Motion correction and the use of motion covariates in multiple-subject fMRI analysis

摘要: The impact of using motion estimates as covariates no interest was examined in general linear modeling (GLM) both block design and rapid event-related functional magnetic resonance imaging (fMRI) data. purpose correction is to identify eliminate artifacts caused by task-correlated while maximizing sensitivity true activations. To optimize this process, a combination approaches applied data from 33 subjects performing block-design an fMRI experiment, including analysis: (1) without correction; (2) with alone; (3) motion-corrected included the GLM; (4) non-motion-corrected GLM. Inclusion found be generally useful for increasing GLM results analysis When parameters were data, it made little difference if actually For design, inclusion had deleterious on when even moderate correlation existed between experimental design. Based these results, we present strategy designs, hybrid designs probable