Alexia without agraphia: a functional assessment of behavior in focal neurologic disease.

There is a substantial literature describing behavioral disorders following damage to the brain (Heilman & Valenstein, 1985; Mesulam, 1985) and the neuroanatomical sites that correlate with these abnormalities (cf. Damasio & Damasio, 1989). Yet the precision with which we can specify the location and etiology of a lesion (via computed tomography, magnetic resonance imaging, cerebral blood flow and metabolism) has not been matched by an equally rigorous examination of the behavior under consideration. The purpose of this paper is to demonstrate the utility of a functional assessment of behavioral dysfunction arising from focal neurologic disease.

Conventional neuropsychological testing (e.g., Lezak, 1983) often provides the context for a rich demonstration of abnormal intellectual performances, such as the inability to communicate (aphasia), to recognize common objects (agnosia), or to draw a clock (apraxia). Despite gross correlations between the results of these tests and neuroimaging techniques, however, it is also common to have different clinical syndromes among patients with ostensibly similar lesions (Mohr & Sidman, 1975). Moreover, different research centers are often unable to replicate one another's findings when studying behavioral aspects of particular illnesses or medications, such as HIV disease (Sidtis & Price, 1990) and anticonvulsant drugs (Trimble, 1987).

One likely reason for the diversity of findings and the failure of replication is the lack of specificity in neuropsychological tests of the stimulus and response variables that comprise these clinical measures. For data concerning brain-behavior relationships to be scientifically meaningful, it is necessary to define disorders operationally, using carefully delineated procedures (Sidman, Stoddard, Mohr, & Leicester, 1971). A functional approach achieves this goal by regarding patients as experiments unto themselves, serving as their own controls, so that deficits are established in the context of an overall pattern of performance.

We describe below the case of a man with a brain tumor who developed difficulty with reading and we use our examination of his deficits to illustrate the application of a functional approach to assessment. All testing occurred at the bedside, using a combination of neuropsychologic tests. All data were recorded by the attending clinician.

A Functional Assessment

Dejerine (1892) described a patient who could not "read" but was able to write. A postmortem examination revealed infarction (stroke) of the left medial occipital lobe and the splenium of the corpus callosum. The syndrome, which has subsequently been reported in a large number of cases (cf. Friedman & Albert, 1985), is usually caused by ischemia of the left posterior cerebral artery territory, and rarely by tumor (e.g., Greenblatt, 1973; Vincent, Sadowsky, Saunders, & Reeves, 1977).

Despite the abundance of reported cases, the data regarding the characteristics of the behavioral disorder are often difficult to interpret because of the limited ways in which reading dysfunction has been measured. If a patient cannot read printed words aloud, for example, it is not possible to determine from that observation alone whether the deficit pertains to printed words, to oral naming, or to the specific linguistic relations between printed stimuli and oral responses. The breakdown, for instance, may occur only when the task is expressive (e.g., naming), with preservation of receptive abilities (e.g., matching). Each stimulus and response, therefore, must also be a component of other tests to assess their multiple roles in the reading repertory (Albert, Yamadori, Gardner, & Howes, 1973). The major aim of a functional assessment, then, is the presentation of stimulus and response conditions which discriminate intact from disrupted abilities.

The diagram in Figure 1 depicts this approach to the assessment of alexia without agraphia. We adapted and extended the testing paradigm used by Sidman (1971) to study equivalence relations associated with reading skills. Each oval represents one category of stimuli; each box represents one modality of responding; and arrows labeled with numbers indicate the categories of relations between them. There are three stimulus categories and two responses modalities. At the extreme left, there are auditory stimuli which are dictated to the patient; they consisted of letter names, syllables, words, and sentences. The oval at the top of the figure denotes pictures and color hues, which were combined here only for purposes of exposition. The inclusion of color-related stimuli is based on the common finding of color anomia (inability to name color hues) and achromatopsia (inability to discriminate hues) in alexic patients (Albert, Goodglass, Helm, Rubens, & Alexander, 1981). Printed stimuli, indicated at the bottom of the diagram, consist of letters, syllables, words, and sentences; and they are the visual equivalents of the auditory stimuli. Printed words and printed sentences also correspond to the pictures and color hues. At the far right, there is written naming (above) and oral naming (below), both performed by the patient.

The lines with arrows, denoted by the numbers 1 through 4, connect the three stimulus ovals and represent matching-to-sample tasks. The patient is presented a sample stimulus and a set of comparison (choice) stimuli; the patient's task is to select the correct comparison. The dictated word "red," for instance, is presented as a sample along with six color hues as choices. The performances represented by Arrows 1-4, pointing from samples to comparisons, are receptive in nature because the patient's behavior demonstrates that there is a relation between two stimuli (e.g., a printed word and a corresponding picture); the choice response, itself, is not part of the relation, but serves as an indicator that a relation exists. There are several reasons for the two arrows between printed stimuli and pictures/hues, associated with Tasks 3 and 4. First, the ability to respond appropriately to the interchangeable roles of samples and comparisons represents a necessary condition for equivalence relations (Sidman, Rauzin, Lazar, Cunningham, Tailby, & Carrigan, 1982). Second, there are possible differences in function of a stimulus as a consequence of its role as a sample or comparison in a matching task (Lazar, Davis-Lang, & Sanchez, 1984). When a visual stimulus is presented as a comparison, it is shown concurrently with other members of the same category (color hues, pictures, etc.). In contrast, a sample stimulus is presented individually on a trial, without simultaneous cues provided by other members in its category. Thus the patient must recall the functional role of each sample stimulus. Tasks 3 and 4 ensure that all visual stimuli serve both sample and comparison functions.

Tasks 5 through 10 are expressive relations; in each instance an explicit naming response is required in the presence of each particular stimulus. Arrows for these relations point from stimulus to response. In Tasks 5, 7, and 9, the patient is asked to write a name; Performances 6, 8, and 10 entail oral naming by the patient.

Examination of the interconnections in Figure 1 reveals the manner in which our analysis permits a systematic approach to the assessment of alexia without agraphia. Each stimulus category is involved in both receptive (matching) and expressive (naming) tasks. Both naming modalities (written and oral) occur in response to each of the three stimulus categories. Oral naming to dictated stimuli (Task 6) and written naming to printed stimuli (Task 9) are not linguistic performances, per se, but serve as other measures of the patient's ability to discriminate among speech sounds and printed stimuli, respectively. Returning to the earlier illustration of the patient who cannot orally read printed words (Task 10), it now becomes feasible to define the reading deficit in terms of all his performances involving printed stimuli and oral naming. Can he match, for example, printed words to dictated names (Task 2) or to pictures/hues (Task 3)? Can he name orally stimuli from other categories, such as pictures (Task 8)?

For purposes of exposition, we have not included in Figure 1 an arrow designating relations between other categories of printed stimuli, such as matching upper and lower case letters, matching synonyms, and matching sentences of comparable meaning but with different syntactic structures. Moreover, we only describe the roles of visual and auditory stimuli and have omitted tactile input. The nature of our analysis, however, does not preclude any of these other possibilities.

Case Study

Clinical history. The patient (O. N.), a 45-year-old man and a university-educated executive in a multinational corporation, was entirely well until February 1983 when he developed episodic hallucinations of flashing lights in the right visual field. The episodes lasted up to a half-hour and were unassociated with other symptoms. In June 1983 he noted the gradual onset of decreased vision to the right of the midline, indicated by an inability to see objects passing from the left to the right visual field. In mid-August he developed mild frontal headache associated with rapid change in posture and he began to note that he had some difficulty remembering names of people he knew well. ln August he sought ophthamologic help, and a right visual defect was found. A CT scan revealed a mass in the left occipital area. He was placed on Dilantin and referred for further treatment.

On neurologic examination he appeared healthy and alert; general physical examination was entirely normal. He was awake, alert, oriented, and could converse easily in four languages. He knew the names of the presidents of the United States, France, and Italy, and performed calculations without difficulty. His speech (in English) was fluent and comprehension appeared intact. He was unable, however, to recall at first attempt any of three words five minutes after he had been asked to repeat them. The patient was able to write without difficulty, but reading was slow and laborious because, as he described it, "I cannot see the whole word." Visual acuity on the left was corrected to 20/40 and on the right to 20/25. There was a complete loss of vision in the right field of both eyes, extending from the midline to the end of the periphery. Pupils were equal and reactive. There were full eye movements without involuntary oscillations. Sensation and motor power in the face were normal, and the rest of the neurologic examination was entirely intact with normal strength, sensation, tone, and deep tendon reflexes. Routine laboratory evaluation was normal except for a CT of the brain revealing a large mass in the left occipital lobe. A left occipital craniotomy was performed with partial removal of a highly malignant, primary brain tumor (glioblastoma multiforme), shown in Figure 2.

Procedure. Patient O. N.'s reading skills were evaluated one day prior to his craniotomy. Table 1 shows that we presented at least one assessment of each of the ten categories of relations in Figure 1. The tests were derived from the Peabody Picture Vocabulary Test - Form M (Dunn & Dunn, 1981), the Boston Diagnostic Aphasia Examination (Goodglass & Kaplan, 1972), the Woodcock Reading Mastery Tests Form A (1973), and the Wide Range Achievement Test (Jastak & Jastak, 1978). It should be noted that with the exception of Task 4, the assessment of each category of relations involved the presentation of standardized, clinical tests. Specific auditory and visual stimuli (and therefore corresponding classes) were not always comparable across tests, numbers of test trials differed, and display characteristics (stimulus size and arrangement, hue properties, etc.) varied considerably. Nevertheless, matching tasks consisted uniformly of the simultaneous presentation of a sample stimulus and at least four comparison stimuli. The patient received verbal instructions prior to each test to either match or to name in accordance with the standardized procedures associated with that task. In none of these tests was the patient given feedback on the appropriateness of his responses.

Task 1 was measured via the Peabody Picture Vocabulary Test, for example, so that the patient was presented with a dictated-word sample (spoken once) and four picture choices on each trial. The ability to match dictated words to printed words (Task 2) was derived from a test in which there was a dictated-word sample (spoken once) and five printed-word comparisons per trial. There was a sample and six choices in printed word to picture/hue matching associated with Task 3. The latter task was reversed so that there could be picture/hue samples and printed-word choices (Task 4).

In the expressive tests, there was a single stimulus on each trial, and the patient was required to make a written (Tasks 5, 7, and 9) or oral (Tasks 6, 8, and 10) response. For Task 5, he had to write individual letters, numerals (not shown in Figure 1), and words in response to dictation. Another word-writing task to dictation was presented (spelling subtest from the Wide Range Achievement Test) in order to document further the absence of a writing disorder. Oral repetition of dictated words and phrases (Task 6) was measured with 10 spoken words and 16 spoken phrases, each presented once per trial. Written TABULAR DATA OMITTED naming to 10 pictures/hues was used for Task 7. Oral naming in response to pictures of objects, geometric forms, pictures of actions, and hues assessed the patient's performance of Task 8. Task 9 was determined by the patient's capacity to copy the printed sentence, "The quick brown fox jumped over the lazy dog." Oral naming of printed stimuli (Task 10) was measured in four tests: Oral naming of printed and script letters, printed words, and printed nonsense syllables from the Woodcock Reading Mastery Tests, and oral naming of printed words on the Wide Range Achievement Test.

Results. From the standpoint of a traditional analysis, the results corroborate the diagnosis of "alexia without agraphia." The patient had difficulty reading printed material aloud (Task 10), but his writing ability in response to dictation (Task 5) was intact. Based on normative data for Task 10, he scored below the 2nd percentile on oral reading of letters and nonsense syllables, and at the 33rd percentile in oral reading of printed words. The reading score on the Wide Range Achievement Test was at the 61 st percentile, which was significantly below expectation for a man of his educational level and occupation. In contrast, his ability to write letters, numerals, and words in response to dictation (Task 5) was perfect when derived from the Boston Diagnostic Aphasia Examination and he performed at the 86th percentile on the spelling portion of the Wide Range Achievement Test.

An examination of the entire test profile, however, suggests a more complex picture, one that entails an analysis of the other functions of the stimulus and response elements that comprise Task 10. It was important to ascertain whether the deficit pertained generically to printed stimuli and/or to oral naming, or only to the specific relations between them. We first asked whether the patient could orally name other stimuli. The data from the tests associated with Task 6 showed nearly perfect repetition of dictated words and phrases, and he was able to name orally almost without error pictures, geometric forms, pictures of actions, and color hues (Task 8). We therefore concluded that oral naming, by itself, was not impaired.

Printed stimuli were involved in four other kinds of relations. The patient made only one error when required to match printed words to dictated-word samples (Task 2). Matching printed words with pictures/color hues (Tasks 3 and 4) was performed with 100% accuracy. Patient O. N. was also able to copy the printed sentence (Task 9) with a single mistake (omission of the letter "i" in quick). Thus the poor performance on Task 10 could not be ascribed to a global disorder with printed material.

As a result of our tests, we were able to rule out independent deficits associated with either oral naming or printed stimuli. The data suggest that the deficit was restricted to the explicit relations between printed stimuli and oral responses. With the aid of an audio-recording of the assessment session, we further analyzed the nature of the responses, and we found two consistent patterns. First, O. N. read each printed stimulus syllable-by-syllable, in a slow, labored manner. A multiple-syllable printed word, for example, usually took several seconds to read aloud. The other response pattern was found in his errors. In most cases, he read the first letters or syllable correctly, but then erred on the remaining portion of the word. Table 2 provides some examples of incorrect oral naming responses to printed words (top panel) and printed syllables (bottom panel). Errors also occurred in the beginning and middle portions of words. In addition, more errors occurred with nonsense syllables than English words, suggesting that stimulus properties from the beginning of printed words provided more linguistic cues than nonwords.

Discussion

A functional approach to the analysis of alexia without agraphia permitted us to establish precisely the conditions under which our patient's performance deteriorated. We examined the variety of functions served by stimuli and responses in the reading/writing repertory to determine whether the deficit was definable in terms of individual stimulus categories, response modalities, or specific relations between these elements. In the case of O. N., we were able to isolate his disorder to the relations between printed stimuli and oral naming by demonstrating that other performances involving printed material and oral responses remained intact. His dysfunction was therefore delineated in the context of an overall evaluation of his reading abilities (c.f. Caramazza, 1984).

Table 2

Representative Errors of Patient's Oral Reading Responses to
Printed Stimuli

Printed Words                   Oral Responses

alertly                            "alert"
alkali                             "alike"
instigator                         "intimator"
unsociable                         "insatiable"
expostulate                        "expotulate"
plagiarism                         "plagaris"

Printed Nonsense Words          Oral Responses

ab                                   "ad"
tash                                 "task"
wips                                 "wiks"
plen                                 "pled"
hets                                 "hast"
plon                                 "plas"
expram                               "expam"
stabe                                "stade"
telequik                             "tetequik"
subscrate                            "lubcrate"

It has been reported previously that matching-to-sample skills can remain intact in the face of dysfunctional oral naming of the same visual stimuli. Kreindler and Ionasescu (1961), employing a technique called "auditory and visual unblocking," found that their patient could not orally name printed words but was able, for example, to match dictated words to printed words. Similar dissociations have also been reported by Albert et al. (1973), Caplan and Hedley-Whyte (1974), Stachowiak and Poeck (1976), and Staller, Buchanon, Singer, Lappin, and Webb (1978), although the selection of tasks in each of these cases was limited and nonsystematic. Nevertheless, the data from these patients in conjunction with ours suggest that alexic syndromes may be task-specific, such as the relation between printed text and oral naming when the lesion is similar to that described by Dejerine (1892).

The measurement of intermediate levels of function remains problematic. In patients who have severe developmental or acquired disabilities, the near-total absence of particular skills makes easier the comparison between intact and disrupted performances. The behavioral manifestations of neurologic disease, however, can range from complete absence of a skill to deficits so subtle that only the patient can detect a change. The use of normative data on neuropsychologic tests helps to interpret partial loss of function to some extent. Nevertheless, we are left with having to quantify the nature and degree of change in individual cases.

An important implication of the functional approach to the study of reading dysfunction concerns the ambiguities inherent in the use of the traditional terms alexia and reading (Mohr, 1976). The present case shows clearly that one kind of reading skill was disrupted, but that others seemed unaffected by the lesion. To some extent, our patient's tendency to incorrectly name the right-hand portions of printed words and syllables resembles the "visual dyslexia" described by Marshall and Newcombe (1973), but this error pattern did not occur in our patient when printed stimuli were components of other tasks. Because reading comprises many performances, it seems reasonable to propose that, rather than relying on broad diagnostic categories, a more powerful methodology would involve correlations between cerebral damage and specific types of reading dysfunction.

It is tempting to draw conclusions from our patient's data regarding the role of naming in equivalence-class formation. In his case, oral naming of printed stimuli was impaired in the context of intact auditory-visual and visual-visual matching, seeming to provide support for the notion that naming is not necessary for the demonstration of equivalence classes (Lazar et al., 1984; Sidman, Willson-Morris, & Kirk, 1986). There are several reasons, however, why O. N.'s skills cannot address this issue. First, we assume that he was capable of naming printed stimuli prior to his illness. We cannot determine the manner in which he acquired stimulus equivalences or naming relations for these stimuli and responses, in particular, or the techniques used to learn reading skills, in general. Could naming have served a mediating role during acquisition of stimulus classes but somehow lost its functional role during maintenance? Further research is needed to determine the consequences of the breakdown of prerequisite tasks or relations for already-established equivalence classes.

Second, English was not our patient's first language. We do not know, for example, whether he was capable of oral naming of printed stimuli in other languages. Could the preservation of matching relations with English stimuli be attributed to mediation by names in another language? Does the importance of naming in the acquisition of equivalences differ between native and second languages?

The ability to address these questions experimentally is a significant advantage of a functional approach to brain-behavior relationships over traditional, neuropsychologic testing. With its emphasis on normative comparisons, psychometric assessment commonly seeks to sample behavior along poorly specified continua in steps that vary simultaneously multiple stimulus and response variables. In contrast, an equivalence-class analysis, for example, permits investigation of a full range of stimulus-stimulus and stimulus-response relations that underlie a complex skill such as reading. A functional view of neuropsychologic dysfunction encompasses a more comprehensive approach to assessment, greater specificity of controlling variables that could potentially be investigated, and a finer-grained evaluation of rehabilitation techniques.

References

ALBERT, M., GOODGLASS, H., HELM, N. A., RUBENS, A. B., ALEXANDER, M. P. (1981). Clinical aspects of dysphasia. New York: Springer-Verlag.

ALBERT, M., YAMADORI, A., GARDNER, H., & HOWES, D. (1973). Comprehension in alexia. Brain, 96, 317-328.

CAPLAN, L. R., & HEDLEY-WHYTE, T. (1974). Cuing and memory dysfunction in alexia without agraphia. Brain, 97, 251-262.

CARAMAZZA, A. (1984). The logic of neuropsychologic research and the problem of patient classification in aphasia. Brain and Language, 21, 9-20.

DAMASIO, H., & DAMASIO, A. R. (1989). Lesion analysis in neuropsychology. New York: Oxford University Press.

DEJERINE, J. (1892). Contribution a l'etude anatomo-pathologique et clinique des differentes varietes de cecite verbale. Memoires de la Societe de Bioloqie, 4, 61-90.

DUNN, L. M., & DUNN, L. M. (1981). Peabody Picture Vocabulary Test-Revised. Circle Pines, MN: American Guidance Service.

FRIEDMAN, R. B., & ALBERT, M. (1985). Alexia. In K. M. Heilman & E. Valenstein (Eds.), Clinical neuropsychology (2nd ed.) (pp. 49-73). New York: Oxford University Press.

GOODGLASS, H., & KAPLAN, E. (1972). Assessment of aphasia and related disorders. Philadelphia: Lea and Febiger.

GREENBLATT, S. H. (1973). Alexia without agraphia or hemianopsia. Brain, 96, 307-316.

HEILMAN, K. M., & VALENSTEIN, E. (1985). Clinical neuropsychology (2nd ed.). New York: Oxford University Press.

JASTAK, J. F., & JASTAK, S. R. (1978). The Wide Range Achievement Test. Wilmington, DE: Jastak.

KREINDLER, A., & IONASESCU, V. (1961). A case of "pure" word blindness. Journal of Neurology, Neurosurgery, and Psychiatry, 24, 275-280.

LAZAR, R. M., DAVIS-LANG, D., & SANCHEZ, L. (1984). Formation of visual stimulus equivalences in children. Journal of the Experimental Analysis of Behavior, 41, 251-266.

LEZAK, M. (1983). Neuropsychological assessment. New York: Oxford University Press.

MARSHALL, J. C., & NEWCOMBE, F. (1973). Patterns of paralexia: A psycholinguistic approach. Journal of Psycholinguistic Research, 2, 175-199.

MESULAM, M-M. (1985). Principles of behavioral neurology. Philadelphia: F. A. Davis.

MOHR, J. P. (1976). An unusual case of dyslexia with agraphia. Brain and Language. 3, 325-334.

MOHR, J. P., & SIDMAN, M. (1975). Aphasia: Behavioral aspects. In M. F. Reiser (Ed.), Organic disorders and psychosomatic medicine. In S. A. Arieti (Ed.), American handbook of psychiatry (VoL 4) (pp. 279-298). New York: Basic Books.

SIDMAN, M. (1971). Reading and auditory-visual equivalences. Journal of Speech and Hearing Research, 14, 5-13.

SIDMAN, M., RAUZIN, R., LAZAR, R., CUNNINGHAM, S., TAILBY, W., & CARRIGAN, P. (1982). A search for symmetry in the conditional discriminations of rhesus monkeys, baboons, and children. Journal of the Experimental Analysis of Behavior, 37, 23-44.

SIDMAN, M., STODDARD, L. T., MOHR, J. P., a LEICESTER, J. (1971). Behavioral studies of aphasia: Methods of investigation and analysis. Neuropsychologia, 9, 119-140.

SIDMAN, M., WILLSON-MORRIS, M., & KIRK, B. (1986). Matching-to-sample procedures and the development of equivalence relations: The role of naming. Analysis and Intervention in Developmental Disabilities 6, 1-19.

SIDTIS, J. J., & PRICE, R. W. (1990). Early HIV-1 infection and the AIDS dementia complex. Neurology, 40, 323-326.

STACHOWIAK, F. J., & POECK, K. (1976). Functional disconnection in pure alexia and color naming deficit demonstrated by facilitation methods. Brain and Language, 3, 135-143.

STALLER, J., BUCHANON, D., SINGER, M., LAPPIN, J., & WEBB, W. (1978). Alexia without agraphia: An experimental case study. Brain and Language, 5, 378-387.

TRIMBLE, M. R. (1987). Anticonvulsant drugs and cognitive function: A review of the literature. Epilepsia, 28(Suppl 3), $37-$45.

VINCENT, F. M., SADOWSKY, C. H., SAUNDERS, R. L., & REEVES, A. G. (1977). Alexia without agraphia, hemianopsia, or color-naming deficit: A disconnection syndrome. Neurology, 27, 689-691.

WOODCOCK, R. W. (1973). Woodcock Reading Mastery Tests. Circle Pines, MN: American Guidance Service.