While the current discussion largely restricts

While the current discussion largely restricts itself to neuroimaging methods that can make spatial inferences, ERP studies of RS have found similar results: while there is evidence of auditory RS involving both temporal and frontal cortices (Dehaene-Lambertz and Dehaene, 1994), ERP studies of visual RS have also largely reported a lack of visual RS in the developing buy Lomustine (see Nordt et al., 2016, for a recent review). To be clear, some fNIRS studies have employed visual stimuli and found clear evidence for RS. However, in these cases low-level visual repetition was controlled in order to examine effects of repetition in domains of higher-cognition such as number (Hyde et al., 2011) or goals (Southgate et al., 2014). Thus, our current finding of a lack of visual RS is not without precedent in the literature while, in contrast, there are clear demonstrations with auditory RS in the infant brain. As reviewed in Nordt et al. (2016), a lack of evidence for visual RS, especially in children, is surprising and in clear contrast to behavioral studies which have established that children readily discriminate between exemplars of faces. There are numerous (not mutually exclusive) possibilities to account for this which are important areas of future research. First, these findings suggest that RS is not the basis of recognition itself, but perhaps the outcome of this recognition process. Second, it could be that requiring infants to direct their gaze consistently to a diversity of visual exemplars could dilute RS to visual stimuli. Because auditory stimuli can be attended to without sustained visual attention, the superior RS to auditory stimuli could be the result of easier access. Third, it could be that early in development, individuals do not exhibit RS unless there is a functional reason to do so. This account would predict that including a task where behavior is benefitted from detecting or anticipating repeats would result in RS. If this were the case, it would support a view of RS as being driven by expectations and/or top-down signals rather than lower-level neural adaptation, and differences in engagement of higher-level cortices when stimuli are repeated would lead to differences in RS in perceptual cortices. This last possibility receives some support from the pattern of frontal lobe involvement that we see in this task. However, it is not immediately clear why there would be differences in the tendency to form expectations in the auditory modality compared to the visual modality and is an important area for future research. Studies implicating frontal lobe involvement in RS suggest an important, functional connection between frontal lobe and perceptual system modulations in response to repetition. The locations of frontal lobe RS are similar to those reported in the current study (Ghuman et al., 2008, e.g., left inferior frontal gyrus for auditory stimuli). Interestingly, numerous studies have found evidence of visual RS in the adult brain but always in combination with RS in the relevant perceptual cortices (e.g., Eriksson et al., 2008; Race et al., 2009; Zago et al., 2005). Ghuman et al. (2008) found that visual stimulus repetition both decreased activity in perceptual and frontal cortices but also increased neural synchrony between these regions. Moreover, the amount of neural synchrony correlates with behavioral facilitation resulting from repetition. There is more indirect evidence that the adult frontal lobe is modulated by repetition from the mismatch negativity (MMN) literature. Specifically, the frontal cortex has been implicated in auditory change detection (i.e., increases in activity with auditory novelty). Again, these frontal lobe findings are found in concert with activity changes in perceptual cortices (e.g., Schönwiesner et al., 2007). Thus, numerous studies have found frontal lobe modulation as a result of repetition (or novelty) and, notably, these frontal lobe findings are always paralleled with modulation of the relevant perceptual cortices.