Motor disruption in the production of time intervals with finger taps.

Abstract

Abstract Two experiments examined the variance associated with the motor component in a time production task involving a series of finger taps. Use of the segmentation strategy described by Grondin (1992) allowed the isolation of a part of the variance associated with the motor component in such a tapping task. It is shown that replacing a finger tap by an internal indicator reduces the coefficient of variation by about .5% per second. As well, these experiments showed that the choice of subintervals, when an interval is segmented during a timing task, is a major source of variance. It is concluded that the segmentation strategy provides a general framework for analyzing the part of the variability provoked by each of two components - the clock and the motor system - involved in a time interval production task.Resume Cet article a pour but d'evaluer la variance provoquee par une composante motrice intervenant lors d'une tache de production d'intervalles temporels a l'aide d'une serie de frappes du doigt. Cette variance est evaluee lors de deux experiences ou est utilisee une strategie de segmentation decrite par Grondin (1992). L'utilisation d'un repere interne, a la place d'une frappe du doigt, permet de reduire le coefficient de variation d'environ .5% par seconde. Aussi, ces experiences demontrent une nouvelle fois que le choix des sous - intervalles, lorsqu'un intervalle est segmente lors d'evaluations temporelles, est un facteur critique dans l'explication de la variance observee lors de ces evaluations. Enfin, la strategie de segmentation fournit un cadre pour interpreter la part de variabilite causee par chacune des deux composantes, l'horloge interne et le systeme moteur, impliquees dans une tache de production d'intervalles temporels.One approach to studying psychological time is to assume that there is an internal clock (Ivry & Keele, 1989). The study of this clock could involve tasks related to time perception or to time production. The present paper is concerned with the latter. When a time production task requires the subjects to produce a series of finger taps marking equal time intervals, it is often assumed that the variability of these intervals depends on two sources of variance, the clock itself and a motor component (Wing & Kristofferson, 1973). The following experiments are designed to extract from the total variance an estimate ofthe cost, in terms of variability, related to this motor component.To estimate the motor - component variability, the segmentation strategy reported in Grondin (1992) was used. The task is to produce a series of time intervals marked by finger taps. The measure of performance is the variability of the inter - tap intervals, this variability being divided by the mean intervals produced. This dependent variable, then, is the coefficient of variation. Before the production of a series of time intervals, a subject is presented the target interval to be produced, t, by way of auditory inputs every t sec. On some trials, these auditory target intervals are subdivided into 2, 3, or 4 subintervals (d) by auditory signals of a different frequency. For example, take cases where a 1.2 - sec interval is not divided, and others where 4.8 - sec intervals are divided into 4 subintervals of 1.2 sec. Whether the task is to produce 1.2 - sec intervals or 4.8 - sec intervals, the same internal duration is kept in active memory as a model (t = 1.2 sec or d = 1.2 sec). But during the production, the subjects tap every 1.2 sec when t = 1.2 sec and only every 4.8 - sec when d = 1.2 sec. The idea is to compare the coefficient of variation observed with the 4.8 - sec productions with the coefficient of variation observed with the 1.2 - sec productions. Because the internal interval is the same in both cases (t = d), and given that the coefficient of variation is a measure taking the total length of the intervals into account, one can predict that these coefficients of variation should be equal. …