Keven & Akins (K&A) propose that spontaneous aerodigestive behaviours may be mistaken for neonatal imitation; however, well-designed neonatal imitation studies already account for reflexive and arousal-driven responses (for a review, see Simpson et al. Reference Simpson, Murray, Paukner and Ferrari2014a). When measuring arousal, either physiologically or behaviourally, and examining its relationship to imitative responding, evidence shows that for humans (e.g., Nagy et al. Reference Nagy, Pilling, Orvos and Molnar2013) and nonhuman primates (NHP; e.g., Paukner et al. Reference Paukner, Pederson and Simpson2017; Simpson et al. Reference Simpson, Paukner, Suomi and Ferrari2014b), changes in arousal alone cannot account for neonatal imitation. In addition, K&A acknowledge that they “have not explained, so far, the differential responses of neonates to specific gestures” (sect. 7.3.1, para. 2). We agree and further argue that differential imitation in neonates is incompatible with aerodigestive or arousal-driven explanations.

An additional misconception is that neonatal imitation is automatic and involuntary. Instead, infants exert active control over imitative responses and “provoke” previously imitated gestures, even after a delay, in both humans (Meltzoff & Moore Reference Meltzoff and Moore1994) and NHP (Paukner et al. Reference Paukner, Ferrari and Suomi2011). Moreover, neonates are sensitive both to the type of action and the identity of the individual who modelled the action, initiating interactions only among social partners with whom they previously interacted (Paukner et al. Reference Paukner, Ferrari and Suomi2011; Simpson et al. Reference Simpson, Paukner, Sclafani, Suomi and Ferrari2013). This ability is remarkable because it indicates that newborns are actively socially engaged (Meltzoff & Moore Reference Meltzoff and Moore1994). Consequently, delayed imitation is inconsistent with the proposal that neonatal imitation is a subcortical automatic response.

The aerodigestive hypothesis claims that imitative responses peak in the first week of life and decline in the following weeks. The data actually show the opposite for facial gestures: Neonatal imitative responses for tongue protrusion steadily increase in frequency from the first week to the ninth week (e.g., Oostenbroek et al. Reference Oostenbroek, Suddendorf, Nielsen, Redshaw, Kennedy-Costantini, Davis, Clark and Slaughter2016; Meltzoff et al. in press Reference Meltzoff, Murray, Simpson, Heimann, Nagy, Nadel, Pederson, Brooks, Messinger, De Pascalis, Subiaul, Paukner and Ferrari2017). Only after 3 months does the frequency of facial gesture imitation decline and infants begin to imitate other actions, such as sounds, vocalizations, and finger movements (Kuhl & Meltzoff Reference Kuhl and Meltzoff1996; Maratos Reference Maratos, Simion and Butterworth1998). Thus, although imitation does undergo changes with development, infants continue to reliably produce matching behaviours (for a review, see Simpson et al. Reference Simpson, Murray, Paukner and Ferrari2014a). These findings support the idea that neonatal responses are not stereotypes but rather intentional, voluntary behaviours.

We agree with K&A that animal studies widen our understanding of various phenomena, including neonatal imitation. Nonetheless, in this instance, K&A neglect to consider animal studies in their potential to inform our understanding of infant social cognitive development (Gerson et al. Reference Gerson, Simpson, Paukner, Sommerville and Decety2016). For example, K&A claim, “neonatal imitation experiments provide the only evidence that mirror neurons are present at birth” (sect. 2, para. 7). Although we agree that neonatal imitation is behavioural evidence of a functioning mirror neuron system, this assertion overlooks NHP studies documenting cortical brain activity through electroencephalography (EEG) and reporting neural evidence of a mirroring system functioning from birth in newborn monkeys (Ferrari et al. Reference Ferrari, Vanderwert, Paukner, Bower, Suomi and Fox2012).

Furthermore, evidence from animal research is consistent with the premise that individual differences in neonatal imitation may reflect individual differences in sociability (Heimann Reference Heimann1989; Heimann et al. Reference Heimann, Nelson and Schaller1989). Neonatal imitation is hypothesized to be a developmental precursor for, and potentially predict, later social cognitive capacities (Heimann Reference Heimann, Archer and Hansen1991; Reference Heimann, Lacerda, von Hofsten and Heimann2001; Reference Heimann, Meltzoff and Prinz2002; Suddendorf et al. Reference Suddendorf, Oostenbroek, Nielsen and Slaughter2013). Although this hypothesis has yet to be fully tested in humans, it has been tested in NHP (see recent review in Simpson et al. Reference Simpson, Miller, Ferrari, Suomi and Paukner2016). Infant monkeys who fail to exhibit neonatal imitation, compared to imitators, are less socially attentive (Simpson et al. Reference Simpson, Paukner, Suomi and Ferrari2014b), look less at faces in general and the eyes in particular (Paukner et al. Reference Paukner, Simpson, Ferrari, Mrozek and Suomi2014), exhibit poorer social cognitive skills such as imitation recognition (Simpson et al. Reference Simpson, Paukner, Suomi, Ferrari, Ferrari and Rizzolatti2015) and gaze following (Simpson et al. Reference Simpson, Miller, Ferrari, Suomi and Paukner2016), exhibit poorer goal-directed motor skills (Ferrari et al. Reference Ferrari, Paukner, Ruggiero, Darcey, Unbehagen and Suomi2009b), play less with peers and exhibit more anxious behaviour at one year of age (Kaburu et al. Reference Kaburu, Paukner, Simpson, Suomi and Ferrari2016). Together, these studies provide a more detailed view of the link between neonatal imitation, early social predispositions, and social development.

As a result of its plasticity, neonatal imitation may also be a fruitful target for intervention, as well as an early marker of sociality. Although we know little about the malleability of neonatal imitation in humans (Jacobson Reference Jacobson1979; Kennedy-Costantini et al. Reference Kennedy-Costantini, Slaughter and Nielsen2016), animal studies enable the manipulation of infants' environments and experiences. In monkeys, across both naturalistic and experimental settings, neonatal experiences impact infants' social capacities (Dettmer et al. Reference Dettmer, Kaburu, Simpson, Paukner, Sclafani, Byers, Murphy, Miller, Marquez, Miller, Suomi and Ferrari2016), including neonatal imitation, which is strengthened by face-to-face interactions in early infancy (Simpson et al. Reference Simpson, Murray, Paukner and Ferrari2014a; Vanderwert et al. Reference Vanderwert, Simpson, Paukner, Suomi, Fox and Ferrari2015). Contrary to the aerodigestive hypothesis, the aforementioned animal research strongly supports the social nature of neonatal imitation.

In sum, evidence to date is inconsistent with the view that neonatal imitation simply reflects spontaneous aerodigestive behaviours. Although we appreciate an approach mindful of the broader context of development, there is nonetheless a wealth of data that directly bear upon K&A's arguments. A narrow focus on arousal effects and reflexes may grossly underestimate neonatal capacities. Recognizing such capacities and establishing neonatal measures of sociality may help identify neonates who fall outside the range of healthy social development and may increase opportunities to intervene and foster positive child outcomes.