Development of ordinal sequence perception in infancy

Abstract

Perception of the ordinal position of a sequence element is critical to many cognitive and motor functions. Here, the prediction that this ability is based on a domain-general perceptual mechanism and, thus, that it emerges prior to the emergence of language was tested. Infants were habituated with sequences of moving/sounding objects and then tested for the ability to perceive the invariant ordinal position of a single element (Experiment 1) or the invariant relative ordinal position of two adjacent elements (Experiment 2). Experiment 1 tested 4- and 6-month-old infants and showed that 4-month-old infants focused on conflicting low-level sequence statistics and, therefore, failed to detect the ordinal position information, but that 6-month-old infants ignored the statistics and detected the ordinal position information. Experiment 2 tested 6-, 8-, and 10-month-old infants and showed that only 10-month-old infants detected relative ordinal position information and that they could only accomplish this with the aid of concurrent statistical cues. Together, these results indicate that a domain-general ability to detect ordinal position information emerges during infancy and that its initial emergence is preceded and facilitated by the earlier emergence of the ability to detect statistical cues.

Figure 1

The sequences and the different objects constituting them presented during the habituation and test phases in Experiment 1 (each unique object had a unique impact sound associated with it). The set of objects depicted for each respective sequence are deliberately presented in different spatial locations so as to capture the fact that the sequences moved down and across the stimulus display screen.

Figure 2

Mean duration of looking in the four types of test trials in Experiment 1. Error bars indicate the standard error of the mean.

Figure 3

The sequences and the different objects constituting them presented during the habituation and test phases in Experiment 2 (each unique object had a unique impact sound associated with it). The set of objects depicted for each respective sequence are deliberately presented in different spatial locations so as to capture the fact that the sequences moved down and across the stimulus display screen.

Figure 4

Mean duration of looking in the test trials in Experiment 2. Error bars indicate the standard error of the mean.