The role of acoustic templates, contextual information and visual representation in the combinatorial communication system of chestnut-crowned babblers

Abstract

Many non-human animals produce vocalisations via the combination of multiple acoustic elements into larger sequences, a property referred to as combinatoriality. Chestnut crowned babblers (Pomatostomus ruficeps) are the only known species other than humans to use phonemic contrasts, i.e. the rearrangement of meaningless acoustic elements into meaningful calls, in the generation of vocal signals. Identifying the similarities and differences in phoneme use between humans and babblers, therefore, can potentially grant valuable insight into the evolution of this linguistic ability in humans. The primary goal of this thesis is to understand the cognitive mechanisms that underlie the reception of vocalisations generated via phonemic contrasts in chestnut crowned babblers. Specifically, I address the following research questions: 1) whether babblers are able to respond appropriately to vocal signals even when the constituent acoustic sub-units have been rearranged; 2) whether individuals are capable of integrating previously acquired contextual information in order to produce appropriate behavioural responses to ambiguous signals; and 3) whether babblers generate a visual mental image of signal referents upon reception of context-specific acoustic signals. I find that subjects respond appropriately to vocal stimuli even when the acoustic sub-units have been rearranged to differ from any existing call template, but only when the modified signal exhibits a high degree of acoustic similarity to the natural call. This suggests that babblers, as opposed to humans, possess flexible acoustic templates for their phoneme-based vocalisations. I also observe that subjects respond appropriately to ambiguous signals if they have been previously exposed to information which may disambiguate the meaning of future, uncertain stimuli. This extends the evidence for the integration of contextual information to an avian species exhibiting a combinatorial repertoire. Finally, I do not find evidence in support of the hypothesis that babblers generate visual mental images of signal referents upon reception of context-specific calls. Thus it remains as of yet unclear whether babblers respond to calls via a process of affect-conditioning, associative learning or conceptual semanticity. The findings of this thesis contribute to the field of comparative and combinatorial communication by describing some of the cognitive mechanisms which underlie the reception of phoneme-based vocalisations in an avian species.