Estimating task-related activity by using first few TRs of a run


I have a data set with 6 runs. In each run the participant are doing a task (resting*2, simulated standing *2, and simulated walking *2) . Participants complete these tasks either in the presence or absence of simulation (30 seconds of no stimulation followed by 30 seconds of stimulation in alternate blocks for the duration of each run). I have been asked to determine the difference in brain activity between tasks when there is no stimulation present. I was thinking about analyzing the first 30 seconds (no stimulation) of each of the two runs with the same task. Is this valid?

Thank you,