Matching-to-Sample as a Model of Symbolic Behavior for Bio-behavioral Investigations

Abstract

This paper discusses the use of matching-to-sample as a procedure to investigate symbolic behavior acquisition when studying the evolution of language. It is argued that generalized identity and stimulus classes formed after learning arbitrary interconnected conditional stimulus relationships may be the result of basic pre-symbolic processes. Such aspects may thus provide relevant information for interdisciplinary research on symbolic behavior and language evolution. An individualized approach when building matching-to-sample repertoires - different from that of traditional training - is described. As an example, the results of a generalized identity matching-to-sample procedure held with an experienced capuchin monkey are presented.