Operationalizing Intentionality in Primate Communication: Social and Ecological Considerations

Abstract

An intentional transfer of information is central to human communication. When comparing nonhuman primate communication systems to language, a critical challenge is to determine whether a signal is used in intentional, goal-oriented ways. As it is not possible to directly observe psychological states in any species, comparative researchers have inferred intentionality via behavioral markers derived from studies on prelinguistic human children. Recent efforts to increase consistency in nonhuman primate communication studies undervalue the effect of possible sources of bias: some behavioral markers are not generalizable across certain signal types (gestures, vocalizations, and facial expressions), contexts, settings, and species. Despite laudable attempts to operationalize first-order intentionality across signal types, a true “multimodal” approach requires integration across their sensory components (visual-silent, contact, audible), as a signal from a certain type can comprise more than one sensory component. Here we discuss how the study of intentional communication in nonlinguistic systems is hampered by issues of reliability, validity, consistency, and generalizability. We then highlight future research avenues that may help to understand the use of goal-oriented communication by opting, whenever possible, for reliable, valid, and consistent behavioral markers, but also taking into account sampling biases and integrating detailed observations of intraspecific communicative interactions.

Social tolerance and interactional opportunities as drivers of gestural redoings in orang-utans

“…To focus on possible cases of repetition and elaboration as forms of gestural redoings, and thus exclude cases of ‘rapid-fire’ sequences that are not contingent on a response [27], our analyses considered only those instances that were preceded by a gesture of the same signaller toward the same recipient with the same presumed goal, but not followed by an apparently satisfactory outcome, and omitted cases for which a response followed immediately. We refrained from applying a specific inter-event interval for this distinction because previous studies diverge regarding a temporal pause that would reflect biologically meaningful ‘response waiting’ in nonhuman primates [3,20]. To ensure inter-observer reliability, we evaluated the coding performance of all observers using the Cohen's Kappa coefficient [48].…”

“…When the response does not or only partly matches the signaller's goal, the latter may repeat the same signal ( repetition ), or modify the type of the signal ( elaboration ) if the apparent goal has not been met [18,19]. These two behaviour patterns have thus become well-established markers of first-order intentionality in nonhuman species [3,17,20], and are thought to constitute an early form of what, more strongly developed, is the typical communicative repair we see in every-day interactions of humans [21]. Repair has been defined as a process by which communicative breakdown is fixed either by own initiative (i.e.…”

“…Persistence and elaboration may thus qualify as simple forms of self-initiated repair (see [21]). The term persistence has sometimes been used to refer to all redoings or acts of communicative perseverance irrespective of whether they were a repetition or elaboration of the initial signal [20], but others adopted a more narrow definition of persistence as repetitions of the same signal [3,19]. To avoid any confusion, we here refer to all iterated uses of the same signal as repetitions , to create a clear contrast to elaboration.…”

“…To eventually receive the desired outcome, great apes are thus capable of ‘repairing’ their communication attempts contingent on their interaction partner's response to their initial signal(s). While elaboration in signalling occurs less frequently than mere repetition [26], it may provide even stronger evidence for communicative redoings involving the monitoring of receiver response, since the change of the signal type seems to require more voluntary control and is thus less likely to be exclusively emotionally driven [3,20,30].…”

Coordinating social action: a primer for the cross-species investigation of communicative repair

“…Although elaboration of signalling occurs less frequently than persistence, it constitutes more solid evidence of other-monitoring (and flexibility in signalling), since the change of the signal type seems to require more voluntary control and is less likely to be purely emotionally driven (e.g. in chimpanzees, repetitions of a ‘pant-grunt’ from a subordinate individual could be explained by high arousal levels, but if the signaller ‘presents the genitals' to the presumed recipient after the ‘pant-grunt’, this may indicate a higher level of cognitive control [2,102,103]).…”