Research Poster Presentations 2012

Business Insurers of the Carolinas
We are grateful for the support of Business Insurers of the Carolinas, who is providing sponsorship for the Best Poster by a Dog Trainer Award for 2012.

Past Awards

Analysis of an Owner Training Package for Integrity of Counterconditioning Targeting Problem Behavior of Dogs

Kristyn Echterling-Savage
Beyond the Dog, LLC

Florence D. DiGennaro Reed and L. Keith Miller
University of Kansas

Sean Savage
Beyond the Dog, LLC

Abstract

This study examined the effects of consumer management procedures on dog owner treatment integrity (i.e., extent to which treatment is implemented as prescribed). In one condition, verbal instruction, modeling, and feedback were used to train owners to implement a counterconditioning procedure to address dog aggression in relatively simple contexts. Generalization Programming consisted of owner plan implementation and performance feedback in more complex contexts. Treatment integrity increased above baseline levels during both conditions. In addition, there was a substantial reduction in dog aggression and a slight reduction in dog precursor behavior (i.e., behavior immediately preceding aggression).  Owners and experts rated the goals, procedures, and effects as highly acceptable. Implications for use are discussed.

top of page

Relative Efficacy of Human Social Interaction and Food as Reinforcers for Domestic Dogs and Hand-reared Wolves

Erica N. Feuerbacher and Clive D. L. Wynne, Department of Psychology, University of Florida, P.O. Box 112250, Gainesville, FL 32611-2250

Despite the intimate relationship dogs share with humans in Western society, we know relatively little about the variables that produce and maintain dog social behavior towards humans. While this is an interesting basic question, it also has direct applications into how we train dogs and how we can establish and maintain effective relationships between dogs and humans. One possibility is that human social interaction is itself a reinforcer for dog behavior. As an initial assessment of the variables that might maintain dog social behavior, we compared the relative efficacy of brief human social interaction to a small piece of food as a reinforcer for an arbitrary response (nose touch). We investigated this in three populations of canids: shelter dogs, owned dogs, and hand-reared wolves. Across all three canid populations, brief social interaction was a relatively ineffective reinforcer compared to food for most canids, producing lower responding and longer latencies than food. Food typically produced rapid and sustained responding, whereas most dogs stopped responding for brief social interaction and responded only after long latencies when they did. However, pilot data suggest that longer durations of social interaction might sustain higher rates of responding, but still fall short of the high, sustained responding produced by food. Our results support the use of food as a reinforcer, especially when sustained responding is required. Our procedures also provide a method for assessing the effectiveness of different reinforcers in individual dogs, which might improve our ability to tailor training to individual dogs.

top of page

Breed Stereotype and Effects of Handler Appearance on Perceptions of Pit Bulls

Lisa Gunter

Previous research has indicated that dog breed stereotypes exist and that the appearance of a human handler alongside a dog can affect perceptions of the dog's temperament. The present study looked at participants' perceptions of a Pit Bull-type dog in comparison to a Labrador Retriever and Border Collie; and whether the addition of a rough adult male, elderly woman or male child influences the dog's perceived characteristics of approachability, aggressiveness, intelligence, friendliness, trainability, or adoptability. The results indicated that participants viewed the Pit Bull least favorably in all six characteristics when evaluated with the other breeds, confirming the presence of a negative stereotype.

The appearance of a handler alongside the Pit Bull influenced participants' impressions of the dog on characteristics of aggressiveness, friendliness, approachability and adoptability. When comparing impressions of the Pit Bull alone versus alongside a handler, perceived intelligence improved across all three conditions. Additionally, perceptions of friendliness and adoptability increased while aggressiveness decreased in both the elderly woman and male child conditions, and the perception of friendliness decreased with the presence of the rough male. These findings demonstrate how the appearance of a human handler in photographs can influence our perceptions of Pit Bulls and suggest possibilities for the use of human handlers to positively affect the perceived qualities of Pit Bulls among the general population and particularly those who are considering adopting a dog.

top of page

Physical prompts to anthropomorphism of the domestic dog

Julie Hecht and Alexandra Horowitz

Humans readily anthropomorphize dogs, assigning them human characteristics. Theories of which physical characteristics prompt anthropomorphism usually invoke neoteny or features involved in the "cute response" (Horowitz & Bekoff, 2007). In this study, those theories are explicitly tested by comparing the inherent "likability" of visible physical features (Linsen et al., 2011). Human subjects (N=124) participated in an aesthetic preference test in which they saw computer-modified image pairs of mixed-breed adult dogs. In each image-pair presentation, one of fourteen features associated with neoteny, the cute response, or other physical characteristics was modified.

The results reveal that some, but not all, characteristics of neoteny were preferred: larger eyes were selected over smaller (p < 0.001); but a larger forehead ("cranial vault") was not (p = n.s.). Nor were other features consistent with the theory of neoteny preferred, such as small nose or big paws. Subjects also evinced a preference for images of dogs with smaller jowls, larger distance between the eyes, distinct and colored irises, and a mouth approximating a smile (p < 0.001). By contrast, symmetry of ears and piebald facial coloration, as well as size of ears, eyebrows, tongue, and nostril, did not lead to uniform subject preference (p = n.s.). Sub-analyses found that subjects' selections varied by life experiences with and perceptions of animals.

These findings demonstrate that this long-theorized human behavior — anthropomorphizing animals — can be analyzed and tested not only in connection with behavior (Mitchell & Hamm, 1997) but also physical features. Neoteny does not consistently explain subjects' preferences, nor are other theories sufficient. A catalogue of physical features which lead to anthropomorphizing could be used to design expressive robots, elicit aid for threatened species, advertise adoptable shelter animals or re-consider dog-breeding practices.

References
Linsen, S., Leyssen, M.H.R., Sammartino, J., Palmer, S.E., 2011. Aesthetic preferences in the size of images of real-world objects. Perception 40, 291-298.

Horowitz, A. C., Bekoff, M., 2007. Naturalizing anthropomorphism: Behavioral prompts to our humanizing of animals. Anthrozoös 20, 23-35.

Mitchell, R.W., Hamm, M., 1997. The interpretation of animal psychology: Anthropomorphism or behavior reading? Behaviour 134, 173-204.

top of page

The correlation of anal and shoulder contact behaviors with social behavior in the temperament tests of shelter dogs

Kathryn Lord1 & Sue Sternberg 2

1 Adjunct Professor, Canisius College, 2001 Main Street, Buffalo, NY 14208-1098
2 Director, Rondout Valley Animals for Adoption, 4628 Route 209, Accord NY 12404

It is estimated that 3-4 million dogs are relinquished to shelters in the US each year.  One of the main reasons for relinquishment is unwanted behavior.  Temperament tests are administered to dogs entering shelters to determine the likelihood that they will exhibit aggressive behavior in the home environment and consequently whether they are safe to be re-homed.  Refining these tests helps to ensure both the safety of adopters and the potential for all adoptable dogs to be re-homed.  The aim of this study is to determine if there is any correlation between four behaviors, not previously used in temperament tests, and sociability scores on the Assess-a-Pet temperament test. The frequency of four contact behaviors, referred to as the anal swipe, anal touch, shoulder stance, and shoulder rub, as well as social interactions were coded from previously recorded temperament tests of 148 dogs from 22 different shelters. The results show that there is a significant negative correlation between these four contact behaviors and sociability scores. These findings suggest that anal touch, anal swipe, shoulder rub, and shoulder stance could be useful diagnostic tools in determining the adoptability of shelter dogs, particularly in cases where the dog is borderline between clearly adoptable and clearly dangerous.

top of page

Canines with Careers:
An Alternative, Relationship-based Approach to Service Dogs

Lisa Lunghofer, Ph.D., and Sherry Woodard, CPDT-KA

Canines with Careers seeks to save dogs that would otherwise be killed in shelters by identifying those that could be trained to become service dogs. One area of focus is training rescue dogs to be psychiatric service dogs for veterans who have Posttraumatic Stress Disorder (PTSD). The Department of Defense estimates that 20-30 percent of the 2 million men and women deployed since 2001 will suffer from PTSD or other psychiatric disorders such as depression. This equates to 400,000-600,000 veterans who could potentially benefit from a service dog.

Statistics from the best-established service dog programs show that, on average, programs place 10-15 dogs per year. These programs rely primarily on a traditional service dog model in which purebred puppies are raised, trained and placed at 18-24 months old. The drawbacks of this model include a high failure rate, very high costs, and long waiting lists. As a result of the small number of dogs placed each year, there is a tremendous unmet need for service dogs.

Canines with Careers, which involves adult rescue dogs, represents an alternative to the traditional service dog model. Model components include rigorous screening procedures, comprehensive matching based on the dog's characteristics and veteran's needs, thorough training using only positive reinforcement, and program evaluation. Key to the success of the model is relationship-based training that fosters a positive and life-long bond between the veteran and the dog.

Findings from an ongoing qualitative study that currently involves interviews with 20 veterans who have PTSD and have partnered with a service dog show the profound effect these dogs are having on veterans' quality of life. Themes emerging across respondents include a renewed sense of purpose in life, increased sense of social support and an expanded network of social connections, improved relationships with family and friends, a sense of security, motivation to make a better life for oneself in order to give the dog a better life, and a desire to give back to the community.

By tracking the implementation of the Canines with Careers program model and demonstrating success, we hope to fundamentally change the service dog field by providing a compassionate, relationship-based, and cost-effective alternative to the standard practice of relying on purebred puppies.

top of page

Can Canines Count?

Krista Macpherson & William A. Roberts
Presenter: Krista Macpherson
University of Western Ontario

Although true counting and arithmetic ability are unique to the human species, an evolutionarily more primitive system for non-verbal numerical discrimination has been shown to exist in humans as well as non-human animals.  These numerical competencies have been thoroughly examined in several species, yet relatively few studies have examined such processes in the domestic dog.  In an initial experiment, procedures from numerical studies of chimpanzees (Beran, 2001; Beran & Beran 2004) were adapted for use with domestic dogs, as well as a small sample of wolves.  Subjects in these experiments watched as different quantities of food were sequentially dropped by a human experimenter into each of 2 containers. The subjects were then allowed to select and consume the contents of 1 of the containers. 

While dogs excelled in a condition in which one bowl contained 1 piece of food and the other bowl contained no food, their performance failed to significantly surpass chance on all other ratios, and their performance on the various ratios (4:1, 2:1, 3:1, 2:3) did not conform to Weber's law.  With limited trials, the performance of one wolf was significantly better than those of the dogs.  The  poor performance of the dogs on this task are interesting because they lie in stark contrast to numerical studies of chimpanzees as well as other species, in which these animals typically perform well in tasks of numerical discrimination.  In a second experiment involving a single subject (a rough collie named Sedona), procedure was revised so that non-food stimuli (circular magnets) were presented simultaneously to the dog on two magnet boards.  If Sedona chose the board with the majority of the magnets, she was rewarded with a piece of food.  If she chose incorrectly, however, she received nothing. Interestingly, Sedona's performance was significantly better than that of subjects in Experiment 1.  Procedural implications of these experiments for the study of domestic dogs will be discussed.

top of page

Using a Canine Companion Training Curriculum to Reduce Return Rates in a No-Kill Animal Shelter Environment

Lisa R. McCluskey, CPDT-KSA, CBCC-KA (lisa@moongazer.com)
Moongazer Canine Companion Training

Alexandra Tellier (Alliewarl@Gmail.Com)
Elizabeth L. Fay (Dogswarl@Gmail.Com)
Jeffrey J. Klunk (warlvolunteer@gmail.com)
Worcester Animal Rescue League

Kevin Small, Ph.D. (kevin.small@tufts.edu)
Tufts Medical Center

While no-kill shelters possess institutional advantages (e.g., increased community support and philanthropy), many challenges arise when balancing individual animal needs with limited sheltering capacities. Solutions often include: restricting intake of new animals, increasing the adoption rate of sheltered animals, and reducing the return rate of adopted animals. In 2011, Worcester Animal Rescue League (WARL)* instituted a canine companion training curriculum (CCTC)[1] to provide enrichment for their shelter dogs and extend support to families of adopted animals.[2] Our credentialed dog training and behavior specialist offered two free, one-hour, on-site, drop-in, group training sessions per week for WARL-affiliated dogs (i.e., volunteers training shelter dogs, owners training adopted dogs, and employees training shelter and/or personal dogs).

This research examines the effect of our CCTC on the WARL adoption return rate via multiple epidemiological analyses.[3] Starting with a matched case-control study (consisting of 245 dogs adopted while the training program was in place from October 2011 through July 2012), we first observed that dogs remaining in their adopted homes were 2.3 times more likely to have attended at least one training session.  When controlling for breed and date of adoption, this increased to 3.1 and 3.6 times respectively. Continuing with a survival analysis, we used a Kaplan-Meier estimate to demonstrate that returned dogs which did not receive training had a median adopted home stay of 120 days versus dogs that attended class had a median stay of 183 days (1.5 times longer).  Furthermore, at eight months post-adoption, dogs that attended at least one training session were 2.9 times more likely to remain in their adoptive home.  Finally, we fit a Cox proportional hazards regression to measure the relative impact of several factors with respect to the duration of adoption including: breed, age at intake, exposure to training, and gender. 

In conclusion, our work provides preliminary, but strong evidence that a CCTC can significantly reduce a shelter's adoption return rate. Additional studies are required to investigate the effect of interventions on adoption/euthanasia rates and the economic and ethical tradeoffs of implementing a CCTC versus accepting higher shelter return rates.
 
[1] L.R. McCluskey and K. Small. An Evidence-Based Canine Companion Training Curriculum Assessment Methodology. The Association of Pet Dog Trainers Annual Educational Conference (Poster), October 2011.
[2] S. Sternberg. A Trainer-Shelter Partnership: How Can Trainers Do More for Dogs in Shelters? The Association of Pet Dog Trainers Annual Educational Conference (Presentation), October 2011.
[3] K.J. Rothman, S. Greenland, and T.L. Lash. Modern Epidemiology. Third Edition. Wolters Kluwer, 2008.

Keywords: Pet Dog Training, Dog Handling, Canine Handler, Shelter Dogs, No-Kill Shelter, Canine Companion Training Curriculum, Canine Enrichment Programs, Adoption Return Rate, Dog Adoptions, Shelter Dog Training

top of page

Extinction – A Preliminary Study

By Lisbeth Plant KPACTP
under the guidance of Terry Ryan CPDT-KA CDBC KPACTP
ABSTRACT

It is sometimes argued that clicker trainers do not always "finish the job", i.e. put behaviors on cue and under stimulus control.

Cues are commonly introduced once a learner routinely earns reinforcement for offering correct behavior. Un-cued responses should then no longer be rewarded, and we may see the extinction burst as the learner "tries harder" to earn reinforcement.

There are examples of leading animal trainers who prefer not to use extinction as a tool.
There is currently no published data with regard to the role of extinction in the chicken model. The chicken is a good model to use as dog trainers typically lack past training experiences with them that can bias their experience.

This preliminary study used the chicken model to examine two out of nine possible options for putting behaviors on cue: #4 Let un-cued behaviors go into extinction, and #8 Give the next cue sooner.
Chickens were trained to peck a target using novel targets and cues.

Cues were withheld according to set schedules and extinction burst behaviors were measured. In option #8, each bird's individual "Natural Wait Time" before pecking was measured so that her wait time for the cue after she had pecked off-cue could be shortened according to schedule.

In option #4, the data show that the chickens learned to wait for the cue three times faster than in option #8. However, this group showed distinct signs of stress, including frequent defecation on the table and 3 of 10 birds shut down or flew away. Option #8 birds did not appear to show overt signs of stress. Instead, data indicate that the chickens learned to shorten the time they had to wait for the next cue by creating a behavior chain of 'peck-to-turn-the-cue-on' followed by an on-cue peck that earned reinforcement.
Given the wide-spread use of these methods and lack of scientific data in non-human animal models, further study into extinction and other methods of putting behaviors on cue and under stimulus control would be helpful.

* WARL (www.worcester-arl.org) is a 100-year-old, no-kill, non-profit animal shelter in central Massachusetts, which also currently serves as the local stray animal impoundment facility. While WARL limits intake of owner-surrendered animals, there is no restriction on stray animals. During this study, 88% of the dogs entering the shelter were strays and WARL never once turned away a single Worcester stray dog in need.

top of page

< Return to Poster information