BRAIN AND COGNITION ARTICLE NO.

33, 135150 (1997) BR970888

Patterns of Language Dominance in Focal Left and Right Hemisphere Epilepsies: Relation to MRI Findings, EEG, Sex, and Age at Onset of Epilepsy
C. HELMSTAEDTER AND M. KURTHEN
Department of Epileptology, University of Bonn, Germany

D. B. LINKE
Department of Neurosurgery, University of Bonn, Germany
AND

C. E. ELGER
Department of Epileptology, University of Bonn, Germany The present study evaluates results of language testing during intracarotid amobarbital procedures in 167 patients with either left (LHE, n 81) or right hemisphere epilepsies (RHE, n 86). In both groups there was a high rate of deviance from complete left hemisphere dominance of 24 and 31%, respectively. Whilst complete right hemisphere or incomplete left hemisphere language dominance were the prominent atypical patterns in LHE, RHE was associated with either bilateral dominance or incomplete left dominance. In LHE, atypical language dominance was frequently associated with an extratemporal localization of lesions or epileptic foci. The age at onset of epilepsy and the degree of right hemisphere language dominance correlated signicantly in LHE but not in RHE. Finally, atypical dominance in LHE but not in RHE was associated with poorer language and nonlanguage functions, the latter being negatively correlated with the degree of right hemisphere language dominance. Conclusions are: (1) The data contradict the assumption of equipotentiality and favor the supposition of a predetermined left hemisphere superiority in language processing. (2) Atypical language dominance in LHE can largely be explained in terms of a plasticity dependent language shift as a consequence of early left hemisphere epilepsies and lesions. (3) Atypical dominance patterns in RHE appear to reect the prevalence of genetically determined variants and the possibility of a language transfer from the right to the left hemisphere. 1997 Academic Press

Address correspondence and reprint requests to C. Helmstaedter, Ph.D., Department of Epileptology, University of Bonn, Sigmund Freud Str. 25, 53105 Bonn, Germany 135 0278-2626/97 $25.00
Copyright 1997 by Academic Press All rights of reproduction in any form reserved.

136

HELMSTAEDTER ET AL.

INTRODUCTION

Since its introduction in 1949 (Wada, 1949) the intracarotid amobarbital procedure (IAP) has become an essential part of the presurgical evaluation of patients with focal epilepsies. Several studies using this technique have shown an increased incidence of atypical language dominance patterns in patients with left hemisphere injury. These have been interpreted in terms of interhemispheric language transfer and cerebral plasticity (Rasmussen & Milner, 1975, 1977). Multiple factors (age at onset of lesion, sex, localization and extent of the lesion, time of recovery) have been identied as contributing to the right hemispheric restitution of cognitive functions originally attributed to the left hemisphere (Ogden, 1988, Satz & Strauss et al., 1988; Dennis, 1980; Strauss et al., 1992). However, basic questions concerning the genetic predisposition of hemisphere dominance, equipotentiality and the participation of the right hemisphere in left sided language dominant subjects, are still under discussion. IAP for language lateralization in presurgical patients is usually indicated when the primary epileptic focus or the associated lesions are suspected to involve language and speech relevant brain regions, or when atypical dominance is suggested by left handedness. Consequently, atypical dominance patterns in patients with right hemisphere lesions or epilepsy have rarely been evaluated or they were excluded from closer examination and discussion (Rasmussen & Milner, 1975, 1977; Loring et al, 1990; Strauss, Satz, & Wada, 1990). Most estimations of the genetic prevalence of right hemisphere language representation stem from studies on patients with cerebrovascular diseases. In a large sample of aphasic patients, Bryden, Hecaen, & DeAgostino (1983) found a right hemisphere language representation in 13.9% right handed and 24% left handed patients. Eight percent were suggested to have a bilateral language representation. The few studies on patients with right hemisphere lesions or epilepsies reported a rate of atypical language dominance patterns ranging between 4% and 20% (Zatorre, 1989; Rey, Dellatolas, Bancaud, & Talairach, 1988; Mateer & Dodrill, 1983). Interestingly, in these patients, language appeared to be distributed bilaterally rather than being restricted to the right hemisphere. In our series of patients undergoing presurgical evaluation, IAP of the right hemisphere was obligatory. Thus, we were able to investigate the occurrence of right hemispheric language in an unselected group of patients with left or right hemispheric focal epilepsies. Complete sets of data of neuroradiological evaluations, invasive EEG monitoring, IAP, and comprehensive neuropsychological evaluations were available in 167 patients. We evaluated the prevalence of language dominance patterns in left and right hemisphere epilepsies, the relation between handedness and language dominance, the inuence of sex, age at onset of epilepsy (with regard to cerebral plasticity),

LANGUAGE DOMINANCE

137

the inuence of the localization and extension of epileptic foci and lesions on dominance patterns, and the question of the cognitive consequences of an atypical language dominance.

METHOD

Subjects The sample consisted of 167 patients with focal epilepsies who were candidates for surgical treatment of epilepsy: 85 patients were men, 82 women; 101 had lesions on MRI/CCT. Postoperative histology revealed tumors (50%), malformations (25%), atrophy (13%), vascular disorders (6%), and other lesions (6% hematoma, abscesses, necroses etc.) in the group of patients with lesions found by imaging techniques. According to semi-invasive and invasive EEG-ECoG recordings, 86 patients (51%) had right hemispheric, and 81 patients had left hemispheric (49%) epilepsies. Patients with left and right hemispheric epilepsies did not differ signicantly in terms of the age at the time of evaluation (LHE mean age 26.9 years, SD 10.2; RHE mean age 29.6 years, SD 10.6; t test: t 1.61, p .10), age of onset of epilepsy (LHE 10.5, SD 7.1; RHE 12.2; SD 9.2; t 1.33, p .18) or the prevalence of MRI/CCT lesions (LHE 66%, RHE 56% 2 1.23, p .26 with Yates correction). The men/ women ratio was 60/40 in the left, and 42/58 in the right hemisphere epilepsy group (2 5.0, p .02 with Yates correction). All patients underwent IAP and extensive neuropsychological examination. Language Dominance as Determined by IAP All patients underwent left and right intracarotid injection of 200 mg amobarbital in a 10% solution on two consecutive days. The procedure was videotaped and monitored with either surface EEG or ECoG in order to monitor the barbiturate effect. Language testing included: 1. Serial speech: counting backward immediately before amobarbital injection, to be continued after the injection. 2. Receptive speech functions I: Comprehension of spoken commands. Four cards showing four objects each were presented, and the patient was asked to point at a named object (Point to the cat). 3. Receptive speech functions II: The patient was asked to perform three actions (press my hand; raise your arm; show your tongue). 4. Naming of four pictorially presented objects. 5. Repetition of two proverbs.

138

HELMSTAEDTER ET AL.

6. Reading of two short sentences and a paragraph consisting of three complex sentences. 7. Spontaneous speech: spontaneous articulation during the procedure. Testing of language functions was performed during the maximal barbiturate effect as indicated by EEG-slowing and paresis. The total performance score in either IAP was dened as the sum of subscores for performance in the individual tasks (1 to 7). Two points were given for a faultless performance of each task, one point for an impaired performance, and no point for a complete failure. Thus, the total score for each IAP ranged between 0 and 14. The lateralization index L was then computed according to the formula L (Score IAP right Score IAP left)/(Score IAP right Score IAP left). L was corrected by multiplication with n/m, with n being the score of the IAP with better language performance, and m being the highest possible score in the IAP (here 14 points). The quotient yielded a value between 1 and 1, 1 expressing complete left hemisphere language dominance and 1 indicating complete right hemisphere language dominance. L-values between 1 and 1 indicate the degree of bilateral language representation, with a maximum at L 0. A more detailed description of the IAP procedure and its evaluation is provided in Kurthen et al. (1992, 1994). Comprehensive Neuropsychological Evaluation All patients received a neuropsychological assessment at the beginning of the presurgical evaluation, when the antiepileptic drug regimen and seizure frequency were stable. The evaluation assessed basic functions of attention and visuomotor speed (letter cancellation/d2-Aufmerksamkeitsbelastungstest; Brickenkamp, 1978), material-specic memory functions (verbal learning and memory/VLMT, Helmstaedter & Durwen, 1990; gural learning and memory/DCS-R, Helmstaedter, Pohl, Hufnagel, & Elger, 1991), language functions (comprehension/Token Test, De Renzi, 1962; word uency/LPS, Horn, 1983) and higher verbal and visuospatial cognitive functions (verbal reasoning/IST, Amthauer, 1973; mental rotation/LPS, Horn, 1983; block design/WAIS, 1955). Vocabulary served as an estimator for education and intelligence (MWT-B, Lehrl, 1978). For more detailed information on the tests see also Helmstaedter, Kurthen, Linke, & Elger, 1994. Most tests were subtests of normed and standardized German intelligence tests (MWT-B, LPS, IST, HAWIE/WAIS). Normative data for the memory tests were acquired in healthy controls (VLMT, n 100, DCS-R, n 57). To get a comprehensive overview, the different tests were semistandardized and rated between 0 and 4, depending on whether the scores were markedly below average ( mean 2 SD 0), below average ( mean 1 SD 1), average (mean 1 SD 2), above average ( mean 1 SD 3) or markedly above average ( mean 2 SD 4) when compared to the normative test data of healthy controls.

LANGUAGE DOMINANCE

139

FIG. 1. Distribution of language lateralization indices in the total group. LD/RD, left/right hemisphere language dominance; BIL, bilateral language representation; ILD/IRD, incomplete left/right hemisphere dominance.

RESULTS

Distribution of Lateralization Indices and Patterns of Language Dominance As can be seen in Fig. 1, lateralization indices were not normally distributed. Their distribution suggested a subdivision into the following groups: left hemisphere language dominance (LD: index 0.79), incomplete left hemisphere language dominance (ILD: index ranges between 0.8 and 0.3), bilateral language representation (BIL: index ranges between 0.29 and 0.29), incomplete right hemisphere language dominance (IRD: index ranges between 0.3 and 0.7), and right hemisphere language dominance (RD: index 0.79). This categorization differs somewhat from that reported in our 1992 study in that patients with indices from 0.8 to 0.9 and from 0.8 to 0.9 are now included in the clearly lateralized groups. These patients deviate from complete left or right dominance with indices of 1 and 1 only by failing on no more than a few test items. The denition of the bilateral group is more strict now, as only those patients are included who have partially lost language functions in one or both IAPs. Patterns of Language Dominance in Left and Right Hemisphere Epilepsy Following the above classication, 24.4% of the patients with right hemisphere (n 21) and 30.8% of the patients with left hemisphere epilepsies (n 25) showed a right hemispheric participation in language processing (2 0.58 p .44; Fig. 2). While the absolute frequency of a right hemisphere participation in language functions was comparable in LHE and RHE, the occurrence of RD and BIL was different in these groups ( 2 13.37, df

140

HELMSTAEDTER ET AL.

FIG. 2. Patterns of atypical language dominance in left (LHE) vs. right (RHE) hemispheric epilepsies. LD/RD, left/right hemisphere language dominance: BIL, bilateral language representation; ILD/IRD, incomplete left/right hemisphere dominance.

3, p .0039). Except for one patient with RHE, RD was observed exclusively in LHE (n 1 vs. n 12). In contrast, BIL was more frequently seen in RHE (n 9 vs. n 4). None of the patients was found to be IRD, while ILD occurred with similar frequency in both groups (n 11 vs. n 9). The performance on language subfunctions during the right and the left IAP in all cases of incomplete dominance, dened as those having indices between 0.8 and 0.8, reected the general predominance of the ILD pattern (Fig. 3). Leaving aside the lateralization of epilepsy, the data also reveal

FIG. 3. Right and left hemisphere performance on language subfunctions of all incomplete dominant patients having lateralization indices between 0.8 and 0.8. RH, right hemisphere performance after left-sided injection; LH, left hemisphere performance after right-sided injection. The score per subtest could range between 0 (no response) and 2 (correct response) in the left or the right IAP. The total points per subtest in the left and the right IAP reect the sum of the scores achieved by all incomplete dominant patients (N 33).

LANGUAGE DOMINANCE

141

a different distribution of language subfunctions within the left and right hemisphere. In contrast to the left hemisphere, which performed equally well on the language subfunctions, the right hemisphere was more able to perform simple input (comprehension of body commands) and output functions (spontaneous speech) than complex functions that require combinations of speech reception and expression (naming, repetition, reading). Based upon the results presented so far, we can conclude that there is no continuum between left and right hemisphere language dominance. The prominent atypical dominance pattern in RHE is BIL or ILD, while RD is mostly observed in LHE. Finally, in contrast to the left hemisphere, which performed equally well on the language subfunctions, the right hemisphere is more able to perform simple than complex expressive and receptive functions. Language Dominance and Sex The distribution of gender was different within the LHE- and RHE-group (LHE 49 men, 32 women; RHE 36 men, 50 women; 2 /Yates corrected 5.0, p .02). Not considering the lateralization of epilepsy, 22 of 85 men and 24 of 82 women were atypically language dominant ( 2/Yates corrected 0.10, p .75). Taking into account the lateralization of epilepsy, within the LHE-group, the incidence of atypical language dominance patterns in women (18 LD, 14 not LD) was higher than in men (38 LD, 11 not LD, 2/Yates corrected 1.26, p .07). In the group with RHE, the sex differences did not reach signicance (women: 40 LD vs. 10 not LD; men: 25 LD vs. 11 not LD; 2/ Yates corrected 0.75, p .38). Hence we nd that, in LHE only, women are more likely to show a right hemisphere participation in language than men. Language Dominance and Handedness (Table 1) Handedness was assessed by the Oldeld Edinburgh Inventory (Oldeld, 1971). We differentiated between three groups, i.e., right-handed, lefthanded, and ambidextrous patients. Cut off scores were 0.6 and 0.6 respectively. Accordingly, 82% of the total group of patients were righthanded, 14% left-handed, and 4% ambidextrous. 18% of the right-handed, 75% of the left-handed and 50% of the ambidextrous patients displayed atypical language dominance patterns ( 2 34.53, df 2, p 0.0001). When considered separately, in LHE, left- and ambidextrous handedness were associated with atypical language dominance in 88% of cases. Alternatively, 64% of the patients with atypical language dominance and 4% of the completely left dominant patients were left-handed or ambidextrous ( 2 36.65, df 2, p .0001). In RHE, however, there was neither an increased occurrence of left-handedness in the presence of atypical language dominance, nor did

142

HELMSTAEDTER ET AL.

TABLE 1 Handedness Right All patients Total Atypical dominance Left dominance 2 34.53, df 2, p Left hemisphere epilepsy Total Atypical dominance Left dominance 2 36.65, df 2, p 137/82% 25 112 .0001 63/78% 9 54 .0001 74/86% 16 58 8/9% 4 4 4/5% 1 3 16/20% 14 2 2/2% 2 0 Left 24/14% 18 6 Ambidextrous 6/4% 3 3

Right hemisphere epilepsy Total Atypical dominance Left dominance 2 3.15, df 2, p .20

left-handedness indicate atypical language dominance (2 3.15, df 2, p .20). Thus we can conclude that only in the presence of LHE is manual preference a good indicator of atypical language dominance. Language Dominance and Age at Onset of Epilepsy The lateralization index was signicantly correlated with the age at onset of epilepsy (time of the rst seizure) in the LHE- (r 0.3 p .001) but not in the RHE-group (r .05 p .6; Table 2). Frequency tabulation gave further information on the relation between presence or degree of atypical language dominance and the age at onset of epilepsy. Six versus 13 years are alternatively discussed as the critical age until which major cerebral plasticity for language is supposed (see Lenneberg, 1978). Within the LHEgroup, 54% of the patients with atypical dominance patterns had an onset of epilepsy before the age of 6 ( 2 4.4, df 1, p .03) and 92% before the age of 14 ( 2 8.5, df 1, p .0035). In contrast, only 19% of the patients with RHE and atypical language dominance had an onset of epilepsy before the age of 6, and only 48% before the age of 14. It should be noted that, leaving language dominance aside, there was no difference between the LHE and RHE-groups in the absolute number of patients with very early, early, and lateonset of epilepsy ( 2 1.38, df 2, p .502). There was no sex difference with respect to the age at onset of epilepsy and atypical language dominance, as had been reported by Strauss, Wada,

LANGUAGE DOMINANCE

143

TABLE 2 Age at epilepsy onset 05 years All patients (correlation r 0.17, p .01) Total* 47/28% Atypical dominance 17 Left dominance 30 2 3.29, df 2, p .19 Left hemisphere epilepsy (correlation r 0.3, p .001)* Total 23/29% Atypical dominance 13 Left dominance 10 2 14.27, df 2, p .0008 Right hemisphere epilepsy (correlation r 0.05, p .60) Total 24/28% Atypical dominance 4 Left dominance 20 2 1.54, df 2, p .46 * One missing value for the left hemisphere epilepsy group. 613 years 56/34% 15 41 13 years 63/38% 13 50

30/37% 9 21

27/34% 2 25

26/30% 6 20

36/42% 11 25

and Goldwater (1992). In the LHE group with atypical language dominance, the mean age at onset of epilepsy was 5 years (SD 3.1) in men and 7.4 (SD 7.4) years in women (t 1.12, p .27). In the RHE-group the mean onset of epilepsy was 14.9 years in men (SD 6.5) and 16.4 (SD 15.1) years in women (t 0.29, p .76). In conclusion, there is a strong relation between right hemisphere language and the age at onset of epilepsy only in LHE, indicating a decreasing plasticity until puberty. In RHE there is an opposite trend of only a few patients with atypical dominance and an early onset of epilepsy and more patients with relatively late onsets. No different periods of cerebral plasticity can be discerned for men versus women. Language Dominance and the Localization of Foci and Lesions According to EEG/ECoG monitoring, 22% of the patients had extratemporal foci or widespread epileptic activity predominantly over one hemisphere (Table 3a). In this group 47% of patients were atypically language dominant, in contrast to only 22% of patients in the group with circumscribed temporal foci ( 2 7.69, df 1, p .005). In the LHE group with atypical dominance, RD was more likely in the presence of extratemporal and extensive foci (8/12, 67%), whereas ILD occurred more often in patients with temporal lobe foci (8/9, 89%) (2 6.93, df 1, p .008). In the RHE group, BIL was mostly associated with temporal foci (8/9, 89%). With regard to radiological localization of lesions, 40% of all patients had

144

HELMSTAEDTER ET AL.

TABLE 3 Localization of Epileptic Foci and Lesions (a) All patients Total Atypical dominance Left dominance 2 7.69, df 1, p .005 Left hemisphere epilepsy Total Atypical dominance Left dominance 2 6.93, df 1, p .008 Subpatterns of atypical dominance RD BIL ILD Right hemisphere epilepsy Total Atypical dominance Left dominance 2 0.56, df 1, p .45 Subpatterns of atypical dominance RD BIL ILD (b) All patients Total Atypical dominance Left dominance 2 11.12, df 2, p .0038 Left hemisphere epilepsy Total Atypical dominance Left dominance 2, p .14 2 3,92, df Subpatterns of atypical dominance RD BIL ILD Right hemisphere epilepsy Total Atypical dominance Left dominance 2, p .018 2 7.99, df Subpatterns of atypical dominance RLD BIL ILD Temporal focus 131/78% 29 102 Extratemporal or extended focus 36/22% 17 19

62/76% 14 48

19/24% 11 8

4 2 8 69/80% 15 54

8 2 1 17/20% 6 11

0 8 7 No lesion 66/40% 20 46 Temporal lesion 66/40% 10 56

1 1 4 Extratemporal/ extended lesion 35/20% 16 19

28/34% 10 18

32/40% 6 26

21/26% 9 12

2 2 6 38/44% 10 28

2 1 3 34/40% 4 30

8 1 0 14/16% 7 7

1 7 2

0 0 4

0 2 5

LANGUAGE DOMINANCE

145

lesions within the temporal region, 20% had extratemporal and/or extensive lesions. In 40% of all patients, no structural lesions could be detected by MRI and/or CCT (Table 3b). Corresponding to ndings on EEG, extratemporal and extensive lesions were more likely to be accompanied by atypical language dominance (16/35, 46%) than in cases without structural lesions 11.12, df (20/66, 30%) or having temporal lesions (10/66, 15%) ( 2 2, p .0038). In both the LHE and RHE groups, atypical dominance was less likely observed in patients with temporal lesions than in patients without a lesion or in those with extratemporal/extensive lesions (LHE: 2 3.92, df 2, p .14; RHE: 2 7.99, df 2, p .018). Concerning the different subpatterns of language dominance in LHE, RD was most frequent in the presence of extratemporal and extensive lesions (8/12, 67%), while ILD was most frequent in nonlesional patients (6/9, 67%). In the RHE group, BIL was predominant in patients without lesions (7/9, 78%), while ILD was more likely in patients with either temporal or extratemporal lesions (9/11/, 82%). Thus, we nd that in LHE the degree of right hemisphere participation in language is signicantly related to the extent and localization of lesions and epileptic foci. In RHE, BIL appears to be associated with temporal foci and the absence of lesions, while ILD is associated with lesions. Cognitive Consequences of Atypical Language Dominance The group differences on cognitive performance were determined by nonparametric Wilcoxon rank tests for pairs. The comparison of patients with left sided dominance and atypical dominance within the LHE group showed that atypical dominance was accompanied by signicantly poorer performance in tests of visuomotor speed (p .007), language (p .02), verbal reasoning (p .01) and visuoconstructional functions (p .001). No differences were observed with respect to vocabulary (as an estimation of education and intelligence) (p .85) and verbal memory performance (p .57). Spearman rank correlations between IAP indices and cognitive performance in the LHE-group showed a signicant correlation between the degree of right hemisphere language participation and poorer performance on attention/visuo-motor speed (r 0.29, p .13), visual/spatial memory (r 0.34, p .003) and visuo-constructional functions (r 0.42, p .001). Performance on language and verbal memory did not correlate with the degree of right hemisphere language participation. The correlation between verbal and nonverbal functions in LHE patients with atypical language dominance was positive (verbal and nonverbal memory, r 0.19, p .39; language and visuo-construction, r 0.61, p .01). In the RHE group, patients with left dominance and patients with atypical dominance did not differ with respect to their cognitive performance, nor did the degree of right hemisphere language participation correlate with any performance. Thus, despite language transfer, patients with LHE and atypical language

146

HELMSTAEDTER ET AL.

dominance displayed a poorer performance on language and nonlanguage functions than patients with LHE who had preserved left hemisphere language dominance. Correlations between verbal and nonverbal functions and IAP indices in atypical dominant LHE patients indicate a suppression effect in terms of a language shift at the cost of nonlanguage functions. However, there was no evidence for a better performance on language at the cost of nonlanguage functions. Instead performance on language and nonlanguage functions were positively correlated.
DISCUSSION

Conditions of Atypical Language Dominance in Patients with Focal Epilepsies The results of the present study largely conrm ndings for atypical language dominance in patients with focal epilepsies reported elsewhere (Rasmussen & Milner, 1975, 1977; Ogden, 1988; Satz, Strauss, Wada, & Orsini, 1988; Dennis, 1980). Leaving aside the lateralization of lesions or epileptic foci, our sample of unselected patients strongly conrms the high incidence of atypical language dominance patterns in focal epilepsies (about 30%). Lesions do not appear to be a necessary precondition for atypical language dominance, thus indicating the signicant inuence of the epileptic dysfunction in the process of language transfer. Equipotentiality versus Left Hemisphere Superiority for Language Dominance The observed patterns of atypical language dominance do not support the assumption of an equipotentiality of both hemispheres (Smith, 1966, 1978). A rst argument against equipotentiality is that there is no continuum from left to right hemisphere language dominance, which has been supposed by Loring et al. (1990). Rather, the distribution of lateralization indices gives evidence of a continuum between left dominance and bilaterality on the one hand, and an all or nothing principle of complete right hemispheric dominance on the other hand. A second argument against equipotentiality can be drawn from estimations of natural atypical dominance on the basis of atypical language dominance beyond the period of plasticity ( 13 years). They indicate complete right hemisphere language dominance to be very rare (about 1%), followed by bilateral patterns (about 5%) and incomplete left dominance (about 10%). Third, we found that in contrast to the left hemisphere, the right hemisphere performed better on simple receptive and expressive language functions than on complex functions that require the interaction of receptive and expressive functions (naming, reading, repetition). Finally,

LANGUAGE DOMINANCE

147

despite language transfer, the right hemisphere could not fully preserve or restitute those functions that were impaired by left hemisphere disorders. Atypical Language Dominance in Left and Right Hemisphere Epilepsies If we take the lateralization of lesions and epileptic foci into consideration, the proportion of atypical dominance patterns in right hemisphere epilepsies was unexpectedly high (24%). However, we obviously have to differentiate between atypical language dominance in the presence of left or right hemisphere injury. Differences between these groups were already indicated by the different prevalence of the subpatterns of atypical language dominance. While RD was typically seen in the group with left hemisphere epilepsy, right hemisphere epilepsy was associated with BIL. Left hemisphere epilepsies. Although a few patients may have had a natural right hemisphere language participation before the onset of epilepsy, the data among LHE clearly indicates that atypical language dominance in this group is largely a consequence of an early left hemisphere disorder. There is a highly signicant negative correlation between IAP index and the age at onset of epilepsy, indicating a decrease of cerebral plasticity with increasing age. Our data strongly support the thesis of Lenneberg (1967) that cerebral plasticity for language lasts until puberty. Strauss et al. (1992) described a sex difference with respect to the period during which reorganization of language is likely to occur; i.e., a right hemispheric reorganization of language in women is most likely to occur very early, while the period during which language reorganization in men is observed may extend beyond puberty. Apart from our observation that women with LHE are more likely to display atypical language dominance than men with LHE, there is no signicant interaction between sex, age at onset of epilepsy and language dominance. However, a closer examination of the data of Strauss et al. (1992) indicates that the results may differ because of the denition of the onset of dysfunction (lesion versus epilepsy) and the different interpretation of a right hemisphere language participation as language shift in patients with right hemisphere epilepsy. As already mentioned above, the cognitive prole of atypically dominant patients with LHE indicates that the right hemispheric reorganization of originally left hemispheric language functions is largely suboptimal. This is supported by the nding that a higher degree of right hemisphere participation in language is not related to better language functioning. Nonlanguage functions are increasingly impaired with an increasing participation of the right hemisphere in language. This latter phenomenon has been interpreted in terms of suppression effects and interference effects, because of limited capacities and/or principal incompatibilities between language and nonlanguage information processing (Ogden, 1989; Strauss, Satz, & Wada, 1990).

148

HELMSTAEDTER ET AL.

However, the intercorrelations of IAP indices and performance in language and nonlanguage tests indicate that the degree of language transfer is associated with suppression, but not the quality, of language restitution. The correlation between the performance in language and nonlanguage tests was positive, a nding which contradicts the supposition of a negative correlation, expounded in the crowding hypothesis. Following the argument of one of our previous studies, we may conclude that the language shift causes a loss of the ability to shift between different modes of information processing; i.e., the right hemisphere mode of information processing becomes replaced by a left hemisphere mode of information processing that applies to both language and nonlanguage tasks (Helmstaedter et al., 1994). This would explain why the group of patients with atypical language dominance and LHE is most severely impaired in those visual-visuoconstructional tasks that cannot be solved solely by use of verbalization or an analytic, focused, and sequential style of information processing. Right hemisphere epilepsies. While the observations of the occurrence of atypical language dominance in left hemisphere epilepsies are largely conclusive, the ndings in right hemisphere epilepsies indicate a quite different situation. A bilateral distribution of language or incomplete left hemispheric dominance were the predominant patterns of atypical language dominance. In contrast to LHE, atypical dominance was mostly accompanied by right handedness, and most patients with atypical language dominance had a comparably late onset of epilepsy. Furthermore, there was no difference in cognitive performance dependent on the presence or degree of right hemisphere language participation in patients with RHE. Three supplementary explanations might hold for these results. First, these patterns may reect the natural occurrence of a right hemisphere language representation. This assumption is supported by the high number of atypical dominant patients with a late onset of epilepsy and the relative independence of the occurrence of atypical dominance patterns from the localization of lesions and foci within the right hemisphere. Second, some of these patterns may reect the result of a language transfer from the right to the left hemisphere as a consequence of a right hemispheric injury. A complete shift of language from the right to the left hemisphere should be possible, especially in patients with an early onset of epilepsy. This might explain the relatively low percentage of atypical dominant RHE patients with an early onset of epilepsy. The fact that partial shifts are possible is indicated by two patients of the bilateral group who showed hemispherically dissociated expressive and receptive language functions in the presence of a frontal and parietal right hemispheric lesion, respectively. These patients have been reported in detail elsewhere (Kurthen et al., 1992). However, neither explanation is completely convincing, because we would expect the same occurrence of natural atypical dominance in the LHE group. Although there were more patients with atypi-

LANGUAGE DOMINANCE

149

cal language dominance in LHE, we miss atypically dominant patients with a late onset of epilepsy in this group. Thus, as a third but very speculative explanation, a higher vulnerability of the right hemisphere to epilepsy in patients with natural atypical dominance might be concluded.
CONCLUSION

In conclusion, the data indicate a predetermined left hemisphere superiority in the treatment of language. The left hemisphere seems to maintain language as far as possible, and right hemispheric language participation in LHE appears to be a makeshift rather than an equipotential solution. Extended dysfunctions and not necessarily structural impairment within the time of cerebral plasticity appear to be the prerequisite for these language shifts. There is evidence that language is transferred at the cost of originally right hemisphere functions, but language transfer and suppression effects do not guarantee a better performance on language. The high incidence of atypical dominance patterns in patients with RHE and the late onset of epilepsy in this group provides evidence of a higher than expected natural incidence of bilateral and incomplete left hemispheric language dominance. Finally, the principal possibility of a language shift from the right to the left hemisphere can be suggested in RHE.
REFERENCES
Amthauer, R. 1973. Intelligenz-struktur-test IST 70. Gottingen/Toronto/Zurich: Hogreve. Brickenkamp, R. 1978. Test d2-Aufmerksamkeits-belastungs-test. Gottingen/Toronto/Zurich: Hogreve. Bryden, M. P., Hecaen, M., & De Agostino, M. 1983. Patterns of cerebral organization. Brain and Language, 20, 249262. Dennis, M. 1980. Capacity and strategy for syntactic comprehension after left and right hemidecortication. Brain & Language, 10, 287317. De Renzi, E., & Vignolo, L. A. 1962. The token test. A sensitive test to detect receptive disturbances in aphasics. Brain, 85, 665678. Helmstaedter, C., & Durwen, H. F. 1990. VLMT-verbaler Lern und Merk-fahigkeitstest. Ein praktikables und differenziertes Instrumentarium zur Prufung der verbalen Gedacht nisleistungen. Schweizer Archiv fu r Neurologie und Psychiatrie, 141, 2130. Helmstaedter, C., Pohl, C., Hufnagel, A., & Elger, C. E. 1991. Visual learning decits in nonresected patients with right temporal lobe epilepsy. Cortex, 27, 547555. Helmstaedter, C., Kurthen, M., Linke, D. B., & Elger, C. E. 1994. Right hemisphere restitution of language functions in right hemisphere language-dominant patients with left temporal lobe epilepsy. Brain, 117, 729737. Horn, W. 1983. Leistungspru fsystem L-P-S. Gottingen/Toronto/Zurich: Hogrefe. Kurthen, M., Helmstaedter, C., Linke, D. B., Solymosi, L., Elger, C. E., & Schramm, J. 1992. Interhemispheric dissociation of expressive and receptive language functions in patients with complex-partial seizures: An amobarbital study. Brain and Language, 43, 694712. Kurthen, M., Helmstaedter, C., Linke, D. B., Hufnagel, A., Elger, C. E., & Schramm, J. 1994. Quantitative and qualitative evaluation of patterns of cerebral language dominance. Brain and Language, 46, 536564.

150

HELMSTAEDTER ET AL.

Lehrl, S. 1978. Mehrfachwahl-Wortschatz-Intelligenztest MWT-B. Erlangen: Verlag Dr.med.Straube. Lenneberg, E. 1967. Biological foundations of language. New York: Wiley. Loring, D. W., Meador, K. J., Lee, G. P., Murro, A. M., Smith, J. R., & Flanigin, H. F. 1990. Cerebral language lateralization: evidence from intracarotid amobarbital testing. Neuropsychologia, 28, 831838. Mateer, C. A., & Dodrill, C. B. 1983. Neuropsychological and linguistic correlates of atypical language lateralization: Evidence from sodium amytal studies. Human Neurobiology, 2, 135157. Ogden, J. A. 1988. Language and memory functions after long recovery periods in left hemispherectomized subjects. Neuropsychologia, 26/5, 645659. Ogden, J. A. 1989. Visuospatial and other right-hemispheric functions after long recovery periods in left-hemispherectomized subjects. Neuropsychologia, 27, 765776. Oldeld, R. C. 1971. The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia, 9, 97113. Rasmussen, T., & Milner, B. 1975. Clinical and surgical studies of the cerebral speech areas in man. In K. J. Zulch, O. Creutzfeldt, & G. C. Galbraith (Eds.), Cerebral localization. Berlin/Heidelberg/New York: Springer Verlag. Rasmussen, T., & Milner, B. 1977. The role of early left-brain injury in determining lateralization of cerebral speech functions. N.Y. Acad. Sci. 29, 355369. Rey, M., Dellatolas, G., Bancaud, C. & Talairach, J. 1988. Hemispheric lateralization of motor and speech functions after early brain lesions: study of 73 epileptic patients with intracarotid amytal test. Neuropsychologia, 26, 167172. Satz, P., Strauss, E., Wada, J., & Orsini, D.L. 1988. Some correlations of intra- and interhemispheric speech organization after focal brain injury. Neuropsychologia, 26/2, 345350. Smith, A. 1966. Speech and other functions after left (dominant) hemispherectomy. Journal of Neurology, Neurosurgery and Psychiatry, 29, 467471. Smith, A. 1978. Dominant and nondominant hemispherectomy. In W. L. Smith (Ed.), Drugs and cerebral function. Springeld: Thomas. Pp. 3768. Strauss, E., Satz, P., & Wada, J. 1990. An examination of the crowding hypothesis in epileptic patients who have undergone the carotid amytal test. Neuropsychologia, 28, 12211227. Strauss, E., Wada, J., & Goldwater, B. 1992. Sex differences in interhemispheric reorganization of speech. Neuropsychologia, 30/4, 353359. Teuber, H. L. 1974. Why two brains? In F.O. Schmitt & Worden, F.G. The neurosciences: Third study program. Cambridge: MIT Press. Pp. 7174. Wada, J. 1949. A new method for the determination of the side of cerebral speech dominance. A preliminary report on the intracarotid injection of sodium amytal in man. Igaku Seibutssugaku, 14, 221222. Wechsler, D. 1955. Wechsler adult intelligence scale Manual. New York: Psychological Corp. Zatorre, R. J. 1989. Perceptual asymmetry on the dichotic fused words test and cerebral speech lateralization determined by the carotid sodium amytal test. Neuropsychologia, 27, 1207 1219.