Keven & Akins (K&A) argue that tongue protrusion and retraction (TP/R) are “orofacial stereotypies crucial to the maturation of aerodigestion in the neonatal period but also unlikely to co-occur with imitative behavior” (target article abstract). Although agreeing that their analysis of TP/R is valuable and helps to explain the phenomenon of imitation “drop out,” I do not agree with their conclusion that TP/R is unlikely to be an example of imitative behavior. K&A view infant behavior at the level of the individual. I present infant behavior in a broader, social and communicative context centered on the phenomenon of spontaneous communication.

Spontaneous communication has been defined to explain the expressive behavior of patients with left and right hemisphere brain damage (LHD, RHD; Buck & Duffy Reference Buck and Duffy1980). Patients were shown emotional images (e.g., of hospital personnel, landscapes, unpleasant scenes, unusual scenes) while being videotaped, and judges viewing them without audio guessed the category of image viewed on each trial. LHD patients, who cannot speak, showed strong expressive responses to the images: laughing and gesturing to images of favorite nurses, and in one case beginning to cry upon seeing an image of a starving child. RHD patients were expressive verbally – one saying in effect, “That is Nurse Jones, we love her, she is wonderful to us” – but in a monotone with no facial expression. LHD patients showed the most accurate nonverbal communication, in that judges could determine what sort of image they were viewing as well as or better than non-brain-damaged comparison patients. In contrast, RHD patients were significantly poorer in nonverbal communication accuracy, not different from Parkinson's patients in whom a lack of facial expression is a key symptom. The kinds of communication that RHD and LHD patients were using were fundamentally different: the communication of RHD patients is symbolic: learned and culturally structured, intentional, consisting of symbols, and propositional in content. In contrast, spontaneous communication is innate in both its sending and receiving aspects, non-intentional, consisting of signs, and non-propositional or emotional in content (Buck Reference Buck1984; Reference Buck2014). A subsequent meta-analysis confirmed that LHD produces deficits in “nonverbal symbolic communication” – pantomime expression and recognition – as well as speech (Buck & Van Lear Reference Buck and Van Lear2002).

Spontaneous communication sheds light on a number of controversies in the field, including but not restricted to infant imitation. Evidence of interpersonal synchrony was demonstrated in the 1970s (Condon Reference Condon and Davis1982; Condon & Sander Reference Condon and Sander1974). Trevarthen et al. suggested that mutually contingent responsiveness between mother and infant associated with such synchrony (e.g., mother and infant responding “online” to the flow of the behavior of the other) naturally affords primary intersubjectivity: infant and mother each automatically attune to the subjective state displayed by the other (Braten & Trevarthen Reference Braten, Trevarthen and Braten2007; Trevarthen Reference Trevarthen and Bullowa1979; Trevarthen & Aitken Reference Trevarthen and Aitken2001). (Trevarthen is cited in the references but not the main text of K&A.) Frans de Waal (Reference de Waal and Braten2007) viewed early spontaneous emotion communication systems to be the basis for more advanced cognitive empathy. His model suggests that advanced forms of empathy involving theory of mind and perspective-taking grow out of more elementary emotional empathy, both evolutionarily and developmentally, analogous to the levels of a Russian matryoshka doll. The ordinarily hidden inner pre-linguistic core of this process involves innate motivational and emotional potential to spontaneously communicate: to form strong attachments, including capacities for imitation and mutual contingent responsiveness (de Waal Reference de Waal and Braten2007). The development of cognitive empathy is seen to depend upon, and involve elaborations of, this hidden inner core of emotional empathy.

There is also evidence that mirror neurons afford the direct detection/pickup of the emotions of others via their displays, and many have suggested that mirror neurons are involved in emotional empathy (e.g., de Waal Reference de Waal and Braten2007; Decety & Jackson Reference Decety and Jackson2004; Keysers Reference Keysers2011). Indeed, the notion of spontaneous communication requires that innate displays be responded to automatically by co-evolved preattunements, and mirror neurons appear to fulfill that requirement nicely. Although the notion of mirror neurons in humans remains controversial (e.g., Heyes Reference Heyes2010; Hickok Reference Hickok2009), there is direct evidence of mirror neurons in human patients undergoing depth electrode study for epilepsy. Mukamel et al. (Reference Mukamel, Ekstrom, Kaplan, Iacoboni and Fried2010) found neurons in humans that acted like classic mirror neurons studied in animals, discharging during both the observation and execution of one specific action but not another. They also found “anti-mirror neurons” in the supplemental motor cortex that increased when a given action was performed but decreased when the same action was observed. Keysers et al. (Reference Keysers, Kaas and Gazzola2010) noted that anti-mirror neurons could disambiguate actions of self from the actions of others and selectively block automatic motor imitation.

Ruth Feldman suggested that there are three prototypes of attachment in human beings: parental, filial, and romantic love; and that these share common brain mechanisms underpinned by oxytocin (OT) in the promotion of trust, emotional empathy, and interpersonal synchrony. Intranasal OT increases synchrony between attached couples in double-blind studies. Interestingly, when a father received OT and played with his infant, OT increased in the infant (Feldman Reference Feldman2012). Although OT has been considered a “cuddle hormone,” its effects are not always positive. It may be better termed a (spontaneous) “communication hormone,” as its effects appear to increase communication, which in insecurely attached persons or vis-à-vis outgroup members may be negative (e.g., Shamay-Tsoory et al. Reference Shamay-Tsoory, Fischer, Dvash, Harari, Perach-Bloom and Levkovitz2009). Thus, neurochemical systems associated with OT may support parental caring, play, friendship, and romantic love; and at the same time they may foster xenophobia: the rejection and ostracism of those not deemed to be within the group.

Spontaneous communication is pervasive, ubiquitous, and omnipresent; and at the same time typically automatic, involuntary, and unconscious. It contributes to syncretic cognition: raw, holistic, direct, immediate, and self-evident knowledge-by-acquaintance; whereas symbolic communication is associated with sequential, linear, propositional, and rational analytic cognition (Tucker Reference Tucker1981). From this point of view, infant imitation may be just another example of spontaneous communication. The infant's tendencies to exhibit TP/R, in addition to being involved in the maturation of aerodigestion, simply affords a way to display and communicate; and when TP/R ceases to occur, it outwardly appears to be “imitation drop out.”