Perceptual narrowing in face and speech perception during infancy

Abstract

Perceptual narrowing is a process through which infants gradually lose sensitivity for stimuli which they have limited exposure to. The present synopsis showcases the findings of four studies carried out with the overall goal of improving our understanding of perceptual narrowing in the domains of face-perception and speech-perception, which typically occurs between the ages of 6 and 9 months. The first two studies were aimed at investigating the hypothesis that the perceptual narrowing processes in face-perception and speech-perception might be driven by shared domain-general mechanisms. In the first study, we used a habituation-dishabituation task to test the ability of 9-month-old infants to discriminate among same-race and other-race faces, as well as among non-native speech tones. We found that infants could discriminate among same-race faces, but not among other-race faces or non-native tones. We also found a significant correlation between their discrimination for other-race faces and non-native tones. In the second study, we used the same task and stimuli to longitudinally test for changes in infants’ discrimination abilities for these stimuli between the ages of 6 and 9 months. We found that while at 6 months infants could discriminate between all three classes of stimuli, by 9 months they could only discriminate among same-race faces, and could no longer discriminate among other-race faces or non-native tones. Thus, the relation between discrimination for other-race faces and non-native tones at 9 months we found in the first study, as well as the parallel decreases in discrimination for these stimuli between 6 and 9 months we found in the second study, support the domain-general theory of perceptual narrowing. In our third study, we investigated whether it is possible to re-sensitive infants to other-race faces even after the onset of perceptual narrowing, by exposing them to specific frequency distributions of such faces. We contrasted exposing 12-month-old infants to either a unimodal distribution of morphs created from two endpoint faces where the midpoint morphs were shown most frequently, or a bimodal distribution where the endpoint morphs were shown the most frequently. We found that while exposure to a unimodal morph distribution did not allow infants to discriminate among between the endpoint faces at test, infants exposed to a bimodal distribution could discriminate between the endpoint faces. Thus, it appears that even after the onset of perceptual narrowing, it is still possible to re-sensitize infants to rarely encountered stimuli using an appropriate frequency distribution. In the fourth study we examine the role of gaze behavior in perceptual narrowing for faces in 9-month-old infants using the same habituation/dishabituation approach described earlier with eye-tracking. We found that infants’ gaze towards the eyes of same-race faces correlated with the discrimination ability for those faces, but the same was not true for other-race faces, suggesting that differences in gaze behavior could be an important factor in perceptual narrowing for faces. In summary, our studies support the domain-general hypothesis of perceptual narrowing, and indicate that gaze behavior and the statistical distribution of encountered faces can affect perceptual narrowing in face perception.