Brain (1992). 115, 87-107
NEUROPSYCHOLOGICAL AND NEUROANATOMICAL DIMENSIONS OF IDEOMOTOR APRAXIA by M. P. ALEXANDER, 1 E. BAKER,2 M. A. NAESER, 2 E. KAPLAN2 and C. PALUMBO2 (From the Aphasia Research Center, Boston University School of Medicine, Department of Neurology, 'Aphasia Programs, Braintree Hospital, 2Psychology Research Center, Boston, USA)
Fifty-five right-handed men with left hemisphere stroke were systematically investigated for ideomotor apraxia of various body parts. Standardized aphasia and apraxia examinations in all cases, and appropriately timed CT in 28 cases, were used for analysis of psychological and anatomical properties of ideomotor apraxia. It is not possible to ignore the modality of eliciting the movement (command or imitation) and the body part being moved (buccofacial, limb or whole body) from any comprehensive theory about ideomotor apraxia. Different levels of performance are seen with different body parts to different modalities in different aphasic groups. Some subtypes of ideomotor apraxia (buccofacial) have highly specified anatomical basis. Some (limb) have apparently distributed anatomies, and others (whole body) have as yet unknown bases.
INTRODUCTION
The first systematic analysis of apraxia was performed by Liepmann (1900, 1906, 1908) and Liepmann and Maas (1907). The terminology and conceptual framework for the apraxic disorders arose from those studies, and both the terminology and the theoretical concepts that it captures remain in use today. The disorder that Liepmann called ideomotor apraxia has been the primary focus of subsequent research. All investigations, including Liepmann's, of ideomotor apraxia have observed a close relationship between apraxia and aphasia. Much analysis of apraxia even today seems motivated by hypotheses about the nature of this relationship. Despite pervasive adherence to Liepmann's schemata (Goodglass and Kaplan, 1963; Geschwind, 1965, 1975; De Renzi et al, 1966, 1980; Heilman et al., 1982; Kertesz and Hooper, 1982; Lehmkuhl et al., 1983) and general, but not complete agreement (Kimura and Archibald, 1974; Kimura, 1977, 1982; Mateer and Kimura, 1977; Watson and Kimura, 1989) that interpretation of at least some of the clinical manifestations of ideomotor apraxia is contaminated by a co-occurring aphasia, a century of research and description has left a number of unresolved questions about ideomotor apraxia. It is not known with certainty if the nature or severity of ideomotor apraxia depend upon the stimulus used to elicit it. This question has often been simply subsumed in a methodological compromise, using imitation to circumvent poor comprehension. The precise relationships between several speech and language deficits and the phenomenology Correspondence to: Dr M. P. Alexander, Braintree Hospital, 250 Pond Street, Braintree, MA 02184, USA. © Oxford University Press 1992
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
SUMMARY
88
M. P. ALEXANDER AND OTHERS
SUBJECTS AND METHODS Subjects Consecutive patients hospitalized for left hemisphere ischaemic stroke were evaluated. All right-handed patients with no prior clinical history of stroke were included, except for very impaired patients who could not be instructed or even modelled into the appropriate set for praxis testing. All patients were male; age and education information are summarized in Table 1. Controls Two control groups were utilized. A group of 23 normal right-handed men matched to the main age cohort of the patient group (45-75) was tested. A group of unilateral right hemisphere ischaemic stroke cases was also tested. This group was also all right-handed men with no history of stroke prior to the index event, distributed across the main age cohort of the left hemisphere-damaged group. Age and educational data for both control groups is summarized in Table 1. For both groups of stroke patients all evaluations were performed between the third and the twelfth week after onset of stroke. Patients were tested as close to the third week post-onset as possible. Aphasia assessment Language was evaluated with the Boston Diagnostic Aphasia Examination (BDAE) (Goodglass and Kaplan, 1983) by experienced speech pathologists unaware of any aspect of the apraxia project. Aphasia diagnoses met the criteria outlined by Goodglass and Kaplan (1983). Four groupings of BDAE subtests were used
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
of apraxia are not established although numerous aspects of this relationship have been investigated. The range of motor systems (body parts) that can be involved in ideomotor apraxia has often not been specified. If ideomotor apraxia can involve buccofacial or axial movements as well as limb, then other potential issues arise, such as, the effects of different lesion sites on ideomotor apraxia of different body parts or preferential sparing of different body parts. There is not a universally accepted, empirically supported unique definition of ideomotor apraxia. All definitions in current use incorporate one or more claims about the nature of ideomotor apraxia that are not completely accepted. There is, however, a thread of general agreement [except Kimura et al. (Kimura and Archibald, 1974; Kimura, 1977, 1982; Mateer and Kimura, 1977; Watson and Kimura, 1989)] around which the definition of ideomotor apraxia for this report was constructed. Ideomotor apraxia is a disorder of 'learned movements' (Geschwind, 1975) or 'skilled movements' without meaning explicitly symbolic movements (Heilman etai, 1982). Ideomotor apraxia cannot be accounted for by any elemental neurological or motivational deficit. It may be most reliably demonstrated either by command (Geschwind, 1975) or by imitation (De Renzi et al., 1980), the former more closely matching the naturalistic setting in which movement is elicited, but the later neutralizing any limitations in language comprehension. We report the results of an investigation of ideomotor apraxia in 55 patients with left hemisphere damage. The investigation was designed to explore three separate, but often entangled, issues: (1) the effect of modality of request, command or imitation, on praxis performance; (2) the relationship between ideomotor apraxia and aphasia, either overall aphasia syndrome or individual components of disordered language; (3) the lesion correlations of apraxic performances of each testable motor subsystem (face, limb, axial, etc.).
IDEOMOTOR APRAXIA
89
TABLE I. CASES AND CONTROLS: DEMOGRAPHIC AND CLINICAL CHARACTERISTICS
Aphasia Transcortical Motor Broca's Global
n 10 13 8
Conduction
6
Wernicke's
9
Anomic
8 18 23
57.9 (49-62) 60.77 (52-73) 52.75 (31-72) 60.17 (48-72) 6.61 (57-78) 59.13 (48-70) 56.28 (24-77) 60.00 (47-79)
Education (yrs) 11.8 (10-16) 11.25 (7-16) 12.14 (8-16) 11.67 (11-12) 10.63 (8-12) 9.25 (7-12) 10.94 (8-14) 12.48 (7-16)
TPO (wks) 5.4
HP 2
HH 0
COMP +0.50
(2-11) 5.46 (3-12)
4
0
+0.18
6
6
-1.40
1
2
+0.75
1
8
-1.03
2
4
+0.56
12
7
—
—
—
—
6.5
(3-10) 3.83 (3-6) 6.67 (3-10) 5.13 (2-10) 5.44 (3-13) —
TPO = time post-onset to initial testing; HP = number of cases with limiting hemiparesis; HH = number of cases with significant visual field deficit; COMP = overall auditory comprehension on BDAE as mean of group z-scorcs.
for statistical studies: articulatory agility; repetition of sentences; overall auditory comprehension; overall severity. Apraxia assessment Apraxia testing was carried out by one of two examiners. The test items and their division by body parts are indicated in the Appendix. Between representational and non-representational tasks, there was as much correspondence of motor complexity as could be devised. For example, touch your nose (nonrepresentational), salute (simple intransitive posture), wipe sweat off of your forehead (simple transitive movement without an object), and pretend to tip your hat (complex transitive movement requiring pretended object use) all demanded bringing the hand to the upper face. This correspondence was maintained as much as possible for all body parts. After presentation of the complete battery to command, there was a rest break, and the entire battery was presented to imitation. The order of modality (imitation or command) was held constant because of concern that imitation would provide a model which might facilitate even a slightly delayed presentation of the task to command. The order of presentation of body parts was counterbalanced across subjects. In the command mode, the patient was instructed to carry out the movement that he heard requested. In the imitation mode, the patient was told to make exactly the same movement that the examiner made. Before each body part grouping, the patient was told which body part was about to be tested and lightly touched in the appropriate area. If no movement was made by the patient, the request (or the gesture) was repeated once. No repetitions were allowed for an incorrect response, only for no response. If after repetition the patient still made no response, the test moved to the next item. Scoring of apraxia The scoring scale was based on qualitative differences in performance observed in the developmental acquisition of praxis in childhood (Kaplan, 1968). Performance was scored from no response to completely correct response (see Table 2). For each body part to each modality of request there were 8 to 12 trial items. The scores given are the means across the trials. Whenever the neurologic condition permitted, both the right and the left arm were tested to both modalities. Inter-rater reliability was assessed. For 10 patients, the two raters simultaneously but independently scored the same test sessions. For overall agreement, for agreement on each body part and for agreement for each modality, the reliability of rating was 91%.
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
Right brain Damaged Normals
Age (yrs)
90
M. P. ALEXANDER AND OTHERS TABLE 2. SCORING OF PRAXIS Score
Achievement
5 4
Completely correct. Correct but only after a self cue or after one or more incorrect movements or with verbalization which directs the movement. For buccofacial and respiratory: onomatopoetic or substituted verbalization. For limb: body part as object or spatial error such as incorrect spacing, rotation or precise limb placement. For axial: appropriate limb movements without axial movement. Repeated undifferentiated movements of correct body part. Perseveration. No movement.
3
2 1 0
RESULTS
Normals Many normals made occasional, minor errors on praxis tests to command. Means and standard deviations for the 23 normals to command are shown in Table 3. A cutoff for normal performance was taken as the mean for each body part minus two standard deviations. Normals perform better on these tasks of gesture when given a model to imitate. To imitation, there were very few errors. No normal made more than one error, so we considered any performance with more than one error to be apraxic. Right hemisphere stroke controls Mean performance to command and to imitation did not differ from the normals. There was some difficulty in this group with one task from the buccofacial tasks; the single command to frown as if angry was incorrectly performed in 13 of the 18 patients. Left hemisphere stroke cases Aphasia group X body part X modality X representationality. Two separate mixed design analyses of variance (ANOVA) were used to assess whether apraxia scores were related to the factors of aphasia type (6 levels), body part (4 levels), modality of presentation (2 levels) or representationality of movement (2 levels, with all of the representational movements for limbs collapsed into one group). The results of the ANOVA are
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
CT studies Non-enhanced CT scans were performed from 2 to 12 wks after onset. Scans were evaluated by two experienced CT researchers who were blind to the apraxia study methodology or hypotheses. The raters used a neuroanatomical checklist (Fig. 1) and rating methodology based on prior research at our unit (Naeser et al., 1990). This method captures partial damage to a rated region. Scan inclinations were determined before lesion ratings were made. A CT atlas (Matsui and Hirano, 1978) was utilized to clarify cortical locations for CTs in anomalous inclinations. For anatomically coherent areas present on more than one CT slice (lower motor cortex, head of caudate nucleus, etc.), the scores for the individual CT slices were summed. For statistical analyses, all scores were rectified to percentage of area with lesion. Only cases with unequivocal single lesions demonstrated in the post-acute interval after 3 wks post-onset were included for anatomical analyses. Not all patients in the praxis studies above had an adequate CT, and others had small (and clinically silent) lesions in addition to the one producing the acute presentation to the hospital. In both events, the cases were included in the praxis study but not the CT lesion analysis.
IDEOMOTOR APRAXIA Slice B
Slice B/W
Slice SM
Slice SM+1
Slice SM+2
FIG. 1. Schematic representation of brain regions rated for lesion site studies. Numbers refer to regions listed in Tables 5A,B.
TABLE 3. APRAXIA SCORES FOR 23 NORMALS. THRESHOLDS FOR ABNORMAL PERFORMANCE TO COMMAND Mean (max. score 5) Standard deviation Lower limit (mean: - 2 x S D )
Buccofacial
Respiratory
4.88 0.28 4.32
4.62 0.42 3.78
Limb 4.%
0.12 4.73
Axial
4.98 0.06 4.86
summarized in Table 4. All pair-wise comparisons of means were carried out using Newman-Keuls post hoc procedure. Significant effects were found for aphasia group, modality of presentation, body part tested and representationality. The conduction aphasic and the anomic aphasic groups were not significantly apraxic when compared with the normal controls. The other 4 groups were significantly apraxic. Some of the effects of aphasia group are displayed in Fig. 2. Global aphasics were significantly more apraxic than the other groups on all body parts and to both modalities. Wernicke's aphasics were significantly more apraxic than transcortical motor aphasics and Broca's aphasics to command, but the differences disappeared to imitation. To imitation, in fact, the Wernicke's group was significantly less apraxic than the Broca's aphasics on the oral and respiratory tasks.
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
Site* W
91
M. P. ALEXANDER AND OTHERS
92
TABLE 4. RESULTS OF ANOVAs OF APRAXIA. APHASIC GROUPS WITH APRAX1A SCORE BY BODY PART AS DEPENDENT MEASURE MS
Anova table—modality Group Modality Body part Group x modality Group x body part Modality x body part Group x modal ity x body part
6 1 3 6 18 3 18
46.43 57.92 20.54 3.01 1.42 0.54 0.38
11.63 43.39 29.69 2.53 2.06 2.39 1.64
NEUROPSYCHOLOGICAL AND NEUROANATOMICAL DIMENSIONS OF IDEOMOTOR APRAXIA by M. P. ALEXANDER, 1 E. BAKER,2 M. A. NAESER, 2 E. KAPLAN2 and C. PALUMBO2 (From the Aphasia Research Center, Boston University School of Medicine, Department of Neurology, 'Aphasia Programs, Braintree Hospital, 2Psychology Research Center, Boston, USA)
Fifty-five right-handed men with left hemisphere stroke were systematically investigated for ideomotor apraxia of various body parts. Standardized aphasia and apraxia examinations in all cases, and appropriately timed CT in 28 cases, were used for analysis of psychological and anatomical properties of ideomotor apraxia. It is not possible to ignore the modality of eliciting the movement (command or imitation) and the body part being moved (buccofacial, limb or whole body) from any comprehensive theory about ideomotor apraxia. Different levels of performance are seen with different body parts to different modalities in different aphasic groups. Some subtypes of ideomotor apraxia (buccofacial) have highly specified anatomical basis. Some (limb) have apparently distributed anatomies, and others (whole body) have as yet unknown bases.
INTRODUCTION
The first systematic analysis of apraxia was performed by Liepmann (1900, 1906, 1908) and Liepmann and Maas (1907). The terminology and conceptual framework for the apraxic disorders arose from those studies, and both the terminology and the theoretical concepts that it captures remain in use today. The disorder that Liepmann called ideomotor apraxia has been the primary focus of subsequent research. All investigations, including Liepmann's, of ideomotor apraxia have observed a close relationship between apraxia and aphasia. Much analysis of apraxia even today seems motivated by hypotheses about the nature of this relationship. Despite pervasive adherence to Liepmann's schemata (Goodglass and Kaplan, 1963; Geschwind, 1965, 1975; De Renzi et al, 1966, 1980; Heilman et al., 1982; Kertesz and Hooper, 1982; Lehmkuhl et al., 1983) and general, but not complete agreement (Kimura and Archibald, 1974; Kimura, 1977, 1982; Mateer and Kimura, 1977; Watson and Kimura, 1989) that interpretation of at least some of the clinical manifestations of ideomotor apraxia is contaminated by a co-occurring aphasia, a century of research and description has left a number of unresolved questions about ideomotor apraxia. It is not known with certainty if the nature or severity of ideomotor apraxia depend upon the stimulus used to elicit it. This question has often been simply subsumed in a methodological compromise, using imitation to circumvent poor comprehension. The precise relationships between several speech and language deficits and the phenomenology Correspondence to: Dr M. P. Alexander, Braintree Hospital, 250 Pond Street, Braintree, MA 02184, USA. © Oxford University Press 1992
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
SUMMARY
88
M. P. ALEXANDER AND OTHERS
SUBJECTS AND METHODS Subjects Consecutive patients hospitalized for left hemisphere ischaemic stroke were evaluated. All right-handed patients with no prior clinical history of stroke were included, except for very impaired patients who could not be instructed or even modelled into the appropriate set for praxis testing. All patients were male; age and education information are summarized in Table 1. Controls Two control groups were utilized. A group of 23 normal right-handed men matched to the main age cohort of the patient group (45-75) was tested. A group of unilateral right hemisphere ischaemic stroke cases was also tested. This group was also all right-handed men with no history of stroke prior to the index event, distributed across the main age cohort of the left hemisphere-damaged group. Age and educational data for both control groups is summarized in Table 1. For both groups of stroke patients all evaluations were performed between the third and the twelfth week after onset of stroke. Patients were tested as close to the third week post-onset as possible. Aphasia assessment Language was evaluated with the Boston Diagnostic Aphasia Examination (BDAE) (Goodglass and Kaplan, 1983) by experienced speech pathologists unaware of any aspect of the apraxia project. Aphasia diagnoses met the criteria outlined by Goodglass and Kaplan (1983). Four groupings of BDAE subtests were used
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
of apraxia are not established although numerous aspects of this relationship have been investigated. The range of motor systems (body parts) that can be involved in ideomotor apraxia has often not been specified. If ideomotor apraxia can involve buccofacial or axial movements as well as limb, then other potential issues arise, such as, the effects of different lesion sites on ideomotor apraxia of different body parts or preferential sparing of different body parts. There is not a universally accepted, empirically supported unique definition of ideomotor apraxia. All definitions in current use incorporate one or more claims about the nature of ideomotor apraxia that are not completely accepted. There is, however, a thread of general agreement [except Kimura et al. (Kimura and Archibald, 1974; Kimura, 1977, 1982; Mateer and Kimura, 1977; Watson and Kimura, 1989)] around which the definition of ideomotor apraxia for this report was constructed. Ideomotor apraxia is a disorder of 'learned movements' (Geschwind, 1975) or 'skilled movements' without meaning explicitly symbolic movements (Heilman etai, 1982). Ideomotor apraxia cannot be accounted for by any elemental neurological or motivational deficit. It may be most reliably demonstrated either by command (Geschwind, 1975) or by imitation (De Renzi et al., 1980), the former more closely matching the naturalistic setting in which movement is elicited, but the later neutralizing any limitations in language comprehension. We report the results of an investigation of ideomotor apraxia in 55 patients with left hemisphere damage. The investigation was designed to explore three separate, but often entangled, issues: (1) the effect of modality of request, command or imitation, on praxis performance; (2) the relationship between ideomotor apraxia and aphasia, either overall aphasia syndrome or individual components of disordered language; (3) the lesion correlations of apraxic performances of each testable motor subsystem (face, limb, axial, etc.).
IDEOMOTOR APRAXIA
89
TABLE I. CASES AND CONTROLS: DEMOGRAPHIC AND CLINICAL CHARACTERISTICS
Aphasia Transcortical Motor Broca's Global
n 10 13 8
Conduction
6
Wernicke's
9
Anomic
8 18 23
57.9 (49-62) 60.77 (52-73) 52.75 (31-72) 60.17 (48-72) 6.61 (57-78) 59.13 (48-70) 56.28 (24-77) 60.00 (47-79)
Education (yrs) 11.8 (10-16) 11.25 (7-16) 12.14 (8-16) 11.67 (11-12) 10.63 (8-12) 9.25 (7-12) 10.94 (8-14) 12.48 (7-16)
TPO (wks) 5.4
HP 2
HH 0
COMP +0.50
(2-11) 5.46 (3-12)
4
0
+0.18
6
6
-1.40
1
2
+0.75
1
8
-1.03
2
4
+0.56
12
7
—
—
—
—
6.5
(3-10) 3.83 (3-6) 6.67 (3-10) 5.13 (2-10) 5.44 (3-13) —
TPO = time post-onset to initial testing; HP = number of cases with limiting hemiparesis; HH = number of cases with significant visual field deficit; COMP = overall auditory comprehension on BDAE as mean of group z-scorcs.
for statistical studies: articulatory agility; repetition of sentences; overall auditory comprehension; overall severity. Apraxia assessment Apraxia testing was carried out by one of two examiners. The test items and their division by body parts are indicated in the Appendix. Between representational and non-representational tasks, there was as much correspondence of motor complexity as could be devised. For example, touch your nose (nonrepresentational), salute (simple intransitive posture), wipe sweat off of your forehead (simple transitive movement without an object), and pretend to tip your hat (complex transitive movement requiring pretended object use) all demanded bringing the hand to the upper face. This correspondence was maintained as much as possible for all body parts. After presentation of the complete battery to command, there was a rest break, and the entire battery was presented to imitation. The order of modality (imitation or command) was held constant because of concern that imitation would provide a model which might facilitate even a slightly delayed presentation of the task to command. The order of presentation of body parts was counterbalanced across subjects. In the command mode, the patient was instructed to carry out the movement that he heard requested. In the imitation mode, the patient was told to make exactly the same movement that the examiner made. Before each body part grouping, the patient was told which body part was about to be tested and lightly touched in the appropriate area. If no movement was made by the patient, the request (or the gesture) was repeated once. No repetitions were allowed for an incorrect response, only for no response. If after repetition the patient still made no response, the test moved to the next item. Scoring of apraxia The scoring scale was based on qualitative differences in performance observed in the developmental acquisition of praxis in childhood (Kaplan, 1968). Performance was scored from no response to completely correct response (see Table 2). For each body part to each modality of request there were 8 to 12 trial items. The scores given are the means across the trials. Whenever the neurologic condition permitted, both the right and the left arm were tested to both modalities. Inter-rater reliability was assessed. For 10 patients, the two raters simultaneously but independently scored the same test sessions. For overall agreement, for agreement on each body part and for agreement for each modality, the reliability of rating was 91%.
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
Right brain Damaged Normals
Age (yrs)
90
M. P. ALEXANDER AND OTHERS TABLE 2. SCORING OF PRAXIS Score
Achievement
5 4
Completely correct. Correct but only after a self cue or after one or more incorrect movements or with verbalization which directs the movement. For buccofacial and respiratory: onomatopoetic or substituted verbalization. For limb: body part as object or spatial error such as incorrect spacing, rotation or precise limb placement. For axial: appropriate limb movements without axial movement. Repeated undifferentiated movements of correct body part. Perseveration. No movement.
3
2 1 0
RESULTS
Normals Many normals made occasional, minor errors on praxis tests to command. Means and standard deviations for the 23 normals to command are shown in Table 3. A cutoff for normal performance was taken as the mean for each body part minus two standard deviations. Normals perform better on these tasks of gesture when given a model to imitate. To imitation, there were very few errors. No normal made more than one error, so we considered any performance with more than one error to be apraxic. Right hemisphere stroke controls Mean performance to command and to imitation did not differ from the normals. There was some difficulty in this group with one task from the buccofacial tasks; the single command to frown as if angry was incorrectly performed in 13 of the 18 patients. Left hemisphere stroke cases Aphasia group X body part X modality X representationality. Two separate mixed design analyses of variance (ANOVA) were used to assess whether apraxia scores were related to the factors of aphasia type (6 levels), body part (4 levels), modality of presentation (2 levels) or representationality of movement (2 levels, with all of the representational movements for limbs collapsed into one group). The results of the ANOVA are
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
CT studies Non-enhanced CT scans were performed from 2 to 12 wks after onset. Scans were evaluated by two experienced CT researchers who were blind to the apraxia study methodology or hypotheses. The raters used a neuroanatomical checklist (Fig. 1) and rating methodology based on prior research at our unit (Naeser et al., 1990). This method captures partial damage to a rated region. Scan inclinations were determined before lesion ratings were made. A CT atlas (Matsui and Hirano, 1978) was utilized to clarify cortical locations for CTs in anomalous inclinations. For anatomically coherent areas present on more than one CT slice (lower motor cortex, head of caudate nucleus, etc.), the scores for the individual CT slices were summed. For statistical analyses, all scores were rectified to percentage of area with lesion. Only cases with unequivocal single lesions demonstrated in the post-acute interval after 3 wks post-onset were included for anatomical analyses. Not all patients in the praxis studies above had an adequate CT, and others had small (and clinically silent) lesions in addition to the one producing the acute presentation to the hospital. In both events, the cases were included in the praxis study but not the CT lesion analysis.
IDEOMOTOR APRAXIA Slice B
Slice B/W
Slice SM
Slice SM+1
Slice SM+2
FIG. 1. Schematic representation of brain regions rated for lesion site studies. Numbers refer to regions listed in Tables 5A,B.
TABLE 3. APRAXIA SCORES FOR 23 NORMALS. THRESHOLDS FOR ABNORMAL PERFORMANCE TO COMMAND Mean (max. score 5) Standard deviation Lower limit (mean: - 2 x S D )
Buccofacial
Respiratory
4.88 0.28 4.32
4.62 0.42 3.78
Limb 4.%
0.12 4.73
Axial
4.98 0.06 4.86
summarized in Table 4. All pair-wise comparisons of means were carried out using Newman-Keuls post hoc procedure. Significant effects were found for aphasia group, modality of presentation, body part tested and representationality. The conduction aphasic and the anomic aphasic groups were not significantly apraxic when compared with the normal controls. The other 4 groups were significantly apraxic. Some of the effects of aphasia group are displayed in Fig. 2. Global aphasics were significantly more apraxic than the other groups on all body parts and to both modalities. Wernicke's aphasics were significantly more apraxic than transcortical motor aphasics and Broca's aphasics to command, but the differences disappeared to imitation. To imitation, in fact, the Wernicke's group was significantly less apraxic than the Broca's aphasics on the oral and respiratory tasks.
Downloaded from http://brain.oxfordjournals.org/ by guest on December 14, 2015
Site* W
91
M. P. ALEXANDER AND OTHERS
92
TABLE 4. RESULTS OF ANOVAs OF APRAXIA. APHASIC GROUPS WITH APRAX1A SCORE BY BODY PART AS DEPENDENT MEASURE MS
Anova table—modality Group Modality Body part Group x modality Group x body part Modality x body part Group x modal ity x body part
6 1 3 6 18 3 18
46.43 57.92 20.54 3.01 1.42 0.54 0.38
11.63 43.39 29.69 2.53 2.06 2.39 1.64
