Cerebral Substrates for Musical Temporal Processes SÉVERINE SAMSON,a,b NATHALIE EHRLÉ,a,b AND MICHEL BAULACb aUniversité
de Lille 3, URECA, UFR de Psychologie, 59653 Villeneuve d’Ascq cedex, France bEpilepsy
Unit, Salpêtrière Hospital and Laboratoire de Neurosciences Cognitives et Imagerie (CNRS UPR 640), Paris, France
ABSTRACT: Music as well as language consists of a succession of auditory events in time, which require elaborate temporal processing. Although several lines of evidence suggest that the left dominant hemisphere is predominantly involved in the processing of rapid temporal changes of speech, very little is known about the cerebral substrates underlying such auditory temporal processes in music. To investigate this issue, we examined epileptic patients with either left (LTL) or right (RTL) temporal lobe lesions as well as normal control subjects (NC) in two different tasks involving the processing of time-related (temporal) information. By manipulating the interonset interval (IOI) in a psychophysical task, as well as in a task of detection of rhythmic changes in real tunes, we studied the processing of temporal microvariations in music. The first task assessed anisochrony (or irregularity) discrimination of sequential information according to different presentation rates (between 80 and 1000 ms IOI). For all subjects, an effect of tempo was obtained; thresholds were lower for the 80 ms IOI than for longer IOIs. Furthermore, there was a specific impairment of rapid anisochronous discrimination (80 ms IOI) for LTL patients as compared to RTL and NC subjects, but no deficit was observed for longer IOIs. These findings suggest the specialization of left temporal lobe structures in processing rapid sequential auditory information. The second task involved the detection of IOI increments in familiar monodic tunes. Performance was measured for two increments (easy vs. difficult to detect according to cognitive expectation) to assess the effect of cognitive expectation using a forced-choice paradigm (changed vs. unchanged melody). The results showed that LTL patients but not RTL were impaired as compared to NC subjects in the increment detection. However, all groups showed differences between the two levels of difficulty, suggesting that top-down processing remains functional. These findings suggest that left temporal lobe structures are predominantly involved in perceiving time-related perturbations in familiar tunes as well as in isochronous sequences, extending to the musical domain findings previously reported in speech. KEYWORDS: Brain; Music; Musical temporal processes
Address for correspondence: Séverine Samson, Université de Lille 3, URECA, UFR de Psychologie, BP 149, 59653 Villeneuve d’Ascq cedex, France. Voice: +33 3 2041 6443; fax: +33 3 2041 6324.
[email protected]
166
SAMSON et al.: MUSICAL TEMPORAL PROCESSES
167
INTRODUCTION Music as well as language consists of a succession of auditory events in time, which require elaborate temporal processing. In the present study, we will consider the temporal coding of musical information that plays an important role in performing and perceiving rhythm. This time-related processing concerns a wide range of time frames that may implicate different forms of information processing. In the neuropsychological domain, simple auditory sequences and musical patterns have been used to investigate the cerebral structures underlying musical temporal processes. Studies reported in the literature will be reviewed first. Then, experimental findings that we obtained in two different studies will be reported to clarify the role of the left as opposed to the right temporal lobe structures in processing subtle temporal variations within a range of ten to hundreds of milliseconds. Temporal Processing in Simple Auditory Sequences Several studies have explored time-related processing by using simple auditory sequences. In this domain, several paradigms classically used in experimental psychology have been adapted to neuropsychology. In a seminal paper, Efron1 investigated the ability of brain-damaged patients to judge the temporal order of two tones of different frequencies separated by a silent interval. The results showed that aphasic patients with left-hemisphere lesions required longer intervals (between 140 and 400 ms) to discriminate temporal order than nonaphasic patients with right-hemisphere lesion or normal subjects (75-ms interval). It was therefore suggested that the left-hemisphere structures generally thought to be involved in language processing may also contribute to the temporal analysis of fast auditory sequential information. Subsequently, Tallal and Newcombe2 provided convergent evidence by demonstrating that selective damage to the left but not to the right hemisphere disrupted the ability to process two tones separated by a short interval (300 ms or less between the tones). Importantly, neither left- nor right-hemisphere–damaged subjects were affected when longer intervals were used. This finding has been reproduced with similar paradigms and generalized to other tasks involving gap detection or perception of simultaneity and succession.3–7 One case report of an amusic patient (H.V.) seems to contradict previous results, since a deficit in temporal order processing of rapid patterns was observed in the presence of a right-hemisphere lesion.8 However, the authors noted that the cortical lesion of this patient was associated with an underlying bilateral white matter lesion, which may have caused bilateral cortical deafferentation. Such a cerebral dysfunction, which presumably involved the left hemisphere, might have been responsible for the disruption of rapid temporal processing, thus explaining the apparent contradiction between the results. However, inconsistent results have been reported in studies investigating the perception of duration. Indeed, results of different studies have shown that impairments in this perceptual ability have been observed in patients with unilateral lesions implicating the right or the left hemisphere9 in a case of auditory agnosia following a bilateral cerebral dysfunction,10 as well as in patients with right temporal lobe lesions.11 These findings suggest that the perception of continuous signals, as opposed to discrete events, depends on different processing involving a distinct neural substrate.
168
ANNALS NEW YORK ACADEMY OF SCIENCES
Except for the perception of duration, the results previously reviewed suggest that left-hemisphere structures are predominantly involved in the processing of rapid sequential information. Recent electrophysiological results indicate that this function can be linked to auditory cortices of the left temporal lobe. By recording intracerebral evoked potentials to syllables in the right and the left human auditory cortices, Liégeois-Chauvel and her collaborators12 demonstrated a specialization of the left auditory cortex for speech perception that depends on rapid temporal coding (within a few tens of milliseconds). If the left temporal cortex is dominant for fine-grained time-related processing of language, it seems plausible to hypothesize that it would also be important for such temporal processing in music. Temporal Processing in Musical Sequences Parallel to the previously reported studies using simple auditory sequences, several studies investigated temporal processes in musical sequences. It has been suggested that temporal processing, of which musical rhythm would be an example, is best performed by the functions of the dominant left hemisphere. However, the evidence reported in the literature provides little support for this hypothesis. The few studies investigating the perception of dichotically presented stimuli in normal listeners usually report a right-ear advantage, which is supposed to reflect left-hemisphere predominance for the perception of temporally complex nonspeech stimuli,13–16 but this perceptual asymmetry has not always been obtained.17 Similarly, studies carried out in brain-damaged subjects have not systematically documented a deficit in rhythmic tasks in the presence of a left-hemisphere lesion. Although evidence supporting lefthemisphere involvement in rhythm has been reported in a few studies carried out in unilateral brain-damaged subjects,18,19 other studies have demonstrated the contribution of right-hemisphere structures and/or a bilateral cerebral involvement in rhythm discrimination.20–25 Two studies have even failed to report deficits in rhythm discrimination after unilateral temporal lobe resection.11,26 However, the use of musical stimuli whose familiarity and rhythmical complexity are extremely variable make the comparison between studies difficult. In light of these seemingly contradictory results obtained in lesion studies, it seems impossible to conclude that left-hemisphere structures are predominantly involved in musical rhythm. One functional neuroimaging study investigated the temporal processing of musical patterns,27 but interpretation of the results in this study is made difficult by methodological factors. In normal subjects, an increase of cerebral blood flow was obtained by positron emission tomography in left inferior Broca’s area and in left insula when rhythmic and pitch judgment conditions were compared. Although this finding was interpreted by the authors as evidence supporting left-hemisphere superiority for temporal processing, it seems impossible to attribute these foci of activation to a purely temporal processing. Indeed, many other nontemporal aspects related to the task demands and to the stimuli that differ between the two compared conditions might have contributed to these metabolic changes. In cognitive psychology, it has been proposed that the perception of rhythmic grouping, consisting of a sequential organization of relative durations of tones and silent intervals, should be differentiated from the perception of meter, referring to the underlying perceived beat marking off equal duration units.28–31 Neuropsychological studies have demonstrated a dissociation between these two components in-
SAMSON et al.: MUSICAL TEMPORAL PROCESSES
169
volved in subjective organization of temporal patterns,23,26,32 but this dissociation has not been systematically observed.20 Moreover, no indication in favor of a lateralized deficit in metrical processing has been reported. On the basis of this review, we can assume that the left dominant hemisphere and more specifically the temporal lobe plays an important role in the processing of simple auditory sequences requiring rapid timing variations ranging from tens to hundred of milliseconds. However, this left hemisphere superiority has not been consistently demonstrated in studies using musical sequences. Although precise timing information is not usually reported in such studies, the description of the stimuli indicates that the manipulated temporal information refers to slower changes resulting, for example, from the permutation of two notes. As previously emphasized, lefthemisphere damage affected the processing of fast but not slow temporal information, therefore explaining why left cerebral lesions do not systematically interfere with time-related judgment involved in musical sequences (which usually concerned longer temporal information greater than 200 ms33). To try to clarify the role of the left temporal lobe in fast temporal coding in music, we examined the consequences of unilateral temporal lobe dysfunction in processing rapid sequential information. In this paper, we present evidence supporting the hypothesis that left temporal lobe structures are predominantly involved in auditory temporal processes within the tens to hundreds of milliseconds time frame. For this purpose, interonset interval (IOI) was manipulated in a psychophysical task of anisochronous discrimination by using simple auditory sequences, as well as in a detection task involving rhythmic changes in real tunes. PERCEPTION OF INTERONSET INCREMENT IN SIMPLE AUDITORY SEQUENCES: ANISOCHRONOUS DISCRIMINATION As frequently suggested, an important component of the rhythmic subjective structure is its underlying isochrony (i.e., a regular beat), which corresponds to musical meter. In its simplest form, the meter can be expressed as a series of regular sounds. In the present study, we designed an experiment based on anisochronous perception, which refers to the ability to differentiate a regular from an irregular sequence of sounds. The goal of this experiment was to test the effect of tempo on anisochronous (or irregularity) discrimination in patients with unilateral temporal lobe dysfunction, using an adaptive procedure. The rate of presentation or tempo (defined by the IOIs separating the sounds in the sequence) was systematically manipulated to compare anisochronous discrimination using different tempi. The anisochronous paradigm is very adequate for looking at temporal processing. As indicated by the results of experiments carried out in normal subjects,34 anisochronous discrimination remains unaffected by duration or number of tones, but it is influenced by tempo or presentation rate. Using sequences presented at various tempi, ranging from 80 ms (fast) to 1000 ms (slow) IOI, allowed us to compare the perception of fast to slow sequential information without changing the task demands. Results obtained in normal listeners also showed that the discrimination thresholds for the anisochronous sequences were proportional to the size of the IOI for tempi between 300 and 1400 ms, whereas sensitivity deteriorated for IOIs of 80 ms as compared to slower tempi. This finding corroborates the assumption of differ-
170
ANNALS NEW YORK ACADEMY OF SCIENCES
FIGURE 1. Example of the two sequences presented at one trial in the anisochronous discrimination task.
ent processes for temporal stimuli less than or equal to 200–300 ms compared to longer ones.35–37 Furthermore, the paradigm that we designed uses an adaptive procedure that offers the opportunity to determine precise and fine individual thresholds that represent sensitive measures of perceptual abilities. In particular, the use of such a method in neuropsychology improves the chance of documenting subtle perceptual difficulties in brain-injured patients that could not be detected by classical paradigms involving fixed levels of stimulation common to all subjects. Two successive sequences of five tones with the same frequency, intensity, duration, and attack and decay time were presented on each trial (one regular and one irregular). The subject had to decide if the irregular sequence was in the first or second position (see FIG . 1). Five conditions with an IOI of 80, 300, 500, 800, or 1000 ms were prepared in order to test the effect of tempo on anisochronous discrimination. The anisochrony was introduced by delaying either the second or the fourth sound of the isochronous sequence. To obtain reliable measures of anisochronous discrimination, a psychophysical procedure developed by Levitt38 was used to determine the minimum temporal shift necessary for each individual to discriminate the regular from the irregular sequence. The size of the first shift corresponded to 10% of the base interval (i.e., 50 ms for an IOI of 500 ms). The threshold of the shift detection was determined by an adaptive procedure in which the subject’s response on one trial determined the size of the shift on the next trial. The perceptual threshold was then expressed as a percentage of the base IOI to allow comparison with other differential thresholds reported in the literature. Further details about the methods are provided in another paper.39 Eighteen patients with medically intractable epilepsy, candidates for surgical treatment at La Salpêtrière Hospital (Paris), as well as a group of normal control subjects (n = 6), participated in this study. They all presented medial temporal lobe epilepsy associated with lateralized hippocampal sclerosis as identified by magnetic resonance imaging (MRI). None of them suffered from language disturbances, and language function was lateralized in the left hemisphere for all subjects. These patients were divided into two groups: those with right hippocampal atrophy (n = 8) and those with left hippocampal atrophy (n = 10). According to the literature, we hypothesized that the thresholds of patients with left-temporal lobe lesions would be significantly higher than the thresholds of patients with right-temporal lobe lesions or normal control subjects for tempi with IOIs at or below 300 ms. By contrast, no specific deficit was predicted for slower tempi with IOIs greater than 300 milliseconds. Finally, we predicted that for all subjects the
SAMSON et al.: MUSICAL TEMPORAL PROCESSES
171
FIGURE 2. Relative thresholds of the three groups of subjects (LHA, left hippocampal atrophy; RHA, right hippocampal atrophy; NC, normal controls) for interonset intervals (IOI) of 80, 300, 500, 800, and 1000 ms. The bars represent the standard error of the mean.
thresholds obtained for the fastest tempo (80 ms) would be higher than those obtained for slower tempi (≥ 300 ms), as indicated by previous results in normal listeners. In keeping with our predictions, the results displayed in FIGURE 2 clearly show that anisochronous discrimination thresholds for the rapid tempo (80 ms IOI) were significantly higher for the patients with left temporal lobe dysfunction (mean threshold: 27.5%) than for the patients with right temporal lobe dysfunction (17.7%) and for the normal control subjects (16.4%). However, there were no differences between the groups for the slower tempi (from 300 ms to 1000 ms IOI). This finding is compatible with results of aphasic patients with left hemisphere lesion1,2,4 and provides strong evidence suggesting that fast auditory sequential information processing depends specifically on the integrity of left temporal lobe structures. In addition, an effect of tempo was systematically obtained for all the subjects. The thresholds for very rapid sequences (80 ms IOI = 21.43%) were always higher than the thresholds obtained with slower tempi, ranging from 300-ms to 1000-ms IOI (7.6% on average), for which the different thresholds remained proportional to IOI. This finding confirmed results of a previous study carried out in healthy subjects34 and is in agreement with the assumption that brief auditory information is processed differently from longer information. 35–37 In a previous paper,39 we proposed that the decreased sensitivity observed for the fastest tempo could be related to an auditory memory saturation. According to echoic memory studies40 (see Ref. 41 for a review), acoustic properties of sounds necessitate a 200- to 300-ms delay following the presentation of a tone to be adequately retained. This time delay was obviously shorter for the fastest tempo (80 ms IOI). Indeed, the interval between the two sequences to be discriminated was twice the base interval, in order to maintain isochrony across the trial. Thus, the intersequence interval for the fastest tempo (80 ms IOI) was 160 ms, whereas a minimum of 600
172
ANNALS NEW YORK ACADEMY OF SCIENCES
ms separated the sequences for the other tempi. Therefore, a saturation of echoic memory could have been responsible for the observed difference between the thresholds of the faster and slower tempi, because the intersequence interval would be too short to allow consolidation of the first sequence. If such an explanation is validated in the future, the deficit documented in patients with left temporal lobe dysfunction in fast sequential auditory perception could be interpreted as an echoic memory dysfunction. Although an eventual relationship between a temporal processing dysfunction and an auditory memory deficit in dysphasic children has already been suggested by Tallal and collaborators,42 the possible link between an echoic memory dysfunction and an anisochronous discrimination deficit of fast auditory information in patients with left temporal lobe lesion remains to be clarified. It is also possible that the decreased temporal sensitivity for the rapid tempo is related to a difficulty in determining if the irregular sequence was the first or the second one, suggesting a deficit in temporal order processing. Although this judgment was not affected when slower tempi were presented, it becomes more difficult to make such a judgment in rapid sequences (subjects sometimes having trouble separating the sequences per se). This additional constraint might have penalized patients with left temporal lobe dysfunction more strongly than the other subjects. In previous studies, rapid temporal processing deficits were usually reported in language-impaired subjects (see Ref. 43 for a review). Results of the present investigation extend this finding to patients presenting a limited left temporal lobe lesion without associated massive language disorders, indicating that time-related disturbances are not necessarily associated with language disturbances. However, the present results differ from the data obtained in language-impaired subjects, since, in this latter case, subjects presented a deficit with a slower tempo (300 ms IOI), while the subjects with left temporal lobe dysfunction did not present any difficulties in this condition. It seems that the severity of the sequential auditory deficit observed in patients with or without verbal deficit varies. Although these divergent results could be explained by methodological differences between the studies, it is also possible that the severity of the temporal deficits could be correlated to the severity of verbal deficits. The patients tested in the present study might be situated at one end of the continuum, showing limited temporal processing deficits and relatively preserved verbal abilities, while dysphasic or dyslexic subjects could be located at the other end, with more severe temporal deficits and more severe language deficits. Conversely, it is also possible that rapid sequential temporal processing can be disturbed without affecting language functions, which would suggest that they depend on distinct neural substrates. Future research will be necessary to determine whether the same underlying mechanism is responsible for the deficit observed in epileptic and dysphasic patients. PERCEPTION OF INTERONSET INCREMENT IN MUSICAL SEQUENCES: TEMPORAL MICROSTRUCTURE OF FAMILIAR TUNES The goal of this study was to extend to the musical domain the role of the left temporal lobe in processing small timing variations. As emphasized in the introduction, it seems that the perception of small timing differences within a musical sequence
SAMSON et al.: MUSICAL TEMPORAL PROCESSES
173
has not been investigated in neuropsychology. To explore this issue, we designed a task involving the detection of IOI increments in familiar monodic tunes, improving therefore the ecological validity of the task. Several descriptive studies showed that performers modify considerably the temporal structure as noted in the score when interpreting a piece of music (for review, see Ref. 44), a piece never being played exactly as it is written in the score. However, music played in this way seems perfect to the listeners, suggesting that we are expecting these temporal changes. Such temporal microvariations occur at specific locations and are supposed to reflect the subjective temporal structure of music. It has been demonstrated that, consequently, a listener is expecting these musical microvariations in perceiving musical excerpts.45 In an experimental study using classical pieces of music (microstructural expectations), Repp showed that increment detection accuracy was correlated with the temporal microstructure profiles of expert performances, suggesting that temporal increments of an interval that is usually lengthened are more difficult to detect than temporal increments of intervals for which no microvariations were introduced. In the cognitive literature, two hypotheses that are not mutually exclusive have been proposed to explain variations in the accuracy detection. A top-down hypothesis suggests that listeners’ expectations reflect expressive performance of temporal microvariations,45 whereas a bottom-up hypothesis indicates that some expectations may be due to psychoacoustical characteristics of the stimulus.44 Based on these observations, we designed an experiment involving the detection of IOI increments introduced in musical sequences to test perception of these temporal microvariations in patients with unilateral temporal lobe lesions. The musical excerpts used were very familiar in order to generate high expectations and to prevent bias due to each listeners’ musical background. In the experimental task, half of the trials consisted of the presentation of a score version (played exactly as it is written in the score) of a familiar tune, whereas the other half consisted of a modified version of the tune in which one IOI was increased by 25%, producing an increment of 10 to 140 ms (mean = 85 ms; standard deviation = 62 ms) depending on the size (or duration) of the modified interval within the tune. The subject’s task was to decide whether the presented trial corresponded to the mechanical or to the modified version of the excerpt using a two-alternative forcedchoice paradigm. Two types of IOIs were introduced. One type corresponded to expected increments, and the other one corresponded to unexpected increments. The level of expectation was determined by results of previous studies detailed elsewhere.46 Basically, the recording of pianists’ performances were compared to a score (or computerized) version of each selected tune to allow the analysis of temporal microstructure. The results of this experiment allowed the identification of one interval that was systematically lengthened in pianists’ recordings as compared to the metrical version, and another interval that was systematically preserved. IOI increments were thus introduced at these specific locations, and it was hypothesized that temporal increments located at “lengthened intervals” would be more difficult to detect than temporal increments located at “preserved intervals” since listeners would be expecting temporal increments in the first but not in the second condition. The ability to distinguish these two types of interonset increments was subsequently tested in a perceptual task, confirming therefore the relevance of this temporal manipulation in normal nonmusician listeners. Based on these results, we designed an
174
ANNALS NEW YORK ACADEMY OF SCIENCES
FIGURE 3. Detection accuracy derived from the area under the receiver operating curve (ROC) of the three groups of subjects (LTR, left temporal resection; RTR, right temporal resection; NC, normal controls) for expected and unexpected interonset increments. The bars represent the standard error of the mean.
experimental task to evaluate the ability to detect subtle temporal changes by taking into account implicit knowledge of the rhythmical structure underlying musical listening. By manipulating temporal increments and cognitive expectations, especially with regard to the temporal (or rhythmic) dimension, it was possible to assess bottom-up as well as top-down processing involved in musical rhythmic perception. Twenty-two patients who had undergone a right (n = 11) or a left (n = 11) temporal lobe resection for the relief of medically intractable epilepsy, as well as 14 normal control subjects, were tested in this experiment. None of the patients presented language disorders or suffered from extra temporal lesions. Language was lateralized on the left side in all subjects. As opposed to the previous study, the patients were tested postoperatively. The temporal resection includes the medial temporal lobe structures (hippocampal and surrounding cortex), and in some patients the excision involves the temporal pole as well. Considering the predominant contribution of left temporal lobe structures in processing rapid sequential information, we expected a deficit in detecting microtemporal variations in familiar tunes in patients with left temporal lobe lesion. In this study, the task of detection of temporal increments required no tonal judgment, for which the right temporal lobe is usually needed.47–50 We therefore postulated adequate performance for patients with right temporal lobe lesion. Moreover, an effect of level of expectation of IOI was predicted for normal subjects and for patients with right temporal lobe lesion. If this effect reflects a perceptually driven processing, it should disappear in patients with left temporal lobe lesion, otherwise it might also be present in this patient group. The results of this study are displayed in FIGURE 3, which represents detection accuracy of the three groups of subjects for expected and unexpected interonset increments. An area under the receiver operating curve (ROC) derived from the number of hits and false alarms was computed for each subject. As predicted, a significant
SAMSON et al.: MUSICAL TEMPORAL PROCESSES
175
deficit in detecting interonset increments was obtained in patients with left temporal lobe lesion as compared to subjects with right temporal lobe lesion and normal subjects, these two latter groups not being different. This deficit characterizing patients with left temporal lobe lesion cannot be attributed to difficulty in identifying the tunes since all the subjects were able to adequately recognize the melodies. The results of this study suggest that the ability to detect an interonset increment introduced in very well-known tunes depends on the integrity of left temporal lobe structures, extending to the musical domain the previously reported role of the left temporal lobe in rapid sequential processing. Moreover, the results showed that it is more difficult to detect expected than unexpected temporal increments for all the subjects, indicating, therefore, that topdown processing remains functional for the three groups of subjects. Moreover, it means that expectations of nonmusician listeners were identical to those of pianists, extending to the simple monodic familiar tunes findings reported with polyphonic classical pieces.45 Although patients with left temporal lobe lesion presented a clear deficit in perceiving temporal microvariations, they seemed to be able to generate temporal expectations to the same extent as healthy subjects. To summarize, the results of this study illustrate a dissociation between a selective deficit in fine-grained temporal sensitivity, whereas temporal expectations seem to be relatively spared in the presence of left temporal lobe lesions, suggesting that top-down (or descending) processing may function independently from bottom-up (or ascending) processing. In addition, these data indicate that top-down processing involved in the temporal perception of musical information can still be preserved despite disrupted bottom-up processing. CONCLUSION The experiments reported in this paper were designed to test the hypothesized role of left temporal lobe structures to musical temporal processing. We focused our interest on the perception of brief sequential information ranging from tens to hundreds of milliseconds by comparing results obtained with very simple auditory sequences and real musical excerpts. In keeping with our predictions, the data reported in this paper suggest that left temporal lobe structures are predominantly involved in perceiving IOI increments in familiar tunes as well as in isochronous sequences, extending to the musical domain findings previously reported in speech.12,43 The patients included in both studies had lesions located within the medial temporal lobe structures and more specifically in the hippocampus. It remains difficult to know if the deficit could be attributed to the hippocampal lesion or to a more global dysfunction of the temporal neocortical structures. Evidence from a functional cerebral imaging study (positron emission tomography with 18 fluorodeoxyglucose, FDG-PET) carried out in a similar population of epileptic subjects indicates that the temporal lobe dysfunction associated with unilateral hippocampal sclerosis is not restricted to the hippocampus but extends to include the anterior part of the temporal neocortex corresponding to the temporal pole.51 Although the structural lesion is apparently limited to the hippocampal formation, it does not mean that the connections between the medial temporal lobe structures and the neocortex or the surrounding and external temporal cortex per se remain functional. The present data does not al-
176
ANNALS NEW YORK ACADEMY OF SCIENCES
low us to differentiate the role of the hippocampus from the contribution of the adjacent cortical areas in auditory sequential information processing. However, it seems relevant to suggest that abnormal metabolism reported in these surrounding cortical structures reflects a temporal lobe dysfunction or an interruption of the hippocampal-neocortical loop that may be responsible for the deficits obtained in our patients with left temporal lobe dysfunction. In the present studies, the time-related deficit following left temporal lobe lesion was reproduced in two different tasks involving purely temporal stimuli as well as multidimensional musical sequences in the context of an adaptive or a fixed level paradigm, respectively. Despite numerous methodological differences, the results of these two studies provide strong support for the selective sensitivity of left temporal lobe structures in processing rapid time-related events. Indeed, the two methodologies require the detection of subtle temporal variations that were made possible by the use of regular sequences or very familiar tunes producing strong temporal expectations. It is therefore tempting to consider that a single mechanism, depending on the processing of very brief sequential events, is responsible for impairments obtained in both tasks. However, it is too early to assume that such impairments can be explained by a unique deficit. We already suggested that deficit in anisochronous discrimination of rapid sequences can also result from the saturation of sensory memory or from an inability to judge the temporal order. Such a hypothesis should be tested in the future to elucidate the nature of the deficit-characterizing subjects with left temporal lobe dysfunction.
ACKNOWLEDGMENTS We are also indebted to Xavier Gangand for programming the experiments. This research was supported by a grant from Contrat d’objectif Région Nord–Pas de Calais: Motricité et Cognition (France), and by GIS Dysfonctionnement de la Cognition. REFERENCES 1. EFRON, R. 1963. Temporal perception, aphasia and déjà vu. Brain 86: 403–424. 2. TALLAL, P. & F. NEWCOMBE. 1978. Impairment of auditory perception and language comprehension in dysphasia. Brain Lang. 5: 13–24. 3. LACKNER, J.R. & H.L. TEUBER. 1973. Alterations in auditory fusion thresholds after cerebral injury in man. Neuropsychologia 11: 409–415. 4. MILLS, L. & G.B. ROLLMAN. 1980. Hemispheric asymmetry for auditory perception of temporal order. Neuropsychologia 18: 41–47. 5. ROBIN, D.A., D. TRANEL & H. DAMASIO. 1990. Auditory perception of temporal and spectral events in patients with focal left and right cerebral lesions. Brain Lang. 39: 539–555. 6. SHERWIN, L. & R. EFRON. 1980. Temporal ordering deficits following anterior temporal lobectomy. Brain Lang. 11: 195–203. 7. SWISHER, L. & I.J. HIRSH. 1972. Brain damage and the ordering of two temporally successive stimuli. Neuropsychologia 10: 137–152. 8. GRIFFITHS, T.D. et al. 1997. Spatial and temporal auditory processing deficits following right hemisphere infarction. Brain 120: 785–794.
SAMSON et al.: MUSICAL TEMPORAL PROCESSES
177
9. MUSIEK, F.E., J.A. BARAN & M.L. PINHEIRO. 1990. Duration pattern recognition in normal subjects and patients with cerebral and cochlear lesions. Audiology 29: 304– 313. 10. ALBERT, M.L. et al. 1972. A case study of auditory agnosia: linguistic and non-linguistic processing. Cortex 8: 427–443. 11. MILNER, B. 1962. Laterality effects in audition. In Interhemispheric Relations and Cerebral Dominance. V. Mountcastle, Ed.: 173–201. John Hopkins University Press. Baltimore, MD. 12. LIÉGEOIS-CHAUVEL, C. et al. 1999. Specialization of left auditory cortex for speech perception in man depends on temporal coding. Cereb. Cortex 9: 484–496. 13. GORDON, H. 1978. Left-hemisphere dominance for rhythmic elements in dichoticallypresented melodies. Cortex 14: 58–70. 14. HALPERIN, Y., I. NACHSHON & A. CARMON. 1973. Shift of ear superiority in dichotic listening to temporally patterned nonverbal stimuli. J. Acoust. Soc. Am. 53: 46–50. 15. PAPÇUN, G. et al. 1974. Is the left hemisphere specialized for speech, language and/or something else? J. Acoust. Soc. Am. 55: 319–327. 16. ROBINSON, G.M. & D.J. SOLOMON. 1974. Rhythm is processed by the speech hemisphere. J. Exp. Psychol. 102: 508–511. 17. PERETZ, I. & J. MORAIS. 1980. Modes of processing melodies and ear asymmetry in non-musicians. Neuropsychologia 18: 447–489. 18. MAVLOV, L. 1980. Amusia due to rhythm agnosia in a musician with left-hemisphere damage: a non-auditory supramodal defect. Cortex 16: 331–338. 19. PRIOR, D.A., G. KINSELLA & J. GIESE. 1990. Assessment of musical processing in brain-damaged patients: implications for laterality of music. J. Clin. Exp. Neuropsychol. 12: 301–312. 20. KESTER, D. et al. 1991. Acute effect of anterior temporal lobectomy on musical processing. Neuropsychologia 29: 703–708. 21. MICHEL, F., F. PERONNET & B. SCHOTT. 1980. A case of cortical deafness: clinical and electrophysiological data. Brain Lang. 10: 367–377. 22. PENHUNE, V.B, R.J. ZATORRE & W.H. FEINDEL. 1999. The role of auditory cortex in the retention of rhythmic patterns as studied in patients with temporal lobe removals including Heschl’s gyrus. Neuropsychologia 37: 315–331. 23. PERETZ, I. 1990. Processing of local and global musical information by unilateral brain-damaged patients. Brain 113: 1185–1205. 24. REITAN, R.M. & D. WOLFSON. 1989. The Seashore rhythm test and brain functions. Clin. Neuropsychol. 3: 70–78. 25. SHAPIRO, B.E., M. GROSSMAN & H. GARDNER. 1981. Selective musical processing deficits in brain damaged populations. Neuropsychologia 19: 161–169. 26. LIÉGEOIS-CHAUVEL, C., et al. 1998. Contribution of different cortical areas in the temporal lobes to music processing. Brain 121: 1853–1867. 27. PLATEL, H. et al. 1997. The structural components of music perception. A functional anatomical study. Brain 120: 229–243. 28. DOWLING, W.J. & D. HARWOOD. 1986. Music cognition. Academic Press. New York. 29. POVEL, D.J. 1981. Internal representation of simple temporal patterns. J. Exp. Psychol.: Hum. Percept. Perform. 7: 3–18. 30. POVEL, D.J. & P. ESSENS. 1985. Perception of temporal patterns. Music Percept. 2: 411–440. 31. REPP, B.H. 1995. Detectability of duration and intensity increments in melody tones: partial connection between music perception and performance. Percept. Psychophys. 57: 1217–1232. 32. POLK, M. & A. KERTESZ. 1993. Music and language in degenerative disease of the brain. Brain Cogn. 22: 98–117. 33. FRAISSE, P. 1982. Rhythm and tempo. In The Psychology of Music. D. Deutsch, Ed.: 149–178. Academic Press. New York. 34. EHRLÉ, N. & S. SAMSON. 1996. Détection de l'irrégularité temporelle: influence de la durée et du tempo. In Sciences Cognitives, Individus et Société. Actes du 6ème Colloque de l'Association pour la Recherche Cognitive. F. Anceaux & J.M. Coquery, Eds.: 209–214. Villeneuve d'Ascq.
178
ANNALS NEW YORK ACADEMY OF SCIENCES
35. MICHON, J.A. 1964. Studies on subjective duration: I. Differential sensitivity in the perception of repeated temporal intervals. Acta Psychol. 22: 441–450. 36. NAKAJIMA, Y., G. TEN HOOPEN & R. VAN DER WILK. 1991. A new illusion of time perception. Music Percept. 8: 431–448. 37. TEN HOOPEN, G. et al. 1995. Auditory isochrony: time shrinking and temporal patterns. Perception 24: 577–593. 38. LEVITT, H. 1971. Transformed up-down methods in psychoacoustics. J. Acoust. Soc. Am. 49: 467–477. 39. EHRLÉ, N., S. SAMSON & M. BAULAC. 2001. Processing of rapid auditory information in epileptic patients with left temporal lobe damage without dysphasia. Neuropsychologia. 39: 525–531. 40. MASSARO, D.W. 1972. Preperceptual images, processing time and perceptual units in auditory perception. Psychol. Rev. 79: 124–145. 41. COWAN, N. 1984. On short and long auditory stores. Psychol. Bull. 96: 341–370. 42. TALLAL, P., R.L. SAINBURG & T. JERNIGAN. 1991. The neuropathology of developmental dysphasia: behavioral, morphological, and physiological evidence for a pervasive temporal processing disorder. Read. Writ. Interdiscip. J. 3: 363–377. 43. TALLAL, P., S. MILLER & R.H. FITCH. 1993. Neurobiological basis of speech: a case for the preeminence of temporal processing. Ann. N.Y. Acad. Sci. 682: 27–47. 44. PENEL, A. & C. DRAKE. 1998. Sources of timing variations in music performance: a psychological segmentation model. Psychol. Res. 61: 12–32. 45. REPP, B.H. 1992. Probing the cognitive representation of musical time: structural constraints on the perception of timing perturbations. Cognition 44: 241–281. 46. EHRLÉ, N. 1998. Traitement temporel de l’information auditive et lobe temporal. Ph.D. thesis. University of Reims, France. 47. SAMSON, S. & R.J. ZATORRE. 1988. Melodic and harmonic discrimination following unilateral cerebral excision. Brain Cogn. 7: 348–360. 48. JOHNSRUDE, I., B.B. PENHUNE & R.J. ZATORRE. 2000. Functional specificity in the right human auditory cortex for perceiving pitch direction. Brain 123: 155–163. 49. ZATORRE, R.J. 1988. Pitch perception of complex tones and human temporal-lobe function. J. Acoust. Soc. Am. 84: 566–572. 50. ZATORRE, R.J. & S. SAMSON. 1991. Role of the right temporal neocortex in retention of pitch in auditory short-term memory. Brain 114: 2403–2417. 51. SEMAH, F. et al. 1995. Is interictal temporal hypometabolism related to mesial temporal sclerosis? A positron emission tomography/magnetic resonance imaging confrontation. Epilepsia 36: 447–456.
