Abstract
Introduction: ‘Whole report’ tests underestimate memory capacity; the crucial innovation was the partial report (Sperling, 1960). However, typical same/different tests of infants' memory - where infants' reactions are tested when a change is made to a set of objects, say, while briefly occluded or invisible - amount to whole report. Purpose: Adapt and apply a partial report test of infants' very short term (VSTM), ‘iconic’ memory span. Method: Over two blocks of 20 trials, ten infants (age range: 5;0–6;30) were presented with a set of 4 identically-shaped, but differently-colored items, spaced symmetrically around central fixation. After a 1 second exposure, a randomly chosen pair of neighboring items disappeared. After 500 msec, the two items reappeared, always with one changed to a new color and the other unchanged. This display remained for 2 seconds (eye-movements were recorded throughout the trial with a Tobii eye tracker). The sudden offset of the two items itself was the partial report post-cue, designed to draw attention and trigger their coding from VSTM into more durable short-term memory. Only by remembering all items in the set would infants be able to distinguish between the changed vs. unchanged items. Results: In what amounts to a 2AFC, we measured which of the two items was 1) fixated more frequently and 2) for a greater total time. For a set size of 4, infants showed positive ‘novelty’ effects, looking more often and longer at the changed item. Preliminary tests with set size 6 suggest positive results as well. Conclusion: We created a partial report paradigm for infants. Infants remembered two items cued randomly from a set of 4 (with suggestive results from set size 6, and work with higher set sizes ongoing), indicating a capacity of VSTM iconic memory of at least 4 items.
Supported by NIH 1R15EY017985-01A1.