Does Hemispheric Asymmetry Reduction in Older Adults in Motor Cortex Reflect Compensation?

Older adults tend to display greater brain activation in the nondominant hemisphere during even basic sensorimotor responses. It is debated whether this hemispheric asymmetry reduction in older adults (HAROLD) reflects a compensatory mechanism. Across two independent fMRI experiments involving adult life span human samples (N = 586 and N = 81, approximately half female) who performed right-hand finger responses, we distinguished between these hypotheses using behavioral and multivariate Bayes (MVB) decoding approaches. Standard univariate analyses replicated a HAROLD pattern in motor cortex, but in and out of scanner behavioral results both demonstrated evidence against a compensatory relationship in that reaction time measures of task performance in older adults did not relate to ipsilateral motor activity. Likewise, MVB showed that this increased ipsilateral activity in older adults did not carry additional information, and if anything, combining ipsilateral with contralateral activity patterns reduced action decoding in older adults (at least in experiment 1). These results contradict the hypothesis that HAROLD is compensatory and instead suggest that the age-related ipsilateral hyperactivation is nonspecific, consistent with alternative hypotheses about age-related reductions in neural efficiency/differentiation or interhemispheric inhibition. SIGNIFICANCE STATEMENT A key goal in the cognitive neuroscience of aging is to provide a mechanistic explanation of how brain–behavior relationships change with age. One interpretation of the common finding that task-based hemispheric activity becomes more symmetrical in older adults is that this shift reflects a compensatory mechanism, with the nondominant hemisphere needing to help out with computations normally performed by the dominant hemisphere. Contrary to this view, our behavioral and brain data indicate that the additional activity in ipsilateral motor cortex in older adults is not reflective of better task performance nor better neural representations of finger actions.