Hirata / Poster

Naive chimpanzees' timing of observation of experienced conspecifics in a honey-fishing task

Satoshi Hirata, Primate Research Institute, Kyoto University and Naruki Morimura, Great Ape Research Institute, Hayashibara Museum of Natural Sciences

     Chimpanzees are known to be good at observational learning of tool use or object manipulation. Laboratory studies have revealed that chimpanzees are capable of imitation or emulation when experimentally exposed to human models. But do chimpanzees really spontaneously come to observe a model in a more naturalistic situation? Do they know when to observe a model? In order to learn how to use a new tool efficiently, they are expected to spontaneously observe a model before their own first attempt or after their own failure. The observation of a model will not occur randomly but will occur depending on the result of their preceding attempts. In order to answer these questions, we investigated the occurrence of naive chimpanzees' spontaneous observation of experienced conspecifics during learning of tool use for a honey-fishing task in which subjects should use a short, slender and flexible tools into a hole to obtain honey inside a bottle. The subjects were presented with twenty kinds of objects, of which eight kinds were "unusable tools" that could not be inserted into the hole. The remaining twelve were "usable tools" of which a certain part could be inserted into the hole. Six pairs of naive and experienced chimpanzees were brought to this honey-fishing situation. The results showed that the naEe individuals spontaneously went to observe the experienced conspecifics when they were unskilled at using tools. The naive subjects frequently observed the partner before their first attempts or after their own failure. Two naive individuals picked up the same tool as the partners' choice after their virtual first observation. In addition, there were some cases in which naive individuals utilized the left-over tools of the experienced, and one of which was active "robbing" of the other's tool. The transmission of tool use was accomplished by active observing of conspecifics in an appropriate timing for observational learning and by utilizing the experienced partners' efficient tools.