Abstract
It is known that observers' preparations for a task cannot be perfect until the task onset, which hinders their RTs (the preparation cost, Rogers & Monsell, 1995). Our study focuses on the nature of the preparation cost to grasp transitions of the state of attention through the task. To examine temporally sensitive profile of attention, the RSVP task would be optimum. In the RSVP task, which requires observers to identify a feature-defining target (e.g., a white letter) inserted in the rapid distractor stream (e.g., blue letters), we can directly observe transient changes in target-detection performance with manipulating when the target appears. Indeed the past RSVP studies have postulated that only a target inserted in the stream can be fairly detected whenever it appears, but the present study tested hypothesis that observers make preparations for the task at the very beginning of the RSVP sequence, which hinders their detection of the target at that time. EXP.1 evidenced that the detection of a target at the beginning of a sequence was dramatically low and recovered as it appeared later, that is, the preparation cost and its recovery process were observed in the RSVP task. EXPs. 2–4 asked what triggers the preparation for the task. The results showed that the preparation was triggered neither by the fixation onset (EXP. 2, in which the duration of the fixation point was extended) nor by mere the beginning of the rapid presentation (EXP. 3, in which the fixation point blinked at the same rate as the task sequence), but was triggered by the beginning of the rapid presentation of the task similar stimuli (EXP. 4, in which some signs, such as “#”, “%”, etc., were sequentially presented just before the task sequence). Our results indicated that the preparation cost that was directly observed in the RSVP task would reflect that observers gradually modulated their temporal attention to a rapid sequence with task relevancy in order to well extract a target from the temporally congested stream.