Maria Dolores de la Rosa and Karin M. Bausenhart
Previous studies demonstrated that perceived duration of visual intervals is strongly influenced by conflicting auditory intervals. However, it remains unclear which mechanisms underlie this multimodal integration of interval duration. To investigate this issue, we employed a reproduction task with empty (Experiment 1) or filled (Experiment 2) intervals, and a paired-comparison task (Experiment 3) to assess perceived duration of visually marked intervals, which could be accompanied by auditory marked intervals with congruent (same) or conflicting (longer and shorter) durations. First, we predicted that conflicting auditory intervals would bias the perceived duration of visual intervals towards the duration of the auditory ones. Second, according to pacemaker-accumulator models, two different mechanisms might contribute to multimodal interval integration: changes in pacemaker rate or changes of the switch component based on temporal ventriloquism effects. In the former case, the multimodal bias effect should increase with increasing interval duration. In the latter case, the effect should remain constant across interval durations. All experiments showed a strong influence of auditory interval duration on perceived visual duration. In Experiments 1 and 2, this bias effect unexpectedly decreased with increasing interval duration, which might be due to limitations of the employed duration reproduction method. In Experiment 3, however, the observed multimodal bias effect remained clearly constant across all interval durations. This finding supports the idea that multimodal integration of conflicting time intervals is mainly determined by a temporal ventriloquism effect, which affects the switch component of a pacemaker-accumulator mechanism. In addition, we demonstrated that bimodal congruent intervals are perceived as longer than unimodal visual ones. For filled (Experiment 2), but not for empty (Experiment 1) intervals, this effect seems to be caused by an increase in pacemaker rate.
Maria D. de la Rosa and Karin M. Bausenhart
In studies on temporal order perception, immediate as well as sustained effects of multisensory integration have been demonstrated repeatedly. Regarding duration perception, the corresponding literature reports clear immediate effects of multisensory integration, but evidence on sustained effects of multisensory duration integration is scarce. In fact, a single study [Heron, J. et al. (2013). A neural hierarchy for illusions of time: Duration adaptation precedes multisensory integration, J. Vis. 13, 1–12.] investigated adaptation to multisensory conflicting intervals, and found no sustained effects of the audiovisual conflict on perceived duration of subsequently presented unimodal visual intervals. In two experiments, we provide independent evidence in support of this finding. In Experiment 1, we demonstrate that adaptation to audiovisual conflict does not alter perceived duration of subsequently presented visual test intervals. Thus, replicating the results of Heron et al. (2013), we observed no sustained effect of multisensory duration integration. However, one might argue that the prolonged exposure to consistent multisensory conflict might have prevented or hampered multisensory integration per se. In Experiment 2, we rule out this alternative explanation by showing that multisensory integration of audiovisual conflicting intervals is still effective after exposure to audiovisual conflict. This further strengthens the conclusion that multisensory integration of interval duration affects perception in an immediate, but not in a sustained manner.