The Perceptual Onset of Musical Tones

The point of P-centres

“…This is a full P-centre setting experiment and a vowel that does not vary in spectral content of amplitude over time, and thus this eect is arguably the demonstration of duration eects alone. In addition, these results are completely congruent with those from the non-speech literature (e.g., Vos & Rasch, 1981) which relate perceptual beats of signals solely to attributes of the onsets of those signals. An exception is the work of Terhardt and SchuÈ tte (1976), who established both eects of onset characteristics and duration of a signal on the``equability'' of tone sequences.…”

“…The P-centre of an event is intrinsically linked to the timing and method of production of that acoustic event, be that an articulatory event, (Fowler, 1979), or a musical tone (Gordon, 1987;Rasch, 1979). Models of acoustic P-centres which relate to onset events alone (Gordon, 1987;Vos & Rasch, 1981;Scott, 1997) explicitly seek to capture this. P-centres are strong candidates for the basis of the perceptual experience of regularity: what needs to be regular in the perception and production of sequences of speech or other discrete events.…”

“…Models of P-centres in musical tones identify the onset of the signal as the principle P-centre determinant (Gordon, 1987;Vos & Rasch 1981, Terhardt & SchuÈ tte, 1976, and Gordon relates this explicitly to the manner in which the sound is made, for example, bowed or plucked. Thus, both when and how the signal is produced in¯uences the P-centre location as a discrete onset event, and such an event-based approach is termed a``local model'' (Lame, 1990).…”

“…Early work by Terhardt and SchuÈ tte had demonstrated eects of both parameters, with onset characteristics and duration aecting the regularity of acoustic sequences (SchuÈ tte, 1978a, b;Terhardt and SchuÈ tte, 1976). Non-speech models (Gordon, 1987;Vos & Rasch, 1981) relate the P-centre to the onset only, with longer signal rise times leading to later P-centres. Models based on speech signals (e.g., Marcus, 1981) consider the role of the duration of segments of the whole signal (con®ned to monosyllabic structures).…”

“…Variations of the dynamic rhythm setting task have been used by a variety of researchers (e.g., Vos & Rasch, 1981;Gordon, 1987;Pompino-Marschall, 1989;Cooper et al, 1986), but few have used the complete task with full comparisons of all stimuli. This is a lengthy procedure but eliminates confounds due to contextual perceptual variation, literally by making all possible comparisons.…”