Concurrent learning of temporal and spatial sequences.

Abstract
In a serial reaction time task, stimulus events simultaneously defined spatial and temporal sequences. Responses were based on the spatial dimension. The temporal sequence was incidental to the task, defined by the response-to-stimulus intervals in Experiment 1 and stimulus onset asynchronies in Experiment 2. The two sequences were either of equal length and correlated or of unequal length. In both experiments, spatial learning occurred regardless of sequence length condition. In contrast, temporal learning occurred only in the correlated condition. These results suggest that timing is an integrated part of action representations and that incidental learning for a temporal pattern does not occur independently from the action. Interestingly, sequence learning was enhanced in the correlated condition, reflecting the integration of spatial-temporal information. Since the seminal article by Lashley (1951) on "the problem of serial order," (p. 112) much research has been directed toward understanding the cognitive mechanisms underlying the acquisi- tion of skill in performing sequential activities. Fluent performance of many sequential activities, such as speech, music, and sports, requires carrying out component actions in the appropriate order. However, just as important as the order of component activities in performing these tasks is the relative timing of these actions. Sometimes, timing is crucial in explicitly defining the task, as in the case of musical performance. In other tasks, timing impacts the smoothness and skillfulness of performance, as in the case of adjusting the speed and direction of a car when driving down a familiar road. How are the order of actions and their temporal relations represented? Whether timing is assumed to be represented explicitly in a motor program (Schmidt, 1980; Viviani & Terzuolo, 1980) or considered an emergent property of the actions themselves (Kelso, 1981; Rumelhart & Norman, 1982), existing theories of timing differ widely in terms of the relationship between sequencing and timing. According to Rosenbaum (1985, 1987), the scheduling of actions is based on the joint specification of the order of actions and their timing. Consistent with this view, there is evidence that information about timing can be integrated with information about the action sequence (Summers, 1975, 1977). Such theories assume sequencing and timing of actions are inseparable. In contrast, the notion of a generalized motor program allows one to postulate abstract representations of temporal structure that can be trans- ferred to different movement rates (Heuer, 1988, 1991; Schmidt, 1980, 1985; Terzuolo & Viviani, 1980; see Gentner, 1987, for a critical review). Similarly, MacKay's (1982, 1987) hierarchical theory of skill acquisition includes temporal representations that are separate from both the representations of the gestures them- selves and the sequential order of successive gestures. The goal of our research was to investigate the relationship between the representation of sequential and temporal information when a new action sequence is learned. Timing in complex actions can be thought of as a sequence of durations or intervals to be learned together with the sequence of gestures. In most situations, sequential and temporal aspects of an action are correlated. For example, a recognizable melody requires that a series of notes be played with a specified temporal relationship across the notes. However, the order and timing of the sequential events can be varied independently; a melody can be played with various rhythms. The experiments reported below were designed to inves- tigate whether a temporal sequence can be learned concurrently with an action sequence. If so, can the temporal sequence be learned independently of the action sequence, or is it learned as an integral part of a specific action sequence?

This publication has 0 references indexed in Scilit: