Poster Session C, Sunday, March 24, 5:00 – 7:00 pm, Pacific Concourse
Characterizing the relationship between working memory capacity and load-related increases in fMRI activity
Catherine R. Walsh1, Jean-Baptiste Pochon1, Kristen D. Enriquez1, Holly Truong1, Agatha Lenartowicz1, Sandra K. Loo1, Catherine A. Sugar1, Carrie E. Bearden1, Robert M. Bilder1, Jesse Rissman1; 1University of California, Los Angeles
A fruitful functional neuroimaging approach for identifying regions involved in working memory (WM) has been to vary the number of to-be-maintained items and examine the degree to which BOLD activity increases as storage demands increase. Studies using delayed-match-to-sample paradigms are capable of isolating WM load effects during both encoding and maintenance, but the relationship of such load effects to trait-level individual differences in WM capacity is under-explored. In the present study, we collected an extensive battery of behavioral measurements of WM capacity from a large sample (N=170) of adult participants, who also underwent fMRI scanning while performing a delayed face recognition task. Each trial of the fMRI task involved the encoding of one face (low load) or three faces (high load), which needed to be maintained across a 7.5 s delay period followed by a match/nonmatch discrimination. Regression analyses identified regions where the magnitude of load-related activity increases was predicted by individual differences in a composite index of WM capacity. Participants with higher capacity estimates showed significantly greater load effects in the intraparietal sulcus and frontal eye fields (regions commonly associated with top-down visuospatial attention) during the encoding-period. However, during the delay-period, capacity was disproportionately correlated with load effects in rostrolateral and dorsolateral prefrontal cortex. Thus, the relationship of fMRI load effects to behavioral differences in WM capacity differs markedly as information processing progresses from stimulus encoding to active maintenance. This distinction may be critical for Research Domains Criteria (RDoC) efforts to relate neurocognitive markers of WM capacity to psychopathology.
Topic Area: EXECUTIVE PROCESSES: Working memory