Visual-preference and operant measures of infant memory

In three experiments with sixty 3- and 6-month-olds, we examined whether operant and visual-preference measures of retention are equivalent. Infants learned to move a mobile by kicking and then received a paired-comparison test with the familiar (training) mobile and a novel one. Kicking above baseline was the direct measure of retention, and longer looking at the novel mobile was the visual-preference or inferred measure. Retention was tested 1 day after training (Experiment 1) or reactivation (Experiment 2) and immediately or 4 days after training (Experiment 3). Despite differences in age and retention interval, infants consistently exhibited retention on the operant measure, but not on the visual-preference measure. These data reveal that the two measures of retention are not equivalent. When expectations are associated with a particular stimulus, infants indicate retention directly and robustly, rarely looking longer at a novel test stimulus. This result is not limited to operant situations.