Perceived empty duration between sounds of different lengths: Possible relation with repetition and rhythmic grouping

Abstract

We investigated how perceived duration of empty time intervals would be modulated by the length of sounds marking those intervals. Three sounds were successively presented in Experiment 1. Each sound was short (S) or long (L), and the temporal position of the middle sound's onset was varied. The lengthening of each sound resulted in delayed perception of the onset; thus, the middle sound's onset had to be presented earlier in the SLS than in the LSL sequence so that participants perceived the three sounds as presented at equal interonset intervals. In Experiment 2, a short sound and a long sound were alternated repeatedly, and the relative duration of the SL interval to the LS interval was varied. This repeated sequence was perceived as consisting of equal interonset intervals when the onsets of all sounds were aligned at physically equal intervals. If the same onset delay as in the preceding experiment had occurred, participants should have perceived equality between the interonset intervals in the repeated sequence when the SL interval was physically shortened relative to the LS interval. The effects of sound length seemed to be canceled out when the presentation of intervals was repeated. Finally, the perceived duration of the interonset intervals in the repeated sequence was not influenced by whether the participant's native language was French or Japanese, or by how the repeated sequence was perceptually segmented into rhythmic groups.

Two-dimensional parsing of the acoustic stream explains the Iambic–Trochaic Law.

“…See the Supplemental Materials for more information. 14 WAGNER that is expected is processed differently from one that cannot be anticipated (Kuroda et al, 2016), as originally hypothesized already in Woodrow (1909). This means that to further tease this apart, more experiments will be necessary, for example, looking at sequences of speech and tones that are less regular.…”

“…For example, in the first study looking at the ITL in speech (Hay & Diehl, 2007), participants listened to sound sequences consisting of repetitions of the syllable ga, and were asked "whether the rhythm consisted of a strong sound followed by a weak sound, or, alternatively, a weak sound followed by a strong sound." Some studies used a visual representation of two stimuli, which represented whether the first or second was more prominent (Bhatara et al, 2013;Boll-Avetisyan et al, 2016Iversen et al, 2008;Kusumoto & Moreton, 1997;Molnar et al, 2016), or asked participants to tap the perceived rhythm with their hands (Rice, 1992). This type of "foot-decision task" conflates the dimensions of prominence and grouping.…”

Discrimination of empty and filled intervals marked by auditory signals with different durations and directions of intensity change

“…For the filled interval (Experiment 2B), a featured finding was that the target interval was perceived as shorter when it was bounded by dd markers compared to when it was bounded by rr markers. In the dd condition, the second marker with decreasing intensity was illusorily grouped with the “tail” of the middle continuous sound, which caused the observers to perceive the “onset” of the second marker as earlier and hence curtail the perceived duration of the middle sound, and with degraded discrimination of larger JNDs (Kuroda et al, ).…”

“…Given the same physical interval enclosed by the short and long auditory markers, the onset and offset starts and ends veridically for the short markers. More recent evidence has shown that the perception of enclosed target interval is dependent on the perceptual grouping (including the rhythmic grouping) between all the sound stimuli (including the markers; Geiser & Gabrieli, ; Klink, Montijn, & van Wezel, ; Kuroda, Hasuo, Labonte, Laflamme, & Grondin, ; Kuroda, Tomimatsu, Grondin, & Miyazaki, ). The to‐be‐timed interval could be perceived earlier or later due to the temporal Gestalt grouping between the time durations and repetitions of all given sound stimuli.…”

“…For instance, when three sounds were successively presented, the middle sound's onset had to be presented earlier in the short (marker)–long (target interval)–short (marker) than in the long (marker)–short (target interval)–long (marker) sequence so that participants perceived the three sounds as presented at equal inter‐onset intervals. The effects of sound length seemed to disappear when the presentation of intervals was repeated (Kuroda et al, ). This line of evidence indicates that discrimination of the target interval is not decided by a simple summation of the effects of the preceding and of the following marker's length.…”