Abstract: Although many studies revealed a subjective shortening in the human’s short-term memory of time, the conclusions remain ambiguous. The present study aimed to explore whether when the different duration (sub- and supra- second intervals) is retained in short-term memory, the duration becomes subjective shortening or lengthening as retention interval increased.
Based on the method of Wearden and Ferrara, Two experiments ( Experiment 1, 2) were used to investigate the retention effect of temporal intervals from 350ms to 650ms and from 1000ms to 2000ms respectively, and each experiment included two sub-experiments which separately examined the duration retention of the auditory presentation (Experiment 1a, 2a) and that of the visual presentation (Experiment 1b, 2b). 2(stimulus type: filled duration and unfilled duration) ×4 (delay time: 1s, 4s, 8s and 12s) ×3 (trial type: short trial, equal trial and long trial) within-subject factorial design was taken separately for every sub-experiment (Experiment 1a, 2a, 1b and 2b). All participants were required to respond by pressing one of the three keys after judging whether the com-parison was longer, shorter, or of the same duration as the standard duration.
The results of the two experiments showed that, the short-term retention of durations from 350ms to 650ms (Experiment 1) indicated subjective shortening both in auditory (Experiment 1a) and visual conditions (Experi-ment 1b), whereas the short-term retention of durations from 1000ms to 2000ms (Experiment 2) showed subjec-tive lengthening both in auditory (Experiment 2a) and visual conditions (Experiment 2b), and there was neither effect of modality nor the effect of stimulus type that influenced the retention effect.
Conclusions were drawn from this study. For Experiment 1 (several hundred milliseconds), the shortening trend resulted from the interaction of shrink of the positive time order error (TOE) and the effect of subjective shortening. By contrast, for Experiment 2 (the retention of durations above 1s), the lengthening trend resulted from the interaction of shrink of the negative TOE effect and the effect of subjective lengthening. According to these results, the present study compensated and perfected the relative duration hypothesis and the partial proc-essing feature of the scalar timing model. However, apart from duration length and stimulus type, other factors such as experimental methods are likely to affect time duration retention as well, and whether these factors are independent or interacted needs to be investigated in future research.

Key words: duration, short-term memory, subjective shortening, subjective lengthening