NOTE A CASE OF AMUSIA CAUSED BY THE INFARCTION OF ANTERIOR PORTION OF BILATERAL TEMPORAL LOBES Masayuki Satoh1,2, Katsuhiko Takeda2, Yasuo Murakami3, Kenji Onouchi2,3, Kiyoharu Inoue3 and Shigeki Kuzuhara1 (1Department of Neurology, Mie University School of Medicine; 2Department of Neurology, Japanese Red Cross Medical Center; 3Department of Neurology, Jikei Medical University) ABSTRACT It remains an unsettled question which brain regions participate in music perception. During singing a familiar song, the retrieval from long-term memory is necessary, but the mechanism of that retrieval is still unclear. We carried out a detailed examination of musical ability in a patient with amusia and control subjects and identified the lesion sites of our patient using MRI. Compared with controls, the patient manifested the following impairments in music perception: (i) the recognition and discrimination of familiar melodies; (ii) the discrimination of unfamiliar phrases; (iii) the discrimination of isolated chords. During singing familiar nursery songs, the patient showed the replacement of one phrase of the melody. In MRI, the patient had old infarction in the anterior portion of the temporal lobes bilaterally. In conclusion, the anterior temporal lobes participate in the perception and expression of music. During singing, the song is retrieved from long-term memory by a unit of one phrase. The dysfunction of that retrieval caused the replacement of the succeeding phrases of the original with the wrong tune, and we named this phenomenon paramelodia. Key words: music processing, amusia, temporal lobe, paramelodia INTRODUCTION Amusia is defined as an acquired deficit of music processing (Loring, 1999). Amusia is generally classified as one of two types, receptive or expressive (Wertheim, 1969). Receptive amusia specifies an inability to read or understand music, while expressive amusia indicates difficulty in singing or writing music, or in instrumental performance (Henson, 1977). In reported cases of receptive amusia, impairment of pitch and rhythm perception differed in severity, suggesting that perception of these two music components occurs independently in the brain (Mavlov, 1980). Studies of expressive amusia have shown differences in the degree to which singing and instrument playing are impaired (McFarland and Fortin, 1982), thus expressive amusia is often subdivided into vocal and instrumental motor amusia. Receptive and expressive amusia are dysfunctions of receptive and expressive systems of music processing, respectively (Wertheim and Botez, 1961). Despite the efforts, through case studies of many authors the localization of lesions that produce amusia remains uncertain. This uncertainty can be attribued to several problems in the reported literatures. First, many reports of amusia lack precise descriptions of the site of lesions, as they were carried out before the advent of neuroimaging techniques such as CT and MRI. Second there have been few reports of ‘pure’ amusia with most amusic patients reported in the literature having accompanying neuropsychological Cortex, (2005) 41, 77-83 deficits such as aphasia or apraxia. In such cases, it is difficult to determine which symptoms were caused solely by amusia. In the present report, we describe a patient with ‘pure’ amusia caused by cerebral infarction. There is evidence that the temporal lobes participate in the perception of music. Based on examinations of patients who had undergone unilateral temporal lobectomy, some authors have concluded that temporal-lobe structures might be important for the memory of unfamiliar (Zatorre, 1985; Samson and Zatorre, 1992) and familiar melodies (Zatorre and Halpern, 1993). But, with respect to the expression of music, it remained an unsettled question how music processing in the brain takes place during singing. When singing a familiar song, it is supposed that one must retrieve the song’s melody from longterm memory. Therefore, deficits in singing performance of amusic patients are possibly caused by impaired memory for melodies or by dysfunction in the retrieval of a melody from longterm memory. In order to clarify the musical abilities in which damaged brain regions participated, we carried out a detailed examination of perception and expression of music, and identified the lesion sites using MRI. Whilst singing a familiar song, our patient revealed the insertion of other melodies into an ongoing song. We analyzed the singing performance of our patient from the view-point of the similarity of the harmonic progression of original and inserted melody. 78 Masayuki Satoh and Others SUBJECTS Case Report History A 70-year-old right-handed woman had an episode of bilateral infarction of the temporal lobes 17 years prior to our study. Acutely, she suffered from right hemiparesis and aphasia, which disappeared after several months. Following her discharge, she found herself unable to recognize familiar music. For example, when she heard her recorded disc to which, before her illness, she had repeatedly listened, she felt as if she were listening to the songs for the first time. She checked the case of the disc, and ascertained that she had been used to listen to those songs. Despite her former experience as a tango dancer, she could no longer distinguish tango from other types of music. In addition she perceived her singing as out of tune, when it was not. These music misperception symptoms remained unchanged for 17 years, and her musical ability had never been tested. On June 19, 1999, the patient developed a sudden mild right hemiparesis and was admitted the Jikei Medical University Hospital in Tokyo. No sensory disturbances, aphasia, ideomotor apraxia, dyscalculia, or visual discrimination difficulties were detected. The pure-tone and speech audiometries were also within normal limits. No apparent deficit was detected in the patient’s receptive and expressive speech, assessed by two aphasic battery sub-tests; notably she obtained a normal score (164/167) on the Token test. The right hemiparesis disappeared in one week. Brain MRI confirmed the existing bilateral temporal lobe lesions, and also revealed a small new infarction at the left coronal radiation. The patient was referred to us for assessment of her impairment in musical ability. We tested her for a week. Her musical background was that of a nonmusician who enjoyed listening to popular and classical music, and regularly attended concerts before her illness. She met criteria for Grison’s third level of musical culture (Grison, 1972). She has completed 14 years of formal education. She has suffered from the arrhythmia and hypertension for twenty years. Lesion Localization Figure 1 shows the brain MRI obtained on June 22, 1999. The atlas of Duvernoy (Duvernoy, 1991) was used to estimate the localization of the lesions. The distribution and intensity of the lesions within each hemisphere were consistent with a preexisting infarction mainly involving a portion of the inferior division of the middle cerebral artery. The superior temporal gyrus and the rostral third of the middle temporal gyrus were damaged bilaterally. The Fig. 1 – Brain MRI of the patient carried out 17 years after the onset of amusia. Upper images show axial T1 weighted images and lower show coronal T2 weighted images. The left side of images shows the right side of the brain. Old infarction lesions are observed bilaterally in the temporal poles including bilateral superior tempopral gyri and the rostral third of the middle temporal gyri. The left insula is also partially damaged. Additional hypointensity was also evident in the left coronal radiation (not shown). lesions included both temporal poles. The left insula was partially damaged. Heschl’s gyrus and the caudal superior temporal gyrus were bilaterally intact. The MRI also showed a new small ischemic lesion at the left coronal radiation (not shown) consistent with the transient right hemiparesis. Thus, the preservation of the patient’s speech and language functions was likely attributable to the preservation of Wernicke’s area. Control Subjects Eight male and four female right-handed subjects with no history of neurological or psychiatric disease served as controls for the tests of musical capacity. Their age range was from 55 to 72 years old (mean: 62 years old). Control subjects were office workers in our labortaory or volunteers with musical background and education similar to those of our patient. METHODS General Procedures The patient and all control subjects gave informed consent to all tests administered. All Amusia following bilateral temporal infarction stimuli were presented in a quiet room at a loudness level that was comfortable and customary for the listeners. Recognition of Environmental Sounds Twenty-four environmental sounds were presented, and subjects were required to say what sound it meant. The materials contained the following sounds: trumpet, phone, cat, craw, drum, male voice, wind, cow, passing train, dog, female voice, car’s engine, clock, waves, music, horse, baby, English, river, firing pistol, chicken, sawing, male laughing, and footstep. PERCEPTION OF MUSIC Pitch Discrimination After presenting a pure-tone of 440Hz as the standard, we presented a second tone and measured the range in which the standard and second tones were heard as the same pitch. The first and second tones, in which the interval was two seconds, were presented for five seconds. Discrimination and Recognition of Familiar Songs Thirty familiar nursery songs were chosen and, in twenty of these songs, a part of the melody was locally altered. The alterations were limited to only one or two tones and consisted of changes in the rhythm and the pitch. Among twenty altered songs, the contour of the melody was preserved in thirteen songs. All tunes were played by piano-tones of a YAMAHA PSR-330 synthesizer and recorded in a mini-disc. Each song lasted for 10~15 seconds, and contained the whole or a half length of original tune. Two different response modes were used: (i) ‘correct-wrong’ discrimination; (ii) ‘familiarunfamiliar’ decision. 79 Meter Discrimination We used items from the meter test of “The test for the assessment of musical capacity for schoolchildren” (Mashino et al., 1979). Following the example of two pieces of music that had double or triple time, the examiner presented four pieces of music that had double or triple time. The subject was then required to answer “two” or “three” for each piece of music. Discrimination of Isolated Chords A pair of chords in which the interval was 1 second was presented and subjects were required to make a ‘same-different’ discrimination. We used items the harmony test of “the test for the assessment of musical capacity for schoolchildren” (Mashino et al., 1979). Among 14 pairs, 7 pairs were consonant and the others dissonant. Expression of Music In the reported literature, the ability of music expression was mainly investigated by singing or playing instrumentals for professional or nonprofessional musicians (McFarland, 1982; Piccirilli et al., 2000). In the case of musically naive patients, it is hard for examiners to know what their singing ability was before illness, so, even if their singing was inaccurate, the examiners could not determine that the ability of singing was impaired. In order to clarify the existence of the impairment of music expression, the easy and simple tasks which every musically naive subjects could readily perform were recommended. If a patient showed any difficulty during the performance of such tasks, we can reasonably conclude that the patient had the impairment of the ability of music expression. Therefore, besides of singing familiar songs, we added two simple tasks of singing: the reproduction of single sounds and the reproduction of the two-notes phrases. Discrimination of Unfamiliar Phrases Singing Familiar Songs We used items from the tonal memory test of “Seashore measures of musical talents” (Seashore et al., 1960), which contained 30 pairs of short phrases consisting of three to five notes. Subjects were presented two serial phrases and asked which tone was altered in the second phrase. The patient and control subjects were asked to sing ten well-known Japanese nursery songs without accompaniment, with the help of written lyrics. The patient had no difficulty in singing, and for later analysis, all performances were recorded onto mini-discs. Rhythm Discrimination Reproduction of Single Sounds From “Seashore measures of musical talents” (Seashore et al., 1960), 30 pairs of rhythm patterns were presented and subjects were asked to perform the ‘same-different’ discrimination. The tasks were divided into 3 groups, referred to as columns A, B, and C, with progressively more complex sequences in the latter two columns. Subjects were required to reproduce vocally single tones presented by the examiner. The pitch of the sound was set within the preferred range of each subject and the sound was presented as a pure-tone using a Yamaha PSR-330 synthesizer. Five tones were tested. The examination was 80 Masayuki Satoh and Others performed under two conditions: (i) while the tone was sounding, and (ii) immediately following the end of the tone presentation. The patient could discriminate pitch-differences as well as normal subjects. Discrimination and Recognition of Familiar Songs Reproduction of the Two-notes Phrases All of the control subjects reported that they were very familiar with all of the songs, and their mean score was 27.7 out of 30 (range: 26-30). The patient stated that she had not heard most of the songs previously, although when presented with the names of the songs, she sometimes reported a nostalgic feeling. The patient’s score was significantly lower than those of the controls (p < 0.0001), but above chance. Therefore, she could correctly discriminate songs that she was unable to recognize accurately, suggesting some ability to make song discriminations at a subconscious level. Among nine songs which she mistook, she stated that she had never listened to five songs so she could not answer the question. She regarded three altered songs as correct, and the alteration consisted of the change of the pitch (two songs) and the rhythm (one song). The contour of the melody was preserved in two of them. Subjects were presented three two-notes phrases and requested asked to sing each phrase. The phrases consisted of two notes whose pitches were separated by 3,4 or 5 degrees (e.g. ‘F-A’, ‘F-Bflat’ and ‘F-C’). Reproduction of the Simple Rhythm Five simple rhythm patterns 2 bars in length in 4/4, were presented and subjects were required to reproduce the rhythm by hand clapping. RESULTS The summary of tests of musical ability is shown in the Table I. For statistical analysis, a general linear model was used. Recognition of Environmental Sounds This test was performed only to three old subjects (aged 69, 70, 72) and our patient. The mean score of old healthy subjects was 22.6 (one 22, two 23). They missed the sounds of a firing pistol and a river. The patient was unable to recognize two of 24 environmental sounds. She answered the sounds of a firing pistol and the river as the sound of engineering works and a storm, respectively. We supposed that this task included some unfamiliar and ambiguous sounds to old Japanese people, so the result of the patient could be regarded as within normal limits. Discrimination of Unfamiliar Phrases Control subjects answered correctly in 21.8 out of 30 tasks (range: 17-27). In the tasks of 3, 4 and 5 notes (columna A, B, and C), their mean score was 9.3 (range: 7-10), 6.4 (range: 5-9), and 6.1 (range: 3-8), respectively. In contrast, the patient was significantly impaired in this test (p < 0.0001): All of the patient’s correct answers were in column A. Moreover she reported hearing both phrases in each pair as the same. Rhythm Discrimination PERCEPTION OF MUSIC The mean score of the control subjects was 24.8 out of 30 (range: 20-29). Scores were 9.4, 8.4 and 7.0 in columns A, B, and C, respectively. The score of the patient was 29, which was excellent. Pitch Discrimination No significant difference was seen between the range of control subjects and that of the patient. TABLE I Results of tests about musical abilities Perception of music Pitch discrimination (Hz) Discrimination and recognition of familiar songs ( /30) Discrimination of unfamiliar phrases ( /30) Rhythm discrimination ( /30) Meter discrimination ( /4) Discrimination of isolated chords ( /14) Expression of music Singing familiar songs Reproduction of single sounds Reproduction of the two-notes phrases Reproduction of the simple rhythm *p < 0.05 **p < 0.01 ***p < 0.001 The range of scores of control subjects are given parenthetically. Patient Control 429-442 21*** 3*** 29 3 impossible 434-446 27.7 (26-30) 21.8 (17-27) 24.8 (20-29) 3.8 (3-4) 12.6 (11-14) paramelodia well well well well well well well Amusia following bilateral temporal infarction 81 Meter Discrimination Reproduction of the Simple Rhythm The mean score of the control subjects was 3.8 (range: 3-4). The patient gave correct responses in 3 out of 4 tasks. The control subjects and the patient could reproduce all patterns correctly. Discrimination of Isolated Chords DISCUSSION The patient could not perform this examination because each chord within the pair sounded the same to her. Her performance was at chance level. Following bilateral infarctions in the anterior portion of the temporal lobes, our patient manifested persistent deficits of the perception and expression of music as follows: (i) impaired recognition and discrimination of familiar melodies; (ii) impaired discrimination of unfamiliar phrases; (iii) impaired discrimination of isolated chords; (iv) the replacement of familiar melodies while singing. The present results suggest that the anterior portions of both temporal lobes might participate in certain aspects of music processing, specifically, in the recognition of familiar melodies, in the discrimination of unfamiliar phrases, and of chords and in the singing of familiar tunes. In the following paragraphs, we discuss the function of these brain regions in relation to the various components of music, including melody, harmony, and rhythm or meter. EXPRESSION OF MUSIC Singing Familiar Songs The patient could correctly sing songs she knew well, such as “The moon”. However, in singing less familiar songs her singing was sometimes out of tune, though the rhythm and contour of the melody were preserved. When she sang an old Japanese tune in quadruple time (“I’m a marine child”), the third phrase of that melody was replaced with a phrase of a tune in triple time (“My home country”). When she sang “New Year’s Day”, the fourth phrase of that melody was replaced with the fourth phrase of another tune (“Urashima Taro”). After singing about two bars of replaced melody, she noticed the replacement and stopped singing. The patient reported that such replacement of melody often occurred when she sang at home. On another day of testingt, we requested her to sing the same songs. In that instance, the second and fourth phrases of “I’m a marine child” were replaced with other melodies. We were not able to identify the songs to which these replaced melodies belonged. We asked two professional musicians to identify these melodies, but they could not. Thus, the possibility remains that the patient improvised these melodies. Twenty minutes after this episode, she was able to sing “I’m a marine child” correctly. In old healthy subjects, such insertion of one melody in another had never been observed, so we could regard the patient’s symptom as pathological. Reproduction of Single Sounds The patient and all subjects could readily reproduce the single tone in both during and following the termination of the tone presentation. Reproduction of the Two-notes Phrases Control and old healthy subjects could discriminate and reproduce these two-notes phrases easily, while the patient was unable to tell the difference between the phrases. She stated that all phrases were heard as similar. Despite her inability to discriminate the phrases, she could correctly sing each phrases. Music Processing and Anterior Temporal Lobes Among the previously published reports of amusia, we were able to find only three patients with ‘pure’ amusia caused by damage to restricted brain regions (McFarland, 1982; Peretz et al., 1994; Piccirilli et al., 2000). The patient reported by McFarland lost the ability to play and identify familiar melodies (McFarland, 1982). Peretz’s patient revealed impairment in recognition of familiar tunes and in discrimination of unfamiliar musical phrases (Peretz, et al., 1994). The patient reported by Piccirilli et al. was impaired in the recognition and production of melodic sequences (Piccirilli et al., 2000). The former two patients’ lesions were localized by CT neuroimaging. Axial images of each patient revealed lesions in the right superior temporal and supramarginal gyri (McFarland, 1982), and bilateral superior/middle temporal gyri and temporal poles (Peretz, 1994). However due to artifact from the pyramidal bones or teeth, the inferior half of the temporal lobe was difficult to investigate. Piccirilli et al. investigated the site of lesions by MRI. Axial and coronal images showed focal damage involving the middle and posterior third of the left superior temporal gyrus and moderate atrophy of the left temporal lobe (Piccirilli et al., 2000). In a PET study investigating the functional neuroanatomy of music processing in nonmusicians, Platel et al. observed bilateral activation of the anterior portion of the superior temporal gyri during a task in which subjects had to answer whether tonal sequences brought to mind something that they knew (Platel 82 Masayuki Satoh and Others et al., 1997). It seems reasonable to suppose, from what has been said above, that superior temporal gyrus, especially its anterior portion, participates in the perception of familiar melodies. This assumption is supported by recent literature which showed the participation of temporal poles in the recognition of familiar objects (Grabowski et al., 2001; Nakamura et al. 2000, 2001). When our patient performed a test of discrimination and recognition of familiar songs, the test melody was compared with the melody retrieved from memory. In discriminating unfamiliar phrases, the first phrase was compared with a second one. Thus, common cognitive processing exists in the discrimination of both familiar melodies and unfamiliar phrases, namely the comparison of melodies. In functional MRI for musical trained individuals, Janata et al. revealed that, when the subjects listened to unfamiliar melodies, superior and middle temporal gyrus, containing their rostral portion, were bilaterally activated (Janata et al., 2002). Therefore, the anterior portion of both temporal lobes may be reasonably assumed to participate in the recognition of not only familiar but also unfamiliar melodies, and in fact our patient was impaired in both tasks. Regarding perception of chords, our patient demonstrated very poor chord discrimination. There are no previous reports in the literature on the impairment of chord perception. Thus the present paper is the first report of a focal brain lesion resulting in impaired chord perception. In our chord discrimination task, the second chord presented was compared with the first one. As mentioned above, the impairment of chord perception might also result from a dysfunction in the comparison between two chords. It is also possible that the anterior portions of the temporal poles are involved in meter processing. This idea is supported by a study of music ability in patients who had undergone unilateral temporal lobectomy, the metric test score was significantly lower in right and left lobectomized patients compared with normal controls (Liégeois-Chauvel et al., 1998). The author of that study suggested that the anterior portion of the superior temporal gyrus bilaterally seems to be critically involved in distinguishing meter. As further evidence, our patient could not distinguish the tango from other forms of music. While she performed well in a rhythm-discrimination task, her score in a meter discrimination task was somewhat, although insignificantly, poorer than controls. Judging from these results, it is possible that the anterior portion of both temporal lobes participates in the perception of meter. It may be worth mentioning the hemispheric lateralization of music processing. Schuppert suggested the complementary contributions of the two cerebral hemispheres in music perception: a hierarchical organization with an initial right- hemisphere recognition of contour and meter followed by identification of interval and rhythm via left-hemisphere subsystems (Schuppert et al., 2000). Our patient had bilateral lesions, and it is possible that she revealed various aspects of impairments of musical ability. There is a wealth of literature regarding amusia whose impairments are incongruent with Schuppert’s hypothesis (Basso, 1998). Much still remains to be done about the hemispheric lateralization in perception and expression of music. Paramelodia The patient often sang another melody while she was singing familiar songs. We have named this phenomenon paramelodia, because we suppose that it is caused by a dysfunction in the retrieval of a melody from long-term memory. The stored memory of the song itself seems not to be impaired, because the patient could sometimes sing the identical songs correctly. We observed that the phenomenon of paramelodia always appeared by the unit of one phrase, namely four bars. This fact suggests that the song is retrieved during singing by a unit of one phrase and not in its entirety. The length of most ancient and modern songs is four, eight, twelve, or sixteen bars, and so on. Thus “four bars” might represent a fundamental unit of melodic organization. Why were phrases of “I’m a marine child” and “New Year’s Day” replaced by phrases from “My home country” and “Urashima Taro”, respectively? The meter seems irrelevant for this phenomenon, because the meter of “I’m a marine child” was 4/4, while that of “My home country” was 3/4. In addition the lyrics and the contour of each melody were quite different. We speculate that the similarity of the harmonic progression caused paramelodia between these tunes. In music theory, the tonic and dominant chord are described as “I” and “V”, since each chord is built on the first and the fifth tone of the scale, respectively. Similarly, the triad built on the fourth tone of the scale is described as “IV”. In harmonic theory, the harmony of nearly every tune can be expressed by three kinds of chords, namely “I”, “IV”, and “V”. We analyzed the harmonic progression of “I’m a marine child”, “New Year’s Day”, “My home country”, and “Urashima Taro”, and found that the second phrase of “I’m a marine child” and “My home country”, and the third phrase of “New Year’s Day” and “Urashima Taro” had an identical harmonic progression: IV-I-V-I in “I’m a marine child” and “My home country”, and I-I-V-V in “New Year’s Day” and “Urashima Taro”. We proposed that this similarity in harmonic progression resulted in the replacement of succeeding phrases, namely the replacement of the third phrase of “I’m a marine child” and of “My home country”, and of the fourth phrase of “New Amusia following bilateral temporal infarction Year’s Day” and of “Urashima Taro”. In other words, when we sing a song, the melody might be retrieved by the unit of one phrase based on the harmonic progression of the preceding phrase. Such a hypothesis makes it possible to interpret why paramelodia occurred between these tunes when sung by our patient. CONCLUSION We reported an amusic patient with an infarction in the anterior portion of bilateral temporal lobes resulting in a persistent deficit in musical ability. The patient manifested the impairments of the recognition of familiar and unfamiliar melodies, and the discrimination of isolated chords. During singing familiar nursery songs, the patient showed the replacement of one phrase of the melody. We named this phenomenon paramelodia, and hypothesized that this was caused by the dysfunction of the song retrieval from long-term memory. We may conclude that anterior temporal lobes participate in music processing. REFERENCES BASSO A. The neuropsychology of music. In G Denes (Eds), Handbook of Clinical and Experimental Neuropsychology. Amsterdam: Elsevier, 1998, pp. 409-418. DUVERNOY H. The Human Brain. Wien: Springer-Verlag, 1991. GRABOWSKI TJ, DAMASIO H, TRANEL D, PONTO LLB, HICHWA RD and DAMASIO AR. A role for left temporal pole in the retrieval of words for unique entities. Huan. Brain Mapping, 13: 199212, 2001. GRISON B. Un étude sur les alterations musicales au cours des lesions hemispheriques [These]. Paris, 1972. [Cited by Benton AL. The amusias. In M Critchley and RA Henson (Eds), Music and the Brain. London: William Heinemann Medical Books Limited, 1977, pp. 378-397.] HENSON RA. Neurological aspects of musical experience. In M Critchley and RA Henson (Eds), Music and the Brain. London: William Heinemann Medical Books, 1977, pp. 3-21. JANATA P, BIRK JL, VAN HORN JD, LEMAN M, TILLMANN B and BHARUCHA JJ. The cortical topography of tonal structures underlying western music. Science, 298: 21672170, 2002. 83 LIÉGEOIS-CHAUVEL C, PERETZ I, BABAÏ M, LAGUITTON V and CHAUVEL P. Contribution of different cortical areas in the temporal lobes to music processing. Brain, 121: 1853-1867, 1998. LORING DW. INS Dictionary of Neuropsychology. New York: Oxford University Press, 1999. MASHINO A, HAMANO M and MOTEGI S. The Test for the Assessment of Musical Ability for Schoolchildren. Tokyo: Nihon-Bunka-Kagakusya, 1979. MAVLOV L. Amusia due to rhythm agnosia in a musician with left hemisphere damage: A non-auditory supramodal defect. Cortex, 16: 331-338, 1980. MCFARLAND HR and FORTIN D. Amusia due to right temporoparietal infarct. Archives of Neurology, 39: 725-727, 1982. NAKAMURA K, KAWASHIMA R, SATO N, NAKAMURA A, KATO T, HATANO K, ITO K, FUKUDA H, SCHORMANN T and ZILLES K. Functional delineation of the human occipitotemporal areas related to face and scene processing. A PET study. Brain, 123: 1903-1912, 2000. NAKAMURA K, KAWASHIMA R, SUGIURA M, KATO T, NAKAMURA A, HATANO K, NAGUMO S, KUBOTA K, FUKUDA H, ITO K and KOJIMA S. Neural substrates for recognition of familiar voices: A PET study. Neuropsychologia, 39: 1047-1054, 2001. PERETZ I, KOLINSKY R, TRAMO M, LABRECQUE R, HUBLET C, DEMEURISSE G and BELLEVILLE S. Functional dissociations following bilateral lesions of cortex. Brain, 117: 1283-1301, 1994. PICCIRILLI M, SCIARMA T and LUZZI S. Modularity of music: Evidence from a case of pure amusia. Journal of Neurology, Neurosurgery and Psychiatry, 69: 541-545, 2000. PLATEL H, PRICE C, BARON JC, WISE R, LAMBERT J, FRACKOWIAK RSJ, LECHEVALIER B and EUSTACHE F. The structural components of music perception. A functional anatomical study. Brain, 120: 229-243, 1997. SAMSON S and ZATORRE RJ. Learning and retention of melodic and verbal information after unilateral temporal lobectomy. Neuropsychologia, 30: 815-26, 1992. SCHUPPERT M, MUNTE T, WIERINGA BM and ALTENMULLER E. Receptive amusia: Evidence for cross-hemispheric neural networks underlying music processing. Brain, 123: 546-559, 2000. SEASHORE CE, LEWIS D and SAETVEIT J. Seashore measures of musical talents. New York: The Psychological Corporation, 1960. WERTHEIM N. The amusias. In PJ Vinken and GW Bruyn (Eds), Handbook of Clinical Neurology, vol. 4. Amsterdam: NorthHolland Publishing, 1969, pp. 195-206. WERTHEIM N and BOTEZ MI. Receptive amusia: A clinical analysis. Brain, 84: 19-31, 1961. ZATORRE RJ. Discrimination and recognition of tonal melodies after unilateral cerebral excisions. Neuropsychologia, 23: 3141, 1985. ZATORRE RJ and HALPERN AR. Effect of unilateral temporal-lobe excision on perception and imagery of songs. Neuropsychologia, 31: 221-32, 1993. M. Satoh, Department of Neurology, Mie University School of Medicine, 2-174 Edobashi, Tsu, Mie-ken, 514-8507 Japan. E-mail: bruckner@clin.medic.mie-u.ac.jp (Received 6 November 2002; reviewed 1 February 2003; revised 4 April 2003; accepted 28 April 2003; Action Editor Henry Buchtel)