The independence of response structure and element production in timing sequences

In three experiments, participants were asked to produce a prescribed temporal sequence of key presses. The number of elements in a key press sequence, the movement time of the elements, and uniformity of the timing elements comprising the sequence were manipulated. If the processing of the sequence structure was independent of the processing of elements comprising the sequence, increasing the number of the elements in the sequence should affect the production of the sequence but not the proficiency with which the individual elements are, produced. Increasing the movement time of the elements, however, should affect the production of the elements but have little, if any, effect on the integrity of the sequence. The findings indicated that increasing the number of elements negatively affected sequence production when the elements were nonuniform but had little or no effect when elements were uniform. Alternatively, element production was affected by movement time but not number of elements. The results of these experiments appear to confirm the independence of sequence and element production. However, Experiment 3, in which the uniformity of the elements was directly contrasted, found strong evidence for coarticulatory influences in the learning processes, whereby the longer and shorter than average elements in the sequence negatively affected the sequence integrity and the production of the individual elements. When faced with a nonuniformn timing pattern, participants appear to elongate or shrink individual elements of a uniform timing pattern to fit the desired sequence requirements. The result is that the longer or shorter than average elements require more practice to produce correctly, tend to regress across retention intervals, and are generally less stable than the elements nearer the average.