Perceptual asymmetry and memory retrieval in 3-month-old infants

Abstract

In visual search tasks with adults, the presence of a feature is detected more rapidly than its absence. In two experiments, we asked if perceptual asymmetry affects the detection of effective retrieval cues. Three-month-olds learned to kick to move a mobile displaying either Rs (feature present) or Ps (feature absent), and these same stimuli were used later as retrieval cues during a long-term retention test. In Experiment 1, infants were tested after 24 hrs with a single P amidst six Rs or vice versa. In concordance with feature integration theory (Treisman, 1986), the feature-present target popped out while the feature-absent target did not. In Experiments 2A-2B, infants were tested after 24 hrs with novel homogeneous mobiles. Although the R mobile was not an effective retrieval cue for P-trained infants, the P mobile was an effective retrieval cue for R-trained infants. Even after only 1 hr, infants failed to discriminate the P test mobile from the R training mobile indicating that they had forgotten the tail of the R as rapidly as this short delay. These findings reveal that perceptual asymmetry alone does not completely determine retrieval from long-term memory. Rather, whether the detail that differentiates the feature-present stimulus from the feature-absent stimulus is accessible to working memory needs to be considered.