Metarepresentation

Cognitive systems are characterized by their ability to construct and process representations of objects and states of affairs. MENTAL REPRESENTATIONS and public representations such as linguistic utterances are themselves objects in the world, and therefore potential objects of second-order representations, or "metarepresentations." Under this or another name, (e.g., "higher-order representations"), metarepresentations are evoked in evolutionary approaches to INTELLIGENCE, in philosophical and developmental approaches to commonsense psychology, in pragmatic approaches to communication, in theories of consciousness, and in the study of reasoning.

It is assumed that the members of most animal species are incapable of recognizing in themselves or attributing to conspecifics mental representations such as beliefs or desires: they utterly lack metarepresentational abilities. Highly intelligent social animals such as primates, on the other hand, are believed to have evolved an ability to interpret and predict the behavior of others by recognizing their mental states. Indeed, Dennett (1987) has described some primates as "second-order intentional systems," capable of having "beliefs and desires about beliefs and desires." Second-order intentional systems are, for instance, capable of deliberate deception. In a population of second-order intentional systems, a third-order intentional system would be at a real advantage, if only because it would be able to see through deception. Similarly, in a population of third-order intentional systems, a fourth-order intentional system would be a greater advantage still, with greater abilities to deceive others and avoid being deceived itself, and so on. Hence the hypothesis, supported by some ethological evidence, that primates have developed a kind of strategic MACHIAVELLIAN INTELLIGENCE (Byrne and Whiten 1988) involving higher-order metarepresentational abilities. These evolutionary and ethological arguments have in part converged, in part conflicted with experimental studies of primates' metarepresentational abilities that have started with Premack and Woodruff's pioneering article "Does the Chimpanzee Have a Theory of Mind?" (1978). Though the level of metarepresentational sophistication of other primates is still disputed, that of human beings is not. The human lineage may be the only one in which a true escalation of metarepresentational abilities has taken place.

Humans are all spontaneous psychologists. They have some understanding of cognitive functions such as perception and MEMORY (see METACOGNITION). They also attribute to one another PROPOSITIONAL ATTITUDES such as beliefs and desires, and do so as a matter of course. While philosophers have described the basic tenets of this commonsense or FOLK PSYCHOLOGY and discussed its empirical adequacy, psychologists have focused on the development of this cognitive ability, often described as a THEORY OF MIND. Philosophers and psychologists have been jointly involved in discussing the mechanism through which humans succeed in metarepresenting other people's thoughts and their own. This investigation has taken the form of a debate between those who believe attribution of mental states to others is done by simulation (e.g., Goldman 1993; Gordon 1986; Harris 1989), and those who believe it is done by inference from principles and evidence (e.g., Gopnik 1993; Leslie 1987; Perner 1991; Wellman 1990). In this debate (see SIMULATION VS. THEORY-THEORY), much attention has been paid to different degrees of metarepresentational competence that may be involved in attributing mental states to others. In particular, the ability to attribute false beliefs is seen as a sufficient, if not necessary, proof of basic metarepresentational competence. This metarepresentational competence can be impaired -- the basis of a new, cognitive approach to AUTISM. Conversely, the study of autism has contributed to the development of a finer-grained understanding of metarepresentations (see Baron-Cohen 1995; Frith 1989).

Cognitive approaches have stressed the metarepresentational complexity of human communication. The very act of communicating involves, on the part of the communicator and addressee, mutual metarepresentations of each other's mental states. In ordinary circumstances, the addressee of a speech act is interested in the linguistic MEANING of the utterance only as a means of discovering the speaker's meaning. Speaker's meaning has been analyzed by the philosopher Paul GRICE (1989) in terms of several layers of metarepresentational intentions, in particular the basic metarepresentational intention to cause in the addressee a certain mental state (e.g., a belief), and the higher-order metarepresentational intention to have that basic intention recognized by the addressee. Grice's analysis of metarepresentational intentions involved in communication has been discussed and developed by numerous philosophers and linguists (e.g., Bach and Harnish 1979; Bennett 1976; Recanati 1986; Schiffer 1972; Searle 1969; Sperber and Wilson 1986).

It has long been observed that human languages have the semantic and syntactic resources to serve as metalanguages. In direct and indirect quotations, utterances and meanings are metarepresented. The study of such metalinguistic devices has been developed in semiotics (see SEMIOTICS AND COGNITION), in philosophy of language, and in PRAGMATICS. In particular, in the study of FIGURATIVE LANGUAGE, irony has been described as a means of distancing oneself from some propositional attitude by metarepresenting it (see Gibbs 1994; Sperber and Wilson 1986).

The ability to metarepresent one's own mental states plays an important role in CONSCIOUSNESS, and may even be seen as defining it. For David Rosenthal (1986; 1997) in particular, a mental state is conscious if it is represented in a higher-order thought. When a thought itself is conscious, then the higher-order thought that represents it is a straightforward metarepresentation. These higher-order thoughts may themselves be the object of yet higher-order thoughts: The reflexive character of consciousness (i.e., that one can be conscious of being conscious) is then explained in terms of a hierarchy of metarepresentations. While many philosophers do not accept this "higher-order thought" theory of consciousness, the role of metarepresentations at least in aspects of consciousness and in related phenomena such as INTROSPECTION is hardly controversial.

Much of spontaneous human reasoning is about states of affairs and how they relate to one another. But some reasoning, especially deliberate reasoning as occurs in science or philosophy, is about hypotheses, theories, or claims -- representations -- and only indirectly about the state of affairs represented in these representations. In the psychology of DEDUCTIVE REASONING, growing attention has been paid to such metarepresentational reasoning, in particular by experimenting with liar and truth teller problems, either from the point of view of "mental logic" (Rips 1994) or from that of MENTAL MODELS (Johnson-Laird and Byrne 1991). In artificial intelligence, too, there is a growing interest in modeling such METAREASONING.

This rapid overview does not exhaust the areas of cognitive science where metarepresentation (whether so named or not) plays an important role. In a great variety of cognitive activities, humans exhibit unique metarepresentational virtuosity. This, together with the possession of language, may be their most distinctive cognitive trait.

References

Bach, K. and R. Harnish. (1979). Linguistic Communication and Speech Acts. Cambridge, MA: MIT Press.

Baron-Cohen, S. (1995). Mindblindness: An Essay on Autism and Theory of Mind. Cambridge, MA: MIT Press.

Bennett, J. (1976). Linguistic Behaviour. Cambridge: Cambridge University Press.

Byrne, R. W., and A. Whiten. (1988). Machiavellian Intelligence: Social Expertise and the Evolution of Intellect in Monkeys, Apes and Humans. Oxford: Oxford University Press.

Dennett, D. (1987). The Intentional Stance. Cambridge, MA: MIT Press.

Frith, U. (1989). Autism: Explaining the Enigma. Oxford: Blackwell.

Gibbs, R. (1994). The Poetics of Mind: Figurative Thought, Language and Understanding. Cambridge: Cambridge University Press.

Goldman, A. (1993). The psychology of folk psychology. Behavioral and Brain Sciences 16:15-28.

Gopnik, A. (1993). How we know our minds: The illusion of first-person knowledge of intentionality. Behavioral and Brain Sciences 16:1-14.

Gordon, R. M. (1986). Folk psychology as simulation. Mind and Language 1:158-171.

Grice, H. P. (1989). Studies in the Way of Words. Cambridge, MA: Harvard University Press.

Harris, P. L. (1989). Children and Emotion: The Development of Psychological Understanding. Oxford: Blackwell.

Johnson-Laird, P. N., and R. M. J. Byrne. (1991). Deduction. Hove: Erlbaum.

Leslie, A. M. (1987). Pretence and representation: The origins of "theory of mind." Psychological Review 94:412-426.

Perner, J. (1991). Understanding the Representational Mind. Cambridge, MA: MIT Press.

Premack, D., and G. Woodruff. (1978). Does the chimpanzee have a theory of mind? Behavioral and Brain Sciences 1:515-526.

Recanati, F. (1986). On defining communicative intentions. Mind and Language 1(3):213-242.

Rips, L. (1994). The Psychology of Proof. Cambridge, MA: MIT Press.

Rosenthal, D. M. (1986). Two concepts of consciousness. Philosophical Studies 49(3):329-359.

Rosenthal, D. M. (1997). A theory of consciousness. In N. Block, O. Flanagan, and G. Güzeldere, Eds., The Nature of Consciousness: Philosophical Debates. Cambridge, MA: MIT Press, pp. 729-753.

Schiffer, S. (1972). Meaning. Oxford: Clarendon Press.

Searle, J. (1969). Speech Acts. Cambridge: Cambridge University Press.

Sperber, D., and D. Wilson. (1986). Relevance: Communication and Cognition. Oxford: Blackwell.

Wellman, H. M. (1990). The Child's Theory of Mind. Cambridge, MA: MIT Press.

Whiten, A. (1991). Natural Theories of Mind: Evolution, Development and Simulation of Everyday Mindreading. Oxford: Blackwell.

Metarepresentation

See also

References

Further Readings