where dogs are designedly subjected to aversive-based methods would raise ethical concerns, as previous studies have already suggested an association between the use of aversive-based methods and indicators of stress in dogs [ 3 , but see 9 ]. Because we did not randomly allocate dogs to the treatments (training methods), we cannot discard the possibility that there are significant differences between dog-owner pairs that led some owners to choose an aversive-based school and others to choose a reward-based school. There were indeed differences among groups in owner gender, in whether or not the household included children and in the information owners relied on for choosing the dog training school. There were also differences among groups in dog age, FCI breed group and age of separation from the mother. The study was not designed to evaluate the effect of these factors and they were therefore treated as potential confounders in the statistical analysis, in order to account for the possibility that they would affect our results. The effects of training method reported in the study are robust to these confounders. We tested for dog age, presence of children in the household and owner gender, factors which have been shown to potentially affect dog stress and welfare [e.g., 41 – 44 ]. The presence of children in the family has been found to be negatively associated with the owners' perception of the relationship with their dogs, in what is to our knowledge the only study addressing how this factor affects dog behavior [ 43 ]. Most research into the relationship between dog age and stress indicators has been conducted in senior dogs and consistently shows higher baseline cortisol and higher cortisol responses to stressful stimuli in aged dogs [ 45 , 46 ]; however, our study did not include any senior dog. Schöberl et al (2017) [ 44 ] found cortisol to decrease with increasing age of the dog in adult dogs, whereas Henessy et al (1998) [ 42 ] found that the juveniles and adults had higher plasma cortisol levels than puppies. Two of the potential confounders were not included in the analysis because of insufficient reliable data: breed (34% mixed breeds, mainly unknown) and age of separation from the mother (22% unknown). Breed differences in behavior are well established [ 43 ] but the classification of breeds into groups has not been found to systematically correlate with behavioral similarities [e.g., 47 ], and the large percentage of mixed breed dogs where the actual breeds were unknown further constrains a meaningful analysis of this factor in our sample. Literature shows that both early [e.g., 48 ] and late [e.g., 49 , 50 ] separation from the mother (before and after 8 weeks-old, respectively) can be associated with stress-related behavioral problems in dogs. Whereas we do not know the animals’ stress levels before the start of training, cortisol data shows no differences between training groups on non-training days.

Secondly, a volunteer bias cannot be excluded and hence any generalization of the present results must take this in account. Finally, this study focused on welfare and did not compare the efficacy of training methods. Presently, the scientific literature on the efficacy of the different methodologies is scarce and inconsistent [ 3 ]. Whereas some studies suggest a higher efficacy of reward methods [ 5 , 12 , 51 – 53 ], one points in the opposite direction [ 31 ] and three show no differences between methods [ 9 , 54 , 55 ]. This limits the extent of evidence-based recommendations. If reward-based methods are, as the current results show, better for dog welfare than aversive-based methods, and also prove to be more or equally effective to aversive-based methods, there is no doubt that owners and dog professionals should use reward-based training practices. If, on the other hand, aversive-based methods prove to be more effective, the recommendation may be to use aversive stimuli as infrequently as possible during training, and use them in combination with reward-based techniques. This applies not only to training in a formal school setting but whenever owners use reinforcement or punishment in their interactions with the dog.

5. Conclusions

Overall, our results show that companion dogs trained with aversive-based methods experienced poorer welfare during training sessions than dogs trained with reward-based methods. Additionally, dogs trained with higher proportions of aversive-based methods experienced poorer welfare outside the training context than dogs trained with reward-based methods. Moreover, whereas different proportions of aversive-based methods did not result in differences in dog welfare outside the training context among aversive-based schools, a higher proportion of aversive-based methods resulted in poorer welfare during training. To our knowledge, this is the first comprehensive and systematic study to evaluate and report the effects of dog training methods on companion dog welfare. Critically, our study points to the fact that the welfare of companion dogs trained with aversive-based methods is at risk, especially if these are used in high proportions.

Supporting information

S1 appendix. proportion (mean ± standard deviation) of intended aversive-based techniques used during the six training sessions analyzed for each training school..

For each training session, the number of intended positive punishments and negative reinforcements was divided by the total number of intended positive punishments, negative reinforcements, positive reinforcements and negative punishments. Schools A and D were categorized as Group Aversive, Schools C and F as Group Mixed and Schools B, E and G as Group Reward.

https://doi.org/10.1371/journal.pone.0225023.s001

S2 Appendix. Raw data underlying all the analyzes performed in the current research paper.

https://doi.org/10.1371/journal.pone.0225023.s002

S3 Appendix. Negative binomial and generalized linear mixed model details.

S3a Table. Analysis of Generalized Estimating Equation for the stress-related behaviors analysis. S3b Table. Analysis of Generalized Estimating Equation for the behavioral state analysis. S3c Table. Analysis of Generalized Estimating Equation for the panting analysis. S3d Table. Solutions for fixed effects from the generalized linear mixed model for the cognitive bias analysis.

https://doi.org/10.1371/journal.pone.0225023.s003

Acknowledgments

We are grateful, first and foremost, to all dogs and their owners who participated in this study; without them this research would never have been possible. A very special acknowledgment to the dog training schools and their trainers that opened their doors for our participant recruitment and data collection.

We would also like to thank Joana Guilherme-Fernandes for the support provided in the development of the setup for the cognitive bias task, for helping with data collection and especially for all the invaluable discussions during study planning and data interpretation. A special acknowledgment also for Margarida Lencastre and Flávia Canastra, who also helped in data collection. We also want to thank Igor M Lopes for the critical help provided with statistical analysis and Jennifer Barrett for input given during data collection and analysis.

Finally, we would like to thank Dr. Carolyn Walsh and the anonymous reviewers for the detailed input on our work, which helped to substantially improve its quality.

1. Reid P. Learning in dogs. In: Jensen P, editor. The Behavioural Biology of Dogs. CABI; 2007. pp. 120–144.
2. Reisner I. The learning dog: A discussion of training methods. In: Serpell J, editor. The Domestic Dog: Its Evolution, Behavior and Interactions with People, 2nd Edition. Cambridge University Press; 2017. pp. 211–226.
View Article
Google Scholar
4. Hineline PN, Rosales-Ruiz J. Behavior in relation to aversive events: Punishment and negative reinforcement. In Madden GJ, Dube WV, Hackenberg TD, Hanley G P, Lattal KA, editors. APA Handbook of Behavior Analysis, Vol. 1. Methods and Principles. APA Handbooks in Psychology®; 2013. pp. 483–512. https://doi.org/10.1158/1078-0432.CCR-12-3764 pmid:23596103
PubMed/NCBI
11. Companion Animal Welfare Council, 2012. The Use of Electric Pulse Training Aids (EPTAs) in Companion Animals. Available from: http://eprints.lincoln.ac.uk/14640/1/CAWC%20ecollar%20report.pdf (Accessed 12 August 2019).
16. Boissy A, Erhard HW. How studying interaction between animal emotion, cognition, and personality can contribute to improve farm animal welfare. In: Grandin T, Deesing MJ, editors. Genetics and the Behavior of Domestic Animals, 2nd Edition. Academic Press; 2014. pp. 81–113.
21. Skinner BF. Science and Human Behavior. Macmillan; 1953.
26. Voith VL, Borchelt PL. Fears and phobias in companion animals. In: Voith VL, Borchelt PL, editors. Readings in companion animal behaviour. Trenton, NJ: Veterinary Learning Systems; 1996. pp. 140–152. https://doi.org/10.1002/eji.1830261009 pmid:8898941

22 August 2024: Due to technical disruption, we are experiencing some delays to publication. We are working to restore services and apologise for the inconvenience. For further updates please visit our website: https://www.cambridge.org/universitypress/about-us/news-and-blogs/cambridge-university-press-publishing-update-following-technical-disruption

We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings .

Login Alert

> Journals
> Animal Welfare
> Volume 13 Issue 1
> Dog training methods: their use, effectiveness and...

Article contents

Dog training methods: their use, effectiveness and interaction with behaviour and welfare.

Published online by Cambridge University Press: 11 January 2023

Historically, pet dogs were trained using mainly negative reinforcement or punishment, but positive reinforcement using rewards has recently become more popular. The methods used may have different impacts on the dogs’ welfare. We distributed a questionnaire to 364 dog owners in order to examine the relative effectiveness of different training methods and their effects upon a pet dog's behaviour. When asked how they trained their dog on seven basic tasks, 66% reported using vocal punishment, 12% used physical punishment, 60% praise (social reward), 51% food rewards and II% play. The owner's ratings for their dog's obedience during eight tasks correlated positively with the number of tasks which they trained using rewards (P< 0.01), but not using punishment (P = 0.5). When asked whether their dog exhibited any of 16 common problematic behaviours, the number of problems reported by the owners correlated with the number of tasks for which their dog was trained using punishment (P< 0.001), but not using rewards (P = 0.17). Exhibition of problematic behaviours may be indicative of compromised welfare, because such behaviours can be caused by— or result in — a state of anxiety and may lead to a dog being relinquished or abandoned. Because punishment was associated with an increased incidence of problematic behaviours, we conclude that it may represent a welfare concern without concurrent benefits in obedience. We suggest that positive training methods may be more useful to the pet-owning community.

Access options

This article has been cited by the following publications. This list is generated based on data provided by Crossref .

Google Scholar

View all Google Scholar citations for this article.

Save article to Kindle

To save this article to your Kindle, first ensure [email protected] is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle .

Note you can select to save to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Volume 13, Issue 1
EF Hiby (a1) , NJ Rooney (a1) and JWS Bradshaw (a1)
DOI: https://doi.org/10.1017/S0962728600026683

Save article to Dropbox

To save this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Dropbox account. Find out more about saving content to Dropbox .

Save article to Google Drive

To save this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Google Drive account. Find out more about saving content to Google Drive .

Reply to: Submit a response

- No HTML tags allowed - Web page URLs will display as text only - Lines and paragraphs break automatically - Attachments, images or tables are not permitted

Your details

Your email address will be used in order to notify you when your comment has been reviewed by the moderator and in case the author(s) of the article or the moderator need to contact you directly.

You have entered the maximum number of contributors

Conflicting interests.

Please list any fees and grants from, employment by, consultancy for, shared ownership in or any close relationship with, at any time over the preceding 36 months, any organisation whose interests may be affected by the publication of the response. Please also list any non-financial associations or interests (personal, professional, political, institutional, religious or other) that a reasonable reader would want to know about in relation to the submitted work. This pertains to all the authors of the piece, their spouses or partners.

How Science is Revolutionizing the World of Dog Training

I was about a month into raising a new border collie puppy, Alsea, when I came to an embarrassing realization: my dog had yet to meet a person who doesn’t look like me.

I’d read several books on raising a dog, and they all agree on at least one thing: proper socialization of a puppy, especially during the critical period from eight to 20 weeks, means introducing her to as many people as I possibly could. Not just people, but diverse people: people with beards and sunglasses; people wearing fedoras and sombreros; people jogging; people in Halloween costumes. And, critically, people of different ethnicities. Fail to do this, and your dog may inexplicably bark at people wearing straw hats or big sunglasses.

This emphasis on socialization is an important element of a new approach to raising the modern dog. It eschews the old, dominating, Cesar Millan–style methods that were based on flawed studies of presumed hierarchies in wolf packs. Those methods made sense when I raised my last dog, Chica, in the early aughts. I read classic dominance-oriented books by the renowned upstate New York trainers The Monks of New Skete , among others, to teach her I was the leader of her pack, even when that meant stern corrections, like shaking her by the scruff of the neck. Chica was a well-behaved dog, but she was easily discouraged when I tried teaching her something new.

I don’t mean to suggest I had no better option; there was then a growing movement to teach dog owners all about early socialization and the value of rewards-based training, and plenty of trainers who employed only positive reinforcement. But in those days, the approach was the subject of debate and derision: treat-trained mongers might do what you want if they know a biscuit is hidden in your palm, but they’d ignore you otherwise. I proudly taught my dog tough love.

This time, with the assistance of a new class of trainers and scientists, I’ve changed my methods entirely, and I have been shocked to discover booming product lines of puzzles, entertaining toys, workshops and “canine enrichment” resources available to the modern dog “parent,” which has helped boost the U.S. pet industry to $86 billion in annual sales. Choke collars, shock collars, even the word no are all-but-verboten. It’s a new day in dog training.

The science upon which these new techniques are based is not exactly new: it’s rooted in learning theory and operant conditioning, which involves positive (the addition of) or negative (the withdrawal of) reinforcement. It also includes the flipside: positive or negative punishment. A brief primer: Petting a dog on the head for fetching the newspaper is positive reinforcement, because you’re taking an action (positive) to encourage (reinforce) a behavior. Scolding a dog to stop an unwanted behavior is positive punishment, because it’s an action to discourage a behavior. A choke collar whose tension is released when the dog stops pulling on it is negative reinforcement, because the dog’s desirable behavior (backing off) results in the removal of an undesirable consequence. Taking away a dog’s frisbee because he’s barking at it is negative punishment, because you’ve withdrawn a stimulus to decrease an unwanted behavior.

Much has changed about the way that science is applied today. As canine training has shifted from the old obedience-driven model directed at show dogs to a more relationship-based approach aimed at companion dogs, trainers have discovered that the use of negative reinforcement and positive punishment actually slow a dog’s progress, because they damage its confidence and, more importantly, its relationship with a handler. Dogs that receive too much correction—especially the harsh physical correction and mean-spirited “Bad dog!” scoldings—begin to retreat from trying new things.

These new methods are backed by a growing body of science—and a rejection of the old thinking, of wolves (and their descendants, dogs) as dominance-oriented creatures. The origin of so-called “alpha theory” comes from a scientist named Rudolph Schenkel, who conducted a study of wolves in 1947 in which animals from different packs were forced into a small enclosure with no prior interaction. They fought, naturally, which Schenkel wrongly interpreted as a battle for dominance. The reality, Schenkel was later forced to admit, was that the wolves were stressed, not striving for alpha status.

A study from Portugal published last fall in the pre-print digital database BioRxiv (meaning it is not yet peer-reviewed) evaluated dozens of dogs selected from schools that either employed the use of shock collars, leash corrections and other aversive techniques or didn’t—sticking entirely or almost entirely to the use of positive reinforcement (treats) to get the behavior they wanted. Dogs from the positive schools universally performed better at tasks the researchers put in front of them, and the dogs from aversive schools displayed considerably more stress, both in observable ways—licking, yawning, pacing, whining—and in cortisol levels measured in saliva swabs.

These new findings are especially relevant this year. Dog adoption in the COVID-19 era has ballooned, arguably because isolated Americans are newly in search of companionship and because working from home makes at least the idea of raising a puppy feasible. Before the pandemic, it was young city dwellers driving the boom in demand for and supply of dog trainers who employ positive methods, and an explosion in the proliferation of professional trainers across the globe. Often because they’ve delayed or decided against having children, millennials and Generation Z are spending lavish amounts of money on pets: toys, food, puzzles, fancy harnesses, rain jackets, life jackets and training. And those professional trainers, from the Guide Dogs for the Blind organization to renowned handler Denise Fenzi , have formed a legion of experimenters. They universally report that the less negativity they use in training, the more quickly their dogs learn.

Over the past 15 years, handlers with Guide Dogs for the Blind, which trains dogs to be aides for sight-impaired people, have extinguished nearly all negative training techniques and with dramatic results. A new dog can now be ready to guide its owner in half the time it once took, and they can remain with an owner for an extra year or two, because they’re so much less stressed out by the job, says Susan Armstrong, the organization’s vice president of client, training and veterinary operations. Even bomb-sniffing and military dogs are seeing more positive reinforcement, which is why you might have noticed that working dogs in even the most serious environments (like airports) seem to be enjoying their jobs more than in the past. “I don’t think you’re imagining that,” Armstrong says. “These dogs love working. They love getting rewards for good behavior. It’s serious, but it can be fun.”

Susan Friedman , a psychology professor at Utah State University, entered the dog-training world after a 20-year career in special education, a field in which she has a doctorate. In the late 1990s, she adopted a parrot, and was shocked to discover that most of the available advice she could find about raising a well-mannered bird involved only harsh corrections: If it bites, abruptly drop the bird on the floor. If it makes too much noise, shroud the cage in complete darkness. If it tries to escape, clip the bird’s flight feathers. Friedman applied her own research and experience to her parrot training, and discovered it all comes down to behavior. “No species on the planet behaves for no reason,” she says. “What’s the function of a parrot biting your hand? Why might a child throw down at the toy aisle? What’s the purpose of the behavior, and how does it open the environment to rewards and also to aversive stimuli?”

Friedman’s early articles about positive-reinforcement animal training met a skeptical audience back in the early aughts. Now, thanks to what she calls a “groundswell from animal trainers” newly concerned about the ethics of animal raising, Friedman is summoned to consult at zoos and aquariums around the world. She emphasizes understanding how a better analysis of an animal’s needs might help trainers punish them less. Last year, she produced a poster called the “hierarchy roadmap” designed to help owners identify underlying causes and conditions of behavior, and address the most likely influencers—illness, for example—before moving on to other assumptions. That’s not to suggest old-school dog trainers might ignore an illness, but they might be too quick to move to punishment before considering causes of unwanted behavior that could be addressed with less-invasive techniques.

The field is changing rapidly, Friedman says. Even in the last year, trainers have discovered new ways to replace an aversive technique with a win: if a dog scratches (instead of politely sitting) at the door to be let out, many trainers would have in recent years advised owners to ignore the scratching so as not to reward the behavior. They would hope for “extinction,” for the dog to eventually stop doing the bad thing that results in no reward. But that’s an inherently negative approach. What if it could be replaced with something positive? Now, most trainers would now recommend redirecting the scratching dog to a better behavior, a come or a sit, rewarded with a treat. The bad behavior not only goes extinct, but the dog learns a better behavior at the same time.

The debate is not entirely quashed. Mark Hines, a trainer with the pet products company Kong who works with dogs across the country, says that while positive reinforcement certainly helps dogs acquire knowledge at the fastest rate, there’s still a feeling among trainers of military and police dogs that some correction is required to get an animal ready for service. “Leash corrections and pinch collars are science-based, as well,” Hines says. “Positive punishment is a part of science.”

The key, Hines says, is to avoid harsh and unnecessary kinds of positive punishment, so as not to damage the relationship between handler and dog. Dogs too often rebuked will steadily narrow the range of things they try, because they figure naturally that might reduce the chance they get yelled at.

The Cesar Millans of the world are not disappearing. But the all- or mostly positive camp is growing faster. Hundreds of trainers attend “ Clicker Expos ,” an annual event put on in various cities by one of the most prominent positivity-based dog-training institutions in the world, the Karen Pryor Academy in Waltham, Mass. And Fenzi, another of the world’s most successful trainers, teaches her positive-reinforcement techniques online to no less than 10,000 students each term.

While there is some lingering argument about how much positivity vs. negativity to introduce into a training regimen, there’s next to zero debate about what may be the most important component of raising a new dog: socialization. Most trainers now teach dog owners about the period between eight and 20 weeks in which it is vital to introduce a dog to all kinds of sights and sounds they may encounter in later life. Most “bad” behavior is really the product of poor early socialization. For two months, I took Alsea to weekly “puppy socials” at Portland’s Doggy Business , where experienced handlers monitor puppies as they interact and play with one another in a romper room filled with ladders and hula hoops and children’s playhouses, strange surfaces that they might otherwise develop fear about encountering. Such classes didn’t exist until a few years ago.

I also took Alsea to dog-training classes, at a different company, Wonder Puppy . At the first session, trainer Kira Moyer reminded her human students that the most important thing we need to do for our dogs is advocate, which is also based in a renewed appreciation of science. Instead of correcting your dog for whining, for example, stop for a moment and think about why that’s happening? What do they want? Can you give that to them, or give them an opportunity to earn the thing they want, and learn good behavior at the same time?

Enrichment is another booming area of the dog-training world. I didn’t feed Alsea out of a regular dog bowl for the first six months she’s been with me, because it was so much more mentally stimulating for her to eat from a food puzzle, a device that makes it just a little bit challenging for an animal to acquire breakfast. These can be as simple as a round plastic plate with kibble dispersed between a set of ridges that have to be navigated, or as complex as the suite of puzzles developed by Swedish entrepreneur Nina Ottosson . At the highest level, a dog might have to move a block, flip the lid up, remove a barrier or spin a wheel to earn food. Another common source of what we consider “bad” behavior in dogs is really just an expression of boredom, of a dog that needs a job and has decided to give himself one: digging through the garbage, barking at the mail carrier. Food puzzles make dinnertime a job. When Ottosson first started, “they called me ‘the crazy dog lady.’ Nobody believed dogs would eat food out of a puzzle,” she says. “Today, nobody calls me that.”

When Alsea was 4 months old (she’s 12 months now), I traveled south of Portland to Oregon’s Willamette Valley to introduce her to Ian Caldicott , a farmer who teaches dogs and handlers how to herd sheep. First we watched one of his students working her own dog. As the border collie made mistakes, the tension in her owner’s voice escalated and her corrections grew increasingly harsh. “Just turn your back and listen,” Caldicott said to me. “You can hear the panic in her voice creeping in.”

Dogs are smart and can read that insecurity. It makes them question their faith in the handler and, in some cases, decide they know better. Raising a good sheepdog is about building trust between the dog and the handler, Caldicott says. That does require some correction—a “Hey!” when the dog goes left instead of right, at times—but what’s most important is confidence, both in the dog and the handler. In the old days, sheepdogs were taught left and right with physical coercion. Now, they’re given just enough guidance to figure out the right track by themselves. “We’re trying to get an animal that thinks for itself. A good herding dog thinks he knows better than you. Your job is to teach him you’re worth listening to,” Caldicott says. “The ones born thinking they’re the king of the universe, all you have to do is not take that away.”

Update, Oct. 27 : This article has been updated to more clearly reflect Susan Friedman’s approach to animal training.

More Must-Reads from TIME

Breaking Down the 2024 Election Calendar
How Ukraine Beat Russia in the Battle of the Black Sea
The Reintroduction of Kamala Harris
Long COVID Looks Different in Kids
What a $129 Frying Pan Says About America’s Eating Habits
The 1 Heart-Health Habit You Should Start When You’re Young
Cuddling Might Help You Get Better Sleep
The 50 Best Romance Novels to Read Right Now

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

We're Hiring!
Help Center

Dog training methods: their use, effectiveness and interaction with behaviour and welfare

Historically, pet dogs were trained using mainly negative reinforcement or punishment, but positive reinforcement using rewards has recently become more popular. The methods used may have different impacts on the dogs' welfare. We distributed a questionnaire to 364 dog owners in order to examine the relative effectiveness of different training methods and their effects upon a pet dog's behaviour. When asked how they trained their dog on seven basic tasks, 66% reported using vocal punishment, 12% used physical punishment, 60% praise (social reward), 51% food rewards and 11% play. The owner's ratings for their dog's obedience during eight tasks correlated positively with the number of tasks which they trained using rewards (P < 0.01), but not using punishment (P = 0.5). When asked whether their dog exhibited any of 16 common problematic behaviours, the number of problems reported by the owners correlated with the number of tasks for which their dog was trained using punishment (P < 0.001), but not using rewards (P = 0.17). Exhibition of problematic behaviours may be indicative of compromised welfare, because such behaviours can be caused by or result in a state of anxiety and may lead to a dog being relinquished or abandoned. Because punishment was associated with an increased incidence of problematic behaviours, we conclude that it may represent a welfare concern without concurrent benefits in obedience. We suggest that positive training methods may be more useful to the pet-owning community.

Related Papers

Michal Piotr Pregowski

Contemporary dog training and the ongoing changes within this field, particularly ones related to perceptions of dogs and their roles, are interesting topics for academic inquiry. Present practices generally rely upon either the pack-and-dominance concept—leading to top-down, discipline-heavy treatment—or behaviorism and operant conditioning, where great emphasis is placed on positive reinforcement. The “positive” approach underlies state-of-the-art training programs of the second decade of the 21st century. Authors of such programs go beyond the limitations of behaviorism, embracing up-to-date information about the emotional and cognitive abilities of dogs—something that trainers strongly attached to behaviorism are prone to overlook. Such a new approach to dog training does not oppose critical anthropomorphism, and it challenges prior understanding of the dog-human relationship. The relationship in question ceases to be unilateral and becomes a bond of mutual benefit, where a force-free, reward-based method of training is in unison with advertising the self-development potential for humans.

Justyna Wlodarczyk

Vivian Zottola

An online survey about puppy training was sent to members of the Center for Canine Behavior Studies and posted on our social media platforms. Six hundred forty-one (641) qualifying owners provided information on 1023 dogs. About half (48%) of the dogs involved in the study attended puppy training and the balance (52%) did not. The goal of the study was to find out whether puppy training at various ages (1–3 months, 4 months, 5–6 months) helped prevent behavior problems later in life (≥1 year). Attending training at 6 months of age or younger resulted in 0.71 the odds of developing aggressive behavior (95% CI: 0.53–0.97; p = 0.030), 0.64 the odds of having a compulsive behavior (95% CI: 0.45–0.92; p = 0.015), 0.60 the odds of exhibiting destructive behavior (95% CI: 0.37–0.96; p = 0.035), 0.68 the odds of excessive barking (95% CI: 0.47–0.99; p = 0.043), and 1.56 the odds of house soiling (95% CI: 1.08–2.27; p = 0.019). Ancillary findings about the entire study population were that d...

Nicola Rooney

Vets, animal welfare charities, and researchers have frequently cited educating owners as a necessity for improving the welfare of companion dogs. The assumption that improving an owner’s knowledge through an education intervention subsequently results in improvements in the welfare of the dog appears reasonable. However, the complexity of dog welfare and dog ownership and the context in which these relationships occur is rapidly changing. Psychology has demonstrated that humans are complex, with values, attitudes, and beliefs influencing our behaviours as much as knowledge and understanding. Equally, the context in which we individuals and our dogs live is rapidly changing and responding to evolving societal and cultural norms. Therefore, we seek to understand education’s effectiveness as an approach to improving welfare through exploring and understanding these complexities, in conjunction with the relevant research from the disciplines of science education and communication. We a...

Javad Khoshnegah

Journal of Veterinary Behavior: Clinical …

Gabriella Tami

Shelly Volsche

The purpose of my Master’s Thesis research is to determine if and what type of relationship exists between owner-to-dog attachment levels and the degree of aversion used in dog training. I also sought to determine if owner characteristics such as parental status, income, gender, and the like influenced the degree of aversion used. My primary hypothesis is that a negative correlation exists between attachment to one’s dog and aversion used in training. That is, as attachment scores increase, aversion scores decrease. In addition to testing this hypothesis, I collected data to determine if there are correlations between the following: gender and aversion used in dog training; parental status and aversion used in dog training; and, socioeconomic status and aversion used in dog training. Data collection took place via online, self-report surveys and included demographic information on owners and their dogs, assessment of owner to dog attachment levels (utilizing the Pet Attachment Life Impact Scale), and a measure of aversion used in training. The survey also included open-ended questions that provided owners with the opportunity to add ethnographic value to the data. To validate the breakdown of training philosophies, I completed a pilot survey of professional trainers in which I asked them to classify training philosophies and assign degrees of aversion to commonly recommended methods and techniques. The results of this dog trainer survey aided in the construction of the aversion measurement tool included in the dog owner survey mentioned above. My recruitment strategy involved multiple Internet outlets along with local canvassing. The sample consisted of 673 respondents from across the United States. Of that population, 90.1% were female and 88.0% identified as white. Additionally, 78.6% were not parents, but a large percentage of the population (65%) considered themselves their dog’s parent or guardian. A weak positive correlation (r=.217, p<.001) was found between participants’ attachment and the reported frequency of aversion used in training their dogs. These results run contrary to my hypothesis. This thesis discusses the interpretations of these findings, including with respect to changing human-dog interactions in the United States. This thesis provides a window into a growing phenomenon of “pet parents,” with data that could drive future research. The human-canine bond is an area of study still in development. In addition, a growing population of individuals identify as “childfree,” choosing not to raise children. Many of these individuals are “parenting” their pets, instead, and the majority of the current sample seems to reflect that population. A large and growing body of research exists concerning the perspective of the dog (cognition, neuroscience, and emotion research in particular), while most data collected from the human perspective focuses on epidemiological and physiological assessments of dog ownership. Future research could focus more on the emotionality of dog ownership to uncover driving factors behind, and to improve, the choices made in care and training.

Daniel Mills

The Veterinary record

David Chamberlain

Applied Animal Behaviour Science

Victoria Voith

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

RELATED PAPERS

Giovanni Sophia

Salla Mikkola

Andrew Jagoe

Claire Diederich

Scientific Reports

Naomi Harvey

Bulletin of University of Agricultural Sciences and Veterinary Medicine Cluj-Napoca. Veterinary Medicine

Marina Spinu

Domestic Dog Cognition and Behavior

Jessica Greenebaum

HANDBOOK OF APPLIED DOG BEHAVIOR AND TRAINING

Mehmet Cihat Keleş

Journal of Veterinary Behavior: Clinical Applications and Research

Journal of Forensic Sciences

Frontiers in Veterinary Science

Aubrey Fine

Javma-journal of The American Veterinary Medical Association

James Serpell

Behavioural Processes

Eniko Kubinyi

Scientific reports

Alexandra Moesta

Kristina Adams

PsycEXTRA Dataset

Journal of Veterinary Behavior-clinical Applications and Research

Fco. Javier Diéguez

Péter Pongrácz

The click is not the trick: the efficacy of clickers and other reinforcement methods in training naïve dogs to perform new tasks

Associated data.

The following information was supplied regarding data availability:

Raw data is available in the Supplemental Files .

A handheld metal noisemaker known as a “clicker” is widely used to train new behaviors in dogs; however, evidence for their superior efficacy compared to providing solely primary reinforcement or other secondary reinforcers in the acquisition of novel behavior in dogs is largely anecdotal.

Three experiments were conducted to determine under what circumstances a clicker secondary reinforcer may result in acquisition of a novel behavior more rapidly or to a higher level compared to other readily available reinforcement methods. In Experiment 1, three groups of 30 dogs each were shaped to emit a novel sit and stay behavior of increasing duration with either the delivery of food alone, a verbal stimulus paired with food, or a clicker with food. The group that received only a primary reinforcer reached a significantly higher criterion of training success than the group trained with a verbal secondary reinforcer. Performance of the group experiencing a clicker as a secondary reinforcer was intermediate between the other two groups, but not significantly different from either. In Experiment 2, three groups of 25 dogs each were shaped to emit a nose targeting behavior and then perform that behavior at increasing distances from the experimenter using the same three methods of positive reinforcement as in Experiment 1. No statistically significant differences between the groups were found. In Experiment 3, three groups of 30 dogs each were shaped to emit a nose-targeting behavior upon an array of wooden blocks with task difficulty increasing throughout testing using the same three methods of positive reinforcement as previously tested. No statistically significant differences between the groups were found.

Overall, the findings suggest that both primary reinforcement alone as well as a verbal or clicker secondary reinforcer can be used successfully in training a dog to perform a novel behavior, but that no positive reinforcement method demonstrated significantly greater efficacy than any other.

Introduction

In positive reinforcement training, a trainer increases the frequency of a target behavior by regularly following it with a primary reinforcer (a stimulus which has reinforcing efficacy without the need for any other behavioral manipulation). A trainer may also present a secondary reinforcer—a stimulus which is initially neutral but which acquires the ability to increase the likelihood of a response by being repeatedly paired with a primary reinforcer ( Domjan, 2005 ). In principle, a secondary reinforcer can be anything the animal perceives—such as a light, a sound, a movement, or a smell. Common secondary reinforcers in dog training are a spoken word, a whistle, or the sound from a clicker device ( Burch & Bailey, 1999 ; Schaefer, 1999 ; Skinner, 1951 ).

Despite the prominence of positive reinforcement-based training methods in the professional dog training community ( Blackwell et al., 2008 ; Hiby, Rooney & Bradshaw, 2004 ), recommendations of how, when, and what method of positive reinforcement should be used are inconsistent (see Browne et al. (2017) , for a review of the general content in best-selling dog training books). Indeed, more generally, the research reveals mixed results as to the best approach to train animals. Studies in other species have shown that, as compared to the use of both a primary and secondary reinforcer, the use of a primary reinforcer alone when establishing a new behavior is more effective in cats ( Willson et al., 2017 ), but less effective in goats ( Langbein et al., 2007 ), and equally effective in horses ( McCall & Burgin, 2002 ).

A widely promulgated form of secondary reinforcement in dog training is the use of a handheld metal noisemaker known as a “clicker” that is first paired with food, after which the “click” can be used to audibly signal the performance of a desired behavior in the animal ( Pryor, 1999 ). There is somewhat more consistency in the training literature with regards to the efficacy of clicker training for behavior acquisition—namely, that it is often no better than other secondary reinforcers or the use of primary reinforcement alone. In some of the earliest studies to compare clicker training to primary reinforcement alone in domesticated animals, similar protocols were used to shape horses ( Williams et al., 2004 ) and dogs ( Smith & Davis, 2008 ) to perform the novel behavior of touching their nose to a cone with either primary reinforcement alone or the sound of the clicker followed by primary reinforcement. The behavior was then placed under extinction, with secondary reinforcement administered to half of the clicker-trained horses and all of the clicker-trained dogs. Neither study found a difference in the number of trials the animals required to reach training criterion across training conditions, and while Williams et al. (2004) found that neither condition nor administration of secondary reinforcement during extinction influenced the total number of trials required to reach extinction, Smith & Davis (2008) found that dogs trained with the clicker required significantly more time and trials to extinguish the behavior when primary reinforcement was discontinued. Although the continued presence of a secondary reinforcer is expected to slow behavioral extinction due to a less pronounced change in the stimulus configuration from training to extinction ( Pearce, 2008 ; Williams, 1994 ), Smith & Davis (2008) suggested that the difference between their findings with dogs and Williams et al. (2004) results with horses could be a true species difference in behavioral processes. Additionally, factors such as heightened arousal to the sound of the mechanical clicker, frustration, or the inability to discriminate between testing and extinction conditions ( Kelleher & Gollub, 1962 ; Williams, 1994 ) could also play a role in increased resistance to extinction for clicker-trained dogs but not horses. Previous studies have compared the relative efficacy of using primary reinforcement alone to the sound of the clicker followed by primary reinforcement when teaching a new behavior, but little is known about the relative efficacy of the clicker compared to other forms of secondary reinforcement.

Two prior studies compared the efficacy of a clicker to a verbal secondary reinforcer in training a novel behavior in dogs and piglets. Thorn et al. (2006) measured the latency to sit upon approach for shelter dogs trained with a clicker compared to a verbal secondary reinforcer. They found no differences in any dependent measures between conditions on the first day of training. On the second day of training, dogs trained with a clicker had a significantly greater latency to sit during their first trial compared to their last trial on the first day and had a significantly greater mean latency to sit on the second day overall. This indicates that a clicker or a verbal secondary reinforcer may be equally efficacious for initially training a behavior, but that clicker-trained dogs have a lower retention of the behavior across days. In a recent study with piglets, Paredes-Ramos et al. (2020) used a clicker or a verbal secondary reinforcer to first teach a “fetch” behavior, and then had the piglets fetch a novel object in a discrimination task. They found that clicker-trained piglets acquired the fetch behavior in significantly fewer trials than the verbally-reinforced piglets, but that verbally-reinforced piglets made significantly more correct choices on the discrimination task. However, both studies omitted a primary reinforcer alone condition, thereby making assessment of putative reinforcing effects difficult.

To date, two studies have investigated the effect of shaping a novel behavior using primary reinforcement alone or in association with a spoken word or a clicker. Chiandetti et al. (2016) trained dogs to open a bread box using a clicker paired with food, the spoken word “bravo” paired with food, or food alone, and then tested whether the dogs could generalize that behavior to a new apparatus. They found that dogs in all three conditions were able to learn and generalize the behavior to an equivalent degree. Using these same three reinforcement types, with the spoken word “next” instead of “bravo” as the verbal reinforcer, Dorey, Blandina & Udell (2020) shaped a stay behavior in one set of puppies, and a wave behavior in second set of puppies. They found that puppies in the stay behavior group trained with food alone progressed significantly further in shaping approximations than those trained with the clicker, with verbally-reinforced puppies reaching an intermediate shaping approximation. When teaching the wave behavior, Dorey, Blandina & Udell (2020) saw no differences across the three conditions.

Several procedural factors may contribute to divergent findings from these studies of primary and putative secondary reinforcement in dogs, and differences in the design of prior studies make it difficult to reconcile their disparate results. For example, Marshall-Pescini et al. (2008) have shown that dogs with more training experience performed significantly better when learning a new task than dogs without prior training. In the studies that have investigated the use of a clicker during dog training, Smith & Davis (2008) and Chiandetti et al. (2016) enrolled dogs 6 months to 12 years old and 8 months to 13 years, respectively. These dogs had prior training experience, but no prior exposure to a clicker. Thorn et al. (2006) utilized dogs over a year old of unknown ownership, and consequently unknown training histories. As dogs in shelters typically have unknown training histories, the best way to reduce the likelihood of such experience influencing dogs’ performance is to test younger dogs. Dorey, Blandina & Udell (2020) addressed this concern by using shelter puppies 8–24 weeks old in their experiments. Additionally, it is possible that the frequency and duration of training sessions may impact behavioral acquisition. Demant et al. (2011) found that dogs trained to sit and stay in a basket once or twice weekly reached a significantly higher acquisition level than those trained daily, as did those trained with only one session as opposed to three consecutive sessions per day. Thorn et al. (2006) conducted testing over 2 days each separated by 2 days. Smith & Davis (2008) used from 2 to 6 days of consecutive training while Chiandetti et al. (2016) tested over “several” days. And while the number of overall trials was controlled by Demant et al. (2011) , Thorn et al. (2006) , and Dorey, Blandina & Udell (2020) , dogs in both Smith & Davis (2008) and Chiandetti et al. (2016) were offered unlimited trials to reach session criteria. Differences in training method efficacy may also relate to the type or complexity of the behavior being trained. Fugazza & Miklósi (2015) found that significantly more dogs learned to open a sliding door when trained with the “Do as I do” method than those shaped with a clicker, but that training method did not influence their ability to learn to jump in the air. Studies that taught a relatively simple behavior like sit and stay found differences in behavior acquisition ( Dorey, Blandina & Udell, 2020 ) and retention ( Thorn et al., 2006 ), whereas no difference in training method efficacy was seen for teaching more complex behaviors like targeting ( Smith & Davis, 2008 ), waving ( Dorey, Blandina & Udell, 2020 ), and object manipulation ( Chiandetti et al., 2016 ).

Given that prior studies varied along multiple procedural dimensions that could have affected acquisition of a novel behavior, we saw the need for a cohesive set of experiments that utilized participants of a similar age, with similar training histories, and that received the same amount of time to complete the testing protocol. By utilizing a single-session design and testing puppies naïve to training, we believe that any indication of greater performance by clicker-trained dogs could be more clearly related to the relationship between the reinforcement method and the behavior being trained.

The aim of the present study was to clarify whether differences exist in the rate of acquisition and (equivalently) terminal level reached in a fixed number of trials when establishing novel behaviors in dogs with one of three commonly used positive reinforcement methods: Primary Alone, Verbal Secondary, or Clicker Secondary. In the first experiment, each dog was taught to sit for increasing periods of time using one of the three reinforcement methods, and the differences in the greatest durations of sit achieved in a fixed number of trials were compared between groups to determine whether one condition resulted in the dogs learning the behavior more rapidly and sitting for a longer duration than the others. In the second experiment, dogs were first shaped to touch a cone next to the trainer, and then required to target the cone at increasing distances. Both the dogs’ progress through the shaping approximations and the greatest distance from the cone that the dogs achieved in a fixed number of trials were compared between groups to determine which condition facilitated the highest level of behavioral acquisition. In the final experiment, dogs were shaped to touch an array of blocks, and then progressed through levels of increasing specificity wherein only targeting specific blocks was reinforced. The shaping approximation that was achieved, the amount of time it took to complete each approximation, and the number of attempts made during those shaping levels were compared between groups to determine which condition resulted in the greatest degree of behavioral specificity in a fixed number of trials.

General design

Testing was conducted inside a conference room at the Arizona Humane Society (AHS: Phoenix, AZ, USA). Metal fencing was used to enclose the testing area in each experiment, with enclosure size varied across experiments. Participants were puppies from AHS, 8–22 weeks old, that came from the shelter’s intake and adoption areas. This age range was chosen so that dogs were physically capable of completing the tasks but unlikely to have experienced prior training. Dogs had to be free of illness, injury, and anesthetics to participate. Per shelter protocol, all dogs received their regular diet in the morning and evening while the studies were being conducted and were not tested immediately after meal consumption. No dog participated in more than one experiment. All procedures in this study were conducted with approval from the Arizona State University Institutional Animal Care and Use Committee (16-1462R RFC2; 19-1668R).

RG served as the trainer throughout the experiment 1 , and all individuals handling the dogs were female. After an opportunity to urinate and defecate, each dog was placed in the testing area and given 4 min to explore. The trainer only interacted with the dog if approached. Next, the trainer stood and placed the clicker on her left middle finger and two treat bags along her shirt collar (Experiments 1 and 2) or a treat bag along her waistband on her lower back (Experiment 3). One bag contained the primary reinforcer (pieces of hot dog), while the other (in Experiments 1 and 2) contained pieces of Pup-peroni® Original Beef Flavor ( Big Heart Pet, Inc. ). The latter treats were tossed into Zone 2 after a dog emitted a behavior in Zone 1 in order to bring the dog back into Zone 2 for the start of each trial. This was done to get the dog out of the reinforced behavior for the experiment (sit or touch) without manhandling it. It also allowed the dog to start every trial in Zone 2, thereby enabling us to assess its motivation to engage in the next trial, which would otherwise have been difficult to do if the dog remained in Zone 1 for the duration of the test.

To ensure the dog did not have prior experience of sitting (Experiment 1) or targeting (Experiments 2 and 3) when prompted, the trainer approached it, said “sit,” (Experiment 1) or presented her empty palm (Experiments 2 and 3), and waited for a response three times. If the dog sat or touched three times, it was excluded from participation. If it sat or touched once or twice, the trainer allowed 30 s to elapse before the dog was again prompted three times. If the dog sat or touched for two or all three prompts, it was removed from the experiment. Dogs that sat or touched only once remained in the experiment.

Dogs were randomly assigned to one of three groups: Primary Alone, Verbal Secondary, or Clicker Secondary. Dogs in the two secondary reinforcement groups received a piece of hot dog paired with the verbal sound “chee” or with the click from a clicker as their reinforcement. Dogs experienced the pairing of the primary reinforcer with the secondary reinforcer 20 times in succession, with each pairing beginning immediately after consumption of the previous piece of food, which is consistent with or exceeds the number of pairings used by Thorn et al. (2006) , Smith & Davis (2008) , Chiandetti et al. (2016) , and Dorey, Blandina & Udell (2020) . To control for the amount of food received prior to the start of training, and any association thereby formed between the experimenter and food, dogs in the primary alone group also received 20 primary reinforcers with the trainer silently thinking “chee” before each presentation of the primary reinforcer to mimic the approximate duration it took to deliver the secondary reinforcers in the other groups. Dogs that did not consume the primary reinforcer were removed from the experiment.

Experiment 1

In the first experiment, we tested whether there were any differences in the maximum duration a dog could be trained to sit for primary reinforcement only, secondary reinforcement from a verbal stimulus, and secondary reinforcement from a clicker in a fixed number of trials.

Setting and subjects

The testing area for Experiment 1 was 157.5 cm by 259.1 cm and was comprised of Zone 1 on the left, where the dog was reinforced for sitting, and Zone 2 on the right, where it was not reinforced for its behavior ( Fig. 1 ). A video camera was set up to record all testing within the fenced area.

An external file that holds a picture, illustration, etc.
Object name is peerj-09-10881-g001.jpg

The 157.5 cm by 259.1 cm enclosed space making up the testing area. A line of tape separating Zone 1 (where sitting was reinforced) from Zone 2 (no reinforced behavior) was placed 127.0 cm from the far end of Zone 1. The star marks where the trainer stood during testing and where reinforcement delivery occurred (the RDS). A research assistant was present on the far right end of Zone 2.

Of the 110 dogs that participated in this study ( Table S1 ), seven sat on command, implying prior training; another 13 did not sit within 25 min of initiating training; and one was later revealed to have been ill during testing, thus excluding it from analysis (see “Procedure” below). The final sample size consisted of 29 dogs in the primary alone group and 30 each in both the verbal and clicker secondary groups. A power sensitivity analysis was conducted using G*Power ( Faul et al., 2007 ) to assess what magnitude of effect could be reasonably detected with this final sample size. These calculations showed that a sample size of 89 dogs has 80% power to detect a medium effect size of f = 0.33 for group differences, controlling for covariates of age, sex, and weight.

After the pairing of secondary reinforcers was complete, the trainer stood at the far-left end of Zone 1 on the Reinforcement Delivery Spot (RDS—see Fig. 1 ). Each dog was given 25 min to enter Zone 1 and sit for one second. If a dog did not perform the behavior within this time, it was excluded from further participation. If the dog did sit within Zone 1 for 1 s, it received its assigned reinforcement, and a Pup-peroni treat was thrown across the line into Zone 2 to encourage the dog to leave Zone 1 before initiating a new trial. If the dog did not follow the treat into Zone 2, the trainer walked to where the treat landed and verbally encouraged the dog to retrieve it, and then returned to the RDS once it left Zone 1 to consume the treat.

All dogs were initially required to sit for 1 s to receive reinforcement, and the sitting criterion increased in 3 s intervals as they performed a sit for the required duration three times in succession. Dogs received reinforcement at the end of each sit that met criterion. If the dog did not sit for the entire duration, it did not receive reinforcement, but still received a treat thrown into Zone 2 to start the next trial. If the dog sat for less than the current duration criterion twice in a row, the sit duration was reduced by 3 s. If the dog alternated performing criterion compliant and noncompliant sits, it remained at its current criterion level until it successfully sat correctly three times in succession or failed to reach criterion twice in succession. Figure 2 diagrams this adaptive schedule of reinforcement. For every 15 s that the dog spent in Zone 2, the trainer would call to it by saying “puppy ba ba ba” and made a kissing sound to encourage it to return to Zone 1. Each sit constituted one trial, and each dog could perform a maximum of 50 trials. Data were retained and analyzed for dogs that completed at least one trial. A dog’s testing ended either when it did not perform another sit within Zone 1 within 2 min from its last reinforced sit, or once it reached 50 trials.

An external file that holds a picture, illustration, etc.
Object name is peerj-09-10881-g002.jpg

Illustration shows an example dog at Level 2, where it was required to sit for 4 s to obtain reinforcement. If it sat for 4 s three times in a row, it would advance to Level 3, where it was expected to sit for 7 s. If, however, at Level 2 the dog did not sit for a full 4 s twice in a row, it would drop down to Level 1, where it was expected to sit for 1 s. If the dog alternated sitting for 4 s and sitting for less than 4 s, it remained at Level 2.

Results and discussion

The final sample size was n = 29 in the primary alone group and n = 30 in each of the other two groups. After verifying the assumptions of normality and homogeneity of variances of residuals with a Shapiro–Wilk test ( W (89) = 0.97, p = 0.054), and a Levene’s test ( F (2, 86) = 2.94, p = 0.06), and confirming that no standard deviation of one group exceeded twice the standard deviation of any other group (SD primary = 11.7 ; SD verbal = 8.8 ; SD clicker = 11.1), we performed a one-factor ANCOVA in SPSS (Version 25) International Business Machines Corp. (2017) with age, sex, and weight as covariates to compare the influence of group on the maximum duration of the dogs’ sitting behavior. Figure 3A provides the means and their standard errors for each group. The ANCOVA revealed a statistically significant difference in durations of sit between groups ( F (2, 83) = 3.95, p = 0.02), with no covariate reaching statistical significance: p = 0.57 for age, p = 0.85 for sex, and p = 0.28 for weight. Furthermore, Tukey post-hoc testing revealed a significant difference between the verbal secondary and primary alone groups ( p = 0.01, d = 0.76, 95% CI [0.23 to 1.28]), while the differences between the clicker secondary and primary alone groups ( p = 0.22, d = 0.33, 95% CI [−0.19 to 0.84]) and clicker and verbal secondary groups ( p = 0.10, d = 0.43, 95% CI [−0.08 to 0.94]) were not statistically significant 2 .

An external file that holds a picture, illustration, etc.
Object name is peerj-09-10881-g003.jpg

(A) Mean duration of a sit (with error bars depicting standard error of the mean) in seconds for each group of dogs in Experiment 1. Bracket with asterisk between groups indicates a significant difference in means on post-hoc pairwise comparisons ( p < 0.05). (B) Mean highest completed step (with error bars depicting standard error of the mean) for each group of dogs in Experiment 2. (C) Mean highest completed level (with error bars depicting standard errors of the mean) for each group of dogs in Experiment 3.

To test whether the superior performance of the primary alone group was due to the lack of an intervening delay imposed by the delivery of the secondary reinforcer, we randomly selected 12 out of 30 dogs in each group and analyzed the first, middle, and last 2-min segments of their testing sessions to assess the delay to reinforcement they experienced using CowLog (Version 3.0.2; Pastell, 2016 ). Coders blind to the hypotheses and purpose of the experiment coded the delays to reinforcement of these dogs. The average delays to reinforcement per condition for the clicker and verbal secondary groups were 2.5 s (SD = 0.7) and 2.6 s (SD = 0.7) respectively, and that of the primary alone group was 3.1 s (SD = 1.5). Thus, the greater efficacy of dogs in the primary alone group cannot be attributed to a shorter delay to reinforcement than the secondary reinforcement groups.

It is possible that dogs in the verbal secondary group performed significantly worse than those in the primary alone group because they had been habituated to the human voice. Although we used the uncommon and likely unfamiliar sound “chee” as our verbal secondary reinforcer to try to control for words that the dogs already had experience with, it is possible that the constant exposure that dogs have to the human voice had desensitized them such that any spoken word could not prove to be a strongly effective secondary reinforcer ( Pearce, 2008 ).

One possible explanation for the failure to detect a benefit of the clicker in this experiment may be that the primary reinforcer was delivered promptly. Pryor (2009) , in arguing for the effectiveness of clickers, gave an example where a click sound was used to signal reinforcement to the animal when it performed a behavior at a distance from the trainer. Under conditions where primary reinforcement cannot be provided quickly, secondary reinforcement can provide feedback to an animal that it has behaved correctly and will receive primary reinforcement upon returning to the trainer. Experiment 2 sought to investigate whether secondary reinforcement could have an impact on the performance of a behavior at a greater distance from the trainer.

Experiment 2

To test whether secondary reinforcement could have a greater impact on the acquisition of novel behavior at a greater distance from the trainer, in Experiment 2 we trained dogs under the same three positive reinforcement methods from Experiment 1, but shaped them to touch a cone at increasing distances from the trainer. This targeting behavior preceded the distance component of testing.

The testing area for Experiment 2 was a 625.0 cm by 157.5 cm area comprising Zone 1, which included the Touch Spot (TS) and the Primary Reinforcer Delivery Spot (PRDS), and Zone 2, which included the Upper Step Markers (USM). The layout of the testing area is shown in Fig. 4 . All reinforced behavior and delivery of primary reinforcement occurred at the TS. The trainer stood on the PRDS when delivering the primary reinforcer over the TS, and both knelt and stood on the PRDS during Steps 1–7. A research assistant sat outside the fencing at the far-right end of Zone 2 during training, and a video camera was set up to record all testing within the fenced area.

An external file that holds a picture, illustration, etc.
Object name is peerj-09-10881-g004.jpg

The star represents where the trainer stood during testing and where primary reinforcement delivery by the trainer occurred (PRDS). The triangle represents the touch spot where all reinforced behavior occurred and where primary reinforcement was delivered (TS). The short lines represent the upper step markers (USM) 9–17 that marked where the trainer stood for Steps 9–17. The USM were placed in line with the PRDS and were spaced 50 cm apart from one another, with USM 9 being 50 cm away from the line. A research assistant was present on the far end of Zone 2.

Of the 84 dogs that participated in this study ( Table S2 ), nine did not touch the trainer’s hand with their nose within 25 min of initiating training (see “Procedure” below) and were excluded from participating further. Twenty-five dogs in each group touched the trainer’s hand with their nose at least once, and their data were analyzed. Power calculations to determine the sensitivity of effect size detection for the final sample size were conducted using G*Power ( Faul et al., 2007 ). The sensitivity analysis showed that a sample size of 75 dogs has 80% power to detect a medium effect size of f = 0.37 for group differences, controlling for covariates of age, sex, and weight.

After the pairing of secondary reinforcers was complete, the trainer walked to the PRDS and knelt holding out a primary reinforcer in her right palm over the TS. The dog had 25 min to approach and nose-touch her palm. If the dog did not do this within the time allotted, it was removed from the study. If the dog touched her palm with its nose, it received its assigned reinforcement, and a treat was thrown into Zone 2 to encourage it to leave the trainer’s side. If the dog did not follow the treat into Zone 2, the trainer tossed up to two more treats, after which point the research assistant would call it toward the far-right side of Zone 2.

To measure the effectiveness of the three reinforcement methods, each dog was first shaped to perform the nose-targeting behavior, and then to perform this behavior at increasing distances from the trainer. Table 1 shows all successive approximations used in the shaping procedure. Dogs received their designated reinforcement when they nosed the object as required by the current step. All dogs were required to complete Steps 1–7 before a distance component was introduced to the testing. At Step 8, the experimenter pivoted on the PRDS and took one step backward from the PRDS to the line. For Steps 8–17, the dog was tasked with nose-touching the tip of the cone that was placed on the TS just as it had done in Steps 6 and 7.

List of each testing step and corresponding reinforced behavior, as well as the trainer’s location and posture at each step.

Step	Reinforced behavior	Trainer location, stature
1	Touch trainer’s palm containing food	PRDS (0 cm from cone), kneeling
2	Touch trainer’s empty palm	PRDS (0 cm from cone), kneeling
3	Touch ball in trainer’s hand	PRDS (0 cm from cone), kneeling
4	Touch ball affixed to cone held in trainer’s hand	PRDS (0 cm from cone), kneeling
5	Touch ball affixed to the cone placed on TS	PRDS (0 cm from cone), kneeling
6	Touch cone placed on TS	PRDS (0 cm from cone), kneeling
7	Touch cone placed on TS	PRDS (0 cm from cone), standing
8	Touch cone placed on TS	Line (127 cm from cone), pivot, standing
9	Touch cone placed on TS	USM 9 (177 cm from cone), standing
10	Touch cone placed on TS	USM 10 (227 cm from cone), standing
11	Touch cone placed on TS	USM 11 (277 cm from cone), standing
12	Touch cone placed on TS	USM 12 (327 cm from cone), standing
13	Touch cone placed on TS	USM 13 (377 cm from cone), standing
14	Touch cone placed on TS	USM 14 (427 cm from cone), standing
15	Touch cone placed on TS	USM 15 (477 cm from cone), standing
16	Touch cone placed on TS	USM 16 (527 cm from cone), standing
17	Touch cone placed on TS	USM 17 (577 cm from cone), standing

Each nose-touch or failure to touch the cone or experimenter’s palm constituted one trial, and each dog was permitted to perform a maximum of fifty trials and a maximum of 17 steps. Data were retained and analyzed for dogs that completed at least one trial. For every 15 s that the dog spent in Zone 2, the trainer called to it by saying “puppy ba ba ba” and making a kissing sound to encourage it to return to Zone 1. If the dog did not enter Zone 1 within 2 min of its last nose-touch, testing ended. If the dog remained inside Zone 1 for 2 min without emitting a nose-touch, it was considered a failed trial and a treat was thrown into Zone 2 to initiate the next trial. If this occurred three times successively, testing ended. If the dog crossed the line with at least one forepaw into Zone 1, but did not nose-touch, it was considered a failed trial. Once a dog reached Step 8, it was not permitted to return to earlier shaping steps after two failed attempts at Step 8, and testing was ended.

Data from 25 dogs in each group were analyzed. After verifying the assumptions of normality and homogeneity of variances of residuals with the Shapiro–Wilk test ( W (75) = 0.971, p = 0.08) and Levene’s test ( F (2, 72) = 1.14, p = 0.33), and confirming that no standard deviation of one group exceeded twice the standard deviation of any other group (SD primary = 3.4 ; SD verbal = 3.5 ; SD clicker = 4.4), we performed a one-factor ANCOVA in SPSS (Version 25) International Business Machines Corp. (2017) with age, sex, and weight as covariates to compare the influence of group on completed levels. Figure 3B provides the means and standard error of the means for each group. The ANCOVA indicated that the difference in number of steps completed between groups was not statistically significant, F (2, 69) = 1.10, p = 0.37, with no covariate reaching statistical significance, p = 0.57 for age, p = 0.71 for sex, and p = 0.07 for weight. Effect sizes for the three pairwise differences in the number of steps completed were d = 0.12 (95% CI [−0.43 to 0.68]), d = 0.09 (95% CI [−0.46 to 0.65]), and d = 0.22 (95% CI [−0.34 to 0.78]), for comparing verbal secondary to primary alone, clicker secondary to primary alone, and clicker secondary to verbal secondary, respectively.

Although we should be careful when making inferences regarding non-significant results, it should be noted that, on average, dogs in the clicker secondary group completed an additional shaping step than those in the verbal secondary group, with the primary alone group achieving an intermediate number of steps. Moreover, the greatest proportion of verbal secondary-group dogs and the smallest proportion of clicker secondary-group dogs dropped out at or before Step 7—the final shaping step before distance was added to the task, with the primary alone group dropping out at an intermediate number of steps.

In order to test for potential differences in delay to reinforcement between the three groups, we randomly selected 12 out of 25 dogs in the primary alone and clicker secondary groups and 10 out of 25 in the verbal secondary group and had coders blind to the study aims and methods analyze the first, middle, and last 2-min segments of their testing videos for delays to primary reinforcement by using BORIS (Version 5.1.0; Friard & Gamba, 2016 ). Average delays to reinforcement per condition for the primary alone, verbal secondary and clicker secondary groups were 3.0 s (SD = 1.4), 2.5 s (SD = 1.4) and 4.0 s (SD = 2.0) respectively. While it is possible that dogs in the clicker secondary group were negatively impacted by the greater delay to reinforcement they experienced compared to the other reinforcement groups, it should be noted that, because, on average they completed more steps in the distance phase of training than dogs in either of the other two groups, they necessarily waited longer as the trainer walked the additional distance to deliver primary reinforcement to them.

In this experiment we did not have enough statistical power to analyze differences between the groups in the shaping of the targeting behavior separately from the distance-related behavior. Future studies with a larger sample size may be able to separately analyze performance on the initial shaping steps and the subsequent steps where the behavior was performed at a distance from the experimenter. Overall there were no statistically significant differences between the groups.

Our aim thus far has been to find the circumstances in which a clicker may result in acquisition of a novel behavior to a higher level than other readily available reinforcement methods. Experiment 1 found no advantage to secondary reinforcement in training a simple sit behavior. Experiment 2 tested the hypothesis that clickers aid in training a behavior performed at a distance from the trainer ( Pryor, 2009 ) and found no one reinforcement group better sustained a targeting behavior over increasing distances. Feng, Howell & Bennett (2018) reported that owners and trainers believe clickers to be more beneficial when teaching discrete behaviors, such as targeting, rather than less specific ones, such as coming when called—a belief also consistent with Pryor (2009) . Consequently, in Experiment 3 we test whether the clicker secondary reinforcer allows dogs to reach a higher criterion of behavioral acquisition in a fixed number of trials when the specificity of the behavior is the focus of training.

Experiment 3

In Experiment 3, dogs were tasked with learning to emit a nose-targeting behavior on an array of alternating yellow and blue wooden blocks. Initially nose-targeting any block in the array was reinforced, but in subsequent phases of training only dogs’ contact with specific blocks was reinforced. Table 2 gives the criteria for progression through the testing levels. Not only did Pryor (2009) propose that clicker training is beneficial due to the speed and precision with which it could be employed, but she also stated that the clicker is intrinsically reinforcing for dogs and aids in keeping them engaged with a task because it activates the SEEKING circuit. Panksepp’s (2010) theory of a SEEKING circuit states that “SEEKING coaxes animals to acquire resources needed for survival. It promotes learning by mediating anticipatory eagerness, partly by coding predictive relationships between events” (p.538). Consequently, Panksepp’s (2010) theory of a SEEKING circuit predicts that dogs trained using a clicker may make more contact with the blocks in this experiment because engagement of the SEEKING circuit increases the reinforcing qualities of the apparatus. Thus, we sought to determine whether dogs in the clicker secondary group attained higher levels of acquisition of this new behavior when compared to those in the primary alone or verbal secondary groups, and whether they attained different rates of responding at each level.

Outline of the shaping and testing phases, as well as all possible levels and their corresponding blocks eligible for reinforcement when touched.

Level	Blocks eligible for reinforcement
Training/Shaping	One randomized block is eligible for reinforcement; all blocks are eligible once
1	All blocks are eligible for reinforcement
2	Blocks 2, 3, and 4 are eligible for reinforcement
3	A randomized pair of blocks (either 1 and 2 or 3 and 4) are eligible for reinforcement
4	A randomized pair of blocks (either 1 and 2 or 3 and 4) are eligible for reinforcement; not the same pair as level 3
5	Blocks 1, 3, and 5 are eligible for reinforcement
6	Blocks 2 and 4 are eligible for reinforcement
7	One randomized block is eligible for reinforcement
8	One randomized block is eligible for reinforcement; not the same block as Level 7
9	One randomized block is eligible for reinforcement; not the same block as Levels 7 or 8
10	One randomized block is eligible for reinforcement; not the same block as Levels 7–9
11	One randomized block is eligible for reinforcement; not the same block as Levels 7–10

The testing area for Experiment 3 was a 157.5 cm by 259.1 cm area comprising the Reinforcement Barrier (RB: two 61.0 cm by 91.4 cm tri-fold cardboard barriers supported by metal table easels), the Block Line (BL), and Zones 1 and 2, as shown in Fig. 5 . The trainer knelt in the 35.6 cm space between the barriers during testing. The BL was comprised of a 94.0 cm by 2.4 cm by 1.3 cm metal bar secured to the floor with Velcro tape, and had five 3.8 cm by 6.4 cm by 9.8 cm magnetic painted wood blocks evenly spaced 18.7 cm apart along the length of the bar. All reinforced behavior and delivery of primary reinforcement occurred along the BL. A research assistant sat outside the fencing at the far-right end of Zone 2 during training, and a video camera was set up to record all testing within the fenced area.

An external file that holds a picture, illustration, etc.
Object name is peerj-09-10881-g005.jpg

The tape line dividing the zones sits 127.0 cm from the far left end of Zone 1. The thin dotted lines represent the Reinforcement Barrier (RB), between which the trainer knelt during testing and behind which sat the primary reinforcement and resetting treats. The black bar represents the Block Line (BL) where all reinforced behavior occurred and where primary reinforcement was delivered, and the arrow from 1 to 5 indicates the numbered direction of the blocks.

One hundred and twelve dogs participated in the third experiment ( Table S3 ). Of these, one was excluded because it touched an empty outstretched palm, implying prior training; another was excluded for vomiting during testing; and one would not eat the primary reinforcer. Nineteen dogs were excluded for not touching all five blocks within 25 min of testing, thus excluding a total of 22 dogs from the analysis. A power sensitivity analysis was conducted using G*Power ( Faul et al., 2007 ), which showed that the final sample size of 90 dogs has 80% power to detect a medium effect size of f = 0.25 for the focal between-group effect, controlling for the covariates of age, sex, and weight.

After the pairing of secondary reinforcers was complete, the trainer led the dog to a research assistant, who faced it away from the testing enclosure while the trainer assembled the testing apparatus. The primary reinforcer and the treat used to move dogs to Zone 2 after completed trials were placed in bowls behind the right RB (from the trainer’s perspective when facing the dog), and a cell phone running a timer was placed behind the left RB. The trainer placed each of the five blocks on the BL, then knelt between the RBs and placed a primary reinforcer on a predetermined, randomly-assigned block. The research assistant returned the dog to the testing enclosure, and it was given 25 min to enter Zone 1 and touch the block with the primary reinforcer on top of it. If the dog did not perform this behavior within the 25 min, it was removed from the experiment. If the dog nose-touched the block with the primary reinforcer on it, it received its assigned reinforcement, and a treat was thrown into Zone 2 to encourage the dog to leave Zone 1 before returning for the next trial. While the dog retrieved this treat, the trainer placed a primary reinforcer on a different predetermined randomly-assigned block, and the process was repeated until all five blocks had been nose-touched once, thus completing the shaping phase.

Once the dog had the experience of touching each block individually, it was then required to perform this behavior on increasingly specific groupings of blocks. The testing phase consisted of 11 levels, each comprising a different arrangement of blocks that were eligible for reinforcement if nose-touched. A dog advanced to the next level only when it touched a correct block with its nose four times successively. Dogs were permitted an unlimited number of attempts to nose-touch the blocks until they touched one eligible for reinforcement, at which time primary reinforcement was delivered directly above the block that it touched. Table 2 and Fig. 6 show details of the progressively more specific testing phases.

An external file that holds a picture, illustration, etc.
Object name is peerj-09-10881-g006.jpg

Blocks are numbered from the trainer’s left to right, beginning at Block 1. Consecutive blocks alternated colors, with a 3 blue/2 yellow and 3 yellow/2 blue arrangement assigned to equal numbers of dogs in each condition. Blocks were evenly spaced 18.7 cm apart.

Each dog had 30 min to complete the shaping and testing phases. Data were retained and analyzed for dogs that completed the shaping phase and were reinforced for at least one nose-touch at Level 1. If a dog remained in Zone 1 for 30 s without completing a nose-touch, including those blocks not eligible for reinforcement, the trainer tossed a treat into Zone 2 to encourage it to leave Zone 1 and then re-enter it for another trial. Training ended if it (a) did not enter Zone 1 within 2 min of its previous nose-touch; (b) completed the shaping phase and all 11 testing levels in less than 30 min; or (c) once 30 min had elapsed—whichever occurred first.

Data from 30 dogs in each group were analyzed. The assumptions of normality and homogeneity of variances of residuals were verified with a Shapiro–Wilk test ( W (90) = 0.98, p = 0.12), a Levene’s test ( F (2, 87) = 0.09, p = 0.91), and we confirmed that no standard deviation of one group exceeded twice the standard deviation of any other group (SD primary = 2.7; SD verbal = 2.4; SD clicker = 3.0). A MANCOVA was used, instead of an ANCOVA as in Experiment 1 and 2, to account for testing levels nearest each other likely correlating more highly than levels further apart. Missing data on the nose-touches-per-level variable were handled using maximum likelihood estimation, which achieved maximum power compared to alternatives such as listwise deletion ( Schafer & Graham, 2002 ). Because two dependent variables were assessed using the same dogs, a Bonferroni adjustment was used to control the family-wise Type I error rate for the two sets of tests, comparing results to a nominal alpha = 0.025. Although effect sizes and confidence intervals for pairwise differences are reported for descriptive purposes only, reported CIs are 97.5%, to be consistent with the Bonferroni adjustment.

Due to violation of the sphericity assumption, a one-factor MANCOVA was performed in SPSS (Version 25) International Business Machines Corp. (2017) controlling for age, sex, and weight to compare the influence of reinforcement group on the levels achieved. Figure 3C provides the means and standard errors of the means for each group. The MANCOVA indicated that the difference in number of levels achieved between groups was not statistically significant, F (2, 86) = 1.25, p = 0.29. The covariates of sex ( p = 0.04), age ( p = 0.14), and weight ( p = 0.06) were also not statistically significant. Effect sizes for the three pairwise differences in the number of levels achieved were d = 0.35 (97.5% CI [−0.24 to 0.95]), d = 0.01 (97.5% CI [−0.57 to 0.59]), and d = 0.33 (97.5% CI [−0.25 to 0.91]), for comparing verbal secondary to primary alone, clicker secondary to primary alone, and clicker secondary to verbal secondary, respectively.

To test whether dogs in the clicker secondary group attained a different response rate at each level than those in the primary alone or verbal secondary groups, we performed a one-factor MANCOVA in SAS (Version 9.4; Cary, NC, USA) controlling for age, sex, and weight comparing the influence of group on the average number of nose-touches made per testing level. PROC MIXED, specifying a random intercept, was used to employ maximum likelihood estimation for missing data. Blind coders recorded the number of nose-touches made by all dogs from their video recordings. The video of one dog in the verbal secondary reinforcement group was lost and unable to be coded. Mean count-per-interval inter-observer agreement (IOA) was tested on 8 out of 30 videos in the primary alone and clicker secondary groups, and on 8 out of 29 videos in the verbal secondary group. IOA was above 90%. The MANCOVA indicated no statistically significant difference between groups in the average number of nose-touches made per level, F (2, 55.4) = 2.21, p = 0.12, and that the level-by-group interaction was also not significant, F (20, 37.8) = 0.59, p = 0.89, with no covariate reaching statistical significance, p = 0.80 for age, p = 0.34 for sex, and p = 0.20 for weight. There was a linear trend to the data when testing the relationship between number of nose-touches and testing level, F (1, 26.1) = 39.64, p < 0.01, indicating that dogs responded more frequently as testing level increased, controlling for age, sex, and weight. Effect sizes for the three pairwise differences in the average number of nose-touches made per level were d = 0.11 (97.5% CI [−0.48 to 0.69]), d = 0.31(97.5% CI [−0.28 to 0.89]), and d = 0.39 (97.5% CI [−0.20 to 0.97]), for comparing verbal secondary to primary alone, clicker secondary to primary alone, and clicker secondary to verbal secondary, respectively.

To investigate the possibility of differences in delay to reinforcement in our groups, we randomly selected 12 dogs from each group and blind coders analyzed the first, middle, and last 2-min segments of their videos and recorded the duration of their delays to primary reinforcement using BORIS (Version 5.1.0; Friard & Gamba, 2016 ). We found that the average delays to reinforcement per condition for the primary alone, verbal secondary, and clicker secondary groups were 1.2 s (SD = 0.6), 1.5 s (SD = 0.5), and 1.3 s (SD = 0.4) respectively. As group performance was ranked by the delay to reinforcement, it is possible that the delay to reinforcement affected the testing level achieved in each of the groups.

Across three experiments, we found no evidence to support the claim that using a clicker as a secondary reinforcer for training dogs results in acquisition of a novel behavior more rapidly or to a higher level compared to primary reinforcement alone or a verbal secondary reinforcer. The only significant difference between reinforcement conditions in our experiments was between the primary alone and verbal secondary groups in Experiment 1, where dogs in the primary alone group attained a significantly longer mean duration of a sit than verbally-reinforced dogs. These results have some points of consistency with both Thorn et al. (2006) and Dorey, Blandina & Udell (2020) in that neither study found clicker-trained dogs learned a sit behavior significantly faster than those trained with primary reinforcement alone or with a verbal secondary reinforcer. The absence of any significant differences between the primary reinforcement and either of the secondary reinforcement groups in both our second and third experiments is consistent with findings by Smith & Davis (2008) , Chiandetti et al. (2016) , Dorey, Blandina & Udell (2020) , and Williams et al. (2004) . Smith & Davis (2008) and Williams et al. (2004) found no difference between primary reinforcement alone and a clicker group when teaching Basenjis and horses respectively a targeting behavior, both using procedures similar to that reported in our second experiment. When teaching behaviors of greater specificity akin to that of our third experiment, Chiandetti et al. (2016) observed no benefit of either form of secondary reinforcement when training dogs to open a bread box, and neither did Dorey, Blandina & Udell (2020) when training dogs to perform a “wave” behavior.

To date, one study has suggested any benefit to a clicker secondary reinforcer. Paredes-Ramos et al. (2020) reported that clicker-trained piglets acquired a novel behavior in significantly fewer trials than did a verbally-reinforced group. A single trainer used a 10-step shaping protocol to teach ten piglets in each group to fetch a novel object for 30 trials a day until all piglets acquired the behavior. A piece of a packaged cookie was the primary reinforcer, which was placed on the ground 1 s after piglets heard their secondary reinforcer. Each day, a different object had to be fetched, and the criterion for the ten shaping levels alternated between five and three consecutive repetitions at every other shaping level before advancing to the next one. Piglets have received little if any prior attention from scientists studying the acquisition of novel behavior, so the particular combination of conditions that led to this result and its broader implications remain unclear.

In the following sections, we address possible explanations for the lack of support for greater efficacy of a clicker over other available reinforcement methods.

Age of the dogs tested

It might be argued that the dogs in our studies were too young to attain the behaviors we attempted to train, however the success of a majority of dogs in each experiment contradicts this contention. We also included age as a covariate in all our analyses and did not find any significant impact on the dogs’ performance. At least one prior study found that dogs as young as 1.5 months can be successfully trained in operant tasks. Lozovskaia (1985) reported that 1.5-month-old pups learned an escape task more rapidly than 7-month-olds.

Statistical power

Another possible explanation for our failure to find a positive impact of using a secondary reinforcer is that our studies did not have sufficient power to detect an effect. Our experiments only had enough power to detect medium and large effects, and thus a small effect of the clicker could have gone undetected. On the other hand, the lack of a clear clicker effect found in our (and others) studies, over several replications, provides evidence that if there is any effect, it may not be large enough to have practical impact under testing conditions explored to date.

Number of primary reinforcement pairings

To test the effect of the clicker on the acquisition of a novel behavior, it must be established as a secondary reinforcer by pairing it sufficiently with the primary prior to testing. In our experiments, the clicker was paired with the primary reinforcer 20 times in order to establish the conditional relationship, which was consistent with both the Smith & Davis (2008) and Dorey, Blandina & Udell (2020) procedures in dogs and Williams et al. (2004) with horses. Under laboratory conditions, Skinner (1938) recommended between 30 (1951) and 60 (1938) pairings of the CS and US for dogs and rats, respectively, while Kelleher & Gollub (1962) indicated that pairings beyond 100 trials no longer strengthened the conditioned reinforcement effect for rats. On the other hand, Pryor (2006) has suggested that only two or three pairings are needed to establish the clicker as a conditioned reinforcer. Similarly, Chiandetti et al. (2016) claimed that two or three pairings were sufficient, and neither Thorn et al. (2006) nor Paredes-Ramos et al. (2020) paired the auditory stimulus with the primary reinforcer at all before initiating testing. While additional pairings would be expected to more strongly establish the stimulus as a secondary reinforcer ( Wike, 1966 ), our procedure was consistent with or more substantial than others in the literature, and we were constrained by the possibility of satiation in our young subjects. It is unlikely that dogs tested in our experiments became satiated prior to the start of testing given the size of each primary reinforcer (approximately 0.63 cm 3 ), and 83% of dogs tested across all experiments continued participating after the pairing protocol, but to provide more than 20 pieces of primary reinforcement before testing began could have affected the motivation of our subjects.

Function of the clicker

Despite the widespread belief that clicker training facilitates faster acquisition of a novel behavior in dogs ( Feng, Howell & Bennett, 2018 ), the peer-reviewed scientific literature has consistently shown that this is not the case. As Feng, Howell & Bennett (2016) and Dorey & Cox (2018) have noted, ambiguity exists in the definitions that scientists and practitioners use when referring to “clicker training,” which is concerning given the apparent disconnect in the two communities’ beliefs in the efficacy of clicker training. Martin & Friedman (2011) speculated that in clicker training, the “click” sound is a secondary reinforcer, a bridging stimulus, and an event marker. A marking stimulus is distinguished from a secondary reinforcer in that it does not provide information about a future opportunity to obtain primary reinforcement. Rather, it is simply a novel and unexpected auditory or visual cue that distinguishes the targeted response from the other behaviors the animal was emitting at that time ( Lieberman, McIntosh & Thomas, 1979 ). If the clicker were a bridging stimulus, it would temporally connect the desired response with the delayed food reinforcement through stimulus-stimulus relationships as opposed to response-stimulus relationships ( Dorey & Cox, 2018 ; Williams, 1994 ). As utilized in the testing phases of our and others’ experiments, the clicker was not a novel stimulus; however it did provide information about a future opportunity to obtain primary reinforcement, and it connected response-stimulus relationships. Although no study has directly tested if the clicker functions as a marking stimulus, bridging stimulus, or secondary reinforcer, by definition alone it does not seem as though the clicker could have been functioning as either a marking or bridging stimulus; instead, the clicker appears to function most similar to a secondary reinforcer because it is deployed immediately following the completion of the desired response and is paired to reliably predict the arrival of the primary reinforcer. Dorey & Cox (2018) suggested that the effectiveness of the clicker as a secondary reinforcer could be tested by presenting the clicker contingent upon the occurrence of a new response, or by comparing resistance to extinction for individuals in which a clicker was used in training and those for whom it was not. While the latter recommendation has been shown to be unreliable in measuring the strength of the secondary reinforcer in laboratory experiments with rats ( Kelleher & Gollub, 1962 ; Williams, 1994 ), implementing a final testing phase in which the clicker is used to train a novel behavior in an applied setting could be used to detect evidence of the clicker’s effect as a secondary reinforcer ( Wike, 1966 ), as was done in Smith & Davis (2008) .

If the clicker was functioning as a secondary reinforcer, we would expect to see any additional reinforcement value of the clicker reflected in an increased rate of performing the desired behavior compared to a control group that only receives primary reinforcement ( Williams, 1991 ). Although we did not find this effect here, standard learning theory states that in a situation in which the primary reinforcer is always promptly available, the influence of a secondary reinforcer on the rate of acquisition is weak compared to that of the primary reinforcer alone ( Rescorla & Wagner, 1972 ).

Laboratory studies dating back to Pavlov (1928) have shown that that more rapid reinforcement leads to better acquisition, thus emphasizing the importance of measuring delays to reinforcement in studies of the efficacy of reinforcement methods. The three experiments reported here show somewhat contradictory findings on this issue, as the rank order of delay to reinforcement did not match performance in the first two experiments, but it did for the third experiment. No prior study comparing clicker training to other positive reinforcement methods has reported the delay to reinforcement for any condition, so the impact of adding a secondary reinforcer on behavioral acquisition in other scenarios is unknown. Clickers or other secondary reinforcers could improve acquisition of novel behavior by reducing the delay to reinforcement in situations where it is not possible to provide immediate primary reinforcement, but they might also have a negative impact on behavioral acquisition in situations where the provision of a secondary reinforcer increases the delay to primary reinforcement, such as when the trainer is already standing next to their dog. Future research should investigate this issue directly by experimentally manipulating the delays to secondary and primary reinforcement.

Applied environment

In applied settings such as where dogs are typically trained, control of environmental stimuli is much less precise and variability in testing conditions is inevitably introduced by the presence of a human in the environment, which leads to uncontrolled factors involved in the relationship of the animal with the human as well as errors in the timing and delivery of stimuli. Our studies and those of Thorn et al. (2006) , Smith & Davis (2008) , Chiandetti et al. (2016) , and Dorey, Blandina & Udell (2020) attempted to maintain controlled experimental environments similar to a laboratory setting, but the environments of these studies are closer in actuality to the real world than a typical animal laboratory. It is quite possible that clicker training functions differently in applied settings than in laboratories, and interactions between dogs and their trainers must play a role in learning ( Feng, Howell & Bennett, 2018 ; Pryor, 1999 ).

Feng et al. (2018) attempted to evaluate the effectiveness of clickers in the home setting by asking owners to teach their dogs new behaviors using either a clicker followed by food or delivering primary reinforcement only. After training, owners were asked to rate their dogs’ and their own perceived difficulty and enjoyment of the task. Feng et al. (2018) found that owners in the clicker group reported a less challenging training experience when teaching a nose-targeting behavior than was reported by owners in the food-only group but observed no differences between training groups for the other five behaviors. No measures were taken of the dogs’ success in learning the different tasks. It is possible that more robust inferences about the training methods could have been made had owners trained their dogs with both methods and then rated the perceived difficulty and enjoyment of each method, or if the dogs’ performances when learning these new behaviors had been objectively measured.

Real-world environments also involve a great diversity of individuals attempting to train their dogs, and future studies should also investigate the impact of variability among trainers on behavioral outcomes. The current study, in line with Thorn et al. (2006) , Smith & Davis (2008) , Chiandetti et al. (2016) , and Dorey, Blandina & Udell (2020) only utilized one trainer throughout testing. This is an advantage for assessment of the impact of a training method but should be broadened in future studies. China, Mills & Cooper (2020) , in a study comparing positive reinforcement to positive punishment in the training of dogs, found significant differences in behavioral acquisition between professional trainers utilizing the same testing protocol.

Conclusions

Of the three experiments reported here investigating the circumstances under which a clicker may result in acquisition of a novel behavior more rapidly or to a higher level of difficulty or specificity compared to other readily available reinforcement methods, only one difference between groups was detected. In Experiment 1, dogs trained to sit for increasing periods of time with primary reinforcement alone reached a significantly higher level of performance than those trained with verbal secondary reinforcement. Clearly, in the present experiments, we have only explored a small subset of the ways in which clickers could be used in dog training. Clickers can certainly be used to train far more complex behaviors than what was studied here—the SPCA Auckland used clicker training to teach dogs how to drive a car ( Wynne, 2012 )—but thus far there is no evidence that such behaviors could not be as efficiently trained using any other reinforcement method. Future investigations should explore how clicker-trained dogs perform on different tasks, with special focus on the role of delay to secondary and primary reinforcement, and how clicker training functions in an applied setting, bearing in mind the need for objective outcome measures which are independent of owners’ and trainers’ preconceptions.

Supplemental Information

Supplemental information 1.

Each A# represents an individual dog, with their performance on each trial reflected in the column below their identifying number. Dogs in Experiments 1 and 2 could complete no more than 50 trials, while dogs in Experiment 3 could complete no more than 49 trials.

Supplemental Information 2

“Longest sit” is the duration in seconds of the longest sit achieved in Experiment 1, except for dogs noted as “sat on command,” or “never sat.” Seven dogs were excluded for sitting on first instruction; 13 for not sitting within 25 minutes of beginning training; and one due to identification of illness. Sex is male (m) or female (f). Age, sex, and weight were determined at the date the dogs were tested. IDs are those noted in shelter records. A dog for which the identification number was not recorded at the time of testing is missing its weight measurement and as such is marked “unknown.”

Supplemental Information 3

“Highest Step” is the number of the highest completed training step achieved during the shaping and distance components of Experiment 2, except for dogs noted as “never touched.” Sex is male (m) or female (f). Age, sex, and weight were determined on the date the dogs were tested. IDs are those noted in shelter records.

Supplemental Information 4

“Highest Level” is the number of the highest level of training the dog completed during testing in Experiment 3. Sex is male (m) or female (f). Age, sex, and weight were determined on the date dogs were tested. IDs are those noted in shelter records.

Funding Statement

This work was supported by the Association of Professional Dog Trainers (No. MGS0477-APDT). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Additional Information and Declarations

The authors declare that they have no competing interests.

Rachel J. Gilchrist conceived and designed the experiments, performed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the paper, and approved the final draft.

Lisa M. Gunter conceived and designed the experiments, analyzed the data, authored or reviewed drafts of the paper, and approved the final draft.

Samantha F. Anderson analyzed the data, authored or reviewed drafts of the paper, and approved the final draft.

Clive D.L. Wynne conceived and designed the experiments, prepared figures and/or tables, authored or reviewed drafts of the paper, and approved the final draft.

The following information was supplied relating to ethical approvals (i.e., approving body and any reference numbers):

The Arizona State University Institutional Animal Care and Use Committee provided full approval for this research (16-1462R RFC 2, 19-1668R).

Cheap and Budget Friendly Recipes
Health Related Illness Diets
Homemade Treat Recipes
Homemade Dog Treats for Health Issues
Seasonal Recipes
Can Dogs Eat…
Dog Diseases & Conditions
Dog Symptoms
Dog Grooming
Caring For Seniors
Dog Loss & Grieving
Dog Reproductive Health
Treatments and Home Remedies
Dry Dog Food
Wet Dog Food
Best Dog Products
Dog Accessories
Dog Health Products
CBD for Dogs
Toy Dog Breeds
Working Dog Breeds
Terrier Dog Breeds
Sporting Dog Breeds
Non-Sporting Dog Breeds
Mixed Breeds
Hound Dog Breeds
Livestock and Herding Dog Breeds

Top Dog Tips - Dog Food Recipes, Care Tips & Best Dog Supplies Reviews

Science-based Dog Training: How Research Influenced Our Approach to Training Dogs

Table of Contents

The Alpha Dog Theory of Training

Classical conditioning: the pavlov's dogs, operant conditioning, punishment, reward-based training, and clicker training, modern science-based dog training methods.

S ince the late Mesolithic period, when humans first began to domesticate wolves, canines have been an integral part of our lives. We are linked to them in so many ways, from using them as working dogs to competing, breeding, and, of course, as companions. And, since man has domesticated dogs and cats ( Driscoll et al. 2009 ), we've discovered some science-based dog training methods that have proven to be effective.

Anthropomorphism is when we attribute human emotions or behaviors to animals, something that so many of us are guilty of. I’m sure at some point, we’ve all thought that our dog was upset with us for leaving them on their own or looked guilty after destroying our favorite slippers.

A study by Alexandra Horowitz, Ph.D. in 2009 found that the ‘guilty look’ appears to be more associated with the owner’s body language and tone of voice , rather than being a display of remorse. Anthropomorphism is a difficult thing to avoid as we have only ever seen the world from a human perspective, and it isn’t easy to put ourselves in their shoes.

There is nothing wrong with adoring our pets and treating them as one of the family, but it is only by learning to think like a dog that we can truly understand these animals whose lives are so inextricably linked with ours. This mindset also aids us in training our dogs and communicating with them effectively. In fact, many science-based dog training methods are based on this theory of thinking like a dog .

In this article, we'll take a look at how science-based dog training methods have progressed over the last century – what we've learned, which methods have been discounted, and which ones we still use today as the most effective approach to training dogs. I'll discuss the following:

The Alpha Dog Theory
Classical Conditioning (Pavlov's Dogs)
Punishment/Reward/Clicker Training

ALSO READ: 12 Tips on How to Mentally Stimulate a Dog (Based on Science)

Science-Based Dog Training: How research influenced our approach to training dogs.

Over the years, an enormous amount of research has been carried out to gain a deeper understanding of canine behavior and psychology, and with this, our approach to dog training has changed a great deal. We have learned to adapt to new findings and have gained a greater appreciation of their mindset. However, opinions vary greatly, and there is still a fair amount of disagreement about how best to get them to perform the behaviors we want.

One of the most debated topics to this day, despite a good amount of evidence , is that of the ‘alpha dog’ or dominance theory. This theory refers to the need for an owner to assert their place as the leader of the pack in order to remain in control of their dog’s behavior.

The Wolf: The Ecology and Behavior of an Endangered Species

L. David Mech has since carried out a vast amount of research on wild wolves. He has now rejected this commonly held belief, as it has become clear that there are enormous differences between the behavior of wild and captive animals. He also talked about how “alpha” and “beta” wolves are scientifically inaccurate terms today.

In this short video, Mech also expands a little more on the terminology of “alpha” as applied to wolves:

More scientists chipped in. A prominent anthrozoologist (read: a clever person who studies human-animal interactions), John Bradshaw, points out other flaws in relating this early research to the behavior of domestic dogs. The ancestors of our modern-day canines were Eurasian Grey wolves, which are only distantly related to the American Timberwolves used in the studies. Several thousands of years of evolution and interaction with humans have changed our pet dogs into a species far removed from their wild ancestors to a point where their behavior cannot be compared to their wild counterparts.

The differences between wolves and dogs were further demonstrated in a series of studies carried out at Eötvös Loránd University, Budapest, at the beginning of this century. Groups of wolf and canine pups were hand-reared under the same conditions, and a number of experiments were conducted to determine differences in behavior between the two species.

When faced with an impossible task, such as trying to obtain food from a bin that was sealed shut, the dogs would look to their owner for help, whereas the wolves would attempt to get to it themselves. They also found that dogs quickly learned to make eye contact with humans for a food reward, while wolves did not.

These studies revealed that dogs were more responsive to their owner than human strangers, but wolves showed no difference in their behavior between the two. The researchers concluded that “selective processes took place in the course of domestication.”

RELATED: Science of Human-Dog Psychology – Are You and Your Dog A Good Match?

No discussion on the history of science-based dog training would be complete without a mention of Ivan Pavlov , the king of classical conditioning. In his well-known experiments in the 1890s, he found that if a bell was rung at the same time as the food was presented, over a period of time, the dogs began to salivate in response to the sound of the bell alone.

This was a revelation at the time and something that has a lot of significance in canine training. Dogs are constantly making these types of associations on a daily basis, with both positive and negative consequences. For instance, if a dog has a bad experience with a man who has a beard, they might develop a fear of all men with beards, which can cause a problem when your bearded friend comes around for dinner.

With an understanding of this process, classical conditioning can be used to our advantage in science-based dog training. We may also be able to limit its adverse influences.

There is little documentation on dog training before WW1, and methods used throughout the first half of the 19 th century are viewed as heavy-handed from today’s standpoint. The book Training Dogs: A Manual by Colonel Konrad Most was published in 1910 and used techniques such as collar corrections and physical punishment of dogs.

Another military man who became well-known for his dog training skills is William Koehler . Working at the War Dog Training Centre during WW2, he then became famous for training dogs for movies and for his book The Koehler Method of Dog Training (1962). Koehler also employed forceful techniques such as pulling hard on choke chains, throwing chains at dogs from a distance, and the ‘alpha roll’: rolling the dog over and pinning it down to show dominance.

The Koehler Method of Dog Training Certified Techniques By Movieland's Most Experienced Dog Trainer

However, Skinner can’t take all the credit for this theory of how effective operant conditioning became as a dog training method, as his work was based on Edward L. Thorndike’s ‘Law of Effect’ from 1905. Using hungry cats, boxes, and a similarly debatable approach to animal well-being, Thorndike found that behaviors followed by desirable consequences are more likely to be repeated , with the opposite also being true.

In the world of behavioral training, the term ‘punishment’ refers merely to something that reduces the likelihood of an action being repeated. However, this term can cause confusion and make it sound like bad behavior needs to be corrected with chastisement or even abuse of the dog.

When it comes to dog training, punishment can be ‘positive’ or ‘negative,’ which can also be misleading. These expressions mean only whether something is added or taken away, not whether they are ’good’ or ‘bad.’ For example, positive punishment could be a verbal scolding, and negative punishment can mean removing your attention by ignoring your dog for a short time.

The same terms are used with reinforcement. For instance, positive reinforcement might be the addition of a food reward, and negative reinforcement can mean the removal of pressure on the lead once your dog is walking nicely by your side.

Heavily influenced by Pavlov and Skinners' research, clicker training came about in the 1940s, thanks to Keller Breland and Marion and Bob Bailey . This science-based dog training method is built upon the combination of both classical and operant conditioning. (I recommend you read this interview with the pair on how it all got started – it's fascinating.)

To start with, the pairing of the clicker with positive reinforcement is classical conditioning, but when the dog learns to perform the desired behavior to receive the ‘click’ alone, operant conditioning is at work. This form of training has proved highly effective with many animal species, including dogs and marine mammals.

Don't Shoot the Dog: The Art of Teaching and Training

Dominance and punishment were common means of science-based dog training until the 1970s when more humane techniques were popularized by people such as Barbara Woodhouse . Although she was an advocate of the dominance theory and the use of choke chains, her techniques were a lot less harsh than those of her predecessors, and the tide began to turn towards less forceful methods of dog training, such as using positive reinforcement.

However, it was also around this time that the Monks of New Skete appeared on the scene. At first, they seemed to promote a positive philosophy of understanding, compassion, and communication, but they were also responsible for popularizing the alpha roll .

Based on a misinterpretation of the way subordinate wolves roll on their backs to appease more dominant members of the pack, it was not realized that this was a voluntary action, not forced by the higher-ranking animal. The monks also supported other forms of physical punishment, such as shaking dogs by the scruff of their necks and hitting them under the chin. Thankfully, it appears that they have changed their tune somewhat, as they no longer promote the use of the alpha roll.

Despite proof of the success of operant conditioning back in the early 1900s, it took until the 1980s for it to become a widespread technique for shaping a dog’s behavior. Once dogs trained using positive methods such as clicker training began to do well in obedience and sports, people started to sit up and take notice.

Realizing that most dog owners were averse to using harsh, traditional techniques to train their pets, animal behaviorist Ian Dunbar began to conduct seminars and release videos encouraging everyday dog owners to train their pets. He promoted the use of friendly, science-based dog training methods such as food rewards as positive reinforcement.

RELATED: 25 Best Dog Training Books for Beginners and Advanced Trainers

Over 100 years after Thorndike's research, and in light of so much scientific and anecdotal evidence, it is surprising that the dominance method is still used. According to dog trainer Prescott Breedon , it may be because of “the very normal human phenomenon of dismissing new information that does not conform to a pre-existing understanding because it is threatening to their world-view .”

Dog-training TV personalities have the ability to be extremely influential, and the danger comes when people take their advice at face value without considering other options or the credentials of the person in question. Just because someone has managed to get a TV show (and you know to whom I'm referring) doesn’t mean they know what they are talking about when it comes to science-based dog training. People have caused more harm than good by attempting to use techniques at home that they don’t fully understand or know how to use correctly.

Thankfully, animal welfare is now extremely important to people, and most owners know that they will form a far more fulfilling and loving bond with their pets through mutual respect and an understanding of animal psychology. We don’t want our dogs to associate us with the fear, mistrust, and pain that can be associated with traditional methods, ones in which science has proved to be flawed and often dangerous or ineffective.

For all of us out there embarking on a dog training journey, it's always best to research and read as much as we can and learn from the experts about science-based dog training , not unsupported claims and opinions of celebrities.

RELATED: Most Effective Dog Training Methods According to Science

LATEST FEATURES

4 Tips & Tricks for Navigating Peak Flea and Tick Season

Have Pet, Will Travel: Guide to Spring or Summer Trips with...

Can I Afford A Dog?

How to Get Rid of a Dog: The Right Way

Why Are Dogs So Loyal?

Why Do Pets Make Us Happy?

Dog Names Starting With Z

How Many Dogs Are Too Many?

Can Dogs Get Sick From Humans?

Dandie Dinmont Terrier Breed Profile

Terms of Use
Privacy Policy
No AI Clause

Search This Blog

Companion animal psychology.

Happy cats. Happy dogs. Thanks to science. By Zazie Todd, PhD, award-winning author of WAG, PURR, and now BARK!. Inspiring people to have happier pets since 2012. Answers to your questions about cat and dog training and behaviour.

Dog Training Science Resources

A Parsons JRT illustrates research on dog training methods

Scientific Research on Dog Training Methods

Covered by Zazie Todd PhD at Companion Animal Psychology: Positive Reinforcement in Dog Training: Little Dogs vs Big Dogs
Also included in the summary of the series on positive reinforcement and dog training
Covered by Zazie Todd PhD at Companion Animal Psychology: How Many People Use Electronic Shock Collars?
Covered by Zazie Todd PhD at Companion Animal Psychology: Positive Reinforcement and Dog Training
Covered by Zazie Todd PhD at Companion Animal Psychology: Dangerous Dogs: Time for a Rethink?
and on Dr. Rachel Casey's blog: Aggressive behaviour in dogs: a survey of UK dog owners
and by Stanley Coren, PhD: Dog aggression is predicted by training methods and breed

Dog training research resources for happy dogs like this border collie

Covered by Zazie Todd PhD at The Pawsitive Post Issue No. 9
Covered by Zazie Todd PhD at Companion Animal Psychology: Positive reinforcement is more effective at training dogs than an electronic collar, study shows
Covered by Linda Lombardi at Fear Free Pets: Electronic collars versus positive reinforcement
Covered by Zazie Todd PhD at Companion Animal Psychology: The End for Shock Collars?
Covered by Zazie Todd PhD at Companion Animal Psychology: Dog Training, Animal Welfare, and the Human-Canine Relationship
and by Stanley Coren, PhD: Is punishment an effective way to change the behavior of dogs?
Covered by Zazie Todd PhD at Psychology Today: Does owner personality affect dog training methods?
Covered by Stanley Coren, PhD: The effect of training methods on the efficiency of learning
Covered by Zazie Todd PhD at Companion Animal Psychology: Positive Reinforcement and Dog Training: Dogs with Behaviour Problems and in the summary of this series on positive reinforcement
and Science Daily: If you're aggressive, your dog will be too
and Dr. Sophia Yin
and Patricia McConnell, PhD, CAAB: Confrontational techniques elicit aggression
Covered by Zazie Todd PhD at Companion Animal Psychology: Positive Reinforcement and Dog Training II
Covered by Zazie Todd PhD at Companion Animal Psychology: Study outlines reasons to ban electronic collars for dogs
Covered by Mark Bekoff, PhD: Is it time to ban shock collars for dogs in all situations?
Covered by Kerry Lengyel at American Veterinarian: Study shows no credible evidence justifying electronic shock collars.
Covered by Zazie Todd PhD at Companion Animal Psychology: Shock collars, regulation and education on alternatives
Covered by Zazie Todd PhD at Companion Animal Psychology: Positive Reinforcement and Dog Training: Learning New Behaviours
Covered by Dr. Sophia Yin: Are shock collars painful or just annoying to dogs? A 2004 study reveals some answers
Covered by Zazie Todd PhD at Companion Animal Psychology: Why don't more people use positive reinforcement to train dogs?
Covered by Zazie Todd PhD at Companion Animal Psychology: Dog training methods affect attachment to the owner
Covered by Stanley Coren, PhD: How training methods affect a dogs attachment to its owner
Covered by Zazie Todd PhD at Companion Animal Psychology: New study shows dogs trained with rewards only are more optimistic
Covered by Stanley Coren, PhD: Does it matter whether dog training is positive or aversive?
Covered by Zazie Todd PhD at Companion Animal Psychology: New Literature Review Recommends Reward-Based Training
Guest post by Zazie Todd, PhD at the Academy for Dog Trainers: T he Double Advantage of Reward-Based Training .

Other Scientific Research on Dog Training

Covered by Zazie Todd PhD at Companion Animal Psychology: Playtime after training improves a dog's memory
Covered by Stanley Coren, PhD: What dogs do after training affects how much they remember
Covered by Julie Hecht at DogSpies: Memory wins when dogs sleep
Covered by Zazie Todd PhD at Companion Animal Psychology: In training, pay your dog with the food or foods they love, science says
Covered by Zazie Todd PhD at Companion Animal Psychology: Can dog training books be trusted?
Covered by Zazie Todd PhD at Companion Animal Psychology: Flat collars risk damage to dogs' necks
Covered by Anne Carter, PhD: Getting hot under the collar--allowing your dog to pull on a collar and lead could be causing damage and increase their risk of heat stroke
Covered by Zazie Todd PhD at Companion Animal Psychology: Dogs' attention declines with age - but training helps
Covered by Zazie Todd PhD at Companion Animal Psychology: Clicker Training vs Treat: Equally Good in Dog Training
Covered by Stanley Coren, PhD: Are voice commands or hand signals more effective for dogs?
Covered by Zazie Todd PhD at Companion Animal Psychology: How Often Should I Train My Dog?
Covered by Zazie Todd PhD at The Pawsitive Post Issue No. 28
Covered by Lynna Feng: The Clicker Training Controversy
Covered by Linda P. Case: Why we click
Covered by Zazie Todd PhD at Companion Animal Psychology: Clicker-plus-food and food-only are equally good dog training methods
Covered by Zazie Todd PhD at Companion Animal Psychology: Now Where's My Treat?
Covered by Claudia Fugazza for Do You Believe in Dog?: Do As I Do: Copy cat social imitation in dog training
Covered by Zazie Todd PhD at Companion Animal Psychology: The Importance of Food in Dog Training
Covered by Zazie Todd PhD at The Pawsitive Post Issue No. 25
Covered by Zazie Todd PhD at Companion Animal Psychology: The Attentive Look of a Dog in Training
Covered by Zazie Todd PhD at Companion Animal Psychology: Frustration in Pet Dog Training
Covered by Zazie Todd PhD at Companion Animal Psychology: I s it Important to Attend Puppy Class?
Covered by Zazie Todd PhD at Fellow Creatures on Psych Today: Training methods affect the service dog-veteran relationship
Covered by Linda Case: Reward-based training and relationship
Covered by Zazie Todd PhD at Companion Animal Psychology: Do Dogs Get That Eureka! Feeling?
Covered by Patricia McConnell, PhD, CAAB: What's a "better learning performance?"
Covered by Zazie Todd PhD at Companion Animal Psychology: The Importance of Food in Dog Training
Covered by Zazie Todd PhD at Companion Animal Psychology: Timing and attention matter in dog training, new study shows
Covered by Zazie Todd PhD at Companion Animal Psychology: Do dogs run faster for more treats or better quality treats?
Covered by Sienna Taylor, MSc, in a guest post for Companion Animal Psychology: To gesture or not to gesture in dog training?
Covered by Stanley Coren, PhD: Do dogs respond more accurately to words or gestures?
Covered by Zazie Todd PhD at Companion Animal Psychology: Confidence and emotions affect people's use of positive reinforcement to train reactive dogs .
Covered by Marc Bekoff, PhD: "Bad dog?" The psychology and importance of using positive reinforcement
Covered by Linda Lombardi at Fear Free: Owner perceptions influence training methods

The Schedule of Events for Bark! Fest, the Book Festival for Animal Lovers

Eight Tips to Help Fearful Dogs Feel Safe (Guide)

What is Negative Reinforcement in Dog Training?

‘This could change everything!’ Nous Research unveils new tool to train powerful AI models with 10,000x efficiency

Share on Facebook
Share on LinkedIn

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More

Nous Research turned heads earlier this month with the release of its permissive, open-source Llama 3.1 variant Hermes 3 .

Now, the small research team dedicated to making “personalized, unrestricted AI” models has announced another seemingly massive breakthrough: DisTrO (Distributed Training Over-the-Internet), a new optimizer that reduces the amount of information that must be sent between various GPUs (graphics processing units) during each step of training an AI model.

Nous’s DisTrO optimizer means powerful AI models can now be trained outside of big companies, across the open web on consumer-grade connections, potentially by individuals or institutions working together from around the world.

DisTrO has already been tested and shown in a Nous Research technical paper to yield an 857 times efficiency increase compared to one popular existing training algorithm, All-Reduce , as well as a massive reduction in the amount of information transmitted during each step of the training process (86.8 megabytes compared to 74.4 gigabytes) while only suffering a slight loss in overall performance. See the results in the table below from the Nous Research technical paper:

Ultimately, the DisTrO method could open the door to many more people being able to train massively powerful AI models as they see fit.

As the firm wrote in a post on X yesterday : “Without relying on a single company to manage and control the training process, researchers and institutions can have more freedom to collaborate and experiment with new techniques, algorithms, and models. This increased competition fosters innovation, drives progress, and ultimately benefits society as a whole.”

What if you could use all the computing power in the world to train a shared, open source AI model? Preliminary report: https://t.co/b1XgJylsnV Nous Research is proud to release a preliminary report on DisTrO (Distributed Training Over-the-Internet) a family of… pic.twitter.com/h2gQJ4m7lB — Nous Research (@NousResearch) August 26, 2024

The problem with AI training: steep hardware requirements

As covered on VentureBeat previously, Nvidia’s GPUs in particular are in high demand in the generative AI era, as the expensive graphics cards’ powerful parallel processing capabilities are needed to train AI models efficiently and (relatively) quickly. This blog post at APNic describes the process well.

A big part of the AI training process relies on GPU clusters — multiple GPUs — exchanging information with one another about the model and the information “learned” within training data sets.

However, this “inter-GPU communication” requires that GPU clusters be architected, or set up, in a precise way in controlled conditions, minimizing latency and maximizing throughput. Hence why companies such as Elon Musk’s Tesla are investing heavily in setting up physical “superclusters” with many thousands (or hundreds of thousands) of GPUs sitting physically side-by-side in the same location — typically a massive airplane hangar-sized warehouse or facility.

Because of these requirements, training generative AI — especially the largest and most powerful models — is typically an extremely capital-heavy endeavor, one that only some of the most well-funded companies can engage in, such as Tesla, Meta, OpenAI, Microsoft, Google, and Anthropic.

The training process for each of these companies looks a little different, of course. But they all follow the same basic steps and use the same basic hardware components. Each of these companies tightly controls its own AI model training processes, and it can be difficult for incumbents, much less laypeople outside of them, to even think of competing by training their own similarly-sized (in terms of parameters, or the settings under the hood) models.

But Nous Research, whose whole approach is essentially the opposite — making the most powerful and capable AI it can on the cheap, openly, freely, for anyone to use and customize as they see fit without many guardrails — has found an alternative.

What DisTrO does differently

While traditional methods of AI training require synchronizing full gradients across all GPUs and rely on extremely high bandwidth connections, DisTrO reduces this communication overhead by four to five orders of magnitude.

The paper authors haven’t fully revealed how their algorithms reduce the amount of information at each step of training while retaining overall model performance, but plan to release more on this soon.

The reduction was achieved without relying on amortized analysis or compromising the convergence rate of the training, allowing large-scale models to be trained over much slower internet connections — 100Mbps download and 10Mbps upload, speeds available to many consumers around the world.

The authors tested DisTrO using the Meta Llama 2, 1.2 billion large language model (LLM) architecture and achieved comparable training performance to conventional methods with significantly less communication overhead.

They note that this is the smallest-size model that worked well with the DisTrO method, and they “do not yet know whether the ratio of bandwidth reduction scales up, down, or stays constant as model size increases.”

Yet, the authors also say that “our preliminary tests indicate that it is possible to get a bandwidth requirements reduction of up to 1000x to 3000x during the pre-training,” phase of LLMs, and “for post-training and fine-tuning, we can achieve up to 10000x without any noticeable degradation in loss.”

They further hypothesize that the research, while initially conducted on LLMs, could be used to train large diffusion models (LDMs) as well: think the Stable Diffusion open source image generation model and popular image generation services derived from it such as Midjourney .

Still need good GPUs

To be clear: DisTrO still relies on GPUs — only instead of clustering them all together in the same location, now they can be spread out across the world and communicate over the consumer internet.

Specifically, DisTrO was evaluated using 32x H100 GPUs, operating under the Distributed Data Parallelism (DDP) strategy, where each GPU had the entire model loaded in VRAM .

This setup allowed the team to rigorously test DisTrO’s capabilities and demonstrate that it can match the convergence rates of AdamW+All-Reduce despite drastically reduced communication requirements.

This result suggests that DisTrO can potentially replace existing training methods without sacrificing model quality, offering a scalable and efficient solution for large-scale distributed training.

By reducing the need for high-speed interconnects DisTrO could enable collaborative model training across decentralized networks, even with participants using consumer-grade internet connections.

The report also explores the implications of DisTrO for various applications, including federated learning and decentralized training.

Additionally, DisTrO’s efficiency could help mitigate the environmental impact of AI training by optimizing the use of existing infrastructure and reducing the need for massive data centers.

Moreover, the breakthroughs could lead to a shift in how large-scale models are trained, moving away from centralized, resource-intensive data centers towards more distributed, collaborative approaches that leverage diverse and geographically dispersed computing resources.

What’s next for the Nous Research team and DisTrO?

The research team invites others to join them in exploring the potential of DisTrO. The preliminary report and supporting materials are available on GitHub , and the team is actively seeking collaborators to help refine and expand this groundbreaking technology.

Already, some AI influencers such as @kimmonismus on X (aka chubby) have praised the research as a huge breakthrough in the field, writing, “This could change everything!”

Wow, amazing! This could change everything! https://t.co/2f0PDSaTSm — Chubby♨️ (@kimmonismus) August 27, 2024

With DisTrO, Nous Research is not only advancing the technical capabilities of AI training but also promoting a more inclusive and resilient research ecosystem that has the potential to unlock unprecedented advancements in AI.

Stay in the know! Get the latest news in your inbox daily

By subscribing, you agree to VentureBeat's Terms of Service.

Thanks for subscribing. Check out more VB newsletters here .

An error occured.

Free worldwide delivery and taxes paid
100% Satisfaction Guarantee
24 Month Warranty
Flexible Payments

Cardiovascular Responses: Flywheel Resistance Training vs. Traditional Resistance Training

Study details.

Banks, N. F., Rogers, E. M., Berry, A. C., & Jenkins, N. D. (2024). Progressive iso-inertial resistance exercise promotes more favorable cardiovascular adaptations than traditional resistance exercise in young adults. American Journal of Physiology-Heart and Circulatory Physiology , 326(1), H32-H43.

Background and Purpose of the Study

The Exerfly Platform was recently used in a study on the cardiovascular adaptations resulting from different high-intensity resistance training methods.

A lot of research has evaluated the effects of both traditional resistance training (TRT) and flywheel resistance training (FRT) on measures of neuromuscular and athletic performance. But much less has been conducted on how high-intensity TRT and FRT influence cardiovascular adaptations, particularly since high intensity strength training is commonly performed in young adults!

To address this gap, researchers compared the cardiovascular effects of TRT, using barbells and cable stack machines, to a FRT group, which used the Exerfly Platform.

A sample of 31 young healthy adults (19 women, 12 men) completed the study. They were randomly assigned to one of three groups:

Control group

All participants completed a battery of assessments before and after the 10-week training period. These included assessments of muscle strength (max torque during an isometric leg extension), body composition, and measures of central arterial stiffness, macrovascular function, and blood pressure reactivity to submaximal exercise.

List of tests performed by the researchers:

Body Composition (body fat %) through a combination of bioelectrical impedance (to measure total water content) and air displacement plethysmography to measure body density.
Resting blood pressure, including systolic, diastolic, and mean arterial pressure.
Resting heart rate
Cardiovagal Baroreflex Sensitivity
Heart Rate Variability (HRV)
Resting blood pressure and heart rate
Flow mediated dilation (FMD) of the brachial artery as a measure of the macrovascular function
Carotid-Femoral Pulse Wave Velocity as a measure of central arterial stiffness
Blood Pressure Reactivity in response to an isometric exercise protocol
Leg extension isometric strength (maximal voluntary isometric torque)

The Training Protocols

Both training groups (TRT and FRT) completed 3 workouts per week, involving 3 sets of 5 exercises using either traditional equipment (barbell and cable stack machine) or on the Exerfly Platform. The primary exercises during each workout were the squat, deadlift, bench press, and row. One accessory exercise (bicep curl, triceps extension, or glute bridge), was added to each session starting in week 6.

The programs progressed using a linear periodization model, where intensity increased over time as volume decreased. For example, after an introductory week of training, the program progressed from 3 sets at 12 repetition maximum (RM) in week 2, to 3 sets at 4 RM during the final week.

Sets 1 and 2 followed the prescribed rep number. But during set 3, they performed reps until technical failure or until a pre-defined threshold. For TRT, this was 4 reps beyond the prescribed number, and for FRT it was when measured outputs from the Exerfly device dropped to 80% of the set 2 average for consecutive reps. The number of successful reps during set 3 was used to determine increases in load for the next session.

The control group did not perform any resistance training and was asked to maintain their current physical activity habits outside of the study. Additionally, all groups were asked to maintain their current dietary habits, to log their food intake, and to complete a physical activity questionnaire at the end of each week.

Both training groups improved isometric leg extension strength compared to baseline and relative to the control group (TRT: +11.4%; FRT: +9.4%), without a statistically significant difference between TRT and FRT.

However, there were significant differences in the cardiovascular adaptations. More specifically, the TRT group had several significant changes which were indicative of negative adaptations to the cardiovascular system, which were not observed when using the FRT instead.

For example, TRT resulted in more exaggerated blood pressure responses to a submaximal isometric exercise protocol compared to FRT and the control group, a change that is often associated with increased risk of cardiovascular issues in otherwise healthy individuals (4,8). This was accompanied by negative changes in cardiovagal reflex sensitivity and heart rate variability in the TRT group, which was suggestive of altered autonomic nervous system responses.

However, there was no evidence of changes to other cardiovascular markers such as the measures of arterial stiffness, resting blood pressure, or macrovascular function.

Conclusions and Takeaways

Overall, it seems that FRT had statistically similar benefits in isometric strength as TRT, but without the same negative changes in certain measures of blood pressure reactivity and autonomic function.

It is not clear why the differences were noted or the long-term implications of such a finding, and more research is needed to determine the hemodynamic effects of FRT compared to other resistance training methods. However, these results provide additional evidence that FRT can be a safe and effective training method for a wide range of individuals.

For example, researchers have previously found FRT to have promising effects for stroke patients (6,7), provide beneficial effects on health and functional markers when combined with aerobic or interval training (2), and for improving neuromuscular performance and functional performance in older adults (5). Additionally, recent studies have found FRT to be well-tolerated by patients with severe hemophilia (3).

So, while FRT has been well-established as an effective method for improving measures of neuromuscular performance such as muscle strength, power, and hypertrophy, it is becoming readily apparent that the benefits of FRT can be safely and effectively applied to a massive variety of populations!

Want to learn more about Exerfly FRT ? Check out our other blogs and content, and feel free to contact us !

Banks, N. F., Rogers, E. M., Berry, A. C., & Jenkins, N. D. (2024). Progressive iso-inertial resistance exercise promotes more favorable cardiovascular adaptations than traditional resistance exercise in young adults. American Journal of Physiology-Heart and Circulatory Physiology, 326(1), H32-H43.
Bruseghini, P., Calabria, E., Tam, E., Milanese, C., Oliboni, E., Pezzato, A., ... & Capelli, C. (2015). Effects of eight weeks of aerobic interval training and of isoinertial resistance training on risk factors of cardiometabolic diseases and exercise capacity in healthy elderly subjects. Oncotarget, 6(19), 16998.
Calatayud, J., Pérez‐Alenda, S., Carrasco, J. J., Cruz‐Montecinos, C., Andersen, L. L., Bonanad, S., ... & Casaña, J. (2021). Feasibility, safety and muscle activity during flywheel vs traditional strength training in adult patients with severe haemophilia. Haemophilia, 27(1), e102-e109.
Caselli, S., Serdoz, A., Mango, F., Lemme, E., Vaquer Seguì, A., Milan, A., ... & Pelliccia, A. (2019). High blood pressure response to exercise predicts future development of hypertension in young athletes. European Heart Journal, 40(1), 62-68.
Čokorilo, N., Horvatin, M., Đorđević, D., Stanković, M., & Pekas, D. (2022). Flywheel Training in Older Adults—A Systematic Review. Sustainability, 14(7), 4137.
Fernandez-Gonzalo, R., Fernandez-Gonzalo, S., Turon, M., Prieto, C., Tesch, P. A., & García-Carreira, M. D. C. (2016). Muscle, functional and cognitive adaptations after flywheel resistance training in stroke patients: a pilot randomized controlled trial. Journal of Neuroengineering and Rehabilitation, 13, 1-11.
Fernandez-Gonzalo, R., Nissemark, C., Åslund, B., Tesch, P. A., & Sojka, P. (2014). Chronic stroke patients show early and robust improvements in muscle and functional performance in response to eccentric-overload flywheel resistance training: a pilot study. Journal of Neuroengineering and Rehabilitation, 11, 1-10.
Manolio, T. A., Burke, G. L., Savage, P. J., Sidney, S., Gardin, J. M., & Oberman, A. (1994). Exercise blood pressure response and 5-year risk of elevated blood pressure in a cohort of young adults: the CARDIA study. American Journal of Hypertension, 7(3), 234-241.

Exerfly Equipment

PNG picture of exerfly flywheel training platform

The Exerfly Platform is a stable, multi-use flywheel training platform built for everything from heavy duty workouts to rehabilitation.

accessories

Don’t know which equipment to choose?

Information

Author Services

Initiatives

You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .

Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.

Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Original Submission Date Received: .

Active Journals
Find a Journal
Proceedings Series
For Authors
For Reviewers
For Editors
For Librarians
For Publishers
For Societies
For Conference Organizers
Open Access Policy
Institutional Open Access Program
Special Issues Guidelines
Editorial Process
Research and Publication Ethics
Article Processing Charges
Testimonials
Preprints.org
SciProfiles
Encyclopedia

Article Menu

Subscribe SciFeed
Recommended Articles
Google Scholar
on Google Scholar
Table of Contents

Find support for a specific problem in the support section of our website.

Please let us know what you think of our products and services.

Visit our dedicated information section to learn more about MDPI.

JSmol Viewer

The effects of blood flow restriction aerobic exercise on body composition, muscle strength, blood biomarkers, and cardiovascular function: a narrative review.

1. Introduction

3. potential physiological and molecular mechanisms of blood flow restriction exercise, 4. effects of blood flow restriction aerobic exercise on body composition.

Author	Subjects	Group	Intervention	Cuff Pressure	Outcomes	p-Value
]	Obese adults (n = 72, <25 yr, BF% > 30%)	CON HIIT HIIT + BFR (during interval) HIIT + BFR (during exercise)	Frequency: 12 wks, 2 times/wk Intensity: 85% VO max Volume: 4 sets (each set 3 min, 3 min rest) Type: HIIT	40% limb occlusive pressure (LOP)	%BF ↓ FM ↓ AVFA ↓	The HIIT + BFR (during interval) and HIIT + BFR (during exercise) groups showed a significant decrease in %BF compared to the HIIT group (p < 0.05). The HIIT + BFR (during interval) group showed a significant decrease in FM and AVFA compared to the HIIT group (p < 0.05).
]	Obese men (n = 40, 18–22 yr, %BF > 25% or BMI > 28 kg/m )	LIT LIT + BFR	Frequency: 12 wks, 2 times/wk Intensity: 40% VO max Volume: 3 sets (each set 15 min, 1 min rest) Type: Cycle	200 mmHg	BW ↓ WC ↓ FM ↓ %BF ↓	The LIT + BFR group showed significant decreases in BW, WC, FM, and %BF compared to the LIT group (p < 0.05).
]	Older women (n = 24, 62.9 ± 3.1 yr)	CON DT DT + BFR	Frequency: 8 wks, 3 times/wk Intensity: 45% HRR Volume: 20 min Type: Walk	150–200 mmHg	BW ↓ Visceral fat ↓	The DT + BFR group showed significant decreases in BW (p = 0.001) and visceral fat (p = 0.003) compared to the DT group.
]	Obese women (n = 11, 44.45 ± 0.8 yr, BMI > 25 kg/m , %BF > 30%)	BFR	Frequency: 4 wks, 3 times/wk Intensity: 4 km/h, 5% grade Volume: 5 sets (each set 2 min) Type: Walk	160–230 mmHg	BW ↓ BMI ↓ FM ↓ TC↓	BW, BMI, FM, and TC decreased after exercise (p < 0.05).
]	Healthy men (n = 50, 18–25 yr)	CON MICT HIIT LICT-BFR	Frequency: 8 wks, 3 times/wk Intensity: 57–63% HRmax Volume: 5 min, warm-up 15 min, 10 min resting phase Type: Walk	200–360 mmHg	%BF ↓ FM ↓ AVFA ↓	The LICT-BFR and MICT groups showed significant decreases in FM and AVFA compared to the CON group (p < 0.05). The exercise groups showed a significant decrease in %BF compared to the CON group (p < 0.05).
]	Obese women (n = 11, 44.45 ± 0.8 yr, BMI > 25 kg/m , %BF > 30%)	BFR	Frequency: 4 wks, 3 times/wk Intensity: 4 km/h, 5% grade Volume: 5 sets (each set 2 min) Type: Walk	160–230 mmHg	BW ↓ BMI ↓ %BF ↔	BW and BMI decreased significantly after exercise (p = 0.022, p = 0.015, respectively). %BF tended to decrease, but this change was not statistically significant (p = 0.07).

5. Effect of Blood Flow Restriction Aerobic Exercise on Muscle Mass and Strength

Author	Subjects	Group	Intervention	Cuff Pressure	Outcomes	p-Value
]	Old adults (n = 19, 60–78 yr)	CON BFR	Frequency: 6 wks, 5 times/wk Intensity: 67 m/min Volume: 20 min Type: Walk	160–200 mmHg	Muscle strength ↑ Mid-thigh CSA ↑ Lower leg CSA ↑ Thigh muscle mass ↑	The BFR group showed significant increases in isometric (11%) and isokinetic (7–16%) knee extension and flexion torque, muscle CSA (5.8% for the thigh and 5.1% for the lower leg), as well as muscle mass (6.0% and 10.7% for total and thigh, respectively) (p < 0.05), but there was no significant difference in the CON group.
]	Old women (n = 24, 62.9 ± 3.1 yr)	CON DT DT + BFR	Frequency: 8 wks, 3 times/wk Intensity: 45% HRR Volume: 20 min Type: Walk	150–200 mmHg	Muscle strength ↑ Muscle quality ↑	The DT + BFR group showed significant increases in muscle strength (p < 0.001) and muscle quality (p < 0.001) compared to both the DT and CON groups.
]	Old adults (n = 23, 57–76 yr)	MIT MIT + BFR	Frequency: 10 wks, 4 times/wk Intensity: 45% HRR Volume: 20 min Type: Walk	140–200 mmHg	Muscle CSA ↑ Muscle strength ↑	Muscle strength (∼15%) and Muscle CSA (3%) increased in the MIT + BFR group, with no significant difference observed in the MIT group.
]	Healthy men (n = 19, 20–26 yr)	LIT LIT + BFR	Frequency: 8 wks, 3 times/wk Intensity: 40% VO max Volume: 15 min Type: Cycle	160–210 mmHg	Muscle CSA ↑ Muscle volume ↑ Isometric muscle strength ↔	Muscle CSA and muscle volume increased by 3.4-5.1% (p < 0.01) in the BFR group, and isometric strength tended to increase by 7.7% (p = 0.10). In contrast, the CON group showed no significant difference in muscle size (~0.6%) and strength (~1.4%).
]	Healthy men (n = 18, 21.2 ± 2.7 yr)	CON BFR	Frequency: 3 wks, 2 times/day Intensity: 50 m/min Volume: 20 min Type: Walk	160–230 mmHg	Muscle CSA ↑ Muscle volume ↑ Muscle strength ↑	The BFR group showed increases in muscle CSA, muscle volume by 4–7%, and isometric strength by 8–10%. In contrast, the CON group showed no significant difference in muscle size or isometric strength.
]	Healthy men (n = 31, 21.2 ± 1.9 yr)	CON BFR	Frequency: 3 wks, 6 times/wk, 2 times/day Intensity: 50 m/min Volume: 5 sets (each set 2 min, 1 min rest) Type: Walk	160–230 mmHg	Muscle volume ↑	The BFR group showed significant increases in upper leg muscle volume (3.8%, p < 0.05) and lower leg muscle volume (3.2%, p < 0.05). In contrast, there was no significant difference in muscle volume in the CON group.
]	Healthy men (n = 31, 22.4 ± 3.0 yr)	CON MIT LIT + BFR	Frequency: 6 wks, 3 times/wk Intensity: MIT: 60–70% HRR LIT + BFR: 30% HRR Volume: 20 min Type: Cycle	160–180 mmHg	Muscle strength ↑ Leg lean mass ↑	The MIT (7.15%) and LIT-BFR (8.90%) groups showed significant increases in muscle strength (p = 0.024 and p = 0.01, respectively). In contrast, there was no difference in muscle strength in the CON group. The LIT-BFR group increased the leg lean mass by 1.15% (p = 0.024), whereas there were no differences in leg lean mass in the MIT (0.44%) and CON (0.89%) groups.
]	Adults (n = 37, 23.8 ± 4.0 yr)	LIT LIT + BFR HIT HIT + BFR	Frequency: 4 wks, 3 times/wk Intensity: LIT + BFR: ~30% P HIT + BFR: ~66% P HIT: ~102% P Volume: 20 min Type: Cycle	140–200 mmHg	Muscle strength ↑	The LIT + BFR group increased isometric strength by 11.4 ± 7.3%, (p = < 0.001). However, there were no differences in isometric strength in other groups: HIT (−0.7 ± 9.9%, p = 0.88), HIT + BFR (−3.5 ± 6.8%, p = 0.32), and LIT (−2.6 ± 6.7%, p = 0.82).
]	Healthy men (n = 30, 30.21 ± 3.0 yr)	LIT LIT + BFR	Frequency: 8 wks, 3 times/wk Intensity: 40% VO max Volume: 15 min Type: Walk	160–240 mmHg	Muscle mass ↑ Muscle volume ↑	The LIT + BFR group showed increases in muscle mass (p <0.001) and right thigh circumference (p = 0.042), whereas no difference were observed in the LIT group.
]	Obese women (n = 11, 44.45 ± 0.8 yr, BMI > 25 kg/m , %BF > 30%)	BFR	Frequency: 4 wks, 3 times/wk Intensity: 4 km/h, 5% grade Volume: 5 sets (each set 2 min) Type: Walk	160–230 mmHg	Muscle strength ↑ Muscle endurance ↑	Muscle strength increased at 60°/s for right and left side extension, left side flexion, and at 180°/s for left side extension after training (p < 0.05).
]	Health men (n = 50, 18–25 yr)	CON MICT HIIT LICT-BFR	Frequency: 8 wks, 3 times/wk Intensity: 57–63% Hrmax Volume: 5 min, warm-up 15 min, 10 min resting phase Type: Walk	200–360 mmHg	Muscle mass ↑	The LICT-BFR and MICT groups showed a significant increase in muscle mass (p < 0.05), while the HIIT group exhibited only a slight increased (p = 0.247) and the CON group showed a decrease in (p = 0.11).

6. Effects of Blood Flow Restriction Aerobic Exercise on Lipid Profiles and Glycemic Metabolism

Author	Subjects	Group	Intervention	Cuff Pressure	Outcomes	p-Value
]	Obese adults (n = 72, < 25 yr, %BF > 30%)	CON HIIT HIIT + BFR (during interval) HIIT + BFR (during exercise)	Frequency: 12 wks, 2 times/wk Intensity: 85% VO max Volume: 4 sets (each set 3 min, 3 min rest) Type: HIIT	40% limb occlusive pressure (LOP)	GLU ↓ Insulin ↓	The HIIT + BFR (during interval) and HIIT + BFR (during exercise) groups showed decreased blood GLU and insulin levels compared to the HIIT group (p < 0.05).
]	Obese men (n = 40, 18–22 yr, %BF > 25% or BMI > 28 kg/m )	LIT LIT + BFR	Frequency: 12 wks, 2 times/wk Intensity: 40% VO max Volume: 3 sets (each set 15 min, 1 min rest) Type: Cycle	200 mmHg	GLU ↓ TC ↓ TG ↔ LDL-C ↓ HDL-C ↑	The LIT + BFR group showed improved GLU, TC, HDL-C and LDL-C compared to the LIT group (p < 0.05).
]	Obese men (n = 18, 37–55 yr, BMI = 27–28 kg/m )	LIT LIT + BFR	Frequency: 8 wks, 3 times/wk Intensity: 3 km/h Volume: 5 sets Type: Walk	140–200 mmHg	TG ↓ LDL-C ↔ HDL-C ↔	The LIT + BFR and LIT groups showed a decrease in TG (p = 0.017). However, there were no differences in LDL-C and HDL-C between the LIT + BFR and LIT groups.
]	Healthy men (n = 13, 25 ± yr)	CON leg BFR leg	Frequency: 6 wks, 3 times/wk Intensity: 60–80% Wmax Volume: 9 sets (each set 2 min, 2 min rest) Type: Cycle	180 mmHg	Glucose uptake ↑	Thigh net glucose uptake was higher in the BFR leg compared to the CON leg (p < 0.01)
]	Obese women (n = 11, 44.45 ± 0.8 yr, BMI > 25 kg/m , %BF > 30%)	BFR	Frequency: 4 wks, 3 times/wk Intensity: 4 km/h, 5% grade Volume: 5 sets (each set 2 min, 1 min rest) Type: Walk	160–230 mmHg	GLU ↔ Insulin ↓ HOMA-IR ↓	Insulin and HOMA-IR decreased after exercise (p = 0.03, p = 0.04, respectively).

7. Effects of Blood Flow Restriction Aerobic Exercise on Cardiovascular Function

Author	Subjects	Group	Intervention	Cuff Pressure	Outcomes	p-Value
]	Healthy men (n = 39, 18–50 yr)	CON HI LI LI + BFR	Frequency: 6 wks, 3 times/wk Intensity: HI: 60–70% VO R LIT: 30–40% VO R LIT + BFR: 30–40% VO R Volume: 30 min Type: Walk	120 individuals’ thigh circumference: <45–50 cm = 120 mmHg 51–55 cm = 150 mmHg 56–59 cm = 180 mmHg ≥65 cm = 210 mmHg	Femoral–tibial PWV ↓ SBP ↔ DBP ↔	The LI + BFR group decreased the femoral–tibial PWV (p < 0.05) from baseline to post-training.
]	Sedentary adults (n = 23, 57–76 yr)	MIT MIT + BFR	Frequency: 10 wks, 4 times/wk Intensity: 45% HRR Volume: 20 min Type: Walk	140–200 mmHg	Carotid arterial compliance ↔	Carotid arterial compliance improved in both the MIT + BFR (50%) and MIT (59%) groups.
]	Men (n = 21, 52.4 ± 3.7 yr)	CON BFR	Frequency: 6 wks, 3 times/wk Intensity: 6 km/h, 5% Volume: 5 sets (each set 3 min, 1 min rest) Type: Walk	80–100 mmHg	SBP ↓ DBP ↔ SDNN ↑ RMSSD ↑	Only the BFR group showed improvements in SDNN (p = 0.002), RMSSD (p = 0.01), and SBP (p = 0.006).
]	Women (n = 16, 59–78 yr)	CON BFR	Frequency: 6 wks, 5 times/wk Intensity: 67 min/m Volume: 20 min Type: Walk	140–200 mmHg	Venous compliance ↑	The BFR group showed improved venous compliance (p < 0.05), while there was no significant difference in the CON group.
]	Healthy men (n = 30, 30.21 ± 3.0 yr)	LIT LIT + BFR	Frequency: 8 wks, 3 times/wk Intensity: 40% VO max Volume: 15 min Type: Walk	160–240 mmHg	FMD ↔ baPWV ↔ SBP ↔ DBP ↔	There were no differences in the time × group interaction effects on FMD, baPWV, SBP, and DBP.

8. Safety and Considerations for Blood Flow Restriction Aerobic Exercise

9. conclusions and future directions, author contributions, conflicts of interest.

Garber, C.E.; Blissmer, B.; Deschenes, M.R.; Franklin, B.A.; Lamonte, M.J.; Lee, I.M.; Nieman, D.C.; Swain, D.P. American College of Sports Medicine position stand. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: Guidance for prescribing exercise. Med. Sci. Sports Exerc. 2011 , 43 , 1334–1359. [ Google Scholar ] [ CrossRef ]
Cognetti, D.J.; Sheean, A.J.; Owens, J.G. Blood Flow Restriction Therapy and Its Use for Rehabilitation and Return to Sport: Physiology, Application, and Guidelines for Implementation. Arthrosc. Sports Med. Rehabil. 2022 , 4 , e71–e76. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Scott, B.R.; Loenneke, J.P.; Slattery, K.M.; Dascombe, B.J. Exercise with blood flow restriction: An updated evidence-based approach for enhanced muscular development. Sports Med. 2015 , 45 , 313–325. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Centner, C.; Wiegel, P.; Gollhofer, A.; Konig, D. Effects of Blood Flow Restriction Training on Muscular Strength and Hypertrophy in Older Individuals: A Systematic Review and Meta-Analysis. Sports Med. 2019 , 49 , 95–108. [ Google Scholar ] [ CrossRef ]
Karabulut, M.; Abe, T.; Sato, Y.; Bemben, M.G. The effects of low-intensity resistance training with vascular restriction on leg muscle strength in older men. Eur. J. Appl. Physiol. 2010 , 108 , 147–155. [ Google Scholar ] [ CrossRef ]
Kim, T.-H.; Lee, S.-H.; Kim, Y.-J.; Kim, S.-J.; Kang, J.-H.; Kwak, H.-B.; Park, D.-H. Effect of Acute Resistance Exercise with Different Level of Blood Flow Restriction on Acute Changes in Muscle Thickness, Blood Lactate, CK, and Oxidative Stress in Male Adults. Exerc. Sci. 2018 , 27 , 50–61. [ Google Scholar ] [ CrossRef ]
Yinghao, L.; Jing, Y.; Yongqi, W.; Jianming, Z.; Zeng, G.; Yiting, T.; Shuoqi, L. Effects of a blood flow restriction exercise under different pressures on testosterone, growth hormone, and insulin-like growth factor levels. J. Int. Med. Res. 2021 , 49 , 3000605211039564. [ Google Scholar ] [ CrossRef ]
Beak, H.J.; Park, W.; Yang, J.H.; Kim, J. Effect of Low-Intensity Aerobic Training Combined with Blood Flow Restriction on Body Composition, Physical Fitness, and Vascular Responses in Recreational Runners. Healthcare 2022 , 10 , 1789. [ Google Scholar ] [ CrossRef ]
Kim, D.; Singh, H.; Loenneke, J.P.; Thiebaud, R.S.; Fahs, C.A.; Rossow, L.M.; Young, K.; Seo, D.I.; Bemben, D.A.; Bemben, M.G. Comparative Effects of Vigorous-Intensity and Low-Intensity Blood Flow Restricted Cycle Training and Detraining on Muscle Mass, Strength, and Aerobic Capacity. J. Strength Cond. Res. 2016 , 30 , 1453–1461. [ Google Scholar ] [ CrossRef ]
Rodrigo-Mallorca, D.; Loaiza-Betancur, A.F.; Monteagudo, P.; Blasco-Lafarga, C.; Chulvi-Medrano, I. Resistance Training with Blood Flow Restriction Compared to Traditional Resistance Training on Strength and Muscle Mass in Non-Active Older Adults: A Systematic Review and Meta-Analysis. Int. J. Environ. Res. Public Health 2021 , 18 , 11441. [ Google Scholar ] [ CrossRef ]
Lixandrao, M.E.; Ugrinowitsch, C.; Berton, R.; Vechin, F.C.; Conceicao, M.S.; Damas, F.; Libardi, C.A.; Roschel, H. Magnitude of Muscle Strength and Mass Adaptations Between High-Load Resistance Training Versus Low-Load Resistance Training Associated with Blood-Flow Restriction: A Systematic Review and Meta-Analysis. Sports Med. 2018 , 48 , 361–378. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Baethge, C.; Goldbeck-Wood, S.; Mertens, S. SANRA-a scale for the quality assessment of narrative review articles. Res. Integr. Peer Rev. 2019 , 4 , 5. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Perera, E.; Zhu, X.M.; Horner, N.S.; Bedi, A.; Ayeni, O.R.; Khan, M. Effects of Blood Flow Restriction Therapy for Muscular Strength, Hypertrophy, and Endurance in Healthy and Special Populations: A Systematic Review and Meta-Analysis. Clin. J. Sport. Med. 2022 , 32 , 531–545. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Queiros, V.S.D.; Rolnick, N.; Schoenfeld, B.J.; Franca, I.M.D.; Vieira, J.G.; Sardeli, A.V.; Kamis, O.; Neto, G.R.; Cabral, B.G.D.T.; Dantas, P.M.S. Hypertrophic effects of low-load blood flow restriction training with different repetition schemes: A systematic review and meta-analysis. PeerJ 2024 , 12 , e17195. [ Google Scholar ] [ CrossRef ]
Pearson, S.J.; Hussain, S.R. A review on the mechanisms of blood-flow restriction resistance training-induced muscle hypertrophy. Sports Med. 2015 , 45 , 187–200. [ Google Scholar ] [ CrossRef ]
Scott, B.R.; Goods, P.S.; Slattery, K.M. High-intensity exercise in hypoxia: Is increased reliance on anaerobic metabolism important? Front. Physiol. 2016 , 7 , 637. [ Google Scholar ] [ CrossRef ]
Kraemer, W.J.; Ratamess, N.A. Hormonal responses and adaptations to resistance exercise and training. Sports Med. 2005 , 35 , 339–361. [ Google Scholar ] [ CrossRef ]
Fink, J.; Schoenfeld, B.J.; Nakazato, K. The role of hormones in muscle hypertrophy. Phys. Sportsmed. 2018 , 46 , 129–134. [ Google Scholar ] [ CrossRef ]
Godfrey, R.J.; Madgwick, Z.; Whyte, G.P. The exercise-induced growth hormone response in athletes. Sports Med. 2003 , 33 , 599–613. [ Google Scholar ] [ CrossRef ]
Reeves, G.V.; Kraemer, R.R.; Hollander, D.B.; Clavier, J.; Thomas, C.; Francois, M.; Castracane, V.D. Comparison of hormone responses following light resistance exercise with partial vascular occlusion and moderately difficult resistance exercise without occlusion. J. Appl. Physiol. 2006 , 101 , 1616–1622. [ Google Scholar ] [ CrossRef ]
Abe, T.; Yasuda, T.; Midorikawa, T.; Sato, Y.; CF, K.; Inoue, K.; Koizumi, K.; Ishii, N. Skeletal muscle size and circulating IGF-1 are increased after two weeks of twice daily “KAATSU” resistance training. Int. J. KAATSU Train. Res. 2005 , 1 , 6–12. [ Google Scholar ] [ CrossRef ]
O’Neil, T.K.; Duffy, L.; Frey, J.; Hornberger, T. The role of phosphoinositide 3-kinase and phosphatidic acid in the regulation of mammalian target of rapamycin following eccentric contractions. J. Physiol. 2009 , 587 , 3691–3701. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Bodine, S.C.; Stitt, T.N.; Gonzalez, M.; Kline, W.O.; Stover, G.L.; Bauerlein, R.; Zlotchenko, E.; Scrimgeour, A.; Lawrence, J.C.; Glass, D.J. Akt/mTOR pathway is a crucial regulator of skeletal muscle hypertrophy and can prevent muscle atrophy in vivo. Nat. Cell Biol. 2001 , 3 , 1014–1019. [ Google Scholar ] [ CrossRef ]
Hodson, N.; West, D.W.; Philp, A.; Burd, N.A.; Moore, D.R. Molecular regulation of human skeletal muscle protein synthesis in response to exercise and nutrients: A compass for overcoming age-related anabolic resistance. Am. J. Physiol. Cell Physiol. 2019 , 317 , C1061–C1078. [ Google Scholar ] [ CrossRef ]
Fry, C.S.; Glynn, E.L.; Drummond, M.J.; Timmerman, K.L.; Fujita, S.; Abe, T.; Dhanani, S.; Volpi, E.; Rasmussen, B.B. Blood flow restriction exercise stimulates mTORC1 signaling and muscle protein synthesis in older men. J. Appl. Physiol. 2010 , 108 , 1199–1209. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Fujita, S.; Abe, T.; Drummond, M.J.; Cadenas, J.G.; Dreyer, H.C.; Sato, Y.; Volpi, E.; Rasmussen, B.B. Blood flow restriction during low-intensity resistance exercise increases S6K1 phosphorylation and muscle protein synthesis. J. Appl. Physiol. 2007 , 103 , 903–910. [ Google Scholar ] [ CrossRef ]
Ito, N.; Ruegg, U.T.; Kudo, A.; Miyagoe-Suzuki, Y.; Takeda, S.i. Activation of calcium signaling through Trpv1 by nNOS and peroxynitrite as a key trigger of skeletal muscle hypertrophy. Nat. Med. 2013 , 19 , 101–106. [ Google Scholar ] [ CrossRef ]
Xu, S.; Liu, X.; Chen, Z.; Li, G.; Chen, Q.; Zhou, G.; Ma, R.; Yao, X.; Huang, X. Transcriptional profiling of rat skeletal muscle hypertrophy under restriction of blood flow. Gene 2016 , 594 , 229–237. [ Google Scholar ] [ CrossRef ]
Häussinger, D.; Gerok, W.; Roth, E.; Lang, F. Cellular hydration state: An important determinant of protein catabolism in health and disease. Lancet 1993 , 341 , 1330–1332. [ Google Scholar ] [ CrossRef ]
Lauber, B.; König, D.; Gollhofer, A.; Centner, C. Isometric blood flow restriction exercise: Acute physiological and neuromuscular responses. BMC Sports Sci. Med. Rehabil. 2021 , 13 , 12. [ Google Scholar ] [ CrossRef ]
Davids, C.J.; Næss, T.C.; Moen, M.; Cumming, K.T.; Horwath, O.; Psilander, N.; Ekblom, B.; Coombes, J.S.; Peake, J.; Raastad, T. Acute cellular and molecular responses and chronic adaptations to low-load blood flow restriction and high-load resistance exercise in trained individuals. J. Appl. Physiol. 2021 , 131 , 1731–1749. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Freitas, E.D.; Poole, C.; Miller, R.M.; Heishman, A.D.; Kaur, J.; Bemben, D.A.; Bemben, M. Time course change in muscle swelling: High-intensity vs. blood flow restriction exercise. Int. J. Sports Med. 2017 , 38 , 1009–1016. [ Google Scholar ] [ CrossRef ]
Manson, J.E.; Colditz, G.A.; Stampfer, M.J.; Willett, W.C.; Rosner, B.; Monson, R.R.; Speizer, F.E.; Hennekens, C.H. A prospective study of obesity and risk of coronary heart disease in women. N. Engl. J. Med. 1990 , 322 , 882–889. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Wang, G.; Singh, G.M.; Danaei, G.; Farzadfar, F.; Stevens, G.A.; Woodward, M.; Wormser, D.; Kaptoge, S.; Whitlock, G.; Qiao, Q.; et al. The Age-Specific Quantitative Effects of Metabolic Risk Factors on Cardiovascular Diseases and Diabetes: A Pooled Analysis. PLoS ONE 2013 , 8 , e65174. [ Google Scholar ] [ CrossRef ]
Calle, E.E.; Rodriguez, C.; Walker-Thurmond, K.; Thun, M.J. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of US adults. N. Engl. J. Med. 2003 , 348 , 1625–1638. [ Google Scholar ] [ CrossRef ]
Ding, C.; Jiang, Y. The Relationship between Body Mass Index and Physical Fitness among Chinese University Students: Results of a Longitudinal Study. Healthcare 2020 , 8 , 570. [ Google Scholar ] [ CrossRef ]
Takarada, Y.; Nakamura, Y.; Aruga, S.; Onda, T.; Miyazaki, S.; Ishii, N. Rapid increase in plasma growth hormone after low-intensity resistance exercise with vascular occlusion. J. Appl. Physiol. 2000 , 88 , 61–65. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Kon, M.; Ikeda, T.; Homma, T.; Suzuki, Y. Effects of low-intensity resistance exercise under acute systemic hypoxia on hormonal responses. J. Strength Cond. Res. 2012 , 26 , 611–617. [ Google Scholar ] [ CrossRef ]
Sugaya, M.; Yasuda, T.; Suga, T.; Okita, K.; Abe, T. Change in intramuscular inorganic phosphate during multiple sets of blood flow-restricted low-intensity exercise. Clin. Physiol. Funct. Imaging 2011 , 31 , 411–413. [ Google Scholar ] [ CrossRef ]
de Freitas, M.C.; Gerosa-Neto, J.; Zanchi, N.E.; Lira, F.S.; Rossi, F.E. Role of metabolic stress for enhancing muscle adaptations: Practical applications. World J. Methodol. 2017 , 7 , 46–54. [ Google Scholar ] [ CrossRef ]
Berneis, K.; Ninnis, R.; Haussinger, D.; Keller, U. Effects of hyper- and hypoosmolality on whole body protein and glucose kinetics in humans. Am. J. Physiol. 1999 , 276 , E188–E195. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Keller, U.; Szinnai, G.; Bilz, S.; Berneis, K. Effects of changes in hydration on protein, glucose and lipid metabolism in man: Impact on health. Eur. J. Clin. Nutr. 2003 , 57 (Suppl. S2), S69–S74. [ Google Scholar ] [ CrossRef ] [ PubMed ]
da Silva, J.C.G.; Freitas, E.D.S.; Aniceto, R.R.; Silva, K.F.; Araujo, J.P.; Bembem, M.G.; Batista, G.R.; Sousa, M.S.C. Aerobic exercise with blood flow restriction: Energy expenditure, excess postexercise oxygen consumption, and respiratory exchange ratio. Clin. Physiol. Funct. Imaging 2022 , 42 , 241–249. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Salzmann, K.; Sanchez, A.M.J.; Borrani, F. Effects of Blood Flow Restriction on O 2 Muscle Extraction and O 2 Pulmonary Uptake Kinetics During Heavy Exercise. Front. Physiol. 2021 , 12 , 722848. [ Google Scholar ] [ CrossRef ]
Belfrage, E.; Hjemdahl, P.; Fredholm, B.B. Metabolic effects of blood flow restriction in adipose tissue. Acta Physiol. Scand. 1979 , 105 , 222–227. [ Google Scholar ] [ CrossRef ]
Conceicao, M.S.; Gaspari, A.F.; Ramkrapes, A.P.B.; Junior, E.M.M.; Bertuzzi, R.; Cavaglieri, C.R.; Chacon-Mikahil, M.P.T. Anaerobic metabolism induces greater total energy expenditure during exercise with blood flow restriction. PLoS ONE 2018 , 13 , e0194776. [ Google Scholar ] [ CrossRef ]
Mora-Rodriguez, R.; Coyle, E.F. Effects of plasma epinephrine on fat metabolism during exercise: Interactions with exercise intensity. Am. J. Physiol. Endocrinol. Metab. 2000 , 278 , E669–E676. [ Google Scholar ] [ CrossRef ]
Li, S.; Guo, R.; Yu, T.; Li, S.; Han, T.; Yu, W. Effect of High-Intensity Interval Training Combined with Blood Flow Restriction at Different Phases on Abdominal Visceral Fat among Obese Adults: A Randomized Controlled Trial. Int. J. Environ. Res. Public Health 2022 , 19 , 11936. [ Google Scholar ] [ CrossRef ]
Ozaki, H.; Loenneke, J.P.; Abe, T. Blood flow-restricted walking in older women: Does the acute hormonal response associate with muscle hypertrophy? Clin. Physiol. Funct. Imaging 2017 , 37 , 379–383. [ Google Scholar ] [ CrossRef ]
Abe, T.; Kearns, C.F.; Sato, Y. Muscle size and strength are increased following walk training with restricted venous blood flow from the leg muscle, Kaatsu-walk training. J. Appl. Physiol. 2006 , 100 , 1460–1466. [ Google Scholar ] [ CrossRef ]
Chen, Y.; Ma, C.; Wang, J.; Gu, Y.; Gao, Y. Effects of 40% of Maximum Oxygen Uptake Intensity Cycling Combined with Blood Flow Restriction Training on Body Composition and Serum Biomarkers of Chinese College Students with Obesity. Int. J. Environ. Res. Public Health 2021 , 19 , 168. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Kargaran, A.; Abedinpour, A.; Saadatmehr, Z.; Yaali, R.; Amani-Shalamzari, S.; Gahreman, D. Effects of dual-task training with blood flow restriction on cognitive functions, muscle quality, and circulatory biomarkers in elderly women. Physiol. Behav. 2021 , 239 , 113500. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Park, M.-S.; Zang, S.-A.; Lee, J.-K. Effect of 4 Weeks’ Walking Exercise with Blood Flow Restriction on Inflammatory index, Isokinetic Muscle Function, and Thigh Circumference in Obese Women. J. Korea Acad. Ind. Coop. Soc. 2017 , 18 , 480–489. [ Google Scholar ]
Lan, C.; Liu, Y.J.; Wang, Y. Effects of different exercise programs on cardiorespiratory fitness and body composition in college students. J. Exerc. Sci. Fit. 2022 , 20 , 62–69. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Oh, D.-H.; Kim, J.-H.; Zhang, S.-A.; Lee, J.-K. Effect of 4 Weeks’ Walking Exercise With Blood Flow Restriction on Insulin Resistance, Adipokines and Gut Hormones in Middle Aged Obese Women. J. Korea Acad. Ind. Coop. Soc. 2018 , 19 , 489–498. [ Google Scholar ]
Kim, G.; Kim, J.H. Impact of Skeletal Muscle Mass on Metabolic Health. Endocrinol. Metab. 2020 , 35 , 1–6. [ Google Scholar ] [ CrossRef ]
Lopez-Jaramillo, P.; Lopez-Lopez, J.P.; Tole, M.C.; Cohen, D.D. Muscular Strength in Risk Factors for Cardiovascular Disease and Mortality: A Narrative Review. Anatol. J. Cardiol. 2022 , 26 , 598–607. [ Google Scholar ] [ CrossRef ]
Kalyani, R.R.; Corriere, M.; Ferrucci, L. Age-related and disease-related muscle loss: The effect of diabetes, obesity, and other diseases. Lancet Diabetes Endocrinol. 2014 , 2 , 819–829. [ Google Scholar ] [ CrossRef ]
Kraemer, W.J.; Patton, J.F.; Gordon, S.E.; Harman, E.A.; Deschenes, M.R.; Reynolds, K.; Newton, R.U.; Triplett, N.T.; Dziados, J.E. Compatibility of high-intensity strength and endurance training on hormonal and skeletal muscle adaptations. J. Appl. Physiol. 1995 , 78 , 976–989. [ Google Scholar ] [ CrossRef ]
Bell, G.J.; Syrotuik, D.; Martin, T.P.; Burnham, R.; Quinney, H.A. Effect of concurrent strength and endurance training on skeletal muscle properties and hormone concentrations in humans. Eur. J. Appl. Physiol. 2000 , 81 , 418–427. [ Google Scholar ] [ CrossRef ]
Frost, R.A.; Lang, C.H. Multifaceted role of insulin-like growth factors and mammalian target of rapamycin in skeletal muscle. Endocrinol. Metab. Clin. 2012 , 41 , 297–322. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Wu, C.W.; Storey, K.B. mTOR Signaling in Metabolic Stress Adaptation. Biomolecules 2021 , 11 , 681. [ Google Scholar ] [ CrossRef ]
Barjaste, A.; Mirzaei, B.; Rahmani-Nia, F.; Haghniyaz, R.; Brocherie, F. Concomitant aerobic-and hypertrophy-related skeletal muscle cell signaling following blood flow-restricted walking. Sci. Sports 2021 , 36 , e51–e58. [ Google Scholar ] [ CrossRef ]
Rossi, F.E.; de Freitas, M.C.; Zanchi, N.E.; Lira, F.S.; Cholewa, J.M. The Role of Inflammation and Immune Cells in Blood Flow Restriction Training Adaptation: A Review. Front. Physiol. 2018 , 9 , 1376. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Moritani, T.; Sherman, W.M.; Shibata, M.; Matsumoto, T.; Shinohara, M. Oxygen availability and motor unit activity in humans. Eur. J. Appl. Physiol. Occup. Physiol. 1992 , 64 , 552–556. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Meyer, R.A. Does blood flow restriction enhance hypertrophic signaling in skeletal muscle? J. Appl. Physiol. 2006 , 100 , 1443–1444. [ Google Scholar ] [ CrossRef ]
Sundberg, C.J. Exercise and training during graded leg ischaemia in healthy man with special reference to effects on skeletal muscle. Acta Physiol. Scand. Suppl. 1994 , 615 , 1–50. [ Google Scholar ]
Loenneke, J.; Fahs, C.; Wilson, J.; Bemben, M. Blood flow restriction: The metabolite/volume threshold theory. Med. Hypotheses 2011 , 77 , 748–752. [ Google Scholar ] [ CrossRef ]
Yasuda, T.; Abe, T.; Brechue, W.F.; Iida, H.; Takano, H.; Meguro, K.; Kurano, M.; Fujita, S.; Nakajima, T. Venous blood gas and metabolite response to low-intensity muscle contractions with external limb compression. Metabolism 2010 , 59 , 1510–1519. [ Google Scholar ] [ CrossRef ]
Takarada, Y.; Takazawa, H.; Sato, Y.; Takebayashi, S.; Tanaka, Y.; Ishii, N. Effects of resistance exercise combined with moderate vascular occlusion on muscular function in humans. J. Appl. Physiol. 2000 , 88 , 2097–2106. [ Google Scholar ] [ CrossRef ]
Takarada, Y.; Sato, Y.; Ishii, N. Effects of resistance exercise combined with vascular occlusion on muscle function in athletes. Eur. J. Appl. Physiol. 2002 , 86 , 308–314. [ Google Scholar ] [ CrossRef ]
Yasuda, T.; Brechue, W.F.; Fujita, T.; Shirakawa, J.; Sato, Y.; Abe, T. Muscle activation during low-intensity muscle contractions with restricted blood flow. J. Sports Sci. 2009 , 27 , 479–489. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Abe, T.; Sakamaki, M.; Fujita, S.; Ozaki, H.; Sugaya, M.; Sato, Y.; Nakajima, T. Effects of low-intensity walk training with restricted leg blood flow on muscle strength and aerobic capacity in older adults. J. Geriatr. Phys. Ther. 2010 , 33 , 34–40. [ Google Scholar ]
Ozaki, H.; Miyachi, M.; Nakajima, T.; Abe, T. Effects of 10 Weeks Walk Training With Leg Blood Flow Reduction on Carotid Arterial Compliance and Muscle Size in the Elderly Adults. Angiology 2010 , 62 , 81–86. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Abe, T.; Fujita, S.; Nakajima, T.; Sakamaki, M.; Ozaki, H.; Ogasawara, R.; Sugaya, M.; Kudo, M.; Kurano, M.; Yasuda, T.; et al. Effects of Low-Intensity Cycle Training with Restricted Leg Blood Flow on Thigh Muscle Volume and VO2MAX in Young Men. J. Sports Sci. Med. 2010 , 9 , 452–458. [ Google Scholar ]
Sakamaki, M.; Bemben, M.G.; Abe, T. Legs and trunk muscle hypertrophy following walk training with restricted leg muscle blood flow. J. Sports Sci. Med. 2011 , 10 , 338–340. [ Google Scholar ]
de Oliveira, M.F.; Caputo, F.; Corvino, R.B.; Denadai, B.S. Short-term low-intensity blood flow restricted interval training improves both aerobic fitness and muscle strength. Scand. J. Med. Sci. Sports 2016 , 26 , 1017–1025. [ Google Scholar ] [ CrossRef ]
Peters, S.A.; Singhateh, Y.; Mackay, D.; Huxley, R.R.; Woodward, M. Total cholesterol as a risk factor for coronary heart disease and stroke in women compared with men: A systematic review and meta-analysis. Atherosclerosis 2016 , 248 , 123–131. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Hokanson, J.E.; Austin, M.A. Plasma Triglyceride Level is a Risk Factor for Cardiovascular Disease Independent of High-Density Lipoprotein Cholesterol Level: A Metaanalysis of Population-Based Prospective Studies. Eur. J. Cardiovasc. Prev. Rehabil. 1996 , 3 , 213–219. [ Google Scholar ] [ CrossRef ]
Adeva-Andany, M.M.; Martinez-Rodriguez, J.; Gonzalez-Lucan, M.; Fernandez-Fernandez, C.; Castro-Quintela, E. Insulin resistance is a cardiovascular risk factor in humans. Diabetes Metab. Syndr. 2019 , 13 , 1449–1455. [ Google Scholar ] [ CrossRef ]
Nahmias, A.; Stahel, P.; Xiao, C.; Lewis, G.F. Glycemia and Atherosclerotic Cardiovascular Disease: Exploring the Gap Between Risk Marker and Risk Factor. Front. Cardiovasc. Med. 2020 , 7 , 100. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Coughlan, K.A.; Valentine, R.J.; Ruderman, N.B.; Saha, A.K. AMPK activation: A therapeutic target for type 2 diabetes? Diabetes Metab. Syndr. Obes. 2014 , 7 , 241–253. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Hardie, D.G.; Pan, D.A. Regulation of fatty acid synthesis and oxidation by the AMP-activated protein kinase. Biochem. Soc. Trans. 2002 , 30 Pt 6 , 1064–1070. [ Google Scholar ] [ CrossRef ]
Rasmussen, B.B.; Hancock, C.R.; Winder, W.W. Postexercise recovery of skeletal muscle malonyl-CoA, acetyl-CoA carboxylase, and AMP-activated protein kinase. J. Appl. Physiol. 1998 , 85 , 1629–1634. [ Google Scholar ] [ CrossRef ]
Wojtaszewski, J.F.; Nielsen, P.; Hansen, B.F.; Richter, E.A.; Kiens, B. Isoform-specific and exercise intensity-dependent activation of 5’-AMP-activated protein kinase in human skeletal muscle. J. Physiol. 2000 , 528 Pt 1 , 221–226. [ Google Scholar ] [ CrossRef ]
Russell, R.R.; Li, J.; Coven, D.L.; Pypaert, M.; Zechner, C.; Palmeri, M.; Giordano, F.J.; Mu, J.; Birnbaum, M.J.; Young, L.H. AMP-activated protein kinase mediates ischemic glucose uptake and prevents postischemic cardiac dysfunction, apoptosis, and injury. J. Clin. Investig. 2004 , 114 , 495–503. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Hu, H.; Li, X.; Ren, D.; Tan, Y.; Chen, J.; Yang, L.; Chen, R.; Li, J.; Zhu, P. The cardioprotective effects of carvedilol on ischemia and reperfusion injury by AMPK signaling pathway. Biomed. Pharmacother. 2019 , 117 , 109106. [ Google Scholar ] [ CrossRef ]
Witczak, C.A.; Sharoff, C.G.; Goodyear, L.J. AMP-activated protein kinase in skeletal muscle: From structure and localization to its role as a master regulator of cellular metabolism. Cell Mol. Life Sci. 2008 , 65 , 3737–3755. [ Google Scholar ] [ CrossRef ]
Entezari, M.; Hashemi, D.; Taheriazam, A.; Zabolian, A.; Mohammadi, S.; Fakhri, F.; Hashemi, M.; Hushmandi, K.; Ashrafizadeh, M.; Zarrabi, A.; et al. AMPK signaling in diabetes mellitus, insulin resistance and diabetic complications: A pre-clinical and clinical investigation. Biomed. Pharmacother. 2022 , 146 , 112563. [ Google Scholar ] [ CrossRef ]
Christiansen, D.; Murphy, R.M.; Bangsbo, J.; Stathis, C.G.; Bishop, D.J. Increased FXYD1 and PGC-1α mRNA after blood flow-restricted running is related to fibre type-specific AMPK signalling and oxidative stress in human muscle. Acta Physiol. 2018 , 223 , e13045. [ Google Scholar ] [ CrossRef ]
Christiansen, D.; Eibye, K.H.; Hostrup, M.; Bangsbo, J. Blood flow-restricted training enhances thigh glucose uptake during exercise and muscle antioxidant function in humans. Metabolism 2019 , 98 , 1–15. [ Google Scholar ] [ CrossRef ]
Tarnopolsky, M.A.; Ruby, B.C. Sex differences in carbohydrate metabolism. Curr. Opin. Clin. Nutr. Metab. Care 2001 , 4 , 521–526. [ Google Scholar ] [ CrossRef ]
Zhao, L.; Hao, F.; Huang, J.; Liu, X.; Ma, X.; Wang, C.; Bao, Y.; Wang, L.; Jia, W.; Zhao, A. Sex-and age-related metabolic characteristics of serum free fatty acids in healthy Chinese adults. J. Proteome Res. 2020 , 19 , 1383–1391. [ Google Scholar ] [ CrossRef ]
Mauvais-Jarvis, F. Gender differences in glucose homeostasis and diabetes. Physiol. Behav. 2018 , 187 , 20–23. [ Google Scholar ] [ CrossRef ]
Razi, O.; Mohammadi, M.; Zamani, N.; Hackney, A.C.; Tourny, C.; Zouita, S.; Laher, I.; Zouhal, H. Walking exercise and lower-body blood flow restriction: Effects on systemic inflammation, lipid profiles and hematological indices in overweight middle-aged males. Res. Sports Med. 2022 , 30 , 41–49. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Vaduganathan, M.; Mensah, G.A.; Turco, J.V.; Fuster, V.; Roth, G.A. The Global Burden of Cardiovascular Diseases and Risk: A Compass for Future Health. J. Am. Coll. Cardiol. 2022 , 80 , 2361–2371. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Ohkuma, T.; Ninomiya, T.; Tomiyama, H.; Kario, K.; Hoshide, S.; Kita, Y.; Inoguchi, T.; Maeda, Y.; Kohara, K.; Tabara, Y.; et al. Brachial-Ankle Pulse Wave Velocity and the Risk Prediction of Cardiovascular Disease: An Individual Participant Data Meta-Analysis. Hypertension 2017 , 69 , 1045–1052. [ Google Scholar ] [ CrossRef ]
Thayer, J.F.; Yamamoto, S.S.; Brosschot, J.F. The relationship of autonomic imbalance, heart rate variability and cardiovascular disease risk factors. Int. J. Cardiol. 2010 , 141 , 122–131. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Heiss, C.; Rodriguez-Mateos, A.; Bapir, M.; Skene, S.S.; Sies, H.; Kelm, M. Flow-mediated dilation reference values for evaluation of endothelial function and cardiovascular health. Cardiovasc. Res. 2023 , 119 , 283–293. [ Google Scholar ] [ CrossRef ]
Staunton, C.A.; May, A.K.; Brandner, C.R.; Warmington, S.A. Haemodynamics of aerobic and resistance blood flow restriction exercise in young and older adults. Eur. J. Appl. Physiol. 2015 , 115 , 2293–2302. [ Google Scholar ] [ CrossRef ]
Ozaki, H.; Brechue, W.F.; Sakamaki, M.; Yasuda, T.; Nishikawa, M.; Aoki, N.; Ogita, F.; Abe, T. Metabolic and cardiovascular responses to upright cycle exercise with leg blood flow reduction. J. Sports Sci. Med. 2010 , 9 , 224. [ Google Scholar ] [ PubMed ]
Kumagai, K.; Kurobe, K.; Zhong, H.; Loenneke, J.; Thiebaud, R.; Ogita, F.; Abe, T. Cardiovascular drift during low intensity exercise with leg blood flow restriction. Acta Physiol. Hung. 2012 , 99 , 392–399. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Sheriff, D.; Wyss, C.R.; Rowell, L.B.; Scher, A.M. Does inadequate oxygen delivery trigger pressor response to muscle hypoperfusion during exercise? Am. J. Physiol. Heart Circ. Physiol. 1987 , 253 , H1199–H1207. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Wyss, C.R.; Ardell, J.L.; Scher, A.M.; Rowell, L.B. Cardiovascular responses to graded reductions in hindlimb perfusion in exercising dogs. Am. J. Physiol. 1983 , 245 , H481–H486. [ Google Scholar ] [ CrossRef ]
Grotle, A.K.; Macefield, V.G.; Farquhar, W.B.; O’Leary, D.S.; Stone, A.J. Recent advances in exercise pressor reflex function in health and disease. Auton. Neurosci. 2020 , 228 , 102698. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Sundblad, P.; Kolegard, R.; Rullman, E.; Gustafsson, T. Effects of training with flow restriction on the exercise pressor reflex. Eur. J. Appl. Physiol. 2018 , 118 , 1903–1909. [ Google Scholar ] [ CrossRef ]
Kolluru, G.K.; Sinha, S.; Majumder, S.; Muley, A.; Siamwala, J.H.; Gupta, R.; Chatterjee, S. Shear stress promotes nitric oxide production in endothelial cells by sub-cellular delocalization of eNOS: A basis for shear stress mediated angiogenesis. Nitric Oxide 2010 , 22 , 304–315. [ Google Scholar ] [ CrossRef ]
Green, D.J.; Smith, K.J. Effects of Exercise on Vascular Function, Structure, and Health in Humans. Cold Spring Harb. Perspect. Med. 2018 , 8 , a029819. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Shimizu, R.; Hotta, K.; Yamamoto, S.; Matsumoto, T.; Kamiya, K.; Kato, M.; Hamazaki, N.; Kamekawa, D.; Akiyama, A.; Kamada, Y.; et al. Low-intensity resistance training with blood flow restriction improves vascular endothelial function and peripheral blood circulation in healthy elderly people. Eur. J. Appl. Physiol. 2016 , 116 , 749–757. [ Google Scholar ] [ CrossRef ]
Rodriguez-Miguelez, P.; Lima-Cabello, E.; Martinez-Florez, S.; Almar, M.; Cuevas, M.J.; Gonzalez-Gallego, J. Hypoxia-inducible factor-1 modulates the expression of vascular endothelial growth factor and endothelial nitric oxide synthase induced by eccentric exercise. J. Appl. Physiol. 2015 , 118 , 1075–1083. [ Google Scholar ] [ CrossRef ]
Ferguson, R.A.; Hunt, J.E.A.; Lewis, M.P.; Martin, N.R.W.; Player, D.J.; Stangier, C.; Taylor, C.W.; Turner, M.C. The acute angiogenic signalling response to low-load resistance exercise with blood flow restriction. Eur. J. Sport. Sci. 2018 , 18 , 397–406. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Joshi, S.; Mahoney, S.; Jahan, J.; Pitts, L.; Hackney, K.J.; Jarajapu, Y.P. Blood flow restriction exercise stimulates mobilization of hematopoietic stem/progenitor cells and increases the circulating ACE2 levels in healthy adults. J. Appl. Physiol. 2020 , 128 , 1423–1431. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Karabulut, M.; Esparza, B.; Dowllah, I.M.; Karabulut, U. The impact of low-intensity blood flow restriction endurance training on aerobic capacity, hemodynamics, and arterial stiffness. J. Sports Med. Phys. Fit. 2021 , 61 , 877–884. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Ferreira Junior, A.; Schamne, J.C.; Altimari, L.R.; Okano, A.H.; Okuno, N.M. Effect of walk training combined with blood flow restriction on resting heart rate variability and resting blood pressure in middle-aged men. Mot. Rev. Educ. Física 2019 , 25 , e101945. [ Google Scholar ] [ CrossRef ]
Iida, H.; Nakajima, T.; Kurano, M.; Yasuda, T.; Sakamaki, M.; Sato, Y.; Yamasoba, T.; Abe, T. Effects of walking with blood flow restriction on limb venous compliance in elderly subjects. Clin. Physiol. Funct. Imaging 2011 , 31 , 472–476. [ Google Scholar ] [ CrossRef ]
Iversen, E.; Rostad, V. Low-load ischemic exercise-induced rhabdomyolysis. Clin. J. Sport. Med. 2010 , 20 , 218–219. [ Google Scholar ] [ CrossRef ]
Tabata, S.; Suzuki, Y.; Azuma, K.; Matsumoto, H. Rhabdomyolysis After Performing Blood Flow Restriction Training: A Case Report. J. Strength Cond. Res. 2016 , 30 , 2064–2068. [ Google Scholar ] [ CrossRef ]
Spranger, M.D.; Krishnan, A.C.; Levy, P.D.; O’Leary, D.S.; Smith, S.A. Blood flow restriction training and the exercise pressor reflex: A call for concern. Am. J. Physiol. Heart Circ. Physiol. 2015 , 309 , H1440–H1452. [ Google Scholar ] [ CrossRef ]
de Queiros, V.S.; Dantas, M.; Neto, G.R.; da Silva, L.F.; Assis, M.G.; Almeida-Neto, P.F.; Dantas, P.M.S.; Cabral, B. Application and side effects of blood flow restriction technique: A cross-sectional questionnaire survey of professionals. Medicine 2021 , 100 , e25794. [ Google Scholar ] [ CrossRef ]
Nakajima, T.; Kurano, M.; Iida, H.; Takano, H.; Oonuma, H.; Morita, T.; Meguro, K.; Sato, Y.; Nagata, T. Use and safety of KAATSU training: Results of a national survey. Int. J. KAATSU Train. Res. 2006 , 2 , 5–13. [ Google Scholar ] [ CrossRef ]
Araujo, J.P.; Silva, E.D.; Silva, J.C.; Souza, T.S.; Lima, E.O.; Guerra, I.; Sousa, M.S. The acute effect of resistance exercise with blood flow restriction with hemodynamic variables on hypertensive subjects. J. Hum. Kinet. 2014 , 43 , 79–85. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Cahalin, L.P.; Formiga, M.F.; Owens, J.; Anderson, B.; Hughes, L. Beneficial Role of Blood Flow Restriction Exercise in Heart Disease and Heart Failure Using the Muscle Hypothesis of Chronic Heart Failure and a Growing Literature. Front. Physiol. 2022 , 13 , 924557. [ Google Scholar ] [ CrossRef ] [ PubMed ]
Silveira, A.L.B.; de Carvalho, L.M.; Di Masi, F.; Pio, T.W.; Bentes, C.M. Blood flow restriction training on hypertesive subjects: A systematic review. Arch. Phys. Med. Rehabil. 2022 , 39 , 101–107. [ Google Scholar ]
Nakajima, T.; Takano, H.; Kurano, M.; Iida, H.; Kubota, N.; Yasuda, T.; Kato, M.; Meguro, K.; Sato, Y.; Yamazaki, Y. Effects of KAATSU training on haemostasis in healthy subjects. Int. J. KAATSU Train. Res. 2007 , 3 , 11–20. [ Google Scholar ] [ CrossRef ]
Nakajima, T.; Morita, T.; Sato, Y. Key considerations when conducting KAATSU training. Int. J. KAATSU Train. Res. 2011 , 7 , 1–6. [ Google Scholar ] [ CrossRef ]
Counts, B.R.; Dankel, S.J.; Barnett, B.E.; Kim, D.; Mouser, J.G.; Allen, K.M.; Thiebaud, R.S.; Abe, T.; Bemben, M.G.; Loenneke, J.P. Influence of relative blood flow restriction pressure on muscle activation and muscle adaptation. Muscle Nerve 2016 , 53 , 438–445. [ Google Scholar ] [ CrossRef ]
Spitz, R.W.; Chatakondi, R.N.; Bell, Z.W.; Wong, V.; Dankel, S.J.; Abe, T.; Loenneke, J.P. The impact of cuff width and biological sex on cuff preference and the perceived discomfort to blood-flow-restricted arm exercise. Physiol. Meas. 2019 , 40 , 055001. [ Google Scholar ] [ CrossRef ]
Mouser, J.G.; Dankel, S.J.; Jessee, M.B.; Mattocks, K.T.; Buckner, S.L.; Counts, B.R.; Loenneke, J.P. A tale of three cuffs: The hemodynamics of blood flow restriction. Eur. J. Appl. Physiol. 2017 , 117 , 1493–1499. [ Google Scholar ] [ CrossRef ]

Click here to enlarge figure

The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

Cho, C.; Lee, S. The Effects of Blood Flow Restriction Aerobic Exercise on Body Composition, Muscle Strength, Blood Biomarkers, and Cardiovascular Function: A Narrative Review. Int. J. Mol. Sci. 2024 , 25 , 9274. https://doi.org/10.3390/ijms25179274

Cho C, Lee S. The Effects of Blood Flow Restriction Aerobic Exercise on Body Composition, Muscle Strength, Blood Biomarkers, and Cardiovascular Function: A Narrative Review. International Journal of Molecular Sciences . 2024; 25(17):9274. https://doi.org/10.3390/ijms25179274

Cho, Chaeeun, and Sewon Lee. 2024. "The Effects of Blood Flow Restriction Aerobic Exercise on Body Composition, Muscle Strength, Blood Biomarkers, and Cardiovascular Function: A Narrative Review" International Journal of Molecular Sciences 25, no. 17: 9274. https://doi.org/10.3390/ijms25179274

Article Metrics

Article access statistics, further information, mdpi initiatives, follow mdpi.

Subscribe to receive issue release notifications and newsletters from MDPI journals

IMAGES

Top 10 Effective Dog Training Methods: A Complete Guide on Dog Training
The Best Training Method for Dogs According to Research
Interesting Dog Training Statistics
Dog Training Science Resources
(PDF) Improving dog training methods: Efficacy and efficiency of reward
Different Training Methods For Your Dog

COMMENTS

Improving dog training methods: Efficacy and efficiency of reward and
The methods used to train dogs range broadly with some using rewards and other non-invasive techniques (reward methods), others using mainly aversive stimuli (aversive methods) and still others using a combination of both (mixed methods). Strong claims have been made for the negative effect of the use of aversive stimuli in training on dog ...
PDF Position Statement on Humane Dog Training
using aversive training methods, they continued to show stress-related behaviors when the cue was presented, suggesting the cue itself had become aversive.5,7,8 In 2020, de Castro et al found that dogs trained with aversive methods were more 'pessimistic' on average compared to dogs trained using reward-based methods.16 Training Effectiveness
Dog training methods: Their use, effectiveness and interaction with
to 364 dog owners in order to examine the relative effectiveness of different training methods and their effects upon a pet dog's. behaviour. When asked how they trained their dog on seven basic ...
Does training method matter? Evidence for the negative impact of
Dog training methods range broadly from those using mostly positive punishment and negative reinforcement (aversive-based) to those using primarily positive reinforcement (reward-based). ... Altogether, previous research suggests that training using positive reinforcement may improve the learning ability of dogs. It remains unclear why a ...
Improving dog training methods: Efficacy and efficiency of ...
Dogs play an important role in our society as companions and work partners, and proper training of these dogs is pivotal. For companion dogs, training helps preventing or managing dog behavioral problems—the most frequently cited reason for relinquishing and euthanasia, and it promotes successful dog-human relationships and thus maximizes benefits humans derive from bonding with dogs.
(PDF) Improving dog training methods: Efficacy and ...
used to examine the frequency of stress-related behaviors (e.g., lip lick, yawn) and the over-. all behavioral state of the dog (e.g., tense, relaxed), and saliva samples were analyzed for ...
PDF Review of dog training methods: welfare, learning ability, and current
rsus aversive-based methods has relied on dog guardians' responses to surveys. The main focus of these surveys has been on the link betwee. training methods and the occurrence of aggression and other problem behaviours. Casey et al. (2014, 2013) conducted a survey of 3897 dog guardians in the UK, asking about the tra.
Improving dog training methods: Efficacy and efficiency of reward and
To that end, we will apply different dog training methods in a population of working dogs and evaluate the outcome after a period of training. The use of working dogs will allow for a rigorous experimental design and control, with randomization of treatments. Military (n = 10) and police (n = 20) dogs will be pseudo-randomly allocated to two ...
Working Dog Training for the Twenty-First Century
Detection Dogs . Training of detection dogs (narcotics and explosives) is a relatively more modern phenomenon with initial research dating back to World War II, but with wide-scale adoption occurring during the Vietnam War ().One of the original manuals for training detection dogs was written during the Vietnam War era as part of a research program by the Southwest Research Institute in San ...
Does training method matter? Evidence for the negative impact of ...
Dog training methods range broadly from those using mostly positive punishment and negative reinforcement (aversive-based) to those using primarily positive reinforcement (reward-based). Although aversive-based training has been strongly criticized for negatively affecting dog welfare, there is no comprehensive research focusing on companion dogs and mainstream techniques, and most studies ...
The effects of using aversive training methods in dogs—A review
The purpose of this study was to review a series of studies (N = 17) regarding the effects of using various methods when training dogs. The reviewed studies examined the differences between training methods (e.g., methods based on positive reinforcement, positive punishment, escape/avoidance, et cetera) on a dog's physiology, welfare, and behavior toward humans and other dogs.
Dog Training Methods: Types, Research, And How to Choose
How To Choose A Dog Training Method. There's no single best training approach for all dogs. "Every dog is an individual and should be trained as such," says Emily Birch, a clinical animal behaviorist.For instance, you might find a confident dog responds well to clicker training, while an anxious dog does better with positive reinforcement alone.
Dog training methods: their use, effectiveness and interaction with
We distributed a questionnaire to 364 dog owners in order to examine the relative effectiveness of different training methods and their effects upon a pet dog's behaviour. When asked how they trained their dog on seven basic tasks, 66% reported using vocal punishment, 12% used physical punishment, 60% praise (social reward), 51% food rewards ...
The New Science of Dog Training
Choke collars, shock collars, even the word no are all-but-verboten. It's a new day in dog training. The science upon which these new techniques are based is not exactly new: it's rooted in ...
Barriers to the adoption of humane dog training methods
Introduction. Despite many studies recommending the use of reward-based training methods for pet dogs, including a recent review (Ziv, 2017), many owners continue to use positive punishment and negative reinforcement.Behavior problems are the leading cause of death of dogs under 3 years old (American Veterinary Society for Animal Behavior [AVSAB], 2008a), and an owner's perception of a dog as ...
Most Effective Dog Training Methods According to Science
Therefore, reward based training has been proven to be one of the most effective dog training methods. Bottom Line: Compared to punishment based training, research shows that reward-based dog ...
Dog training methods: their use, effectiveness and interaction with
Because satisfied owners are less likely to relinquish or abandon their dogs (Arkow & Dow 1984), training methods that produce an obedient dog may exert a secondary welfare benefit. Problematic behaviours are common within the general dog population (Voith et al 1992; Clark & Boyer 1993; O Farrell 1997).
The Best Training Method for Dogs According to Research
3. Electronic Training. Electronic training, also known as shock-collar training, on the other hand, is based on the use of corrections through the use of an electric collar that delivers a shock when the dog fails to perform a desired behavior. Electronic training falls under the category of aversion-based methods.
Training methods based on punishment compromise dog welfare, study
After aversive training, dogs had a lower behavioral state (higher stress and anxiety), a new study has found. If aversive methods were used in high proportions, that persisted even in other contexts.
The click is not the trick: the efficacy of clickers and other
Despite the prominence of positive reinforcement-based training methods in the professional dog training community (Blackwell et al., 2008; Hiby, Rooney & Bradshaw, 2004), recommendations of how, when, and what method of positive reinforcement should be used are inconsistent (see Browne et al. (2017), for a review of the general content in best ...
Science Based Dog Training: Training Dogs Research & History
Heavily influenced by Pavlov and Skinners' research, clicker training came about in the 1940s, thanks to Keller Breland and Marion and Bob Bailey. This science-based dog training method is built ...
Positive Reinforcement and Dog Training VII: Summary and Conclusions
In conclusion, these studies show that the use of positive reinforcement only is the best way to train a dog. Sadly, this research also shows that just 16-20% of owners take this approach. Most dog owners continue to use punishment, with about 50% using punishment more often than rewards. In fact, a new 2017 literature review (that includes the ...
Dog Training Science Resources
The first section lists scientific research on dog training methods ( positive reinforcement vs aversives, shock collars etc). The second section looks at other aspects of the science of dog training (body language, preference for types of positive reinforcement, etc). This page is updated regularly. If you have suggestions for additions ...
Nous Research unveils powerful new AI training optimizer DisTrO
Nous Research unveils new tool to train powerful AI models with 10,000x efficiency. ... While traditional methods of AI training require synchronizing full gradients across all GPUs, and rely on ...
Cardiovascular Responses: Flywheel Resistance Training vs. Traditional
The Exerfly Platform was recently used in a study on the cardiovascular adaptations resulting from different high-intensity resistance training methods. A lot of research has evaluated the effects of both traditional resistance training (TRT) and flywheel resistance training (FRT) on measures of neuromuscular and athletic performance. But much ...
IJMS
Blood flow restriction exercise has emerged as a promising alternative, particularly for elderly individuals and those unable to participate in high-intensity exercise. However, existing research has predominantly focused on blood flow restriction resistance exercise. There remains a notable gap in understanding the comprehensive effects of blood flow restriction aerobic exercise (BFRAE) on ...

Save citation to file

Add to My Bibliography

Improving dog training methods: Efficacy and efficiency of reward and mixed training methods

Conflict of interest statement

Similar articles

Grants and funding

Other Literature Sources

Miscellaneous

Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare

1. Introduction

2. Materials and methods

2.2. Training schools

2.2.2. Classification of training methods.

2.3. Subjects

2.4. Data collection

2.4.1. Phase 1 –Evaluating welfare within the training context.

2.4.2. Phase 2—Evaluating welfare outside the training context.

2.5. Questionnaire

2.6. Data analysis

2.6.2. Phase 2 –Evaluating welfare outside the training context.

2.7. Statistical analyses

3.1. Questionnaire

3.1.2. Owner demographics, experience with dogs and dog training.

3.2. Phase 1 –Evaluating welfare within the training context

3.2.2. Physiological data.

3.3. Phase 2 –Evaluating welfare outside the training context

3.3.1. Training phase.

3.3.2. Test phase.

4. Discussion

5. Conclusions

Supporting information

S2 Appendix. Raw data underlying all the analyzes performed in the current research paper.

S3 Appendix. Negative binomial and generalized linear mixed model details.

Acknowledgments

Login Alert

Article contents

Access options

This article has been cited by the following publications. This list is generated based on data provided by Crossref .

Save article to Kindle

Save article to Dropbox

Save article to Google Drive

Reply to: Submit a response

Your details

You have entered the maximum number of contributors

How Science is Revolutionizing the World of Dog Training

More Must-Reads from TIME

Dog training methods: their use, effectiveness and interaction with behaviour and welfare

Related Papers

RELATED PAPERS

RELATED TOPICS

The click is not the trick: the efficacy of clickers and other reinforcement methods in training naïve dogs to perform new tasks

Introduction

General design

Experiment 1

Setting and subjects

Results and discussion

Experiment 2

Experiment 3

Age of the dogs tested

Statistical power

Number of primary reinforcement pairings

Function of the clicker

Applied environment

Conclusions

Supplemental Information

Supplemental Information 2

Supplemental Information 3

Supplemental Information 4

Funding Statement

Additional Information and Declarations

Science-based Dog Training: How Research Influenced Our Approach to Training Dogs

The Alpha Dog Theory of Training

Science-Based Dog Training: How research influenced our approach to training dogs.

LATEST FEATURES

Search This Blog

Dog Training Science Resources

Scientific Research on Dog Training Methods

Other Scientific Research on Dog Training

Popular posts from this blog

The Schedule of Events for Bark! Fest, the Book Festival for Animal Lovers