Influence of the interstimulus interval on temporal processing and learning: testing the state-dependent network model

Abstract

The ability to determine the interval and duration of sensory events is fundamental to most forms of sensory processing, including speech and music perception. Recent experimental data support the notion that different mechanisms underlie temporal processing in the subsecond and suprasecond range. Here, we examine the predictions of one class of subsecond timing models: state-dependent networks. We establish that the interval between the comparison and the test interval, interstimulus interval (ISI), in a two-interval forced-choice discrimination task, alters the accuracy of interval discrimination but not the point of subjective equality-i.e. while timing was impaired, subjective time contraction or expansion was not observed. We also examined whether the deficit in temporal processing produced by short ISIs can be reduced by learning, and determined the generalization patterns. These results show that training subjects on a task using a short or long ISI produces dramatically different generalization patterns, suggesting different forms of perceptual learning are being engaged. Together, our results are consistent with the notion that timing in the range of hundreds of milliseconds is local as opposed to centralized, and that rapid stimulus presentation rates impair temporal discrimination. This interference is, however, decreased if the stimuli are presented to different sensory channels.

Figure 1

ISI impairs interval discrimination for the same frequency task. (a) Fitted psychometric functions for all 19 subjects on the Short_ISI-Same_Fr (blue), Long_ISI-Same_Fr (red), Short_ISI-Diff_Fr (green) and Long_ISI-Diff_Fr (cyan) conditions. (b) Mean data for the difference limens (DLs) and (c) PSE values. The interaction for the DLs was significant (F_1,18=33, p<0.0005). Error bars represent the standard-error of the mean. Open bars, short ISI; filled bars, long ISI.

Figure 2

Effects of the ISI on interval and frequency discrimination. Different thresholds for interval (open bars) and frequency discrimination (filled bars) tasks. The repeated ANOVA was significant for the interval thresholds (F_4,56=6.6, p<0.001) but not for the frequency thresholds (see text).

Figure 3

Generalization after training on different ISI intervals. (a) Individual learning curves for all subjects in the (i) Short_ISI-Same_Fr and (ii) Long_ISI-Same_Fr conditions. Subjects that exhibited a significant learning curve over 8 days of training are plotted in red. The data from two ‘non-learners’ in figures (i) and (ii) are not shown because their performance was off the chart. (b) Mean pre- (open bars) and post-test (filled bars) values on the four conditions for the subjects trained on the (i) Short_ISI-Same_Fr and (ii) Long_ISI-Same_Fr conditions (hatched bars show the trained conditions). In both experiments, there was a significant learning effect of the trained conditions. The subjects trained on (i) Short_ISI-Same_Fr exhibited no transfer to the three untrained conditions as indicated by the absence of main effect of training (F_1,7=0.82, p=0.39) and interaction (F_2,14=0.58, p=0.57). By contrast, for the subjects trained on the (ii) Long_ISI-Same_Fr stimulus, there was a significant generalization as indicated by the significant main effect of training (F_1,7=25.6, p<0.002). The interaction was not significant (F_2,14=0.68, p=0.52), indicating similar degrees of generalization to the three untrained conditions.