The Paced Auditory Serial Addition test (PASAT) is a sensitive task for evaluating cognitive impairment in patients with diffuse brain disorders, such as multiple sclerosis patients. Brain areas involved in this task have been investigated in diverse fMRI studies using different methodologies to control the subjects' responses during scanning. Here, we examined the possible differences between overt and covert responses during the PASAT task in 13 volunteers. Results showed similar activations in parietal and frontal brain areas during both versions of the task. The contrast between the two conditions (overt and covert) indicated that differences in these two methodologies were minimal. Unlike the covert condition, the overt version of the task obtained significant activations in the left superior and inferior frontal gyrus, bilateral occipital cortex, caudate nucleus and cerebellum. As expected, no significant overactivations were observed in the covert when compared with the overt condition. Discussion focuses on the lower cost of using verbal responses to monitor performance during the PASAT task, which might be generalisable to other frontal lobe tasks requiring discrete responses. Hum Brain Mapp, 2008. © 2007 Wiley-Liss, Inc.