Order information in short-term memory and time estimation

Abstract

In previous experiments, the amount of interference between time production and visual or memory search tasks was shown not to be related to the level of difficulty of the search task per se, but instead to the amount of processing in short-term memory required in the search task. The first experiment of the present study verified whether the amount of interference between time production and a short-term memory task may be related to the level of difficulty of the short-term memory task. Two versions of a memory task, with and without processing of order information, were combined with a temporal interval production task in a concurrent processing condition. As is shown in a control reaction time task, processing order information increased the level of difficulty of the memory search task. In the concurrent processing condition, the interference between short-term memory processing and time production was stronger when the level of difficulty of the short-term memory search task was increased by requiring that order information be processed. The results of Experiment 2 showed that the amount of interference between a similar short-term memory task and time production seems not to be related to the amount of order information that must be maintained during the time production task. This dissociation between the effects of processing and the maintenance of order information is compatible with a similar dissociation, observed in previous experiments, between the effects of processing and those of maintaining item information in short-term memory on concurrent time production.A current view, in time estimation research, holds that short-term or working memory is involved in performing the time estimation tasks used in most time research paradigms (Zakay, 1990). For example, a well-known psychophysical model of time estimation assumes that, to reproduce a brieftime interval (e.g., 2 sec), temporal information would be accumulated that could then be stored in a short-term memory store during the temporal reproduction. This information would be compared with a criterion amount of temporal information, corresponding to the to-be-reproduced interval, stored in long-term memory (Church, 1984; Gibbon, Church, & Meek, 1984). Some recent research with human subjects suggests that shortterm memory would also be involved in the comparison of brief time intervals (Wearden & Ferrara, 1993).Fortin and Rousseau (1987) combined a short-termmemory search task with time interval production. In this experiment, the subject memorized a set ofdigits and then produced a subjective 2-sec interval. During the temporal production, a probe digit was presented. At the end of the subjective 2-sec interval, the subject was asked to press one oftwo response buttons: One was used for answering This research was supported by a grant from the Natural Science and Engineering Research Council of Canada to the first author. We would like to thank Y. Lacouture for helpful suggestions regarding this research and M. Poirier for useful comments on an ea...

Temporal order in memory and interval timing: An interference analysis

“…Whereas processing temporal information interfered with temporal production in Fortin & Massé's (1999) study, produced intervals were not affected by increasing the number of items to be Temporal order in memory and timing 8 maintained in memory in their temporal order of presentation. This dissociation between the effects of processing and maintenance of information in memory on concurrent time production was investigated thoroughly in a subsequent study; results showed , in which there was no effect of that increasing the number of colors or tone timbres to retain during a time production task had no effect (Field & Groeger, 2004, Experiments 2 and 3) whereas retention of pitch information or tone durations showed load-related effects (Experiments 1 and 4).…”

“…Whereas adding digits during time discrimination disturbed timing performance, letter or visual pattern recognition did not. As stated by the authors, the absence of interference from retention tasks in Experiments 2 and 3 may be explained by the fact that these tasks involved passive storage of information essentially, which was shown in previous studies not to affect concurrent timing (Fortin & Massé, 1999; see also Field & Groeger, 2004); , and possibly also of athe relatively low level of task difficulty in those experiments may also have been a factor. There was a slight effect of performing concurrent timing concurrently on recognition errors in the spatial memory task, but disruption of recognition in the visuospatial task was clearly weaker than the corresponding effect of arithmetic on timing.…”

“…In the second experiment of the present study, the number of items to be maintained in their correct temporal order was varied during a time interval production task, and this condition was compared directly with a similar load manipulation involving the retrieval of retrieving information on temporal order. Although previous results (Field & Groeger, 2004;Fortin & Massé, 1999;Rammsayer & Ulrich, 2005) suggest that a dissociation should be obtained such that produced intervals would lengthen with increasing load in the processing task but not in the retention task, increasing load in the passive retention condition might also affect time production because the memory and timing tasks both require to the maintenance of maintain temporal information.…”

“…A shortening of perceived time with increasing difficulty of concurrent processing was obtained with a wide range of tasks involving perceptual (Brown, 1985;Casini & Macar, 1997;Coull, Vidal, Nazarian, & Macar, 2004;Field & Groeger, 2004;Macar, 2002;Zakay, 1993), memory (e.g., Fortin & Couture, 2002;Fortin & Massé, 1999;Hicks & Brundige, 1974;Rammsayer & Ulrich, 2005), and verbal (McClain, 1983;Miller, Hicks, & Willette, 1978;Zakay, 1989) processing.…”

“…Over a certain period of time, missing cues will lead to a general decrease in the number of accumulated temporal cues, hence shorter perceived duration. This attentional allocation model accounts for numerous results in time estimation research, where underestimation is directly related to the level of difficulty of concurrent nontemporal tasks (Fortin & Massé, 1999;Sawyer, Meyer, & Hauser, 1994;Zakay, Nitzan & Temporal order in memory and timing 4 Glickson, 1983) or to the proportion of attention deliberately allocated to nontemporal features of a stimulus (Casini & Macar, 1997;Coull, Vidal, Nazarian, & Macar, 2004).…”