Cortical Thickness Predictors of Performance-Based Functional Task Variability in the Alzheimer Disease Spectrum.

Purpose: Determine whether regional measures of cortical thickness are associated with functional task performance.

Patients: One hundred six older adults who were classified as either cognitively unimpaired (CU), mild cognitive impairment (MCI), or Alzheimer disease (AD) completed a performance-based functional task.

Methods: The SD of completion times of 6 consecutive trials within a session of the functional task was used as the primary measure for each participant, reflecting intraindividual variability. Regression tree analysis identified cortical gray matter thickness measures that best predicted intraindividual variability on the functional task.

Results: Cortical thickness measures from temporal, parietal, frontal, and occipital regions best predicted intraindividual variability on the task, which are cortical regions associated with learning, executive function, and visuospatial function. Specifically, the fusiform gyrus was featured prominently in these and prior regression tree results, suggesting its possible involvement in this behavioral task.

Conclusion: These analyses suggest a mechanistic focus of variability on this functional task, which could serve as an outcome in clinical trials.