Fluctuations in sustained attention explain moment-to-moment shifts in children's memory formation

Abstract

Children get a lot of attention for being powerful “learning machines” in the popular press, but the truth is they remember much less than adults. Even still, many of children’s individual memories are rich and complex and similar in quality to adults’, suggesting what improves in development is not just the quality of our memories, but the frequency with which we form them. Here, we ask why children form memories less often than adults; instead of focusing on memory mechanisms, we focus on an entirely separate aspect of cognition: sustained attention. In adults, sustained attention fluctuates to shape memory in each moment, but we know little about how attention fluctuates in childhood to shape memory formation. To address this gap, 7–10-year-old children and adults (n=120) completed a sustained attention task in which they classified trial-unique images as living or nonliving. We then tested memory for each image, and related attentional fluctuations during classification to subsequent memory. We found that attention fluctuated between states more frequently in children than adults, and that across children, attentional lapse rates correlated with lower memory performance. Within participants, attentional fluctuations shaped the fate of individual memories, such that lapses predicted memory failures. While these fluctuations shaped memory for expected events in both children and adults, they only shaped memory for unexpected events in children, highlighting their particularly detrimental and pervasive influence in development. Our findings raise the possibility that broad developmental differences in cognitive performance reflect the ability to sustain attention.

Memory’s pulse: episodic memory formation is theta-rhythmic

“…Regardless of the cause, the dissociations we observe between attentional and mnemonic rhythms paints a more complex picture of attention-memory interactions than have been previously considered 38 . On a trial-by-trial basis we confirmed that people were better at forming memories when in a good attentional state 41,42,47 . Yet, on a millisecond-by-millisecond basis, the best moments for memory formation were not those when participants were best able to process the images.…”

“…In a pre-registered exploratory analysis, we examined whether slow fluctuations in sustained attention that occur across seconds and minutes modulate rapid sub-second fluctuations in memory formation. Based on prior research [40][41][42]47 , we classified periods of low RT deviance as an 'in the zone' or good attentional state, and periods of higher RT deviance as an 'out of the zone' or bad attentional state. Prior to computing attentional states, we regressed out the influence of stimulus-dependent, time-dependent, and trial-specific effects on RT, which systematically biased classification RT, but were unrelated to attentional state.…”

“…Lastly, in an exploratory analysis, we evaluated if nicotine users spent less time in a good attentional state, as would be expected if nicotine use reduced sensitivity to endogenous ACh, given links between acetylcholine and sustained attention 27 . To test this, we computed the proportion of time each individual spent in the zone using the procedures described above, except this time we assigned trials to "in the zone" and "out of the zone" using the group median, as in prior work characterizing individual differences 42,47 . Specifically, we calculated the median absolute deviance of RT residuals for each participant, then used the mean of these values as the group threshold.…”