Preferential Looking and Intermodal Perception in Infancy: Comment on Lewkowicz (1992)

Lewkowicz (1992a) reported two experiments that used a visual preference method to investigate 4and 8-month-old infants’ detection of two auditory-visual temporal invariants: rate and synchrony. Infants showed no visual preference for an event occurring at the same rate as a concurrent sound; they showed a weak visual preference for an event occurring in synchrony with a concurrent sound. Based on further experiments using habituation methods, Lewkowicz concluded that infants are sensitive to temporal invariants in bimodally specified events (Allen, Walker, Symonds, & Marcell, 1977; Humphrey & Tees, 1980; Lewkowicz, 1992a, 1992b). Nevertheless, he concludes, infants do not express this sensitivity by looking preferentially at an acoustically specified visual display. The stimuli and the procedure of Lewkowicz’s preference experiments were designed to resemble closely those of experiments reported in Spelke (1979), and a major focus of Lewkowicz (1992a) concerns the relation between the findings of these two sets of studies. According to Lewkowicz (1992a), “Spelke (1979) and Spelke, Born, and Chu (1983) reported that infants prefer to look at a bouncing object whose rate of impact with a surface corresponds to the rate at which a sound is made . . .” (p. 298); the findings reported in Lewkowicz (1992a) therefore “differ from the findings reported by Spelke (1979)” (p. 303). The general discussion of the article focuses on possible reasons for this difference. In fact, the findings of the studies reported in Lewkowicz (1992a) agree quite closely with the findings reported in Spelke (1979), and they are consistent with the find-