- Email: [email protected]
Music Alexia in a Patient with Mild Pure Alexia: Disturbed Visual Perception of Nonverbal Meaningful Figures
NOTE MUSIC ALEXIA IN A PATIENT WITH MILD PURE ALEXIA: DISTURBED VISUAL PERCEPTION OF NONVERBAL MEANINGFUL FIGURES* Toru Horikoshi, Yasuhiro Asari, Arata Watanabe, Yoshishige Nagaseki, Hideaki Nukui, Hideo Sasaki l and Keiji Komiya2 (Department of Neurosurgery, Yamanashi Medical University, Yamanashi, Japan; IDepartment of Neurosurgery, and 2Speech Therapy Service, Kofu-Johnan Hospital, Yamanashi, Japan)
ABSTRACT
A 26-year-old female pianist suffered from an intracerebral hematoma caused by an arteriovenous malformation of the left occipital parasplenial region, which was operated on seven months after the onset. Incomplete right hemianopsia, mild pure alexia, and partially disturbed naming of visual objects persisted several months after the removal of the malformation. Evaluation of musical ability one and three months after surgery showed that her auditory recognition of music was intact. She could sing and play melodies already learned and could dictate well the notes after hearing tones. However, she had difficulty in reading music, especially the pitch of notes, even for simple sequences of 4 notes. In contrast, her rhythm reading was fairly good. Her visual recognition of other symbolic figures like road signs was also markedly impaired. These results suggest that her visual recognition of written music as well as of other symbolic figures underwent a preliminary verbal decoding in the left hemisphere and that pitch reading was more dependent on verbal processing than rhythm reading.
INTRODUCTION
Written music consists of many types of symbols. Notes are the main component and simply express the pitch and duration of each tone in combination with rests, clefts, bemolle, or diesis, whereas words such as forte, piano, and other musical symbols including numerals indicate speed, loudness, timbre and other musical values. There is an analogy between musical symbols and written letters because both are conventional symbols of the actual sound or articulation. In contrast to letters, written music offers temporal and quantitative information of sound. This time sequence information is essential for the performance of music. In the history of neuropsychology, studies concerning the disruption of musical ability have mainly focused on auditory perception of melodies and singing ability rather than on reading or writing music, because very few patients had this kind of skills. Research into impaired music reading and writing can provide useful information about how nonverbal meaningful figures related to sound are recognized and about the difference between music and language in visual processing. We present the case of a professional pianist who suffered from pure alexia following a hemorrhage caused by an arteriovenous malformation.
* This
paper was briefly presented at the 19th annual meeting of the Japanese Society of Aphasiology on 16. Nov. 1995. Tokyo.
Cortex, (1997) 33, 187-194
188
Toru Horikoshi and Others CASE REPORT
A 26-year-old female had studied piano for 15 years from the age of 9 years. After graduating from college at age 20 years, she worked as a kindergarten teacher for one year and then as a piano teacher for five years after establishing her own private piano school. In March 1994, she complained of the sudden onset of severe headache and vomiting and became unconscious. On admission to the emergency service of a local hospital, CT revealed an intracerebral hematoma in the left occipital lobe and intraventricular hemorrhage combined with acute hydrocephalus. The intracerebral hematoma involved the splenium of the corpus callosum. An emergency cerebrospinal fluid drainage from the bilateral anterior horns was performed for relieving the acute hydrocephalus. She gradually regained consciousness after the procedure, but was aphasic. She underwent speech training, and was transferred to our hospital for further treatment of the malformation 7 months after the onset. No motor or somatosensory impairments were detected and she did not require any aid in daily life, though incomplete right homonymous hemianopsia and mild speech disturbance were still present 7 months after the onset. Cerebral angiography showed an arteriovenous malformation of 4 cm length and 2 cm width in the white matter of the interhemispheric part of the left occipital lobe adjacent to the splenium of the corpus callosum. Surgical excision of the lesion through the occipital approach was successfully performed without causing additional neurological deficits. MR image 5 months after surgery showed damage to the occipito-temporal deep white matter surrounding the trigone, the posterior horn of the left lateral ventricle and the splenium of the corpus callosum (Figure 1). SPECT using 99m-Tcethyl-cysteinate-dimer demonstrated that decreased cerebral blood flow in the left occipital lobe and posterior part of temporoparietal lobe still persisted 5 months after surgery (Figure 2).
Fig. 1 - Postoperative T2 weighted MR images jive months after surgery showing tissue damage and hemosiderin deposits in the left posterior deep white matter around the trigone of the lateral ventricle and the splenium of the corpus callosum (arrows).
Music alexia and pure alexia
189
\ Fig. 2 - SPECT images indicating decreased cerebral blood flow in the left occipital lobe, posterior part of the temporoparietal lobe, and the thalamus (arrows).
Neuropsychological Evaluation Her intelligence was evaluated using the Japanese version of W AIS-R one month after surgery (Table I). Her verbal IQ was 57 and her performance IQ was 52. Re-evaluation three months after surgery using the same scale showed improvement with a verbal IQ of 72 and a performance IQ of 81. In particular, the scores of picture completion, picture arrangement, block design, and object assembly subtests were within the normal range, and the Kohs cube test showed a performance IQ of 98. The Standard Language Test for Aphasia of the Janapese Society of Aphasiology showed that one month after surgery her oral language disability was of the sensory dominant type and was combined with agraphia and alexia (Table II). Her performance was defective in spoken order comprehension, visual naming, oral explanation, word reading and sentence reading, written sentence comprehension, and spontaneous writing. Her language disability rapidly improved three months after surgery. TABLE I
Wechsler Intelligence Scale: Standardized Scores Evaluated One Month (Score 1) and Three Months (Score 2) after Surgery Score I Verbal tests Information Digit span Vocabulary Arithmetic Comprehension Similarities Performance tests Picture arrangement Picture completion Block design Object assembly Digit symbol Verbal IQ Performance IQ TotalIQ
Score 2
3 8 2 3 2 2
5 8 4 6 5 6
1 7 7 3
9 9
1
57 52 51
10 8 2 72 81 74
Toru Horikoshi and Others
190
TABLE II
Results of Standard Language Test of Aphasia Score I
Listening Word comprehension Sentence comprehension Execution of spoken orders Syllable comprehension Speaking Visual object naming Word repetition Explanation of picture Recall words Reading kanji word aloud Reading kana character aloud Reading kana word aloud Reading sentence aloud Reading Kanji word comprehension Kana word comprehension Sentence comprehension Execution of written orders Writing Kanji word Kana word Explanation of picture Dictation with kana character Dictation of kanji word Dictation of kana word Dictation of sentence Calculation
Score 2
Control mean
SD
100 100 50 100
100 100 100 100
100 95 96 100
2 8 7 1
35 50 100 80 60 100 100 20
50 90 100 80 100 100 100 100
98 99 90 84 100 100 100 98
4 4 16 30 8 2 2 6
70 90 10 0
100 100 80 100
99 100 96 94
8 1 10 15
60 80 20 100 100 100 100 90
100 100 100 100 100 100 100 100
84 22 14 96 86.7 18 97 II 86 20 96 16 80 30 81.5 21
Note: Score I (one month) and Score 2 (three months after surgery) presented as percentages. Kanji is a pictorial symbol in written Japanese, and kana a phonetic symbol.
However, naming of visual objects was still defective and comprehension of written sentences was below the normal average. It was more difficult for her to read kanji, the pictorial form of written Japanese, than kana, the phonetic form, and the alphabet. Three months after surgery, her color discrimination was intact but she was slow in naming colors. She could recognize categories of figures, and correctly chose a figure standing out from five alternatives in the differentiation task, e.g., a rectangular triangle among isosceles triangles. She could also match to sample meaningless complex figures. There was no difficulty in copying figures, and no spatial agnosia or left-right disorientation. In contrast, her recognition of meaningful figures was markedly impaired. She could explain verbally only 60% of map symbols, and 24% of road signs, even though she had a driver's license. She also found it difficult to indicate the meaning of the road signs by gesture or using a toy car. It seemed to be more difficult for her to recognize abstract signs than realistic ones. She made mistakes in naming 25% of visual objects, like toothbrush or shoes, and had some difficulty in explaining how to use them. Verbal recognition of pictures showing daily activities, like fire fighting or medical services in a hospital, was also difficult.
Examination of Musical Ability Her musical ability was examined during the post operative course, one and three months after surgery. All subtests were performed using a piano, because of her greater familiarity with this instrument. At first, her general music skills, hearing and singing, were examined. When the examiner sounded two tones or two sequences of tones on the piano, she could
Music alexia and pure alexia
(a)
191
(b)
(d)
7Fig. 3 - (a) Examples of 4 quarter note sequences used for testing reading ability. (b) Dotted quarter notes were also included in some sequences. (c) Notes written by the patient in the copying task. (d) Notes correctly dictated by the patient.
easily discriminate and indicate which of the two tones was higher, louder, or longer and which of two tone sequences had a faster rhythm. She also recognized and gave the names of melodies which were familiar to her. She seemed to enjoy listening to music by herself. She could sing easy popular folk songs without any mistake in pitch or rhythm. When she was asked to play the piano, she could play practice pieces which she had already learned by heart. The simple sequences of 4 quarter notes or dotted quarter notes in the C major scale, as shown in Figure 3 (a) (b), were used for further examinations, especially of reading and writing skills. Four famous nursery songs with 8 bars length were also used in some subtests. The patient was requested to listen to the four tones and the names of four notes, to read the names of four notes, and to read four notes or music. Then, she was asked to choose the appropriate sequence of four notes from six written alternatives, to write four notes on five lines, to play them on the piano, to give their names, to give the name of the song and to choose the written name of the song from four alternatives. All of these subtests consisted of 4 to 40 problems each. The same note sequences and songs were used in all subtests. In the reading test, more complex phrases with 16th notes, as shown in Figure 4, and practice pieces (familiar to her, or unlearned) were also used. In some subtests the patient was asked
Ie JJJ :rJJJ ~J33 't J Fig. 4 - Examples of sequences with 16th notes used for evaluation of rhythm reading.
192
Toru Horikoshi and Others
to play the piano with her right and left hand separately. The results of each subtest were classified into three categories; normal: all problems were correctly answered within a reasonable time; delayed : all correct but they took much time; and incorrect: one or more wrong answers. Three female volunteer piano teachers of the same age with similar experience were used as controls for response time in some of the subtests. The results are shown in Table III. The tests showed that her reading ability was significantly impaired. All tasks requiring the reading of notes or music, except copying tasks (Figure 3c), were impaired. The results of subtest requiring the choice of written notes or music were also impaired. She could not play practice pieces by sight reading. Even in simple 4 note sequences, she frequently made pitch mistakes when she played the piano or named notes. Playing with the el ft hand took the same or more time than with the right hand. Compared to pitch discrimination, her rhythm recognition was much better preserved even for complex rhythms, and she made no mistakes in the second testing session. Reading names of notes also took much time because of alexia but was better than reading notes. Her music writing skill was very good, and she could dictate notes correctly after listening to them and could copy notes without hesitation (Figure 3d). The same tendency was seen at the second evaluation, though some improvement had occurred.
DISCUSSION
Most studies have focused on localization of musical skills, especially in relationship to their hemisphere representation (Brust, 1980). While impairment of hearing and singing skills are not closely related to hemispheric dominance, music alexia and music agraphia are TABLE III
Results of Music Skills Test
Task and response required Listen to tones or songs Choose a series of notes Write notes Play on piano Give name of notes aloud Give name of song ' aloud Listen to name of notes Write notes Play on piano Read name of notes Choose a seri es of notes Write notes Play on piano Give name of notes aloud Read notes Write notes Play on piano Give name of notes aloud Read nursery songs Play on piano Give name of song aloud Choose written name of song Additional tasks Read and play on piano Familiar practice pieces Unfamiliar practice pieces Rhythm tasks
Score I 29/40 717 717 717 414
(incorrect) (normal) (normal) (normal) (normal)
Score 2 36/40 717 717 717 4/4
(incorrect) (normal) (normal) (normal) (normal)
4/4 717
(normal) (normal)
4/4 717
(normal) (normal)
717 4/4 717 4/4
(delayed) 31" «I") (delayed) 27" (4") (delayed) 10" (3") (normal)
617 4/4 717 4/4
(incorrect) (delayed) 15" (4") (delayed) 10" (3") (normal)
4/4 16/20 6112
(normal) (incorrect) (incorrect)
4/4 16120 7112
2/4 3/4 2/4
(incorrect) (incorrect) (incorrect)
2/4 4/4 4/4
(incorrect) (delayed) 54" (3") (delayed) 22" (3")
1/2
(incorrect) (incorrect) (incorrect)
212 012 6/6
(normal) (incorrect) (normal)
0/2 5/6
(normal) (incorrect) (incorrect)
NOle: Score I (one month) and Score 2 (three months after surgery). The scores are the correct response out of total questions.
Mean reaction times of the patient and controls (in parentheses) are shown for delayed responses.
Music alexia and pure alexia
193
generally found in association with language disturbance (Benton, 1977). Music alexia is usually associated to verbal alexia, and music agraphia to verbal agraphia (Brust, 1980), but cases of verbal alexia without music alexia (Basso and Capitani, 1985; Gates and Bradshaw, 1977; Luria, 1965) and of music alexia without verbal alexia have also been reported (Gates and Bradshaw, 1977). Our patient had pure music alexia combined with pure verbal alexia, similary to the case reported by AssaI (1973). Levin and Rose (1979) also reported a case of music alexia, but they did not mention music writing skill. In contrast to language, which is spontaneously learned in early life, musical skills, especially writing and reading music, require special training. Therefore, the music process in the brain may be variable among individuals, due to differences in personal experience, method used in learning music, age when learning started, or type of musical instrument. In our patient, the disturbance of visual perception involved recognition of informative figures such as road signs, map symbols, national flags and other well-known symbolic figures as well as musical notes. The meanings of these abstract symbols can be verbally decoded. Actually, many former studies suggested that nonverbal visual stimulus recognition can be mediated by verbal encoding in the left hemisphere (Boller and De Renzi, 1967; Milner, 1962; Rubino, 1970; Brewer, 1969). According to Faglioni, Scotti and Spinnler (1968), abstract symbols may lend themselves to semantic processing. Music notes, especially the pitch dimension, are very similar to abstract signs, because they are symbolic and can be represented by words like do, re, mi, etc. Our patient confessed that she had always used to translate the pitch of notes into an inner representation of sound, i.e., do, re, mi, when she read music. Thus, music symbols, even if apparently nonverbal, may undergo verbal processing. The recognition of rhythm and pitch of notes in music reading dissociated in our patient. She could recognize the rhythm of phrases fairly well in spite of significant impairment of pitch discrimination. This pattern is similar to that of Brust's (1980) patient who had damage to the left temporal and parietal lobes. In contrast, AssaI reported (1973) a patient with Wernicke's aphasia and mild music dyslexia, who had much more difficulty in rhythm reading than in pitch. Such discrepancies have already been described in other types of amusia (Brust, 1980; Shapiro, Grossman and Gardner, 1981; Peretz, 1990), and various proposals have been made about hemisphere representation of musical skills and dissociation in music processing (Hachinski, 1989). Bever and Chiarello (1974) suggested that the left hemisphere is dominant for analytic processing and the right hemisphere for holistic processing. Since melodies are composed of an ordered series of pitches, they should be mainly processed by the left hemisphere. Barbizet et al. (1972, cited by Benton, 1977) thought that the right hemisphere participation in musical activity is primarily at the perceptual and executive level, while the left hemisphere mediates the recognition and memory of musical structures, the symbolic process of reading and writing music and the higher level integrated functions involved in musical composition. In the studies of hemisphere depression following intracarotid sodium amy tal injection (Gordon and Bogen, 1974), the rhythmic aspect was unaffected by the depression of either hemisphere, while pitch discrimination in singing was disturbed by depression of the right hemisphere. Gordon and Bogen (1974) suggested that, though the left hemisphere may be superior at rhythmic perception, the right side is still quite capable of rhythm processing. Mostafa, Kotby, Barakah et al. (1989) demonstrated deterioration of all musical abilities in patients with right-side brain damage, but only rhythm disturbance in patients with left-side brain damage. Grossman, Shapiro and Gardner (1981) reported that the musical judgments of patients with right anterior brain damage relied on the mood of the phrases, while patients with Broca's aphasia relied on a linear sequential musical processing strategy. These data suggest that each hemisphere may make a distinctive contribution to pitch and rhythm in receptive and expressive musical performance, but the nature of each hemisphere contribution is still controversial. Our results seem to be consistent with suggestions by Bever and Chiarello (1974) and Barbizet et al. (1972, cited by Benton, 1977), but do not permit to conclude whether the discrepancy in reading pitch and rhythm is due to their different hemisphere representation. What we can say is that rhythm discrimination was less affected than pitch recognition, when the patient decoded music symbols. Another interesting finding of this study is the greater improvement of word reading disturbance compared to music reading disturbance. Language skills rapidly improved in the
194
Toru Horikoshi and Others
observation period, while music readingi~pairment persisted. This difference may be due to the differential training-that the two skills received. There is usually no stimulus requiring music reading in daily life, while there is abundant opportunity to read words in television, magazines, or newspapers. The speech training undergone in the former hospital could also have promoted the improvement of word reading. Alternatively, music reading skills may be similar to the second language of bilingual individuals, which is generally more severely disturbed than the mother language following brain damage. The pattern of the patient's neurological deficits was typical for pure alexia with hemianopsia. Visual information projected to the right visual cortex was difficult to analyze verbally in the left posterior hemisphere, because of the splenial interruption of the interhemisperic fibers, and/or dysfunction of the posterior part of the left hemisphere, as shown by MRI and the SPECT study. Piano playing with the left hand was disturbed to the same degree or even more than with the right hand, which suggests that the information from the right visual cortex had to be first processed in the left temporoparietal lobe before being transmitted to the right motor area via the body of the corpus callosum.
Acknowledgement. We would like to express our thanks to Ms. Kumiko Nozawa of Yamanashi Symphony Orchestra for her technical assistance and advice. REFERENCES ASSAL, G. Aphasie de Wernicke sans amusie chez un pianiste. Revue Neurologique, 129: 251-255, 1973. BASSO, A., and CAPITAN!, E. Spared musical abilities in a conductor with global aphasia and ideomotor apraxia. Journal of Neurology, Neurosurgery and Psychiatry, 48: 407-412, 1985. BENTON, A.L. The amusia. In M. Critchley and R.A. Henson (Eds.), Music and the Brain. London: William Heinemann Medical Books, 1977, pp. 378-397. BEVER, T.G, and CHIARIELLO, R.J. Cerebral dominance in musicians and nonmusicians. Science, 185: 537-539, 1974. BOLLER, F., and DE RENZI, E. Relationship between visual memory defects and hemispheric locus of lesion. Neurology, 17: 1052-1058, 1967. BREWER, W.F. Visual memory, verbal encoding and hemispheric localization. Cortex,S: 145-151, 1969. BRUST, J.e.M. Music and language. Musical alexia and agraphia. Brain, 103: 367-392, 1980. FAGLlONI, P., SCOTTI, G., and SPINNLER, H. Impaired recognition of written letters following unilateral hemispheric damage. Cortex,S: 120-133, 1968. GATES, A., and BRADSHAW, L. The role of the hemispheres in music. Brain and Language, 4: 403431, 1977. GORDON, H.W., and BOGEN, J.E. Hemispheric lateralization of singing after intracarotid sodium amylobarbitone. Journal of Neurology, Neurosurgery and Psychiatry, 37: 727-738, 1974. GROSSMAN, M., SHAPIRO, B.E., and GARDNER, H. Dissociable musical processing strategies after localized brain damage. Neuropsychologia, 19: 425-433, 1981. HACH!NSKI, V.C. Effect of strokes on musical ability and performance. Seminars in Neurology, 9: 159162, 1989. LEVIN, H.S., and ROSE, J.E. Alexia without agraphia in a musician after transcallosal removal of a left intraventricular meningioma. Neurosurgery, 4: 168-173, 1979. LURIA, A.E., TSVETKOVA, L.S., and FUTER, D.S. Aphasia in composer. Journal of the Neurological Sciences, 2: 288-292, 1965. MAZZIOTTA, J.e., PHELPS, M.E., CARSON, R.E., and KUHL, D.E. Tomographic mapping of human cerebral metabolism; Auditory stimulation. Neurology, 32: 921-937, 1982. MILNER, B. Laterality effects in audition. In V.B. Mountcastle (Eds.), Interhemispheric Relations and Cerebral Dominance. Baltimore: Johns Hopkins Press, 1962, pp. 177-195. MOSTAFA, M., KOTBY, M.N., BARAKAH, M., EL-SADY, S., ALLOSH, T., ELSHOBARY, A., and SALEH, M. Dominant functions of right versus left hemisphere. Acta Oto-laryngologica (Stockholm), 107: 479-484, 1989. PERETZ, T. Processing of local and global musical information by unilateral brain-damaged patients. Brain, 113: 1185-1205, 1990. RUBINO, e.A. Hemispheric lateralization of visual perception. Cortex, 6: 102-130, 1970. SHAPIRO, B.E., GROSSMAN, M., and GARDNER, H. Selective musical processing deficits in brain damaged populations. Neuropsychologia, 19: 161-169, 1981. Toru Horikoshi, M.D., Department of Neurosurgery, Yarnanashi Medical University, 1110 Shimokato, Tamaho-machi, Yamanashi, Japan, 409·38.
(Received 24 October 1995; accepted 19 April 1996)
ABSTRACT
A 26-year-old female pianist suffered from an intracerebral hematoma caused by an arteriovenous malformation of the left occipital parasplenial region, which was operated on seven months after the onset. Incomplete right hemianopsia, mild pure alexia, and partially disturbed naming of visual objects persisted several months after the removal of the malformation. Evaluation of musical ability one and three months after surgery showed that her auditory recognition of music was intact. She could sing and play melodies already learned and could dictate well the notes after hearing tones. However, she had difficulty in reading music, especially the pitch of notes, even for simple sequences of 4 notes. In contrast, her rhythm reading was fairly good. Her visual recognition of other symbolic figures like road signs was also markedly impaired. These results suggest that her visual recognition of written music as well as of other symbolic figures underwent a preliminary verbal decoding in the left hemisphere and that pitch reading was more dependent on verbal processing than rhythm reading.
INTRODUCTION
Written music consists of many types of symbols. Notes are the main component and simply express the pitch and duration of each tone in combination with rests, clefts, bemolle, or diesis, whereas words such as forte, piano, and other musical symbols including numerals indicate speed, loudness, timbre and other musical values. There is an analogy between musical symbols and written letters because both are conventional symbols of the actual sound or articulation. In contrast to letters, written music offers temporal and quantitative information of sound. This time sequence information is essential for the performance of music. In the history of neuropsychology, studies concerning the disruption of musical ability have mainly focused on auditory perception of melodies and singing ability rather than on reading or writing music, because very few patients had this kind of skills. Research into impaired music reading and writing can provide useful information about how nonverbal meaningful figures related to sound are recognized and about the difference between music and language in visual processing. We present the case of a professional pianist who suffered from pure alexia following a hemorrhage caused by an arteriovenous malformation.
* This
paper was briefly presented at the 19th annual meeting of the Japanese Society of Aphasiology on 16. Nov. 1995. Tokyo.
Cortex, (1997) 33, 187-194
188
Toru Horikoshi and Others CASE REPORT
A 26-year-old female had studied piano for 15 years from the age of 9 years. After graduating from college at age 20 years, she worked as a kindergarten teacher for one year and then as a piano teacher for five years after establishing her own private piano school. In March 1994, she complained of the sudden onset of severe headache and vomiting and became unconscious. On admission to the emergency service of a local hospital, CT revealed an intracerebral hematoma in the left occipital lobe and intraventricular hemorrhage combined with acute hydrocephalus. The intracerebral hematoma involved the splenium of the corpus callosum. An emergency cerebrospinal fluid drainage from the bilateral anterior horns was performed for relieving the acute hydrocephalus. She gradually regained consciousness after the procedure, but was aphasic. She underwent speech training, and was transferred to our hospital for further treatment of the malformation 7 months after the onset. No motor or somatosensory impairments were detected and she did not require any aid in daily life, though incomplete right homonymous hemianopsia and mild speech disturbance were still present 7 months after the onset. Cerebral angiography showed an arteriovenous malformation of 4 cm length and 2 cm width in the white matter of the interhemispheric part of the left occipital lobe adjacent to the splenium of the corpus callosum. Surgical excision of the lesion through the occipital approach was successfully performed without causing additional neurological deficits. MR image 5 months after surgery showed damage to the occipito-temporal deep white matter surrounding the trigone, the posterior horn of the left lateral ventricle and the splenium of the corpus callosum (Figure 1). SPECT using 99m-Tcethyl-cysteinate-dimer demonstrated that decreased cerebral blood flow in the left occipital lobe and posterior part of temporoparietal lobe still persisted 5 months after surgery (Figure 2).
Fig. 1 - Postoperative T2 weighted MR images jive months after surgery showing tissue damage and hemosiderin deposits in the left posterior deep white matter around the trigone of the lateral ventricle and the splenium of the corpus callosum (arrows).
Music alexia and pure alexia
189
\ Fig. 2 - SPECT images indicating decreased cerebral blood flow in the left occipital lobe, posterior part of the temporoparietal lobe, and the thalamus (arrows).
Neuropsychological Evaluation Her intelligence was evaluated using the Japanese version of W AIS-R one month after surgery (Table I). Her verbal IQ was 57 and her performance IQ was 52. Re-evaluation three months after surgery using the same scale showed improvement with a verbal IQ of 72 and a performance IQ of 81. In particular, the scores of picture completion, picture arrangement, block design, and object assembly subtests were within the normal range, and the Kohs cube test showed a performance IQ of 98. The Standard Language Test for Aphasia of the Janapese Society of Aphasiology showed that one month after surgery her oral language disability was of the sensory dominant type and was combined with agraphia and alexia (Table II). Her performance was defective in spoken order comprehension, visual naming, oral explanation, word reading and sentence reading, written sentence comprehension, and spontaneous writing. Her language disability rapidly improved three months after surgery. TABLE I
Wechsler Intelligence Scale: Standardized Scores Evaluated One Month (Score 1) and Three Months (Score 2) after Surgery Score I Verbal tests Information Digit span Vocabulary Arithmetic Comprehension Similarities Performance tests Picture arrangement Picture completion Block design Object assembly Digit symbol Verbal IQ Performance IQ TotalIQ
Score 2
3 8 2 3 2 2
5 8 4 6 5 6
1 7 7 3
9 9
1
57 52 51
10 8 2 72 81 74
Toru Horikoshi and Others
190
TABLE II
Results of Standard Language Test of Aphasia Score I
Listening Word comprehension Sentence comprehension Execution of spoken orders Syllable comprehension Speaking Visual object naming Word repetition Explanation of picture Recall words Reading kanji word aloud Reading kana character aloud Reading kana word aloud Reading sentence aloud Reading Kanji word comprehension Kana word comprehension Sentence comprehension Execution of written orders Writing Kanji word Kana word Explanation of picture Dictation with kana character Dictation of kanji word Dictation of kana word Dictation of sentence Calculation
Score 2
Control mean
SD
100 100 50 100
100 100 100 100
100 95 96 100
2 8 7 1
35 50 100 80 60 100 100 20
50 90 100 80 100 100 100 100
98 99 90 84 100 100 100 98
4 4 16 30 8 2 2 6
70 90 10 0
100 100 80 100
99 100 96 94
8 1 10 15
60 80 20 100 100 100 100 90
100 100 100 100 100 100 100 100
84 22 14 96 86.7 18 97 II 86 20 96 16 80 30 81.5 21
Note: Score I (one month) and Score 2 (three months after surgery) presented as percentages. Kanji is a pictorial symbol in written Japanese, and kana a phonetic symbol.
However, naming of visual objects was still defective and comprehension of written sentences was below the normal average. It was more difficult for her to read kanji, the pictorial form of written Japanese, than kana, the phonetic form, and the alphabet. Three months after surgery, her color discrimination was intact but she was slow in naming colors. She could recognize categories of figures, and correctly chose a figure standing out from five alternatives in the differentiation task, e.g., a rectangular triangle among isosceles triangles. She could also match to sample meaningless complex figures. There was no difficulty in copying figures, and no spatial agnosia or left-right disorientation. In contrast, her recognition of meaningful figures was markedly impaired. She could explain verbally only 60% of map symbols, and 24% of road signs, even though she had a driver's license. She also found it difficult to indicate the meaning of the road signs by gesture or using a toy car. It seemed to be more difficult for her to recognize abstract signs than realistic ones. She made mistakes in naming 25% of visual objects, like toothbrush or shoes, and had some difficulty in explaining how to use them. Verbal recognition of pictures showing daily activities, like fire fighting or medical services in a hospital, was also difficult.
Examination of Musical Ability Her musical ability was examined during the post operative course, one and three months after surgery. All subtests were performed using a piano, because of her greater familiarity with this instrument. At first, her general music skills, hearing and singing, were examined. When the examiner sounded two tones or two sequences of tones on the piano, she could
Music alexia and pure alexia
(a)
191
(b)
(d)
7Fig. 3 - (a) Examples of 4 quarter note sequences used for testing reading ability. (b) Dotted quarter notes were also included in some sequences. (c) Notes written by the patient in the copying task. (d) Notes correctly dictated by the patient.
easily discriminate and indicate which of the two tones was higher, louder, or longer and which of two tone sequences had a faster rhythm. She also recognized and gave the names of melodies which were familiar to her. She seemed to enjoy listening to music by herself. She could sing easy popular folk songs without any mistake in pitch or rhythm. When she was asked to play the piano, she could play practice pieces which she had already learned by heart. The simple sequences of 4 quarter notes or dotted quarter notes in the C major scale, as shown in Figure 3 (a) (b), were used for further examinations, especially of reading and writing skills. Four famous nursery songs with 8 bars length were also used in some subtests. The patient was requested to listen to the four tones and the names of four notes, to read the names of four notes, and to read four notes or music. Then, she was asked to choose the appropriate sequence of four notes from six written alternatives, to write four notes on five lines, to play them on the piano, to give their names, to give the name of the song and to choose the written name of the song from four alternatives. All of these subtests consisted of 4 to 40 problems each. The same note sequences and songs were used in all subtests. In the reading test, more complex phrases with 16th notes, as shown in Figure 4, and practice pieces (familiar to her, or unlearned) were also used. In some subtests the patient was asked
Ie JJJ :rJJJ ~J33 't J Fig. 4 - Examples of sequences with 16th notes used for evaluation of rhythm reading.
192
Toru Horikoshi and Others
to play the piano with her right and left hand separately. The results of each subtest were classified into three categories; normal: all problems were correctly answered within a reasonable time; delayed : all correct but they took much time; and incorrect: one or more wrong answers. Three female volunteer piano teachers of the same age with similar experience were used as controls for response time in some of the subtests. The results are shown in Table III. The tests showed that her reading ability was significantly impaired. All tasks requiring the reading of notes or music, except copying tasks (Figure 3c), were impaired. The results of subtest requiring the choice of written notes or music were also impaired. She could not play practice pieces by sight reading. Even in simple 4 note sequences, she frequently made pitch mistakes when she played the piano or named notes. Playing with the el ft hand took the same or more time than with the right hand. Compared to pitch discrimination, her rhythm recognition was much better preserved even for complex rhythms, and she made no mistakes in the second testing session. Reading names of notes also took much time because of alexia but was better than reading notes. Her music writing skill was very good, and she could dictate notes correctly after listening to them and could copy notes without hesitation (Figure 3d). The same tendency was seen at the second evaluation, though some improvement had occurred.
DISCUSSION
Most studies have focused on localization of musical skills, especially in relationship to their hemisphere representation (Brust, 1980). While impairment of hearing and singing skills are not closely related to hemispheric dominance, music alexia and music agraphia are TABLE III
Results of Music Skills Test
Task and response required Listen to tones or songs Choose a series of notes Write notes Play on piano Give name of notes aloud Give name of song ' aloud Listen to name of notes Write notes Play on piano Read name of notes Choose a seri es of notes Write notes Play on piano Give name of notes aloud Read notes Write notes Play on piano Give name of notes aloud Read nursery songs Play on piano Give name of song aloud Choose written name of song Additional tasks Read and play on piano Familiar practice pieces Unfamiliar practice pieces Rhythm tasks
Score I 29/40 717 717 717 414
(incorrect) (normal) (normal) (normal) (normal)
Score 2 36/40 717 717 717 4/4
(incorrect) (normal) (normal) (normal) (normal)
4/4 717
(normal) (normal)
4/4 717
(normal) (normal)
717 4/4 717 4/4
(delayed) 31" «I") (delayed) 27" (4") (delayed) 10" (3") (normal)
617 4/4 717 4/4
(incorrect) (delayed) 15" (4") (delayed) 10" (3") (normal)
4/4 16/20 6112
(normal) (incorrect) (incorrect)
4/4 16120 7112
2/4 3/4 2/4
(incorrect) (incorrect) (incorrect)
2/4 4/4 4/4
(incorrect) (delayed) 54" (3") (delayed) 22" (3")
1/2
(incorrect) (incorrect) (incorrect)
212 012 6/6
(normal) (incorrect) (normal)
0/2 5/6
(normal) (incorrect) (incorrect)
NOle: Score I (one month) and Score 2 (three months after surgery). The scores are the correct response out of total questions.
Mean reaction times of the patient and controls (in parentheses) are shown for delayed responses.
Music alexia and pure alexia
193
generally found in association with language disturbance (Benton, 1977). Music alexia is usually associated to verbal alexia, and music agraphia to verbal agraphia (Brust, 1980), but cases of verbal alexia without music alexia (Basso and Capitani, 1985; Gates and Bradshaw, 1977; Luria, 1965) and of music alexia without verbal alexia have also been reported (Gates and Bradshaw, 1977). Our patient had pure music alexia combined with pure verbal alexia, similary to the case reported by AssaI (1973). Levin and Rose (1979) also reported a case of music alexia, but they did not mention music writing skill. In contrast to language, which is spontaneously learned in early life, musical skills, especially writing and reading music, require special training. Therefore, the music process in the brain may be variable among individuals, due to differences in personal experience, method used in learning music, age when learning started, or type of musical instrument. In our patient, the disturbance of visual perception involved recognition of informative figures such as road signs, map symbols, national flags and other well-known symbolic figures as well as musical notes. The meanings of these abstract symbols can be verbally decoded. Actually, many former studies suggested that nonverbal visual stimulus recognition can be mediated by verbal encoding in the left hemisphere (Boller and De Renzi, 1967; Milner, 1962; Rubino, 1970; Brewer, 1969). According to Faglioni, Scotti and Spinnler (1968), abstract symbols may lend themselves to semantic processing. Music notes, especially the pitch dimension, are very similar to abstract signs, because they are symbolic and can be represented by words like do, re, mi, etc. Our patient confessed that she had always used to translate the pitch of notes into an inner representation of sound, i.e., do, re, mi, when she read music. Thus, music symbols, even if apparently nonverbal, may undergo verbal processing. The recognition of rhythm and pitch of notes in music reading dissociated in our patient. She could recognize the rhythm of phrases fairly well in spite of significant impairment of pitch discrimination. This pattern is similar to that of Brust's (1980) patient who had damage to the left temporal and parietal lobes. In contrast, AssaI reported (1973) a patient with Wernicke's aphasia and mild music dyslexia, who had much more difficulty in rhythm reading than in pitch. Such discrepancies have already been described in other types of amusia (Brust, 1980; Shapiro, Grossman and Gardner, 1981; Peretz, 1990), and various proposals have been made about hemisphere representation of musical skills and dissociation in music processing (Hachinski, 1989). Bever and Chiarello (1974) suggested that the left hemisphere is dominant for analytic processing and the right hemisphere for holistic processing. Since melodies are composed of an ordered series of pitches, they should be mainly processed by the left hemisphere. Barbizet et al. (1972, cited by Benton, 1977) thought that the right hemisphere participation in musical activity is primarily at the perceptual and executive level, while the left hemisphere mediates the recognition and memory of musical structures, the symbolic process of reading and writing music and the higher level integrated functions involved in musical composition. In the studies of hemisphere depression following intracarotid sodium amy tal injection (Gordon and Bogen, 1974), the rhythmic aspect was unaffected by the depression of either hemisphere, while pitch discrimination in singing was disturbed by depression of the right hemisphere. Gordon and Bogen (1974) suggested that, though the left hemisphere may be superior at rhythmic perception, the right side is still quite capable of rhythm processing. Mostafa, Kotby, Barakah et al. (1989) demonstrated deterioration of all musical abilities in patients with right-side brain damage, but only rhythm disturbance in patients with left-side brain damage. Grossman, Shapiro and Gardner (1981) reported that the musical judgments of patients with right anterior brain damage relied on the mood of the phrases, while patients with Broca's aphasia relied on a linear sequential musical processing strategy. These data suggest that each hemisphere may make a distinctive contribution to pitch and rhythm in receptive and expressive musical performance, but the nature of each hemisphere contribution is still controversial. Our results seem to be consistent with suggestions by Bever and Chiarello (1974) and Barbizet et al. (1972, cited by Benton, 1977), but do not permit to conclude whether the discrepancy in reading pitch and rhythm is due to their different hemisphere representation. What we can say is that rhythm discrimination was less affected than pitch recognition, when the patient decoded music symbols. Another interesting finding of this study is the greater improvement of word reading disturbance compared to music reading disturbance. Language skills rapidly improved in the
194
Toru Horikoshi and Others
observation period, while music readingi~pairment persisted. This difference may be due to the differential training-that the two skills received. There is usually no stimulus requiring music reading in daily life, while there is abundant opportunity to read words in television, magazines, or newspapers. The speech training undergone in the former hospital could also have promoted the improvement of word reading. Alternatively, music reading skills may be similar to the second language of bilingual individuals, which is generally more severely disturbed than the mother language following brain damage. The pattern of the patient's neurological deficits was typical for pure alexia with hemianopsia. Visual information projected to the right visual cortex was difficult to analyze verbally in the left posterior hemisphere, because of the splenial interruption of the interhemisperic fibers, and/or dysfunction of the posterior part of the left hemisphere, as shown by MRI and the SPECT study. Piano playing with the left hand was disturbed to the same degree or even more than with the right hand, which suggests that the information from the right visual cortex had to be first processed in the left temporoparietal lobe before being transmitted to the right motor area via the body of the corpus callosum.
Acknowledgement. We would like to express our thanks to Ms. Kumiko Nozawa of Yamanashi Symphony Orchestra for her technical assistance and advice. REFERENCES ASSAL, G. Aphasie de Wernicke sans amusie chez un pianiste. Revue Neurologique, 129: 251-255, 1973. BASSO, A., and CAPITAN!, E. Spared musical abilities in a conductor with global aphasia and ideomotor apraxia. Journal of Neurology, Neurosurgery and Psychiatry, 48: 407-412, 1985. BENTON, A.L. The amusia. In M. Critchley and R.A. Henson (Eds.), Music and the Brain. London: William Heinemann Medical Books, 1977, pp. 378-397. BEVER, T.G, and CHIARIELLO, R.J. Cerebral dominance in musicians and nonmusicians. Science, 185: 537-539, 1974. BOLLER, F., and DE RENZI, E. Relationship between visual memory defects and hemispheric locus of lesion. Neurology, 17: 1052-1058, 1967. BREWER, W.F. Visual memory, verbal encoding and hemispheric localization. Cortex,S: 145-151, 1969. BRUST, J.e.M. Music and language. Musical alexia and agraphia. Brain, 103: 367-392, 1980. FAGLlONI, P., SCOTTI, G., and SPINNLER, H. Impaired recognition of written letters following unilateral hemispheric damage. Cortex,S: 120-133, 1968. GATES, A., and BRADSHAW, L. The role of the hemispheres in music. Brain and Language, 4: 403431, 1977. GORDON, H.W., and BOGEN, J.E. Hemispheric lateralization of singing after intracarotid sodium amylobarbitone. Journal of Neurology, Neurosurgery and Psychiatry, 37: 727-738, 1974. GROSSMAN, M., SHAPIRO, B.E., and GARDNER, H. Dissociable musical processing strategies after localized brain damage. Neuropsychologia, 19: 425-433, 1981. HACH!NSKI, V.C. Effect of strokes on musical ability and performance. Seminars in Neurology, 9: 159162, 1989. LEVIN, H.S., and ROSE, J.E. Alexia without agraphia in a musician after transcallosal removal of a left intraventricular meningioma. Neurosurgery, 4: 168-173, 1979. LURIA, A.E., TSVETKOVA, L.S., and FUTER, D.S. Aphasia in composer. Journal of the Neurological Sciences, 2: 288-292, 1965. MAZZIOTTA, J.e., PHELPS, M.E., CARSON, R.E., and KUHL, D.E. Tomographic mapping of human cerebral metabolism; Auditory stimulation. Neurology, 32: 921-937, 1982. MILNER, B. Laterality effects in audition. In V.B. Mountcastle (Eds.), Interhemispheric Relations and Cerebral Dominance. Baltimore: Johns Hopkins Press, 1962, pp. 177-195. MOSTAFA, M., KOTBY, M.N., BARAKAH, M., EL-SADY, S., ALLOSH, T., ELSHOBARY, A., and SALEH, M. Dominant functions of right versus left hemisphere. Acta Oto-laryngologica (Stockholm), 107: 479-484, 1989. PERETZ, T. Processing of local and global musical information by unilateral brain-damaged patients. Brain, 113: 1185-1205, 1990. RUBINO, e.A. Hemispheric lateralization of visual perception. Cortex, 6: 102-130, 1970. SHAPIRO, B.E., GROSSMAN, M., and GARDNER, H. Selective musical processing deficits in brain damaged populations. Neuropsychologia, 19: 161-169, 1981. Toru Horikoshi, M.D., Department of Neurosurgery, Yarnanashi Medical University, 1110 Shimokato, Tamaho-machi, Yamanashi, Japan, 409·38.
(Received 24 October 1995; accepted 19 April 1996)