Courage, M., Reynolds, G.D., & Richards, J.E. (2004). Developmental and individual differences in infants' attention as a function of stimulus characteristics. International Society for Infancy Studies, Chicago, IL.
There is evidence that the duration of infants' looks (e.g., mean or peak look length) during habituation and other familiarization procedures decreases with age across the first year of life. This has generally been attributed to increased speed and/or efficiency in information processing with age. Moreover, individual differences in these same look-duration measures within age have been observed in several research labs (e.g., Colombo and colleagues; Rose and colleagues). These data indicate that "short" lookers compared to "long" lookers encode information more quickly (i.e., with briefer, more broadly distributed fixations), show higher novelty preferences (i.e., better recognition) on immediate and delayed tests, require less time to shift from processing global to local pattern information, disengage fixation from a stimulus more quickly, and show higher performance on tests of language and cognition in childhood. Longer look duration is ostensibly less mature and is typical of very young infants, as well as those who are at risk for developmental delay (e.g., preterm infants, infants with Down syndrome).
However, most of these data have been based in infants' attention to static, two-dimensional, black and white patterned stimuli. In contrast, research in which infants' attention to dynamic, colorful, and complex stimuli (e.g., television material, novel toys) has been observed indicates that look duration often increases with age (e.g., Richards and colleagues; Ruff, Oakes and colleagues). This is especially likely when infants' heart rate and/or facial _expression and manual examination indicate that they are engaged in sustained or focused attention. One goal of the study reported here was to examine developmental and individual differences in infants' look duration to stimuli that vary in their configuration, complexity, and movement characteristics.
Five groups of infants aged 14, 20, 26, 39, and 52 weeks were the participants. The infants were presented with 8 stimuli for 20 seconds of accumulated looking to each one. The stimuli were: a still frame of a woman's face, the same face in conversation (no sound), static and dynamic versions of a black and white matrix of dots, static and dynamic versions of a black and white geometric pattern, a single still frame from a Sesame Street video and a dynamic clip from the same video. The series of stimuli were presented twice. Each infant's heart rate was assessed during a 2 minute baseline periods immediately before the first and second series of stimulus presentations. Heart rate was also measured throughout the stimulus presentations.
The results showed that for all stimulus types there was a significant decrease in look duration (peak look, mean look) from 14 to 26 weeks followed by one of two general patterns of look duration from 39 to 52 months (a) a continued decrease or a plateau -- for the achromatic geometric and the face stimuli or (b) a significant increase -- for the Sesame Street stimuli. This latter pattern is inconsistent with the traditional model of a linear decrease in look duration across the first postnatal year but is consistent with models of infant attention in which measures of look duration have been shown to increase with age sometime after 6 months. Patterns of individual differences in look duration and heart rate variability were complexly interrelated with these general developmental trends. In conclusion, the results of this study support the view that the development of infants' attention to visual stimuli is not a unitary process but may follow a multiphasic course (e.g., Colombo et al., 1999) in which look duration reflects the maturation of different cognitive and attentional processes over time.