Professional Documents
Culture Documents
Domestication Does Not Explain The Presence of Inequity Aversion in Dogs
Domestication Does Not Explain The Presence of Inequity Aversion in Dogs
Report
*Correspondence: jennifer.essler@vetmeduni.ac.at
http://dx.doi.org/10.1016/j.cub.2017.05.061
Current Biology 27, 1861–1865, June 19, 2017 ª 2017 Elsevier Ltd. 1861
Figure 1. Set-Up View from One Enclosure
Fencing between the two enclosures allowed for
each individual to fully see their partner’s behavior.
The same setup was used for the dogs (left) and the
wolves (right). See also Figures S1 and S2.
STAR+METHODS
4. Brucks, D., Essler, J.L., Marshall-Pescini, S., and Range, F. (2016). 19. Topál, J., Miklósi, Á., Csányi, V., and Dóka, A. (1998). Attachment behavior
Inequity aversion negatively affects tolerance and contact-seeking behav- in dogs (Canis familiaris): a new application of Ainsworth’s (1969) Strange
iors towards partner and experimenter. PLoS ONE 11, e0153799. Situation Test. J. Comp. Psychol. 112, 219–229.
5. Range, F., Leitner, K., and Virányi, Z. (2012). The influence of the relation- 20. Topál, J., Gácsi, M., Miklósi, Á., Virányi, Z., Kubinyi, E., and Csányi, V.
ship and motivation on inequity aversion in dogs. Soc. Justice Res. 25, (2005). Attachment to humans: A comparative study on hand-reared
170–194. wolves and differently socialized dog puppies. Anim. Behav. 70, 1367–
1375.
6. Range, F., Horn, L., Virányi, Z., and Huber, L. (2009). The absence of
reward induces inequity aversion in dogs. Proc. Natl. Acad. Sci. USA 21. R Core Team (2016). R: a language and environment for statistical
106, 340–345. computing (R Foundation for Statistical Computing). https://www.
r-project.org/.
7. Hare, B., and Tomasello, M. (2005). Human-like social skills in dogs?
Trends Cogn. Sci. 9, 439–444. 22. Venables, W.N., and Ripley, B.D. (2002). Modern Applied Statistics with S,
Fourth Edition (Springer).
8. Hare, B., Wobber, V., and Wrangham, R. (2012). The self-domestication
hypothesis: evolution of bonobo psychology is due to selection against 23. Bates, D., Maechler, M., Bolker, B.M., and Walker, S. (2015). Fitting linear
aggression. Anim. Behav. 83, 573–585. mixed-effects models using lme4. J. Stat. Softw. 67, 1–48.
9. Range, F., and Virányi, Z. (2015). Tracking the evolutionary origins of dog- 24. Wickham, H. (2009). ggplot2: Elegant Graphics for Data Analysis
human cooperation: the ‘‘canine cooperation hypothesis.’’ Front. Psychol. (Springer).
5, 1–10. 25. Fox, J., and Weisberg, S. (2011). An {R} Companion to Applied Regression,
10. Mech, L.D. (1970). The Wolf: Ecology and Behavior (Natural History Press, Second Edition (Sage).
Doubleday). 26. Kuznetsova, A., Brockhoff, P.B., and Christensen, R.H.B. (2016). lmerTest:
11. MacNulty, D.R., Tallian, A., Stahler, D.R., and Smith, D.W. (2014). tests in linear mixed effects models. https://cran.r-project.org/web/
Influence of group size on the success of wolves hunting bison. PLoS packages/lmerTest/index.html
ONE 9, e112884. 27. Range, F., and Virányi, Z. (2013). Social learning from humans or conspe-
12. Mech, L.D., and Boitani, L. (2003). Wolves: Behavior, Ecology, and cifics: differences and similarities between wolves and dogs. Front.
Conservation, Second Edition (Chicago: University of Chicago Press). Psychol. 4, 1–10.
13. Clutton-Brock, J. (1996). Origins of the dog: domestication and early his- 28. Gammell, M.P., and de Vries, H. (2003). David’s score: a more appropriate
tory. In The Domestic Dog: Its Evolution, Behaviour and Interactions with dominance ranking method than Clutton-Brock et al.’s index. Anim.
People, J. Serpell, ed. (Cambridge University Press), pp. 7–20. Behav. 66, 601–605.
Further information and requests for resources should be directed to and will be fulfilled by the Lead Contact, Jennifer L. Essler
(jennifer.essler@vetmeduni.ac.at).
SUBJECT DETAILS
Ten dogs and nine timber wolves were tested at the Wolf Science Center (WSC) in Ernstbrunn, Austria. All animals except those dogs
in 2014 were brought to the center at ten days of age and hand-raised by professional experimenters, spending all of their time in the
presence of a human from the age of ten days to four months. After four months, individuals were introduced into previously estab-
lished packs of adult wolves or dogs, respectively. Dogs born in 2014 were born at the center, and spent at least four hours per day
with a trainer and other puppies (without the mother) before being left in their pack full-time at the age of four months (see [27] for
further information on hand-raising methods). Dogs and wolves are kept in a similar way and have the same life experiences in order
to compare the two species without major differences in life experience. This study was discussed and approved by the institutional
ethics and animal welfare committee at the University of Veterinary Medicine Vienna in accordance with Good Scientific Practice
guidelines and national legislation (Unser Zeichen #10/09/97/2013).
Subject Partner Subject Sex Partner Sex Pack Birth Year LVR HVR
Wolves
Kaspar Aragorn M M 1 2008 Kibble Meat
Aragorn Kaspar M M 1 2008 Kibble Meat
Chitto Shima M F 1 2012 Kibble Meat
Shima Chitto F M 1 2008 Kibble Meat
Amarok Kenai M M 2 2012 Kibble Meat
Kenai Amarok M M 2 2011 Kibble Meat
Nanuk Una M F 3 2009 Kibble Meat
Geronimo Yukon M F 4 2009 Kibble Meat
Wamblee Yukon M F 4 2012 Kibble Meat
(Continued on next page)
METHOD DETAILS
Pre-Testing
We carried out a food preference test to establish a low- and high-value (LVR and HVR, respectively) reward to be used during testing.
A reward was considered to be of high value if they chose the HVR over the LVR at least 85% of the time, in two consecutive sessions
on different days. To be used in the test, the LVR had to be consumed by the animals at least ten times in a row.
Training
Prior to the test, all animals were trained to place their paw on a buzzer placed on a wooden apparatus (Figure S1) to obtain a reward
in response to an experimenter pointing at the apparatus/buzzer and giving the command ‘press.’ The buzzer was then moved to the
full-apparatus setup, and animals were considered ready for testing when they pressed the buzzer with their paw, in response to an
experimenter pointing and giving the command ‘press,’ ten times in a row. Reward used in the training were not used in the testing.
Experimental Setup
Individuals were placed in two familiar adjacent enclosures (7.3 m2 each). These enclosures were made with large wire mesh. On the
outside of each enclosure, one larger hole (18 cm x 18 cm) was cut into the fence for the buzzer (Eaton FAK-S/KC11/I) to be moved in
and out. Two buzzers were positioned each on a wooden table, so that the action of pressing the buzzer required the animals to lift a
paw upward in order to complete the task. Behind each table, but further obscured from the vision of the animals by opaque panels,
was one helper person per table who inserted and removed the buzzer on each trial. Between the two tables, and fully visible to the
animals, was the experimenter. Food reward for the test were placed in a bowl in front of the experimenter, where half the bowl held
the HVR and the other half held the LVR. Both food reward were present and visible in every condition, and the HVR was always closer
to the subjects than the LVR (see Figure S2). Outside of each enclosure was a camera to film every condition for later coding.
Each subject participated in six conditions across separate sessions on separate days (see Table 1). To test whether subjects
respond to receiving a different reward from their partner, we compared the equity baseline condition (ET), where both individuals
received the LVR, to the quality inequity (QI) condition (subject: LVR, partner: HVR) and the reward inequity (RI) condition (subject:
nothing, partner: HVR). To ensure that any differences in the previous tests were not due to a frustration effect of simply seeing
but not receiving the HVR, we further compared the ET baseline to a food control (FC) test, where both individuals were shown
the HVR but then given the LVR for completing the task. Further, to determine if any differences found between the ET and RI con-
dition were indeed due to not receiving a reward in the presence of a rewarded partner, and not simply due to not receiving any
reward (regardless of the partner), we compared the RI condition to a no reward (NR) solo condition, where the subject was tested
alone and received no reward, while the HVR was shown to the partner’s empty enclosure (controlling for the movement of the food).
In practice, the NR condition was the same as the RI condition, except that there was no animal in the partner’s enclosure.
Test Procedure
For each session for the buzzer task, individuals were placed in the two adjacent enclosures, with the sides randomly assigned. The
experimenter kneeled in-between the tables, with the bowl of food reward situated in front of her. Two helpers, one behind each table,
were responsible for moving the buzzers in and out of the enclosures for each trial. To start a session, the partner’s buzzer was moved
into his/her enclosure, while simultaneously the experimenter pointed to the buzzer and said ‘‘press.’’ Once the individual pressed the
buzzer, the experimenter picked up the food reward for that condition and held it between the two enclosures so both animals had the
opportunity to see the reward. The experimenter then gave the food to the partner animal. This was then repeated for the subject
animal, then continued back and forth until the session was over, with either 60 trials being reached (30 per animal) or one animal
refusing to continue.
Behavioral Coding
Each condition, both during the Buzzer Test and afterward in the Interaction Test, was recoded with video cameras (Sony HDR-
CX220E Camcorder) and later coded in Observer XT 10.5 (Noldus Information Technologies). Behaviors coded included buzzer
presses, number of commands given and stress behaviors in the Buzzer Test, and time spent near partner/experimenter in the Inter-
action Test (see Table S1). 20% of the videos were coded by an independent outside individual (all behaviors above k = 0.85).
Social Relationships
Behavioral observations from the WSC were used to assess the social relationships for the analyses. These daily observations are
done by long-term students at the WSC who have previously shown interrater-reliability with established observers at the lab.
Observations are done when the animals are in their home enclosures with their full pack, where no individual is missing. Each obser-
vation is a ten minute focal on one individual where their social behaviors with their pack members are recorded by instantaneous
sampling (see Table S2). Data from these focals were used to calculate a David Score [28] for each individual and then subtracted
individual A’s score from individual B’s in order to obtain a value of rank distance for each pair.
Statistical Analyses
For all models in this study, we used a p value reduction method where we removed the least-significant variable from each model,
fully reducing it until only significant factors were left. Variables are listed in both the order they were removed (first to last) and with the
values they had in the model when they were removed. ‘‘FINAL’’ under ‘Order Removed’ denotes that the following variables were
present in the final model for that particular model. In all models, ‘‘:’’ between two variables denotes an interaction in the model. See
Data S1, Data S2.
Generalized linear mixed models (GLMMs) were run for the Buzzer Task using the total number of trials completed as the response
variable, and condition, rank distance, and species as predictor variables, with species by condition and by rank distance, run as
possible interactions. Additionally, GLMMs were run for both the number of commands (controlling for test duration) and the rate
of stress behaviors (together, whine, yawn, and lips licking). For the Interaction Test behaviors, we ran Linear Mixed Models
(LMMs) for time spent with experimenter/latency to experimenter and time spent with partner/latency to partner (where applicable)
as response variables and with the same predictor variables as in the Buzzer Task models. Analyses were run in R version 3.2.2 [21]
with the packages MASS [22], lme4 [23], ggplot2 [24], car [25], lmerTest [26].
DATA AVAILABILITY
Supplemental Information
Figure S2. Overhead view of the buzzer task setup, Related to Figure 1.
Behavior Definition
Buzzer Test Behaviors
Press/Trial Completed Actor (A) lifts paw up and places paw on buzzer
Experimenter command Experimenter gives A a command, where command is either “come,” “A
name”, or “press”
In sight A is in sight of the camera
Out of sight A is out of sight of the camera
Lips Licking A moves tongue outside of mouth and runs along the side of the mouth or
nose (not coded within 2s of receiving food)
Offensive Behavior A gives an offensive behaviour (growling, barking, snapping) directed
through the fence at R (receiver)
Whine A makes a low cry
Yawn A opens mouth with a long inhalation
Muzzle bite A places mouth around R’s muzzle
Interaction Test Behaviors
Time to partner Latency from when A is let into the large enclosure to when A is within 1
meter of R
Time with partner A is within 1 meter of R
Time to experimenter Latency from when A is let out into the large enclosure to when A is within
1 meter of the experimenter
Time with experimenter A is within 1 meter of the experimenter
Table S1. Behaviours coded in both the Buzzer Test and the Interaction Test, Related to STAR Methods.
Behavior Definition
Dominance Behavior
Stand tall Subject straightens up to full height, with a rigid posture and tail, may include
raised hackles, ears erect and tail perpendicular or above the back.
Paw on To place one or both forepaws on the other’s back.
Ride up To mount another one from behind or from the side, exhibiting a thrusting
motion with the hips.
Head on The subject approaches another’s shoulder/back and puts its heat on it.
Muzzle Bite To graph the muzzle of another subject either softly or with enough pressure to
make the other whimper.
Approach dominant To approach another subject within one body lenght for at least 5 seconds, with
the tail perpendicular or above the plane of the back and the ears erect and
pointed forward.
Submissive Behavior
Crouch Lowering the head, sometimes bending the legs, arching the back, lowering the
tail between the hindlegs, and avoiding eye-contact.
Passive submission To lie on the back showing the stomach and holding the tail between the legs.
The ears are held back and close to the head and the subject raises a hind leg for
inguinal presentation.
Active submission The subject has its tail tucked between the hind legs sometimes wagging it while
he is in a crouched position (with hindquarters lowered) and may attempt to paw
and lick the side of actors’/aggressor’s muzzle. The behavior may also include
urination
Withdrawing The subject withdraws from another moving away slowly in the opposite
direction, displaying a submissive posture. It occurs when a subject has been
threatened or attacked by another, or a fight has taken place.
Flee To run away from another with tail tucked between the legs and body ducked. It
occurs when a subject has been threatened or attacked by another, or after a fight.
Avoidance In response to another reducing the distance towards it, the subject moves away
displaying a submissive posture. The subject may also look at the individual he
is trying to avoid.
Approach submissive To slowly approach another within one body lenght remaining within that
distance for at least 5 seconds. The approach is characterized by a ducked
posture and tail between the legs. Subject can also be moving in a wavy line
and in a hesitant (stop-start) manner.
Table S2 – Behaviors from WSC Ethogram used to calculate the David’s Score and rank distance for dyads