The three task- or set-switching tasks all required participants to switch back and forth between two subtasks. In each trial of the number–letter, participants saw a number-letter or letter-number pair in one quadrant of a box and categorized the number (2-9) as odd or even if the pair appeared in one of the top two quadrants, but categorized the letter (A, E, I, U, G, K, M, or R) as consonant or vowel if the pair appeared in one of the bottom two quadrants. In each trial of the color–shape task, participants saw a colored (red or green) shape (circle or triangle) and categorized the shape or color depending on a cue (C or S) that appeared above the stimulus. In each trial of the category-switch task, participants saw a word (alligator, bicycle, cloud, coat, goldfish, knob, lion, lizard, marble, mushroom, oak, pebble, shark, snowflake, sparrow, or table) and categorized it as describing something that is smaller or bigger than a soccer ball or living or nonliving, depending on a symbol (heart or crossed arrows) that appeared above the stimulus. In each trial (except for the number–letter single-task and predictable-switch blocks), the cue (darkening of the quadrant, letter, or symbol) started 350 ms before the target stimulus appeared. The cue and stimulus remained on the screen until the participant responded with one of two button presses (left indicated odd or consonant, red or circle, and small or nonliving, and right indicated even or vowel, green or triangle, and big or living), which triggered the next trial after a 350 ms response-to-cue interval. A 200 ms buzz sounded for errors.
Each task began with single-task blocks in which only one task was required (number then letter; color then shape; and animacy then size) for 24 trials each in the color–shape task and 32 trials each in the number–letter and category-switch tasks (block length differed slightly due to counterbalancing considerations). Each of these single-trial blocks was preceded by a 12-trial practice block and included two extra "warm-up" trials that were not analyzed. After these single-task blocks, participants completed two mixed blocks (56 trials each for color–shape, and 64 trials each for the other tasks), in which the two subtasks were pseudo-randomly mixed such that half the trials required switching subtasks. In the number–letter task they also completed two 64-trial predictable-switch blocks (not analyzed here) prior to the random mixed blocks, in which the stimuli circled the box in a clockwise pattern. The first of each type of switch block was preceded by a 24-trial practice block, and each switch block included four extra "warm-up" trials. The dependent measure for each task was the local switch cost: the difference between average RTs for switch trials and repeat trials in the random mixed blocks.
The primary change from the Wave 1 versions was the addition of the error signal (to further improve accuracy), the single-task blocks, the use of a 350-ms cue-to-stimulus interval throughout the mixed blocks, and the elimination of longer cue-to-stimulus interval blocks. In the Wave 1 version, participants completed four blocks that alternated between a 150 ms cue-to-stimulus interval and a 1,500 ms cue-to-stimulus interval (used to calculate residual switch costs); however, the blocks with the longer interval were not analyzed for the primary report (Friedman et al., 2008 (link)). The single-task blocks gave participants practice with the subtasks and response mappings and provided a baseline for calculating global switch costs and mixing costs (not analyzed here).
Each task began with single-task blocks in which only one task was required (number then letter; color then shape; and animacy then size) for 24 trials each in the color–shape task and 32 trials each in the number–letter and category-switch tasks (block length differed slightly due to counterbalancing considerations). Each of these single-trial blocks was preceded by a 12-trial practice block and included two extra "warm-up" trials that were not analyzed. After these single-task blocks, participants completed two mixed blocks (56 trials each for color–shape, and 64 trials each for the other tasks), in which the two subtasks were pseudo-randomly mixed such that half the trials required switching subtasks. In the number–letter task they also completed two 64-trial predictable-switch blocks (not analyzed here) prior to the random mixed blocks, in which the stimuli circled the box in a clockwise pattern. The first of each type of switch block was preceded by a 24-trial practice block, and each switch block included four extra "warm-up" trials. The dependent measure for each task was the local switch cost: the difference between average RTs for switch trials and repeat trials in the random mixed blocks.
The primary change from the Wave 1 versions was the addition of the error signal (to further improve accuracy), the single-task blocks, the use of a 350-ms cue-to-stimulus interval throughout the mixed blocks, and the elimination of longer cue-to-stimulus interval blocks. In the Wave 1 version, participants completed four blocks that alternated between a 150 ms cue-to-stimulus interval and a 1,500 ms cue-to-stimulus interval (used to calculate residual switch costs); however, the blocks with the longer interval were not analyzed for the primary report (Friedman et al., 2008 (link)). The single-task blocks gave participants practice with the subtasks and response mappings and provided a baseline for calculating global switch costs and mixing costs (not analyzed here).
