Dirk-Jan Povel

Perception of Temporal Patterns

To gain insight into the internal representation of temporal patterns, we studied the perception and reproduction of tone sequences in which only the tone-onset intervals were varied. A theory of the processing of such sequences, partly implemented as a computer program, is presented. A basic assumption of the theory is that perceivers try to generate an internal clock while listening to a temporal pattern. This internal clock is of a flexible nature that adapts itself to certain characteristics of the pattern under consideration. The distribution of accented events perceived in the sequence is supposed to determine whether a clock can (and which clock will) be generated internally. Further it is assumed that if a clock is induced in the perceiver, it will be used as a measuring device to specify the temporal structure of the pattern. The nature of this specification is formalized in a tentative coding model. Three experiments are reported that test different aspects of the model. In Experiment 1, subjects reproduced various temporal patterns that only differed structurally in order to test the hypothesis that patterns more readily inducing an internal clock will give rise to more accurate percepts. In Experiment 2, clock induction is manipulated experimentally to test the clock notion more directly. Experiment 3 tests the coding portion of the model by correlating theoretical complexity of temporal patterns based on the coding model with complexity judgments. The experiments yield data that support the theoretical ideas.

Accents in equitone sequences.

Equitone sequences are defined as sequences of tones that are identical in all respects: frequency, spectral composition, intensity, and duration. The only parameter varied in these sequences is the time-interval between tones. In such sequences, clear accents are perceived. This paper describes accent perception in equitone sequences containing two alternating intervals; such sequences are perceived as consisting of repeating groups of two tones. An accent is heard on the first tone of a group if the difference between the intervals is about 5% to 10%. If the difference is made bigger, the accent is heard on the last tone of the group; this latter accent is considerably stronger than the accent previously heard on the first tone. In a number of experiments, the conditions under which the two types of accents occur were investigated. From these experiments, it was tentatively concluded that the accent on the last tone is heard because that tone, since it is followed by a longer interval, can be processed more completely. This “intervalproduced” accent indeed occurs only if the between-group interval is considerably longer than the within-group interval and if the latter does not exceed a duration of about 250 msec. The effect is slightly dependent on tone duration. The interval-produced accent can be balanced if the nonaccented tone is increased by about 4 dB in intensity. This shows that the effect is quite robust. The specific type of accentuation reported here might explain some rhythmical phenomena, examples of which are given.

Structural factors in patterned finger tapping

Abstract The relationship between the internal representation and the performance of serial tasks is studied. An experiment is reported in which subjects tapped as fast as possible serial ‘finger patterns’: sequences of taps made with different fingers. The patterns are assumed to be represented internally as hierarchies of operations. The latencies between successive taps are predicted from a production model — the Tree Traversal interpreter — which specifies the necessary actions to produce the finger pattern from a hierarchical memory representation. It was found that varying the internal representation of one and the same sequence of taps affects the performance in a predictable way. The relation of the proposed model to general theories of complex motor behavior is discussed.

A theoretical framework for rhythm perception.

SummaryThis study relates to the perception of simple rhythmical patterns. A theoretical framework is presented that aims at predicting the perceived organization, the judged complexity, and the experienced rhythmical value of temporal sequences. Two simple assumptions form the basis of the framework. The notion of the ‘temporal grid’ is proposed to specify the temporal structure of a sequence. Such a grid is a time scale consisting of isochronic intervals. Since a rhythmical pattern generally allows for several different possible grids, an ‘economy principle’ is employed for selecting the most efficient grid. Economy of description is determined by the number of tones covered by the grid and the ease of specifying the noncovered tones. Since this basic model cannot explain all relevant phenomena, it is extended in order to incorporate three additional factors, namely the starting point of a sequence, subjective accents, and tempo.

Temporal structure of performed music: Some preliminary observations

Abstract After some preliminary remarks concerning the scope and aim of our study, the present paper describes the temporal analysis of a piece of music (the first prelude of J.S. Bachs Wohltemperiertes Clavier) as performed by three professional musicians. A number of temporal characteristics are derived that possibly play a role in the perception and appreciation of musical performances.

Speech planning and execution in stutterers

Abstract In this study, the maximum speaking rates of 19 stutterers and 19 nonstutterers were measured for three speech conditions: silent, lipped, and overt. Two types of stimulus sentences were used: tongue twisters and matched control sentences. The data show that stutterers are slower than nonstutterers for each combination of stimulus type and speech condition. The difference between stutterers and nonstutterers is larger for lipped speech than for silent speech and is strongest in the overt condition. These results suggest that speech planning is impaired in stutterers. Speech execution may be independently affected, or, alternatively, the planning impairment may have stronger repercussions with actual speech motor execution.

The effect of segmental and suprasegmental corrections on the intelligibility of deaf speech.

Three experiments were conducted to study the effect of segmental and suprasegmental corrections on the intelligibility and judged quality of deaf speech. By means of digital signal processing techniques, including LPC analysis, transformations of separate speech sounds, temporal structure, and intonation were carried out on 30 Dutch sentences spoken by ten deaf children. The transformed sentences were tested for intelligibility and acceptability by presenting them to inexperienced listeners. In experiment 1, LPC based reflection coefficients describing segmental characteristics of deaf speakers were replaced by those of hearing speakers. A complete segmental correction caused a dramatic increase in intelligibility from 24% to 72%, which, for a major part, was due to correction of vowels. Experiment 2 revealed that correction of temporal structure and intonation caused only a small improvement from 24% to about 34%. Combination of segmental and suprasegmental corrections yielded almost perfectly understandable sentences, due to a more than additive effect of the two corrections. Quality judgments, collected in experiment 3, were in close agreement with the intelligibility measures. The results show that, in order for these speakers to become more intelligible, improving their articulation is more important than improving their production of temporal structure and intonation.

On The Relation among Speech Errors, Disfluencies, and Self-Repairs:

In this study the relationship between speech errors (deviations from a speech plan), disfluencies (interruptions in the execution of a speech plan), and self-repairs (corrections of speech errors) was examined. Two hypotheses were formulated: Either disfluencies are special types of speech errors, or they resemble self-repairs (i.e., they are corrective actions applied to anticipated, internal errors). To test these two hypotheses, patterns of speech errors, disfluencies, and self-repairs were compared in a task in which speakers recited stimulus sentences four times in succession under time pressure. Subjects in one condition were explicitly instructed to pay close attention to accuracy of speaking. in another condition subjects were told that speech accuracy was not important. A much lower speech error rate was found in the higher-accuracy condition, but rates of disfluencies and self-repairs did not differ significantly between the two accuracy conditions. This is regarded as support for the self-repair account of disfluencies. When accuracy of speaking is stressed, speakers tend to avoid and repair speech errors at the cost of reduced speech fluency.

Stuttering as an attentional phenomenon

Abstract In a dual-task paradigm in which subjects performed a tracking task in combination with speech tasks of varying complexity, we tested two hypotheses concerning stuttering involving the concept of attention. The “overload” hypothesis assumes that stuttering is related to insufficient automatization of the speech production process, which consequently relies too much on attentional control. The “regression” hypothesis, in contrast, presupposes that speech production is normally automatized, but assumes that under certain circumstances the stutterer regresses to a controlled way of speech production, which results in stuttering. Eleven stutterers and 11 nonstutterers participated in the experiment. In general, stutterers showed fewer disfluencies while simultaneously speaking and performing a tracking task than they showed when performing the speech tasks only. This effect was strongest when spontaneous speech was to be generated. This result supports the “regression” hypothesis. However, the finding that a more complex speech task yielded more disfluencies supports the “overload” hypothesis.

The influence of height and key on the perceptual similarity of transposed melodies

In two experiments, the perceptual similarity between a strong tonal melody and various transpositions was investigated using a paradigm in which listeners compared the perceptual similarity of a melody and its transposition with that of the same melody and another transposition. The paradigm has the advantage that it provides a direct judgment regarding the similarity of transposed melodies. The experimental results indicate that the perceptual similarity of a strong tonal melody and its transposition is mainly determined by two factors: (1) the distance on the height dimension between the original melody and its transposition (pitch distance), and (2) the distance between keys as inferred from the circle of fifths (key distance). The major part of the variance is explained by the factor pitch distance, whereas key distance explains only a small part.