On the relationship between communication and cognition

Some birds and non-human primates give characteristic calls in response to the discovery of food or the approach of a particular type of predator. Companions respond with food-searching behaviour or by adopting appropriate escape responses. Such referential signals are controversial because they challenge the special status of language, which has long been considered the principal exception to the otherwise clear pattern of evolutionary continuity between humans and non-human animals. The belief that language has no clear precursors in the natural behaviour of non-human animals is a vestige of Cartesian dualism quite out of step with the modern synthesis, but it remains the consensus view among linguists and psychologists. Until now, there have been no empirical data that would force us to reject this position. Previous attempts to explore the meaning of referential signals such as food calls and alarm calls have concentrated particularly on social primates. Playback experiments have generated results that, while suggestive, can readily be accommodated by behaviorist models that invoke only the reflexive expression of a particular motor patterns. We have adopted a new approach that relies instead on selectively manipulating an animal's experience with the putative referent. Our results demonstrate for the first time that the response to food calls is mediated by a mental representation of food (ie these sounds stimulate retrieval of specific stored information, which then determines receiver response). Animal signals thus engage cognitive processes like those that underpin the comprehension of language. It is perhaps startling that these data have been obtained with chickens - a species not widely credited with sophisticated cognitive function. This implies that the precursors for language are not unique to our closes phylogenetic relatives; they may instead be widely shared among social vertebrates.