BRAIN AND LANGUAGE IO, 120-131 (1980) Semantic Aphasia: A Neglected Entity DANIEL B. HIER, SHELLEY I. MOGIL, NAN P. RUBIN, AND GAIL R. KOMROS Departments of Neurology and Speech Pathology, Massachusetts Rehabilitation Hospital, and the Neurology Service, Massachusetts General Hospital Three cases of semantic aphasia are reported. Computerized brain tomography showed bilateral temporo-parieto-occipital junction hemorrhages in one patient, and left parieto-occipital junction infarctions in the other two patients. The auditory comprehension defect of the three patients was characterized by preserved understanding of single words and impaired understanding of grammatically complex constructions. It is suggested that this comprehension defect reflects an inability to fully grasp the meaning of words and grammatical constructions imbued with spatial or quasi-spatial signiticance. Each of the three patients showed a complex spatial disorder that included constructional apraxia, spatial agnosia, and elements of Gerstmann’s syndrome. The aphasic as well as the spatial disorder of semantic aphasics may be manifestations of a common defect in the perception of spatial relationships produced by left temporo-parieto-occipital region damage. Head (1920) introduced the term semantic aphasia to describe patients who could understand isolated words, yet were unable to “grasp the ultimate meaning” of a sentence. However, with the exception of Luria (1970) and Brown (1972), recent taxonomists of the aphasias have not included semantic aphasia as a diagnostic entity. Nonetheless, semantic aphasia is of theoretical interest, for its existence suggests that focal brain injuries may selectively impair the understanding of syntax (grammatical structure) while sparing the understanding of lexicon (vocabulary). Different psychological operations (possible occurring at anatomically discrete sites) may underlie the comprehension of lexicon and syntax. The three patients described below showed the comprehension deficits characteristic of semantic aphasia. Supported in part by NINCDS Fellowship NS05917. Address reprint requests to Daniel B. Hier, M.D., Department of Neurology, Michael Reese Hospital, Chicago, IL 60616. 120 0093-934x/80/03012012$02.00/0 Copyright All ri@ts @ 1980 by Academic Press, Inc. of reproduction in any form reserved. SEMANTIC APHASIA REPORT OF CASES 121 Case V.S. This 62-year-old right-handed woman has a 3-year history of lymphoma treated intermittently with chemotherapy and radiation. On March 5, 1978 she suddenly developed right hemiplegia and aphasia after undergoing a surgical procedure for removal of a ureteral stone. Computerized brain tomography showed hemorrhages in both the left and right parietal lobes. On neurological examination one month later, she was noted to have minimal right arm drift, a right homonymous inferior quadrantanopia, hyperreflexia on the right, and a right hemisensory loss to pinprick, vibration, and proprioception. Stereognosis was impaired bilaterally, more so in the right than left hand. Marked ideomotor apraxia was present. A repeat computerized brain tomogram showed areas of resolving hemorrhage at the left and right temporo-parieto-occipital junctions (Figs. 1 and 2). Case H.B. This X-year-old right-handed man with a long history of FIG. 1. hemorrhages. Computerized brain tomogram There are areas of resolving of V.S. obtained 5 weeks after onset of cerebral hemorrhage in both parietal lobes. 122 HIER ET AL. FIG. 2. Lateral views of left and right hemispheres detailing approximate intrace :rebral hemmorhages in VS. extent of hypertension and peripheral vascular disease suddenly developed difficulty in speaking and an unsteady gait on September 2, 1978. Neurological examination revealed a right homonymous inferior quadrantanopia, but no weakness, sensory loss, or reflex asymmetry. There was no ideomotor or ideational apraxia. Computerized brain tomography showed an area of infarction at the left parieto-occipital junction (Figs. 3 and 4). Case AN. This 57-year-old right-handed woman has a long history of both hypertension and diabetes mellitus. She completed nine school FIG. 3. Computerized brain tomogram of H.B. obtained 1 week after onset of stroke. There is a focal infarction at the left parieto-occipital junction. SEMANTIC APHASIA 123 FIG. 4. Lateral view of left hemisphere of H.B. detailing extent of left parieto- .occipital infarct. grades and her premorbid intelligence is estimated to have been low normal to borderline. On April 25, 1979 she suddenly developed headache, nausea, and vomiting. On immediate neurological examination, she was noted to ignore her right side. A right homonymous hemianopia was present, as was a mild right arm drift and hyperrflexia on the right. Sensation was intact and there was no ideomotor or ideational apraxia. Two days later she sustained several brief generalized seizures that did not recur on anticonvulsant medication. A computerized brain tomogram showed an area of low density at the left temporo-parieto-occipital junction compatible with early infarction (Figs. 5 and 6). INVESTIGATIONS OF HIGHER CORTICAL FUNCTION Aphasia. All three patients spoke fluently and effortlessly, without evidence of agrammatism, dysprosody, or literal paraphasia. Each exhibited occasional verbal pat-aphasias and word-finding difficulties. No comprehension defect was apparent in casual conversation. The quantity of speech output by V.S. and A.N. was within normal limits whereas H.B. tended to be terse and laconic. The performance of the three patients on the Boston Diagnositc Aphasia Examination (BDAE) was compared to mean performance of the test standardization sample of 207 aphasics (Table 1; Goodglass & Kaplan, 1972). On the BDAE V.S. and H.B. were unimpaired with respect to fluency, repetition, and automatic speech production, whereas A.N. showed some impairment in repetition and automatic speech production. All three patients showed alexia, agraphia, and mild anomia. On the BDAE subtests of auditory comprehension, all three patients did substantially better in comprehending single words (V.S. 100% correct, H.B. 86% correct, A.N. 93% correct) than in comprehending complex ideational material (V.S. 50% correct, H.B. 50% correct, A.N. 75% correct), suggesting difficulties in dealing with syntactical complexity. The auditory comprehension disturbance was further delineated as described below. Auditory comprehension. Supplementary auditory comprehension tests included the Quick Test (Ammons & Ammons, 1962), the Token Test (DeRenzi & Vignolo, 1962), and the Logical-Grammatical Sentence Comprehension Test (Wiig & Semel, 1974). The Quick Test is a measure of 124 HIER ET AL. FIG . 5. Computerized brain tomogram of A.N. obtained 1 week after onset of str.oke. There is 2m area of low density at the left parieto-occipital junction compatible with new infarc, tion (arrows). receptive vocabulary (auditory comprehension at the single word level). Form 1 of the test consists of 50 vocabulary items of increasing lexical difficulty (e.g., belt through cacophony). After dictation of each vocabulary item, a pointing response is required to one of four picture plates. On the Quick Test, V.S. gave 48 of 50 correct responses (90th percentile for normal adults), H.B. gave 47 of 50 correct responses (85th percentile for adults), and A.N. 32 of 50 correct responses (10th percentile for adults). A.N.‘s lower score is compatible with premorbid estimates of her intelligence and probably does not reflect new deficits. The Token Test and Logical-Grammatical Sentence Comprehension Test were administered to assess comprehension of grammatically com- FIG. 6. Lateral view of left hemisphere of A.N. detailing extent of infarction. SEMANTIC TABLE PERFORMANCE OF THE BOSTON 125 APHASIA 1 DIAGNOSTIC APHASIA EXAMINATION Correct responses (%) Section Fluency Naming Auditory comprehension Repetition Automatic speech Oral reading Reading comprehension Writing Standardization sample of 207 aphasics V.S. H.B. A.N. 66 58 100 83 96 77 93 71 72 53 83 100 81 100 94 54 63 57 100 95 100 92 75 80 69 54 61 28 83 68 68 69 plex constructions. The Token Test consists of 61 instructions for the manipulation of chips of various shape and color. Lexical complexity remains low and approximately constant throughout the series, whereas information content and syntactical complexity of the instructions increase progressively (Table 2). All three patients had considerable difficulty with Part V, the part with highest syntactical complexity and the only part to utilize prepositions. Difficulties in the comprehension of syntactically complex constructions were confirmed by administation of the Logical-Grammatical Sentence Comprehension Test (Table 3). Although all three patients showed good comprehension of comparative relationships, each had great difficulty in grasping temporal, spatial, or passive relationships. TABLE PERFORMANCE 2 ON THE TOKEN TEST Correct responses Part Sample item V.S. H.B. A.N. I. Touch the red circle. Touch the small yellow circle. Touch the yellow circle and the red rectangle. Touch the small yellow circle and the large green rectangle. Put the blue circle under the white rectangle. 9110 lo/lo lO/lO lo/lo 9110 7110 4110 lO/lO 5110 4110 9110 o/10 7121 32l61 12121 51/61 4121 25161 II. III. IV. V. Total 126 HIER ET AL. TABLE PERFORMANCE 3 OF THE LOGICAL-GRAMMATICAL SENTENCE COMPREHENSION TEST Correct responses Part Comparative Temporal Passive Spatial Sample item V.S. H.B. A.N. Are watermelons bigger than apples? Does lunch come after dinner? John was hit by Eric. Was Eric hit? The elephant sat on the mouse. Was the mouse on top? lO/lO 7110 5110” lO/lO 6/10 S/10” 8110 4110’ 5110’ 5110” 3110” 3110” a Correct responses at chance level or below. Intelligence. The Wechsler Adult Intelligence Scale (WAIS) was administered to all three patients (Table 4; Wechsler, 1955). V.S. and H.B. performed similarly on the WAIS, obtaining substantially higher verbal than performance IQs. A.N. scored lower on all scales, at least partially reflecting her lower premorbid intelligence. All three patients did poorly on the subtests requiring analysis of spatial relationships (Block Design and Object Assembly) or the comprehension of the figurative and metaphorical aspects of language (Similarities and Comprehension). V.S. and H.B. performed normally on the Vocabulary and Digit Span subtests, whereas A.N. showed considerable weakness on these subtests. Constructional apraxia. All three patients showed a striking inability to manipulate objects in space (WAIS Block Design and Object Assembly TABLE PERFORMANCE OF THE WECHSLER 4 ADULT INTELLIGENCE SCALES V.S. H.B. A.N. Full scale IQ Verbal IQ Performance IQ 90 99 76 85 94 74 62 64 65 Information Comprehension Arithmetic Similarities Vocabulary Digit Span 10 7 3 8 11 14 10 7 5 7 11 10 Digit symbol Picture completion Block design Picture arrangement Object assembly 0 7 0 6 3 0 7 3 4 3 a Subtest scores are scaled scores. SEMANTIC vs 127 APHASIA HB AN FIG. 7. Attempts of three patients to copy a cube. subtests). None of the patients was able to find hidden figures (Within et al., 1971) or to copy geometric shapes (Fig. 7). Finger agnosia. V.S. and H.B. were unable to reliably identify on dictated command fingers on either their own hands or the examiner’s hands. A.N. showed no finger agnosia. In contrast, all three patients could accurately name fingers on their own hands. Right-left orientation. V.S. was unable to discriminate right from left on her own body or the examiner’s body. H.B. identified right and left correctly on his own body, but was unable to execute crossed-body commands or to discriminate right from left on the examiner’s body. A.N. showed no right-left confusion. Culculutions. Although all three patients could count serially without difkulty and understood the numerical value of numbers, each experienced severe difficulties in adding or subtracting except when the numbers were limited to single digits. Dyscalculia was confirmed by their low arithmetic subtest scores dn the WAIS (Table 4). Writing. All three patients wrote with their preferred right hands. Their writing was characterized by poor legibility, improper spacing, and mis- FIG. 8. Writing to dictation: “She can’t see them” (from the Boston Diagnostic Aphasia Examination). From top to bottom: VS., H.B., and A.N. 128 HIER ET AL. spellings (letter distortions, letter reversals, and letter substitutions, see Fig. 8). Each patient was more adept at oral spelling than at written spelling (cf., patient reported by Kinsbourne & Rosenfield, 1974). Upon dictation of 10 words from the BDAE, correct spelling scores were as follows: Oral spelling Written spelling V.S. H.B. A.N. 9/10 3110 7110 4110 6/10 3110 The superiority of oral over written spelling suggests that the etiology of the writing disturbance in the three patients may have been primarily spatial or apraxic rather than linguistic. DISCUSSION Nature of the Syndrome The three patients described above exemplify semantic aphasia, a syndrome that Head (1923) described as “a want of recognition of the full significance of words and phrases.” Luria et al. (1969) have noted that affected patients “understand ordinary speech addressed to them, they answer questions adequately, and their only difficulty in spoken speech is a slight tendency to forget words. However, detailed investigations reveal a gross disturbance of the logical structure of speech; of the whole complex system all that remains is mere vocabulary, and the understanding of speech does not extend beyond the limits of simple phrases, while perception and understanding of complex logico-grammatical relationships in a phrase are disturbed.” Semantic aphasics show little impairment in many areas of language function. The three patients described above did not manifest notable deficiencies in articulation, auditory discrimination, fluency, or prosody of speech. None spoke agrammatically and none showed gross comprehension deficits in casual conversation. All three patients showed an intact ability to comprehend single words as expressed by their performance on the Quick Test. In contrast, each patient showed a disturbed comprehension of syntactically complex constructions as illustrated by their difhculties on the Logical-Grammatical Sentence Comprehension Test and Part V of the Token Test. This dissociation between the comprehension of lexicon (vocabulary) and that of syntax (grammatical constructions) typifies semantic aphasia. The Accompanying Spatial Disorder Along with their difficulties in comprehending syntax, the three patients described above manifested a complex spatial disorder characterized by constructional apraxia and elements of the Gerstmann’s syndrome. Luria SEMANTIC APHASIA 129 (1973) has suggested that this spatial disorder is not simply an incidental accompaniment of semantic aphasia, but rather that semantic aphasia is “the same defect of perception of simultaneous spatial structures, but transferred to a higher (symbolic) level.” Consistent with Luria’s hypothesis, semantic aphasics have extraordinary difficulty in grasping the meaning of words imbued with spatial or quasi-spatial significance. For example, the three patients described above were able to use correctly on the Quick Test the words crystallized, saccharin, immature, and decisive. Yet, none of them was able to follow correctly commands on the Token Test that involved the spatial prepositions beside, under, behind, before, or away from. Furthermore, both V.S. and H.B. (but not A.N.) experienced difficulty in discriminating right from left and in identifying fingers. Just as right and left are spatial concepts, so do the names of fingers carry spatial significance. Thus, words devoid of spatial significance pose little difficulty for semantic aphasics, whereas words denoting spatial or quasi-spatial relationships befuddle them. Not only are semantic aphasics impaired in their comprehension of words imbued with spatial significance, they are also impaired in their comprehension of grammatical structure. This impairment in the analysis of grammatical structure is a linguistic analog to their impairment in the analysis of spatial structures. The inability to analyze grammatical structures leaves semantic aphasics confused as to the difference in meaning between sentences such as “the lion kills the man” and “the lion is killed by the man.” Even though the semantic aphasic understands the meaning of each of the major lexical items (lion, kill, man), his inability to analyze grammatical structure according to the principles of syntax leaves him confused as to the ultimate meaning of the sentence. Just as semantic aphasics are inept at the manipulation of objects in space, so are they be impaired in the manipulations of grammatical structure that are requisite to the comprehension of syntax. As Luria (1970) suggests, the semantic aphasic is unable to “manipulate the inner schemata of statements” and thus is “uable to process or decode incoming information according to the logico-grammatical code . . . of language.” Dissociated Comprehension of Syntax and Lexicon The observation that semantic aphasics are disproportionately impaired in the comprehension of syntax supports the hypothesis of Caramazza and Berndt (1978) that psychologically distinct mechanisms, possibly occurring at discrete anatomic sites, subserve the comprehension of syntax and lexicon. Patients with damage to the left temporal lobe (anemic and Wemicke’ aphasics) appear to have particular dficulty understanding the major lexical items of a sentence (Goodglass et al., 1970; Heilman & Scholes, 1976; Coughlan & Warrington, 1978). Patients with damage to the left temporo-parieto-occipital junction, like the three patients de- 130 HIER ET AL. scribed above,’ show particular diaculty understanding the grammatical structure of a sentence (Head, 1920; Luria, 1970). These observations suggest an intriguing dichotomy: left temporal lobe damage producing impairment in the comprehension of lexicon, left temporo-parietooccipital junction damage producing impairment in the comprehension of syntax. It should be noted that Broca’s and conduction aphasics share with semantic aphasics difficulties in dealing with syntactical complexity (Caramazza & Berndt, 1978; Heilman & Scholes, 1976). However, the dissociation between the comprehension of syntax and lexicon is far less dramatic in cases of Broca’s or conduction aphasia. Additional clinicalanatomical studies should resolve the extent to which disorders in the comprehension of grammatical structure show anatomical correlation with lesions of the left temporo-parieto-occipital region and psychological correlation with spatial disorders. Similarities to the Language Deficit of Left Hemidecorticates Finally, a similarity may be noted between the language impairment of children undergoing left hemidecortication and that of semantic aphasics. Like semantic aphasics, children with left hemidecortications may exhibit normal verbal intelligence, normal articulation, normal expressive and receptive vocabularies, and normal auditory discrimination (Dennis & Whitaker, 1976). However, children with left hemidecortication, unlike right hemidecorticates, have difficulty grasping complex grammatical structure. Like the three semantic aphasics described above, left hemidecorticates do poorly on Part V of the Token Test and have difficulty in understanding passive voice constructions. As Dermis and Whitaker (1976) note, “syntactic abilities have not been acquired as well by the isolated right hemisphere as by the left.” If the above hypothesis that lexical operations are localizable to the vicinity of the left temporal lobe and syntactical operations are localizable to the vicinity of the left temporo-parieto-occipital junction is correct, then the experience with left hemidecorticates suggests that whereas the right temporal lobe can effectively supstitute for the left with respect to lexical operations, the right temporo-parieto-occipital junction cannot replace the left with respect to syntactic operations. REFERENCES Ammons, R. B., & Ammons, C. H. 1%2. The quick Specialists. rest. Missoula, MT: Psychological Test 1 Patient VS. had a small hemorrhage in the right parietal lobe as well as a larger hemorrhage in the left parietal lobe (Figs. 1 and 2). Although it seems parsimonious to attribute her semantic aphasia to the larger left hemisphere hemorrhage, it should be cautioned that subtle verbal comprehension deficits have been reported to occur after right hemisphere injury. SEMANTIC APHASIA 131 Brown, J. W. 1972. Aphasia, upruxia, and agnosia. Springfield, Ill.: Thomas, Pp. 31-55. Caramazza, A., & Bemdt, R. S. 1978. Semantic and syntactic processes in aphasia: A review of the literature. Psychological Bulletin, 85, 898-918. Coughlan, A. K., & Wanington, E. K. 1978. Word-comprehension and word-retrieval in patients with localized cerebral lesions. Brain, 101, 163-185. Dennis, M., & Whitaker, H. A. 1976. Language acquisition following hemidecortication: Linguistic superiority of the left over the right hemisphere. Brain and Language, 3, 404-433. DeRenzi, E., & Vignolo, L. A. 1962. The token test: A sensitive test to detect receptive disturbances in aphasics. Bruin, 85, 665-678. Goodglass, H., & Kaplan, E. 1972. The assessment of aphnsiu and related disorders. Philadelphia: Lea and Febiger. Goodglass, H., Gleason, J. B., & Hyde, M. R. 1970. Some dimensions of auditory comprehension in aphasia. Journal of Speech and Hearing Research, 13, 595406. Head, H. 1920. Aphasia and kindred disorders of speech. Brain, 43, 87-165. Head, H. 1923. Speech and cerebral localization. Brain, 46, 355-528. Heilman, K. M., & Scholes, R. J. 1976. The nature of comprehension errors in Broca’s, conduction, and Wemicke’s aphasics. Cortex, 12, 258-265. Kinsbourne, M., & Rosenfield, D. B. 1974. Agraphia selective for written spelling. Brain and Language, 1, 215-225. Luria, A. R. 1970. Traumatic aphasia. The Hague: Mouton. Luria, A. R. 1973. The working brain. New York: Basic Books. Luria, A. R., Naydin, V. L., Tsvetkova, L. S., & Vinarskaya, E. N. Restoration of higher cortical function following local brain damage. In P. J. Vinken and G. W. Bruyn (Eds.), Handbook of clinical neurology. Amsterdam: North-Holland Vol. 3, pp. 368-433. Wechsler, D. 1955. Manual for the Wechsler Adult Intelligence Scale. New York: Psychological Corporation. Wiig, E. H., & Semel, E. M. 1974. Development of comprehensions of logical-grammatical sentences by grade school children. Perceptual and Motor Skills, 38, 171-176. Witkin, H. A., Oltman, P. K., Raskin, E., & Karp, S. A. 1971. Manual for the Embedded Figures Test. Palo Alto, CA: Consulting Psychologists Press.