Gestural communication in a new world parrot

Intentional communication: solving methodological issues to assigning first-order intentional signalling

Intentional signalling plays a fundamental role in human communication. Mapping the taxonomic distribution of comparable capacities may thus shed light on the selective pressures that enabled the evolution of human communication. Nonetheless, severe methodological issues undermine comparisons among studies, species and communicative modalities. Here, we discuss three main obstacles that hinder comparative research of 'first-order' intentional signalling (i.e. voluntary signalling in pursuit of a cognitively represented goal): (i) inconsistency in how behavioural hallmarks are defined and operationalised, (ii) testing of behavioural hallmarks without statistical comparison to control conditions, and (iii) bias against the publication of negative results. To address these obstacles, we present a four-step scheme with 20 statistical operational criteria to distinguish between non-intentional and first-order intentional signalling. Our unified scheme applies to visual and audible signals, thereby validating comparison across communicative modalities and species. This, in turn, promotes the generation and testing of hypotheses about the evolution of intentional communication.

Bonobos modify communication signals according to recipient familiarity

Human and nonhuman primate communication differs in various ways. In particular, humans base communicative efforts on mutual knowledge and conventions shared between interlocutors. In this study, we experimentally tested whether bonobos (Pan paniscus), a close relative to humans, are able to take into account the familiarity, i.e. the shared interaction history, when communicating with a human partner. In five experimental conditions we found that subjects took the recipients' attentional state and their own communicative effectiveness into account by adjusting signal production accordingly. More importantly, in case of communicative failure, subjects repeated previously successful signals more often with a familiar than unfamiliar recipient, with whom they had no previous interactions, and elaborated by switching to new signals more with the unfamiliar than the familiar one, similar to what has previously been found in two year-old children. We discuss these findings in relation to the human capacity to establish common ground between interlocutors, a crucial aspect of human cooperative communication.