Translating between orthogonally oriented stimulus and response arrays in four-choice reaction tasks

Three experiments examined performance of four-choice reaction tasks using stimulus and response arrays oriented along parallel or orthogonal axes. All used a procedure in which pairs of locations were precued in advance of the target stimulus. Responses were slower for orthogonal than for parallel stimulus-response sets, but the pattern of relative precuing benefits was similar. Complete transfer occurred when the stimulus array was changed from an orthogonal to a parallel orientation with respect to the response array after three sessions of practice. Transfer was also evident when the orientation of the response array was changed from orthogonal to parallel with respect to the stimulus array, as long as the assignment of stimulus locations to fingers was not altered. The results suggest that coding in the four-choice task is by relative location regardless of whether the stimulus and response sets are oriented orthogonally, and that an additional transformation operation to align the frames of reference is performed for orthogonal orientations.