doi:10.1016/j.anbehav.2006.04.007 
Copyright © 2006 The Association for the Study of Animal Behaviour Published by Elsevier Ltd.
Development of tool use in New Caledonian crows: inherited action patterns and social influences
Ben Kenward1, a, Christian Rutza, Alex A.S. Weira and Alex Kacelnik
, a, 
aBehavioural Ecology Research Group, Department of Zoology, University of Oxford, U.K
Received 10 August 2005;
revised 8 November 2005;
accepted 10 April 2006.
MS. number: 8654.
Available online 18 October 2006.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
New Caledonian crows, Corvus moneduloides, are the most advanced avian tool makers and tool users. We previously reported that captive-bred isolated New Caledonian crows spontaneously use twig tools and cut tools out of Pandanus spp. tree leaves, an activity possibly under cultural influence in the wild. However, what aspects of these behaviours are inherited and how they interact with individual and social experience remained unknown. To examine the interaction between inherited traits, individual learning and social transmission, we observed the ontogeny of twig tool use in hand-reared juveniles. Successful food retrieval was preceded by stereotyped object manipulation action patterns that resembled components of the mature behaviour, demonstrating that tool-oriented behaviours in this species are an evolved specialization. However, there was also an effect of social learning: juveniles that had received demonstrations of twig tool use by their human foster parent showed higher levels of handling and insertion of twigs than did their naïve counterparts; a choice experiment showed that they preferred to handle objects that they had seen being manipulated by their human foster parent. Our observations are consistent with the hypothesis that individual learning, cultural transmission and creative problem solving all contribute to the acquisition of the tool-oriented behaviours in the wild, but inherited species-typical action patterns have a greater role than has been recognized.
Figure 1. Photographs of (a) a block hole and (b) a crevice platform, both with a twig being inserted by Uék.
Figure 2. Mean ± SE percentage of time spent in precursor actions (□) and insertion (■) by four juvenile New Caledonian crows as a function of age. Asterisk indicates the week when successful food retrieval was first observed. To show general development, tutored and untutored birds were pooled. Data points in each week are slightly offset to avoid overlap. See Methods for explanation of ‘branching’.
Figure 3. Mean ± SE percentage of time spent in precursor actions shown by four juvenile New Caledonian crows as a function of age, showing rubbing (○), protoprobing (■), poking (●) and wrong-angle probing (□). Data pooled as in Fig. 2. Data points in each week are slightly offset to avoid overlap. See Methods for explanation of ‘branching’.
Figure 4. Mean ± SE percentage of time spent in different activities by four juvenile New Caledonian crows, showing locomotion (■), object touching excluding carrying (○) and object carrying (□). Data points in each week are slightly offset to avoid overlap. See Methods for explanation of ‘branching’.
Figure 5. Object categories touched by four juvenile New Caledonian crows. Categories are self-explanatory or defined as follows. Hole: the lip or cavity of any natural or artificial holes and crevices in the aviary. Perch: any wooden part of the aviary fixtures. Aviary fixture: any part of the aviary not covered by other categories. Other portable: any nonfood, nontwig item that the bird could carry (e.g. toys). See Methods for explanation of ‘branching’.
Figure 6. Mean ± SE percentage of time spent in different activities by four juvenile tutored and untutored New Caledonian crows. Solid lines: tutored group, Uék (●) and Nalik (■); dashed lines: untutored group, Oiseau (○) and Corbeau (□). Inset panels show the raw data and model fit for the two groups. Dependent variables and significant predictors are as follows. (a) Carrying twigs: Age, Group, Age*Individual, Age*Group. (b) Inserting twigs: Age, Group, Age*Individual, Age*Group. (c) Carrying nonfood, nontwig items: Age, Age*Individual. (d) Inserting nonfood, nontwig items: Age, Age*Group. (e) Inserting food: Age, Age*Group. (f) Locomotion: Age, Individual, Age*Individual. See Methods for explanation of ‘branching’.
Figure 7. Juvenile New Caledonian crow object preference after witnessing human foster parent manipulating one of a pair of novel objects. The categories are the phases within a trial, in sequence. Bars indicate how often the subject manipulated demonstration target first (■), and most often (□). Dashed line indicates results expected under the null hypothesis of no effect of demonstration. N = 32 (16 trials each for Uék and Nalik, pooled). P values are calculated against the chance binomial distribution.
Table 1.
Ethogram definitions of object-oriented behaviours in juvenile New Caledonian crows
* Possible only during carrying.
Table 2.
Effects of age, treatment group and individual difference on behaviours in juvenile New Caledonian crows
Significant P values shown in bold. See Methods for GLM details and Fig. 6 for visualization.
Abstract
Purchase PDF (584 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (7)
Purchase PDF (447 K)
