Tempo sensitivity in auditory sequences

Abstract

Differential thresholds for 11 tempi (ranging from 100 to 1500 ms between successive onsets) were measured for four subjects using a 2AFC paradigm. In a first experiment, the number of events in the sequence was varied to test whether sensitivity is greater in regularsequences than in simple duration discrimination tasks (only two events). Relative jnd were: (1) optimal at intermediate tempi (as low as 1.5% in the range between 300–800 ms), and (2) decreased as the number of events increased (2 events=6%, 3 events=4%, 5 events=3.2%, 7 events=3%). A second experiment tested whether this higher sensitivity was due to the fact that the sequences were regular or not by measuring differential tempo thresholds for irregular sequences of five events. Globally, sensitivity for these irregular sequences was of an intermediate level between that of the simple duration task and the sequences with five regular events. The results are discussed in terms of the hypothesized ‘‘regularity detectors.’’

Timing and aging: Slowing of fastest regular tapping rate with preserved timing error detection and correction.

“…This is consistent with gapdetection results of a reduced temporal sensitivity among older listeners for brief stimuli (Abel, Krever, & Alberti, 1990;Fitzgibbons, Gordon-Salant, & Barrett, 2007;Grose, Hall, & Buss, 2006;Schneider & Hamstra, 1999). For the 600-ms IOI, they found a lower sensitivity for increased tempo than positive PS in both young and older listeners that they attributed to tempo changes providing the brain with 'multiple looks' of the same interval (Drake & Botte, 1993). In the current study, the multiple-look model would predict an even more precise representation of the repetition period as there were five to nine IOIs before each PS.…”

Temporal predictions based on a gradual change in tempo

“…Where a tempo progression was present, the duration of early intervals would differ substantially from that of the final interval; it would seem reasonable to assume that these early intervals contribute less information about the duration of the final interval than the early intervals of isochronous sequences. Drake and colleagues (Drake et al, 1992;Drake and Botte, 1993) demonstrated that even irregular sequences contribute some information to the listener to improve discrimination compared to that of single intervals with no context. It necessarily follows that when the gradual tempo change is small, some information is imparted by intervals earlier than the penultimate.…”

“…IOIs were calculated such that the final IOI in a sequence was 530 ms for sequences decreasing in tempo and 250 ms for those increasing in tempo; isochronous sequences were tested at both periodicities. The parameters were chosen to produce sequences in which all intervals were between 200 and 800 ms, an area of high temporal sensitivity (Drake et al, 1992;Drake and Botte, 1993;Mauk and Buonomano, 2004;Grondin, 2010b) below the range where counting becomes a beneficial strategy (Grondin et al, 1999), while aiming to minimize total testing time. Overall sequence tempo was randomly varied from trial to trial within a range of 6 20% (fixed within trials).…”

“…Presentation of the reference or comparison interval more than once, in particular when done in a repeated cyclic fashion, has been demonstrated to improve performance. Drake and colleagues (Drake et al, 1992;Drake and Botte, 1993) presented subjects with isochronous sequences of differing tempi, comprising one to six intervals. They found that discrimination thresholds were inversely related to the square root of the number of intervals in each sequence and proposed that duration discrimination could be viewed as a "multiple-look" process in which temporal estimates are refined by the provision of additional information.…”

Sonification design for complex work domains: Dimensions and distractors.

“…The psychoacoustic literature provides basic guidance about the sensitivity of the auditory system to changes in specific auditory dimensions (e.g., Drake, Botte, & Baruch, 1992;Handel, 1989;Hirsh, Monahan, Grant, & Singh, 1990;Moore, 1997). These data may be of limited relevance, however, to the problems of sonification design (Anderson & Sanderson, 2004;Walker & Kramer, 2004) for several reasons.…”