Inter-onset Intervals Research Articles

The Effects of Stimulus Rate and Tapping Rate on Tapping Performance

when finger taps are synchronized with an isochronous click, it is known that tap-click asynchrony and its variability increase with the interonset interval (IOI). It remains unclear whether these results are due to the IOI or the intertap interval (ITI) duration. The present study examines how these two factors influence tapping performance by altering the tap-click ratio (i.e., 1:n tapping). It has been shown that holding the ITI constant while decreasing the IOI—so that extra clicks subdivide each tap—results in a reduction of tapping variability, described as a subdivision benefit (Repp, 2003). Two questions remain: Does asynchrony and variability increase with the ITI while holding the IOI constant? Does asynchrony decrease with the IOI while holding ITI constant? Using linear regression, both asynchrony and variability decreased with the IOI, with little additional effect of ITI. In contrast, when using ITI as a predictor, the contribution of IOI was significant, suggesting that IOI is the main determinant of tapping performance. In addition, an ANOVA revealed a disadvantage for 1:3 tapping, supporting a categorical distinction between duple and triple meters since 1:n tapping can engender the subjective feel of different metric structures.

Modification of simple beat tracking algorithm

A new efficient modification of the known heuristic algorithm for real-time musical beat tracking is suggested. The modification uses an improved formula for updating relative frequencies of inter-onset time intervals onsets. The algorithm demonstrates a good efficiency when tested on the MIREX Beat Tracking evaluation set.

Tapping in Synchrony With a Perturbed Metronome: The Phase Correction Response to Small and Large Phase Shifts as a Function of Tempo

ABSTRACT When tapping is paced by an auditory sequence containing small phase shift (PS) perturbations, the phase correction response (PCR) of the tap following a PS increases with the baseline interonset interval (IOI), leading eventually to overcorrection (B. H. Repp, 2008). Experiment 1 shows that this holds even for fixed-size PSs that become imperceptible as the IOI increases (here, from 400 to 1200 ms). Earlier research has also shown (but only for IOI = 500 ms) that the PCR is proportionally smaller for large than for small PSs (B. H. Repp, 2002a, 2002b). Experiment 2 introduced large PSs and found smaller PCRs than in Experiment 1, at all of the same IOIs. In Experiments 3A and 3B, the author investigated whether the change in slope of the sigmoid function relating PCR and PS magnitudes occurs at a fixed absolute or relative PS magnitude across different IOIs (600, 1000, 1400 ms). The results suggest no clear answer; the exact shape of the function may depend on the range of PSs used in an experiment. Experiment 4 examined the PCR in the IOI range from 1000 to 2000 ms and found overcorrection throughout, but with the PCR increasing much more gradually than in Experiment 1. These results provide important new information about the phase correction process and pose challenges for models of sensorimotor synchronization, which presently cannot explain nonlinear PCR functions and overcorrection.

The role of temporal prediction abilities in interpersonal sensorimotor synchronization

Musical ensemble performance is a form of joint action that requires highly precise yet flexible interpersonal action coordination. To maintain synchrony during expressive passages that contain tempo variations, musicians presumably anticipate the sounds that will be produced by their co-performers. Our previous studies revealed that individuals differ in their ability to predict upcoming event timing when finger tapping in synchrony with tempo-changing pacing signals (i.e., the degree to which inter-tap intervals match vs. lag behind inter-onset intervals in the pacing signal varies between individuals). The current study examines the influence of these individual differences on synchronization performance in a dyadic tapping task. In addition, the stability of individual prediction tendencies across time is tested. Individuals with high or low prediction tendencies were invited to participate in two experimental sessions. In both sessions, participants were asked (1) to tap alone with a tempo-changing pacing signal and (2) to tap synchronously in dyads comprising individuals with similar or different prediction tendencies. Results indicated that individual differences in prediction tendencies were stable over several months and played a significant role in dyadic synchronization. Dyads composed of two high-predicting individuals tapped with higher accuracy and less variability than low-predicting dyads, while mixed dyads were intermediate. Prediction tendencies explained variance in dyadic synchronization performance over and above individual synchronization ability. These findings suggest that individual differences in temporal prediction ability may potentially mediate the interaction of cognitive, motor, and social processes underlying musical joint action.

Age effects in discrimination of repeating sequence intervals

The study measured listener sensitivity to increments in the inter-onset intervals (IOIs) of successive 20-ms 4000-Hz tone bursts in isochronous sequences. The stimulus sequences contained two-six tone bursts, separated equally by silent intervals, with tonal IOIs ranging from 25 to 100 ms. Difference limens (DLs) for increments of the tonal IOIs were measured to assess listener sensitivity to changes of sequence rate. Comparative DLs were also measured for increments of a single interval located within six-tone isochronous sequences with different tone rates. Listeners included younger normal-hearing adults and two groups of older adults with and without high-frequency sensorineural hearing loss. The results, expressed as Weber fractions (DL/IOI), revealed that discrimination improved as the sequence tone rate decreased and the number of tonal components increased. Discrimination of a single sequence interval also improved as the number of sequence components increased from two to six but only for brief intervals and fast sequence rates. Discrimination performance of the older listeners with and without hearing loss was equivalent and significantly poorer than that of the younger listeners. The discrimination results are examined and discussed within the context of multiple-look mechanisms and possible age-related differences in the sensory coding of signal onsets.

The Edit Distance as a Measure of Perceived Rhythmic Similarity

The 'edit distance' (or 'Levenshtein distance') measure of distance between two data sets is defined as the minimum number of editing operations - insertions, deletions, and substitutions - that are required to transform one data set to the other (Orpen and Huron, 1992). This measure of distance has been applied frequently and successfully in music information retrieval, but rarely in predicting human perception of distance. In this study, we investigate the effectiveness of the edit distance as a predictor of perceived rhythmic dissimilarity under simple rhythmic alterations. Approaching rhythms as a set of pulses that are either onsets or silences, we study two types of alterations. The first experiment is designed to test the model's accuracy for rhythms that are relatively similar; whether rhythmic variations with the same edit distance to a source rhythm are also perceived as relatively similar by human subjects. In addition, we observe whether the salience of an edit operation is affected by its metric placement in the rhythm. Instead of using a rhythm that regularly subdivides a 4/4 meter, our source rhythm is a syncopated 16-pulse rhythm, the son. Results show a high correlation between the predictions by the edit distance model and human similarity judgments (r = 0.87); a higher correlation than for the well-known generative theory of tonal music (r = 0.64). In the second experiment, we seek to assess the accuracy of the edit distance model in predicting relatively dissimilar rhythms. The stimuli used are random permutations of the son's inter-onset intervals: 3-3-4-2-4. The results again indicate that the edit distance correlates well with the perceived rhythmic dissimilarity judgments of the subjects (r = 0.76). To gain insight in the relationships between the individual rhythms, the results are also presented by means of graphic phylogenetic trees.

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

This study assessed motor limits of regular tapping, timing error detection, and correction in 60 participants aged from 19 to 98 years. Rate limitations on motor production were estimated from the average inter-tap interval when tapping as fast as possible for 30 s. Timing error detection required participants to judge whether a sound sequence presented at a slow, intermediate, or fast speed contained an irregularity because of phase shift. This was performed with or without synchronizing to the sounds. On the basis of the just-detectable positive phase shift (JND), participants synchronized with sequences containing phase shifts that were subliminal, just detectable or supraliminal. On average, JNDs were 9% of the inter-onset interval and by and large were not affected by synchronization tapping. Speed of error correction was estimated from the number of tones to return within 20% of the preshift synchronization error. Consistent with previous findings of motor slowing with aging, the fastest inter-tap interval increased with age. However, there was no age-related decline in JNDs or speed of error correction, both of which reflect predictive abilities for intervals within the motor repertoire of human adults. These results point towards intact timing error processing up to an advanced age. In assessing timing abilities in the brain of older adults, it is important to differentiate between motor slowing and its impact on rhythmic behavior (e.g., walking pace) from anticipatory mechanisms ('what to expect when') and how these are used to adjust the timing of actions ('what to do when').

Strong anticipation: Multifractal cascade dynamics modulate scaling in synchronization behaviors

Previous research on anticipatory behaviors has found that the fractal scaling of human behavior may attune to the fractal scaling of an unpredictable signal [Stephen DG, Stepp N, Dixon JA, Turvey MT. Strong anticipation: Sensitivity to long-range correlations in synchronization behavior. Physica A 2008;387:5271–8]. We propose to explain this attunement as a case of multifractal cascade dynamics [Schertzer D, Lovejoy S. Generalised scale invariance in turbulent phenomena. Physico-Chem Hydrodyn J 1985;6:623–5] in which perceptual-motor fluctuations are coordinated across multiple time scales. This account will serve to sharpen the contrast between strong and weak anticipation: whereas the former entails a sensitivity to the intermittent temporal structure of an unpredictable signal, the latter simply predicts sensitivity to an aggregate description of an unpredictable signal irrespective of actual sequence. We pursue this distinction through a reanalysis of Stephen et al.’s data by examining the relationship between the widths of singularity spectra for intertap interval time series and for each corresponding interonset interval time series. We find that the attunement of fractal scaling reported by Stephen et al. was not the trivial result of sensitivity to temporal structure in aggregate but reflected a subtle sensitivity to the coordination across multiple time scales of fluctuation in the unpredictable signal.

No Sustained Sound Illusion in Rhythmic Sequences

Recent research has shown that subdivision of intervals between beats makes the beat tempo seem slower—a "divided time illusion" (DTI) in music. Another temporal illusion described in the psychophysical literature is that a sustained sound seems longer than a silent interval of the same duration. This "sustained sound illusion" (SSI) may be due to acceleration of an internal pacemaker by continuous sound, or it may result from slower perception of sound offsets than of sound onsets. Experiment 1 tested the pacemaker acceleration hypothesis in a rhythmic context by asking musicians to compare or reproduce the tempi of isochronous tone sequences played legato ("filled") or staccato ("unfilled"). There was no indication that legato sequences were perceived as slower than staccato sequences. Experiment 2 tested the delayed offset perception hypothesis by asking musicians to judge the relative time of occurrence of abrupt or decaying tone offsets in the interonset intervals of isochronous sequences. There was no evidence of delayed perception of abrupt offsets, and decaying offsets were perceived only slightly late. These results suggest that the SSI, unlike the DTI, does not occur in rhythmic contexts and thus is probably not of musical relevance. More generally, the results challenge some proposed explanations of this illusion and call for further research on the conditions under which it does occur.

No Sustained Sound Illusion in Rhythmic Sequences

Recent Research has shown that subdivision of intervals between beats makes the beat tempo seem slower—a "divided time illusion" (DTI) in music. Another temporal illusion described in the psychophysical literature is that a sustained sound seems longer than a silent interval of the same duration. This "sustained sound illusion" (SSI) may be due to acceleration of an internal pacemaker by continuous sound, or it may result from slower perception of sound offsets than of sound onsets. Experiment 1 tested the pacemaker acceleration hypothesis in a rhythmic context by asking musicians to compare or reproduce the tempi of isochronous tone sequences played legato ("filled") or staccato ("unfilled"). There was no indication that legato sequences were perceived as slower than staccato sequences. Experiment 2 tested the delayed offset perception hypothesis by asking musicians to judge the relative time of occurrence of abrupt or decaying tone offsets in the interonset intervals of isochronous sequences. There was no evidence of delayed perception of abrupt offsets, and decaying offsets were perceived only slightly late. These results suggest that the SSI, unlike the DTI, does not occur in rhythmic contexts and thus is probably not of musical relevance. More generally, the results challenge some proposed explanations of this illusion and call for further research on the conditions under which it does occur.

MMN evidence for asymmetry in detection of IOI shortening and lengthening at behavioral indifference tempo

Most behavioral investigations indicated an indifference interval of 500–700ms, at which the detection of inter-onset interval (IOI) shortening and lengthening are equally difficult and no perceptual bias would be expected. However, some event-related potential (ERPs) studies showed that even at this behavioral indifference time, the detection of shortening and lengthening might rely on different aspects of information processing. This work was aimed to investigate whether the pre-attentive processing of local tempo perturbations, i.e., IOI shortening and lengthening, are different when the basic tempo is at the behavioral indifference area. Tempo perturbations were introduced by shortening or lengthening the third IOI by 10% of the base IOI of the 5-beat isochronous sequence. ERPs recorded in a passive experiment showed that both tempo perturbations elicited a distinct frontal mismatch negativity (MMN). The low resolution electromagnetic tomography (LORETA) source estimation indicated a left prefrontal predominance activity around the MMN peak, implicating an important role of the frontal lobe in the processing of local tempo perturbations. Statistical analysis revealed that the MMN to IOI shortening had an earlier and greater peak than that to IOI lengthening, implying that IOI shortening might be more easily to be detected than IOI lengthening even at indifference tempo. Our results suggested that local IOI perturbations at behavioral indifference area have an asymmetric effect on the pre-attentive processing of temporal variation detection.

F14 Speeded tapping assesses progression of huntington's disease within one year—results from the track-HD study

BackgroundSpeeded (ST) and metronome (MT) tapping have been shown to be sensitive in distinguishing premanifest (preHD) and manifest (HD) stages of Huntington's disease (HD) in the cross-sectional analysis of the...

Spatiotemporal Relations and Movement Trajectories in Visuomotor Synchronization

This work investigates how spatial and temporal aspects of rhythmic movements are affected by spatial and temporal components of a visual pacing sequence. Participants synchronized finger taps with three visual pacing sequences (Flash, High Finger, and Low Finger) at two tempi (500 ms and 400 ms interonset interval). The Flash sequence contained only temporal information, whereas the two Finger sequences contained apparent motion of different amplitudes. Participants' finger movements were recorded with a motion capture system, and movement trajectories and timing accuracy were analyzed. Results indicated that apparent motion facilitates visuomotor synchronization accuracy, which likely stems from tight perception-action links. Stimulus amplitude modulated tap amplitudes in the same direction, but this spatial assimilation did not adversely affect timing accuracy. Flexion times toward the target were significantly shorter than extension or dwell times, and could indicate a relatively ballistic movement trajectory. Local deviations of tap timing correlated with the movement trajectories from the preceding and following movement cycle. For example, after a late tap, the following movement cycle had lower amplitude and shorter extension and dwell times. This could signify the workings of error correction mechanisms that ensure stable synchronization.

Discrimination of repetitive intervals by younger and older listeners.

The study measured listener sensitivity to increments in the inter-onset intervals (IOIs) separating successive 20-ms 4000-Hz tone bursts in isochronous sequences. Stimulus sequences contained from 2–6 tone bursts, separated equally by IOIs in the range of 25–100 ms across stimulus conditions. Difference limens (DLs) for increments of all tonal IOIs were measured to assess listener sensitivity to changes in sequence rate. A DL was also measured for increments of a single IOI located at a fixed position within 6-tone sequences. Listeners included younger and older normal-hearing adults and older adults with high-frequency hearing loss. The results revealed that the relative DLs for sequence rate decreased as the magnitude of the reference IOI and the number of sequence components increased. The relative DL for a single interval embedded within a six-tone sequence was smaller than corresponding DLs measured with two-tone sequences, but only for brief reference IOIs. The discrimination performance of the older listeners was poorer than that of the younger listeners, especially for two-tone sequences with the shortest reference IOIs. The findings are interpreted within the context of multiple-look mechanisms and possible age-related differences in the sensory coding of signal onsets. [Research supported by the National Institute on Aging, NIH.]

Age-related differences in discrimination of temporal intervals in accented tone sequences

The study measured listener sensitivity to increments of a target inter-onset interval (IOI) embedded within isochronous tone sequences that featured a single accented tonal component. The sequences consisted of six 1000-Hz tone bursts separated by silent intervals to establish equal tonal IOIs of 200 ms within the sequence. Tone burst durations within the sequences were 50 ms, except one tone had a longer duration of 100 ms to produce a perception of accent. Duration DLs in ms for increments of a single sequence IOI were measured adaptively by adjusting the duration of the silent interval between two tones. Sequence position of the target IOI differed across conditions. Listeners included young normal-hearing adults and older adults with and without hearing loss. Discrimination performance of the two older listener groups was equivalent and significantly poorer than that of the younger listeners in each discrimination condition. The age-related discrimination deficits were independent of sequence locations of both the target interval and the accented tonal component. Comparative DLs collected for target intervals in unaccented tone sequences with equal tone durations revealed that the detrimental effects of accent on temporal discrimination were primarily restricted to the older listeners.

What/when causal expectation modelling applied to audio signals

A causal system to represent a stream of music into musical events, and to generate further expected events, is presented. Starting from an auditory front-end that extracts low-level (i.e. MFCC) and mid-level features such as onsets and beats, an unsupervised clustering process builds and maintains a set of symbols aimed at representing musical stream events using both timbre and time descriptions. The time events are represented using inter-onset intervals relative to the beats. These symbols are then processed by an expectation module using Predictive Partial Match, a multiscale technique based on N-grams. To characterise the ability of the system to generate an expectation that matches both ground truth and system transcription, we introduce several measures that take into account the uncertainty associated with the unsupervised encoding of the musical sequence. The system is evaluated using a subset of the ENST-drums database of annotated drum recordings. We compare three approaches to combine timing (when) and timbre (what) expectation. In our experiments, we show that the induced representation is useful for generating expectation patterns in a causal way.

A FEATURE SELECTION APPROACH FOR AUTOMATIC MUSIC GENRE CLASSIFICATION

In this paper we present an analysis of the suitability of four different feature sets which are currently employed to represent music signals in the context of the automatic music genre classification. To such an aim, feature selection is carried out through genetic algorithms, and it is applied to multiple feature vectors generated from different segments of the music signal. The feature sets used in this paper, which encompass time-domain and frequency-domain characteristics of the music signal, comprise: short-time Fourier transform, Mel frequency cepstral coefficient, beat-related features, pitch-related features, inter-onset interval histogram coefficients, rhythm histograms and statistical spectrum descriptors. The classification is based on the use of multiple feature vectors and an ensemble approach, according to time and space decomposition strategies. Feature vectors are extracted from music segments from the beginning, middle and end parts of the music signal (time-decomposition). Despite music genre classification being a multi-class problem, we accomplish the task using a combination of binary classifiers, whose results are merged to produce the final music genre label (space decomposition). Experiments were carried out on two databases: the Latin Music Database, which contains 3,227 music pieces categorized into ten musical genres; the ISMIR'2004 genre contest database which contains 1,458 music pieces categorized into six popular western musical genres. The experimental results have shown that the feature sets have different importance according to the part of the music signal from where the feature vectors are extracted. Furthermore, the ensemble approach provides better results than the individual segments in most cases. For high-dimensional feature sets, the feature selection provides a compact but discriminative feature subset which has an interesting trade-off between classification accuracy and computational effort.

Neuroticism is correlated with drift in serial time interval production

Low neuroticism and high intelligence are both related to lower intertrial variability in reaction time tasks. However, intelligence and neuroticism are weakly related traits, which suggests that they may be related to different sources of timing variability. The relation between intelligence and timing variability has recently been investigated using isochronous serial interval production (ISIP). This is a simple, automatic timing task where participants first synchronize movements with an isochronous sound sequence and then continue with self-paced production of a sequence of intervals with the same inter-onset interval (IOI). For all IOIs, local interval-to-interval variability correlated strongest with intelligence. The purpose of the present study was to test whether neuroticism, in contrast, is related to the non-local component of ISIP variability, i.e. drift or gradual changes in response IOI. We found a significant correlation of r = 0.42 between drift and neuroticism, thereby confirming the hypothesis. We suggest that this finding reflects that individuals high on neuroticism have more frequent slips in top–down cognitive control mechanisms. These cognitive failures may in turn interfere with the processing of previously produced intervals in short-term memory, which gives an unstable IOI in the ISIP task, i.e. drift.

Auditory feedback affects the long-range correlation of isochronous serial interval production: support for a closed-loop or memory model of timing

Long-range dependence is a characteristic property of successively produced time intervals, such as in un-paced or continuation tapping. We hypothesise in the present paper that serial dependence in such tasks could be related to a closed-loop regulation process, in which the current interval is determined by preceding ones. As a consequence, the quality of sensory feedback is likely to affect serial dependence. An experiment with human participants shows that diminished sensory information tends to increase the Hurst exponent for short inter-onset intervals and tends to decrease it for long intervals. A simulation shows that a simple auto-regressive model, whose order depends on the ratio between the inter-onset interval and an assumed temporal integration span, is able to account for most of our empirical results, including the duration specificity of long-range correlation.

Correlations between intelligence and components of serial timing variability

Psychometric intelligence correlates with reaction time in elementary cognitive tasks, as well as with performance in time discrimination and judgment tasks. It has remained unclear, however, to what extent these correlations are due to top–down mechanisms, such as attention, and bottom–up mechanisms, i.e. basic neural properties that influence both temporal accuracy and cognitive processes. Here, we assessed correlations between intelligence (Raven SPM Plus) and performance in isochronous serial interval production, a simple, automatic timing task where participants first make movements in synchrony with an isochronous sequence of sounds and then continue with self-paced production to produce a sequence of intervals with the same inter-onset interval (IOI). The target IOI varied across trials. A number of different measures of timing variability were considered, all negatively correlated with intelligence. Across all stimulus IOIs, local interval-to-interval variability correlated more strongly with intelligence than drift, i.e. gradual changes in response IOI. The strongest correlations with intelligence were found for IOIs between 400 and 900 ms, rather than above 1 s, which is typically considered a lower limit for cognitive timing. Furthermore, poor trials, i.e. trials arguably most affected by lapses in attention, did not predict intelligence better than the most accurate trials. We discuss these results in relation to the human timing literature, and argue that they support a bottom–up model of the relation between temporal variability of neural activity and intelligence.