Wagers for work: Decomposing the costs of cognitive effort
Sarah L. Master,Clayton E. Curtis,Peter Dayan
DOI: https://doi.org/10.1371/journal.pcbi.1012060
2024-04-30
PLoS Computational Biology
Abstract:Some aspects of cognition are more taxing than others. Accordingly, many people will avoid cognitively demanding tasks in favor of simpler alternatives. Which components of these tasks are costly, and how much, remains unknown. Here, we use a novel task design in which subjects request wages for completing cognitive tasks and a computational modeling procedure that decomposes their wages into the costs driving them. Using working memory as a test case, our approach revealed that gating new information into memory and protecting against interference are costly. Critically, other factors, like memory load, appeared less costly. Other key factors which may drive effort costs, such as error avoidance, had minimal influence on wage requests. Our approach is sensitive to individual differences, and could be used in psychiatric populations to understand the true underlying nature of apparent cognitive deficits. Anyone who has tried to mentally calculate how much to tip at a restaurant knows that cognitive effort can feel aversive. Doing math in your head, like most high-level cognitive abilities, depends critically on working memory (WM). We know that WM is sometimes effortful to use, but we don't know which aspects of WM use drive these effort costs. To address this question, we had participants request wages in exchange for performing various tasks that differed in their specific WM demands. Using computational models of their wage demands, we demonstrated that some aspects of WM are costly, such as bringing new information into memory and preventing interference. Other factors, like the amount of information in memory and attempts to avoid mistakes, were less costly. Our approach identified which specific subcomponents of WM are aversive. Future research could use these methods to test theories about how motivational problems might be masquerading as cognitive deficits in psychiatric populations.
biochemical research methods,mathematical & computational biology