Choice reaction times for structural relationships between possible response sequences.

Abstract

To investigate human motor programming, choice reaction times were measured on tasks for which subjects made choices between two alternative finger-tapping-movement sequences. The total-number-of-responses and the hierarchical editor models were tested. In Exp. 1 the choice was carried on the situations with the same total numbers of possible responses and different structural relations between alternative sequences. The right-hand reaction times in mirror choice (e.g., subject chose between the middle, index, and ring finger sequences of the left or right hand) were shorter than those in nonmirror choice (e.g., subject chose between the middle, index, and ring finger sequence on one hand and the middle, ring, and index finger sequence on the other hand); the total-number-of-responses model was not supported. In Exp. 2 two conditions had the same operation numbers of the hierarchical editor model. In Condition 1 subjects chose between the index finger of the right hand and the ring, index, and middle finger sequence of the left hand. In Condition 2 subjects chose between the index, ring, and middle finger sequences of the left right hand. The reaction time in the former condition was shorter than that in the latter condition. Exp. 2 exhibited a counterexample of the hierarchical editor model that had been fairly robust in previous studies.