Elsevier

Applied Animal Behaviour Science

Volume 163, February 2015, Pages 158-166
Applied Animal Behaviour Science

Consistency of shelter dogs’ behavior toward a fake versus real stimulus dog during a behavior evaluation

https://doi.org/10.1016/j.applanim.2014.12.001Get rights and content

Highlights

  • We examined the consistency of shelter dogs’ reactions toward fake and real dogs.

  • We found a high degree of agreement for the friendly trait.

  • Aggressive trait did not demonstrate adequate agreement.

  • Fearful trait demonstrated moderate agreement.

Abstract

Behavior evaluations are widely used by animal shelters and other organizations that rehome dogs. The dog-to-dog subtest is a common feature of most canine behavior evaluations. The use of model devices such as a stuffed dog during this subtest could be convenient for shelters and increase safety. However, there is little research indicating if a fake dog can be reliably used instead of a live dog. In this study, the consistency of shelter dogs’ reactions toward a fake and a real dog during the dog-to-dog subtest was investigated. Forty-five shelter dogs were evaluated using two different stimulus conditions. In one condition, the test dog was confronted with a single plush dog (the same plush dog for all test dogs), and in the other, with a single live dog (the same live dog for all test dogs). A standardized list of behaviors was recorded as observed or absent for both conditions with each dog serving as its own control. To calculate the agreement of individual behaviors between the two conditions, Cohen's Kappa was used. However, since many of the behaviors occurred at very low or high frequency rates, Prevalence-Adjusted, Bias-Adjusted Kappa (PABAK) was used along with Cohen's Kappa due to Cohen's Kappa's sensitivity to high or low prevalence, for which PABAK adjusts. For the purposes of this study, PABAK or Kappa scores greater than 0.61 were considered an indicator of a good degree of agreement between reactions toward the fake and the real dogs. The degree of agreement varied widely across individual behaviors with, Kappa ranging from −0.04 to 0.75 and PABAK from 0.29 to 1. Collapsing individual behaviors into behavior traits (e.g., friendly, aggressive, fearful) revealed a high degree of agreement for the friendly trait (Kappa = 0.60, PABAK = 0.69). However, the aggressive trait did not demonstrate adequate agreement (Kappa = 0.11 and PABAK = 0.38) and the fearful trait demonstrated only moderate agreement between the two stimulus conditions (Kappa = 0.50 and PABAK = 0.51). These results suggest that, while it may be possible to use a fake dog for the dog-to-dog subtest to assess friendly behavior toward other dogs, fearful and aggressive behaviors may not be consistent between the fake and real dogs, thus limiting the usefulness of the fake dog during behavior evaluations. In addition, the results of this study suggest more research is needed into the predictive validity of both fake and real dogs, since it appears the stimulus dog, whether fake or real, can influence the subtest's results.

Introduction

Behavior evaluations play a critical role for shelter and rescue dogs, and are used to identify behavior tendencies in order to rehome an animal into an appropriate home (D’Arpino et al., 2012, Dowling-Guyer et al., 2011, Ledger and Baxter, 1997, Reid and Collins, 2012, Van der Borg et al., 1991). Research has shown that 28% of sheltering organizations use a standardized behavior evaluation (D’Arpino et al., 2012), with 63% of higher volume organizations (with annual intake more than 1000 dogs) using one. According to Taylor and Mills (2006), the dog-to-dog subtest is one of the 20 most commonly used subtests included in most standardized behavior evaluations, such as: Assess-A-Pet™ (Bollen and Horowitz, 2008), Match-Up II Shelter Dog Rehoming Program™ (Center, 2013, Dowling-Guyer et al., 2011), SAFER® Aggression Assessment (Weiss, 2007). This subtest is used to gain insight into shelter dog behavior toward conspecifics. Since many shelter dogs are strays or transfers from external organizations, with no behavioral history to consult, the dog-to-dog subtest is a key instrument to gather information about a dog's behavior toward other dogs. Although common, it can be a difficult subtest to implement because it depends entirely on what other dogs are available in the shelter for use as the stimulus dog. Therefore, the use of model devices such as a stuffed dog could be beneficial for many shelters. Model devices are convenient to use and reduce the risk of injury to the dog, and might be especially useful in large organizations where staff has limited time to perform evaluations. Moreover, use of a fake dog instead of a real stimulus dog during a dog-to-dog evaluation allows direct interaction that otherwise could be uncomfortable or dangerous for both the dog and handler (Reid and Collins, 2012).

Although the use of a life-like artificial dog substitute in the evaluation of shelter dogs is not a new idea, there is little research indicating if a fake dog can be reliably used instead of a live dog in shelter behavior evaluations. Previous studies have used plastic dogs (Barnard et al., 2012, Reid and Collins, 2012) or stuffed models (Leaver and Reimchen, 2008) to learn how dogs behave toward these models. In a study by Leaver and Reimchen (2008), a fake dog was fitted with several varying-length tails and introduced to several real dogs to learn more about canine communication. Two other studies evaluated the validity of using a fake dog during a behavior evaluation. Barnard et al. (2012) used the C-BARQ questionnaire to assess the dogs’ behavior history toward conspecifics, then the same dogs were evaluated using a plastic dog. Although the fake dog elicited many social reactions that were not observed when an ambiguous object (black plastic garbage bag filled with crumpled newspaper) was shown to the dogs, the reactions to the device were only partially consistent with the dogs’ aggressive history. Only two dogs from the group of 12 dogs who had a history of aggression toward conspecifics displayed some levels of aggressive behavior (bark, growl, snaps) toward the fake dog. Reid and Collins (2012) evaluated Pit Bull Terrier type fighting dogs. They found that the dogs’ responses to the fake dog were similar to a real dog using the same evaluation. However, incidences of aggression toward conspecifics in dogs bred for fighting are significantly higher than in most shelter dogs (Capra et al., 2009, Reid and Collins, 2012). Consequently, Reid and Collins's study results might not apply to a shelter population which is more diverse than the dogs used in that study.

This study investigated the subject further using shelter dogs, a controlled environment and a standardized dog-to-dog interaction assessment to examine the consistency of reactions toward a fake and a real dog. We hypothesized that dogs would react similarly to a fake dog as to a real dog. This study was not designed to investigate whether information gathered during the dog-to-dog subtest accurately predicted future behavior toward conspecifics; rather, this study examined the consistency of behavior between the two conditions. The results could advise shelters whether they could reliably use a fake dog instead of a live dog in dog-to-dog subtests already being conducted in shelters. In addition, information about the effectiveness of the fake dog as a proxy for live dogs may help shelters to accelerate the evaluation process, as well as make it safer and more comfortable for both humans and animals.

Section snippets

Materials and methods

The behavior evaluation and the subsequent analysis of test results received full ethical approval from participating shelter organizations’ senior management. The protocol for this study was approved by the Center for Shelter Dogs’ internal review panel prior to initiation of the project.

Behavior frequencies

Fig. 2 presents the percent of dogs who exhibited each behavior toward the fake and the real dog separately. A wide range of behaviors was observed, although ‘approach’ and ‘sniff dog’ were the most common behaviors, displayed by more than 90% of the sample toward both the fake and the real dog. The second most common behaviors were ‘wag tail’ and ‘lip lick’, displayed by more than 50% of the sample, toward both the fake and the real dog. However, some behaviors were not observed at all, among

Discussion

Our hypothesis that dogs would react similarly to a fake dog as to a real dog was partially supported by the results of this study. Although the observed agreement was very high for all behaviors, it was mainly due to the absence of behaviors rather than to their presence. Because so few instances of many behaviors were observed, those behaviors reached a very high degree of negative agreement, meaning that when the behavior was not present in one condition, it most likely was not present in

Acknowledgments

This study was supported by a grant from The Stanton Foundation. The funder had no involvement with the conduct of this study, interpretation of the results, or writing and submission of this manuscript.

The authors thank Kim Melanson, Carol Ahearn, Laney Nee, and Christopher D’Arpino and his dog Ferris for helping with technical preparation and implementation of the study, Dr. Amy Marder for her professional advice, and Liz Fay for her great onsite support during the study implementation as

References (26)

  • W.J. Netto et al.

    Behavioral testing for aggression in the domestic dog

    Appl. Anim. Behav. Sci.

    (1997)
  • K.D. Taylor et al.

    The development and assessment of temperament tests for adult companion dogs

    J. Vet. Behav.

    (2006)
  • J.A.M. Van der Borg et al.

    Behavioral testing of dogs in animal shelters to predict problem behavior

    Appl. Anim. Behav. Sci.

    (1991)
  • Cited by (12)

    • Demographic characteristics of shelter dogs predict performance on tests of a behavioral evaluation and overall performance predicts adoption success

      2023, Journal of Veterinary Behavior
      Citation Excerpt :

      However, of the 1,104 dogs tested during our study period, evaluators used a fake dog with only 6 dogs during the seeing subtest and 12 dogs during the meeting subtest, so use of the fake dog probably had little effect on our findings for these subtests. Additionally, fake dogs have been found to be useful in assessing friendly behavior toward other dogs (Shabelansky et al., 2015) as well as aggressive behavior in fighting dogs (Reid et al., 2022). Fourth, the small number of dogs assessed as showing dangerous behaviors meant that for most analyses, we combined their data with data from dogs showing concerning behaviors (measures of prevalence were an exception), thereby losing information on the demographics and likelihood of return of dogs showing different degrees of behavioral responses.

    • The utility of model dogs for assessing conspecific aggression in fighting dogs

      2022, Applied Animal Behaviour Science
      Citation Excerpt :

      However, within individuals, the type of the social behavior exhibited varied between the two stimuli. The authors concluded that a model dog can be useful for identifying friendly behavior toward other dogs, but its ability to identify aggressive or fearful behavior is limited (Shabelansky et al., 2015). More recent shelter-based research supported Barnard and colleagues (2012), reporting a high correlation between the social behavior directed at a model and that directed at live conspecifics (Barnard et al., 2019).

    • Dogs’ responses to visual, auditory, and olfactory cat-related cues

      2017, Applied Animal Behaviour Science
      Citation Excerpt :

      Other shelters may resort to using live cats to conduct informal assessments of dogs’ behaviors around cats, but such assessments have not been validated and may be stressful and potentially harmful for the cats employed. Given that previous studies have shown that dog models and dolls do provide some indication of how dogs respond to living dogs and children (Barnard et al., 2012; Shabelansky et al., 2015), it is possible that a dog’s response to a cat model may predict how a dog will respond to cats or other small animals. Previous research examining dogs’ responses to dogs and children have relied upon visual representations of dogs and small children even though the dog’s umwelt differs substantially from our own.

    • What is the evidence for reliability and validity of behavior evaluations for shelter dogs? A prequel to “No better than flipping a coin”

      2019, Journal of Veterinary Behavior
      Citation Excerpt :

      Most of these provocations, the justifications behind them, and their subsequent interpretations have not been empirically evaluated. However, data have shed doubt on the ability of model devices such as a human-like doll (Barnard et al., 2012) and a fake dog to assess responses to real children and dogs (Shebelansky et al., 2015). Both of these are used in some behavior evaluations.

    • Assessing the Behavior of Shelter Dogs

      2022, Animal Behavior for Shelter Veterinarians and Staff
    View all citing articles on Scopus
    View full text