1. Relationship between aerodigestive and affiliative behaviours

Observational studies of human and nonhuman primate (NHP) infants are remarkably consistent in showing that mouth movements originating in ingestion nevertheless have distinct, communicative significance (Trevarthen Reference Trevarthen and Bullowa1979; Van Hooff Reference Van Hooff1962). These include TP in humans and lip smacking (LS) in NHPs, including rhesus macaques. In each case, the behaviour rapidly becomes prominent in early parent-infant interactions (Ferrari et al. Reference Ferrari, Paukner, Ionica and Suomi2009a; Trevarthen Reference Trevarthen1974; Murray et al. Reference Murray, De Pascalis, Bozicevic, Hawkins, Sclafani and Ferrari2016), and is highly organized, systematically co-occurring with other, clearly affiliative behaviors. For instance, in humans, TP in the first two months is associated with smiling, wide mouth-opening, and positive vocalizations – a cluster of expressions termed “prespeech” (Fig. 1; Murray et al. Reference Murray, De Pascalis, Bozicevic, Hawkins, Sclafani and Ferrari2016). These expressions occur during direct gaze to the adult's face, often accompanied by arm waving with open hand movements (Fig. 1a) (Lavelli & Fogel Reference Lavelli and Fogel2002; Reference Lavelli and Fogel2005; Reference Lavelli and Fogel2013; Supplement SI1; Trevarthen Reference Trevarthen1974; Reference Trevarthen and Bullowa1979).

Figure 1a and 1b. Infant tongue protrusion, with and without arm/hand movements, during face-to-face interaction.

In macaques, LS similarly appears as part of social encounters, and co-occurs with direct gaze and proximal contact with the parent; even newborns actively solicit their mothers to interact using LS (Dettmer et al. Reference Dettmer, Kaburu, Simpson, Paukner, Sclafani, Byers, Murphy, Miller, Marquez, Miller, Suomi and Ferrari2016; Ferrari et al. Reference Ferrari, Paukner, Ionica and Suomi2009a). The time course of TP and LS is strikingly similar in humans and monkeys, increasing in frequency over the first few weeks, and then declining with the reduction in face-to-face interactions and the infant's growing exploration (Ferrari et al. Reference Ferrari, Paukner, Ionica and Suomi2009a; Trevarthen & Aitken Reference Trevarthen and Aitken2001). Notably, each behavior is functionally autonomous with respect to digestive chewing – whether in terms of its co-occurrence (TP [Murray et al. Reference Murray, De Pascalis, Bozicevic, Hawkins, Sclafani and Ferrari2016]; [Supplement SI1]) or developmental trajectory (LS [Ghazanfar & Takahashi Reference Ghazanfar and Takahashi2014]).

2. Sensitivity to social context

Infant TP and LS are highly sensitive and responsive to others' interactive behavior. In addition to the consistent evidence from well-conducted studies for neonatal imitation of these gestures (Simpson et al. Reference Simpson, Murray, Paukner and Ferrari2014a), human research using experimental perturbations shows that if normal face-to-face contact is broken by the parent adopting a still, blank face, infants show less positive social engagement (Mesman et al. Reference Mesman, van IJzendoorn and Bakermans-Kranenburg2009), including reduced TP (Murray & Trevarthen Reference Murray, Trevarthen, Field and Fox1985). This effect is not simply a function of lack of parental stimulation lowering infant arousal, because similar reductions in engagement (and TP) occur when infants see their parent in a noncontingent versus identical contingent face-to-face interaction (Murray & Trevarthen Reference Murray, Trevarthen, Field and Fox1985; Nadel et al. Reference Nadel, Carchon, Kervella, Marcelli and Reserbat-Plantey1999). Disturbances in clinical populations also demonstrate the influence of variations in face-to-face interactions on human infants' social responses. For example, 2-month-olds of socially anxious mothers show low levels of social engagement themselves, including TP, an effect that is mediated by their mothers' reduced positive social signals (Murray et al. Reference Murray, Cooper, Creswell, Schofield and Sack2007). Similar effects are found for infants of depressed mothers (Murray et al. Reference Murray, Fiori-Cowley, Hooper and Cooper1996). Parallel findings to the human experimental studies emerge from NHP research: specifically, infant macaques reduce their LS and social attention when presented with a still face versus an active interactive experimenter; and they show more LS and attention when an experimenter interacts with them using contingent, imitative mouth responses rather than similarly prominent, but noncontingent, repetitive mouth movements, despite the latter condition providing more overall stimulation (Sclafani et al. Reference Sclafani, Paukner, Suomi and Ferrari2014).

3. Role of infant TP and LS in later social development

Human observational studies show that parents respond positively to early infant signs of social engagement or “prespeech,” imitating and affirming them, and according them communicative and playful significance (Lavelli & Fogel Reference Lavelli and Fogel2002; Trevarthen Reference Trevarthen and Bullowa1979; Murray et al. Reference Murray, De Pascalis, Bozicevic, Hawkins, Sclafani and Ferrari2016; Supplement SI2). In turn, the further development of these infant behaviours is promoted by parental facial responsiveness (Murray et al. Reference Murray, De Pascalis, Bozicevic, Hawkins, Sclafani and Ferrari2016), so that cultural differences in its prevalence (e.g., as between U.S./European and some African populations) predict somewhat different infant social trajectories (Kärtner et al. Reference Kärtner, Keller and Yovsi2010; Wörmann et al. Reference Wörmann, Holodynski, Kärtner and Keller2012). Face-to-face interactions between mother and infant macaques also influence the development of infant social functioning: Monkey neonates that receive more facial responsiveness from their mothers spend more time in social contact with other monkeys at 2 months of age, and they initiate more social interactions at 5 months (Dettmer et al. Reference Dettmer, Kaburu, Simpson, Paukner, Sclafani, Byers, Murphy, Miller, Marquez, Miller, Suomi and Ferrari2016). That this effect is driven by experience of face-to-face interactions, rather than physical contact, is indicated by the finding that nursery-reared infants receiving mutual gaze and LS from a human caregiver subsequently show more social interest and social contact with peers than infants receiving either handling without gaze and LS, or standard nursery care. Notably, infant experience of early social interactions influences putative mirror neuron system responses, with infant monkeys raised with their biological mothers already demonstrating more mu desynchronization during observation of LS at 3 days postpartum compared to those raised apart (Vanderwert et al. Reference Vanderwert, Simpson, Paukner, Suomi, Fox and Ferrari2015). Such impact suggests a preparedness of the neonate brain to respond to social cues by harnessing aerodigestive behaviours, with a rudimentary mirror system rapidly refined by early experience. This may increase neural sensitivity to socially relevant stimuli such as LS, and thereby confer significant benefits for infants' navigation of the complex social world into which they are born (Vanderwert et al. Reference Vanderwert, Simpson, Paukner, Suomi, Fox and Ferrari2015). Recent research with human children similarly suggests mirror system involvement in the processing of facial expressions from a young age (Rayson et al. Reference Rayson, Bonaiuto, Ferrari and Murray2016; Reference Rayson, Bonaiuto, Ferrari and Murray2017).

In sum, we provide evidence from naturalistic, experimental, and clinical studies to show that infant mouth movements like TP/LS are complex and sensitive to context, and are fundamentally embedded in social interactions early in development, with longer-term significance for social functioning. Therefore, while both TP and LS may indeed have aerodigestive origins, they have also been exapted for uniquely social purposes.