Abstract # 64:

Scheduled for Saturday, September 19, 2009 06:30 PM-09:00 PM: Session 5 (Mission Bay Ballroom CDE) Poster Presentation


C. F. Martin1, B. Dora2 and T. Matsuzawa1
1Primate Research Insistute of Kyoto University, Kanrin, Inuyama City, Aichi 484-8506, Japan, 2Department of Zoology, University of Oxford

This study examined the ability of a chimpanzee (Pan troglodytes) named Ai to make symbolic associations based on a conspecific’s actions in a matching-to-sample task. A model chimpanzee touched a stimulus on a touch-panel, and Ai observed the sample stimulus being touched before matching it on a second panel. Two task types were given: in the first, the sample was presented alone on the model’s panel, and Ai could determine it without paying attention to the model’s touch. In the second, the sample appeared alongside a distracter stimulus, and Ai had to use the model’s touch as cue to determine the sample. Two matching conditions were presented: identity matching, where the set of stimuli (colors or Japanese kanji characters) was the same throughout the task, and symbolic matching, in which Ai matched the sample to a symbolically related stimulus. Ai performed above chance level both tasks [Binomial p<0.001]. Her performance was significantly lower when use of a social cue was required versus not required, and significantly lower for symbolic matching versus identity matching [two-way ANOVA α=0.05]. There was near significant interaction between the task type and symbolic versus identity matching [F(1,5)=4.60, p=0.09], showing that the compound effect of using social cues while making symbolic associations was greater than the sum of their individual effects on performance.