Review of Natural theories of mind: Evolution, development and simulation of everyday mindreading.

Abstract

In recent years there has been a phenomenal growth in interest and research directed at what has become known as a Theory of Mind (ToM) and its development. Among the many edited books recently made available on the topic, Whiten's Natural theories of is unique in the eclectic, multidisciplinary approach it brings to this vital, yet fledgling area. As suggested by the subtitle; Evolution, development and simulation of everyday mindreading, Whiten has brought together researchers involved in everything from ethological approaches to mind - reading in primates and birds, to those developing formal mathematical models of planning and plan recognition.This interdisciplinary approach, which also includes a chapter by Carrithers placing the development of a theory of within the broader context of sociology and anthropology, is at the same time both the strength of this volume and it's limitation. strength comes from connecting what to an outsider at times seem like an esoteric and philosophical debate to the modelling of cognitive science, the careful ethological studies of primates, and experimental and descriptive studies of cognitive development in infants and children. necessary and negative corollary of taking such a broad strategy in a 362 page volume is that no single approach is developed in much depth. With the number of other volumes (Astington, Harris & Olson, 1988; Frye & Moore, 1991; Perner, 1991; Wellman, 1990) on theory of and it's development in humans, however, the niche for a volume on developing theories of in children is quite full.A key issue throughout this general area of research and this volume specifically (but only rarely made explicit, e.g., by Premack & Dasser and by Cheney & Seyfarth), is whether asking when (in the case of humans) or if (in other animals) a theory of exists is an empirical question. What constitutes sufficient evidence to conclude an individual has a theory of mind? Several chapters (most notably those by Leslie, Baron - Cohen, Bennett, Wellman, Astington & Gopnik) in this book suggest a number of gradations toward or distinctions (such as between desire - based and belief - based thoughts and actions) that contribute to making hypotheses in theory of testable.These gradations in mindreading also serve to tie in the continuity issues (from both ontogenetic and phylogenetic time frames) that underlie much of this book. (They also bog the reader down somewhat with different terminologies for what appear to be similar, if not identical concepts.) For now, at least, it still comes down to what Chandler, Fritz and Hala (1989) have termed the distinction between 'boosters' -- those willing to attribute a theory of to a child or animal without conclusive evidence -- and 'scoffers' -- those hesitant to attribute such a capability when alternate interpretations are available. distinction seems to be based primarily on: 1) whether one focusses on the experimental evidence (which tends to find evidence of a theory of only at later ages in humans and not at all in other species) or the anecdotal evidence; and 2) what one takes as the most parsimonious explanation of behaviour (if one cares for parsimony at all, cf. Bennett's chapter and below). chapters by Dunn, Reddy and Leekam describing teasing, joking, lying and pretence in children observed under natural conditions argue strongly for abilities yet to be captured in laboratory tasks, while Bennett argues that accepting the most parsimonious explanation of behaviour, at least for human behaviour, may be a flawed strategy. Perhaps the relative parsimony of behaviourist and mindreading explanations for animal behaviour was best expressed in the closing sentences of Premack and Woodruff (1978). The ape could only be a mentalist. Unless we are badly mistaken, he is not intelligent enough to be a behaviourist. (p. 526). (Would we say the same about a 2 - year - old? …