Decreasing task-related brain activity over repeated functional MRI scans and sessions with no change in performance: implications for serial investigations.

For prospective functional imaging studies of learning and for clinical studies of recovery or disease progression, it is important that the magnitude of brain activity does not exhibit a trend over repeated sessions in the absence of changes in task performance. This may confuse the interpretation of proposed mechanisms. The objective of this study was to use functional magnetic resonance imaging to determine if a linear trend in brain activity was present for simple and commonly used motor and cognitive tasks. Fourteen healthy individuals participated in three sessions on different days during which four scans each of a finger flexion task and a working memory task were performed in a block design. The general linear model was used to determine brain regions exhibiting activity differences between sessions conducted on different days, as well as between scans performed within the same session. Task-related brain activity decreased over sessions and scans in prefrontal and frontal cortices for both tasks. No increases, nor quadratic trends, were detected. Activity within premotor and ipsilateral primary somatosensory cortex decreased over scans for externally cued finger flexion and over sessions for self-paced finger flexion. Activity within parietal cortex and contralateral supplementary motor area decreased over sessions for all forms of finger flexion. These results suggest that motor planning and sensory regions, as well as frontal and parietal cortices, exhibit linear decreasing brain activity over repeated sessions in the absence of changes in task performance for even the simplest block design paradigms.