Presentation Details
| Social Learning in Dogs (Canis familiaris): A systematic review and meta- analysis Anya E Parks1, Francys Subiaul1, Jeffery Stevens2, Colleen Buckley3, Yasamin Buckley, Colleen- Saberzadeh Ardest1. 1George Washington University, Washington, District of Columbia, USA.2University of Nebraska-Lincoln, Lincoln, Nebraska, USA.3Canine Companions for Independence, Santa Rosa, California, USA |
Abstract
The dog, Canis familiaris, has a long co-evolutionary relationship with humans.
This history has resulted in dogs exhibiting social cognitive abilities originally
thought to be exclusive to our species. Directly quantifying
social learning in dogs is an important step in contextualizing their utility as models for social learning. Hence the value of our meta-analysis studying social learning in dogs. To better assess null results and to average across unknown parameters, we used hierarchical, robust
Bayesian modeling. Using the PRISMA method, we returned 31 studies, including multiple
experiments from each study. Bayesian hierarchical modeling returned an effect size of Cohen’s d, 0.53, 95% CI [0.22,0.83] collapsing across all forms of social learning, with a bayes factor of 75. Goal emulation had an effect size of Cohen's d = 0.785, 95% CI [0.481, 1.094]. Novel imitation had an effect size of Cohen’s d = 0.35, 95% CI [0.15, 0.53], and familiar imitation had a Cohen's d of 0.87, 95% CI [0.468, 1.278]. When combined imitation had an effect size of Cohen's d = 0.52, 95%
CI [0.23, 0.79]."
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
This history has resulted in dogs exhibiting social cognitive abilities originally
thought to be exclusive to our species. Directly quantifying
social learning in dogs is an important step in contextualizing their utility as models for social learning. Hence the value of our meta-analysis studying social learning in dogs. To better assess null results and to average across unknown parameters, we used hierarchical, robust
Bayesian modeling. Using the PRISMA method, we returned 31 studies, including multiple
experiments from each study. Bayesian hierarchical modeling returned an effect size of Cohen’s d, 0.53, 95% CI [0.22,0.83] collapsing across all forms of social learning, with a bayes factor of 75. Goal emulation had an effect size of Cohen's d = 0.785, 95% CI [0.481, 1.094]. Novel imitation had an effect size of Cohen’s d = 0.35, 95% CI [0.15, 0.53], and familiar imitation had a Cohen's d of 0.87, 95% CI [0.468, 1.278]. When combined imitation had an effect size of Cohen's d = 0.52, 95%
CI [0.23, 0.79]."
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
Content Locked. Log into a registered attendee account to access this presentation.