BRA,N

AND

LANGUAGE

23, 337-348 (1984)

Residual Reading Capability in a Patient with Alexia
without Agraphia
DARIO GROSSI, NINA A. FRAGASSI, ARTURO ORSINI,
FABRIZIO A. DE FALCO,* AND ORNELLA SEPE*
C/inica

Neurologica,

II Facolta
di Medicina,
di Loreto Nuovo,

Naples,
Naples

and *Ospedale

S. Maria

A case of pure alexia due to an ischemic lesion of the occipital temporal region
is described. Written words could be matched but not read. Immediate memory
span for graphemes was defective. The reading defect probably depends on the
inability to modify the written word “globally”; the phonological process was
intact, but the memory disturbance impeded reading. The dissociation is explained
by the preservation of word forms, which are linked to the semantic stage. Nonwritten stimuli trigger a “meaning” which evokes the word form and so the
written word is recognized even though it cannot be read. o 1984 Academic PRSS,
Inc.

Alexia without agraphia is a relatively usual clinical syndrome. It was
first described by Dejerine in 1892 and has been associated with involvement
of the medial occipital region in the dominant hemisphere (Benson, Brown,
& Tomlinson, 1971), often related to occlusion of the left posterior cerebral
artery (Gloning, Gloning, & Hoff, 1968; Benson & Geshwind, 1969).
A residual reading capability has been found in patients with alexia.
Kreinder and Ionasescu (1961) used facilitating (“unblocking”)
auditory
and visual tests to permit the reading of certain words and sentences in
a patient with “pure” alexia. Albert, Yamadori, Gardner, and Howes
(1973), in a patient with a particular form of alexia with agraphia, found
that perception of written words was intact and that semantic understanding
of them was relatively well preserved. Caplan and Hedley-White
(1974)
in a patient with alexia without agraphia demonstrated that reading was
possible only when naming was not required or when cueing was introduced. Rothi and Heilman (1981) reported a patient with alexia and
The authors are grateful to Professor Hans Spinnler of the Centro di Neuropsicologia
of Milan for advice concerning investigation methods, and to Professor Virgilio Agnetti of
the Neurological Department, University of Sassari for providing the map of the lesion.
Address correspondence to Dr. Dario Grossi, Clinical Neurologica, 11 Falcolta di Medicina,
Via Pansini 5, 80131 Naples, Italy.
0093-934X/84 $3.00
Copyright
0 1984 by Academic Press, Inc.
All rights of reproduction
in any form reserved.

338

GROSS1

ET AL.

agraphia who had intact spelling and comprehension of spelled words
and who used “ideographic
letter-naming strategy” to read.
We report a clinical neuropsychological
investigation of a patient with
pure alexia. We found a dissociation between naming and recognition
of written words, and an impaired memory span for graphemes.
This study highlights the role played by the internal representation of
written words in first recognizing, then connecting written words to
meaning (semantic stage), and the role of memory in letter-by-letter
reading.
CASE REPORT
Our patient is a 64-year-old, right-handed decorator and sign painter,
with primary school education only. He had a lo-year history of diabetes,
treated with oral hypoglycemic agents.
In April 1980, at the age of 62, he suddenly developed loss of strength
and a sensation of cold in the lower limbs, derangement, and spacetime disorientation.
Upon hospitalization,
examination
showed blood
pressure of 200 mm Hg, disorientation (which improved after a few days),
anomia, right hemianopsia and alexia, and a mild loss of strength in the
right upper limb. A radioisotope brain scan showed increased uptake in
the left medial occipital area.
Laboratory tests and radiological examination were within normal limits.
In 1982 he was admitted to the Neurological
Section of S. Maria di
Loreto Hospital because of the persistence of reading disturbances. Neurological examination showed right homonimous hemianopsia (confirmed,
by Goldman perimetry) and alexia. Sensation was unimpaired. An EEG
demonstrated
moderate left parieto-occipital
slowing. Laboratory and
radiological examinations were within normal limits. A C.T. scan showed
a normal fourth ventricle, and a clearly outlined area of decreased density
in the occipital and temporo-occipital
region with mild enlargement of
the lateral ventricle in the left hemisphere. Left thalamus and splenium
were also damaged. There was no cerebral cortical atrophy. Figures la
and b show the C.T. scan and an anatomical reconstruction of the damage.
A map of the lesion is illustrated on a drawing of the medial brain face
in Fig. 2.
NEUROPSYCHOLOGICAL

STUDY

(1) Agnosia
(a) Color agnosia (Table I). (al). Color matching test. The patient
was requested to pair equally colored labels within two equal series of
15 differently colored labels. Score 15/15.
(a2) Zschihara test. Score 21121.
(a3) Color naming. The patient was requested to name 12 colored
labels. Score 12112.

READING

CAPACITY

FIG. 1.

IN ALEXIA

339

340

GROSS ET AL.

SUPERFICIAL
DEEP
+
SUPERFICIAL

FIG.

2.

(a4) Color recognition. The patient was requested to identify a color

named by the examiner

among 12 colored labels. Score 12/12.

correlation. The
patient was shown a picture of an object. There were five objects on a
table: one was the object represented in the picture (e.g., a pipe), two
were objects of the same category (e.g., a cigarette and an ashtray), and
two were objects of different categories (e.g., a glass and a pen). The
patient was requested to select the object represented in the picture
among others. All object names had a low frequency (Bortolini, Tagliavini,
& Zampolli,
1972). Score 1000.
(b2) Naming simple drawings. The patient was shown 15 pictures
(5 of objects in everyday use, 5 of common flowers, and 5 of common
fruits) one at time. He was asked to name them. Score 15/15.
(b) Picture agnosia (Table 2). (61) Picture-object

TABLE
COLOR

1

AGNOSIA

Task

Score

Matching
Naming
Recognition
Ischiara

D/15
12/12
12/12
21121

READING

CAPACITY

341

IN ALEXIA

TABLE

2

PICTURE AGNOS~A

Task

Score

Picture-object
Naming
Describing

lO/lO
15/15
14115

(b3) Describing complex pictures. The patient was shown the following
pictures: five common objects in unusual perspectives (e.g., a glass shown
from above); five architectonic perspectives generally used by architects
to study a place (e.g., a bedroom shown from a corner); five b&are
pictures (e.g., a centaur with an elegant dress or a man with enormous
hands and very small feet). The patient was requested to describe the
pictures; he had to identify the principal subjects and the particularity
of the pictures. Score 14115.
(2) Memory
(a) Immediate memory span (Table 3.) (al) Corsi’s spatial span test

(Grossi, Matarese,

& Orsini,

1980). Score 4.

(a2) Wechsler Digit Forward test. Score 4.
(a3) Word verbal span. Score 2.
(a4) Phoneme span. The patient was asked to repeat a phoneme

string (consonant and vowels). The test began with a sequence of one
unit and if the patient succeeded, increasingly long sequences up to a
possible maximum of eight were presented. Each time a maximum of
five different sequences was given. The score was the highest string
reproduced accurately three times out of five. The rate of presentation
was one phoneme per second. Score 5.
(a5) Grapheme span. The patient was asked to repeat a grapheme
string (consonant and vowels) orally. Each grapheme of one string was
given, one at time, on a label in printed capitals. The patient was invited
to read aloud each grapheme and then to repeat the entire string. The
scoring and the rate of presentation was the same of the phoneme span
test. Score 1.
TABLE
IMMEDIATE

3

MEMORY

Task
Corsi’s spatial span test
Digit forward test
Word verbal span
Phoneme span
Grapheme span

SPAN

Score
4
4
2
5
1

342

GROSS1 ET AL.

(b) Learning. (bl) Spatial learning at Corsi’s test (Orsini, Schiappa,
Chiacchio, & Grossi, 1982) of a span series plus two digits. The patient
did not reach the criterion (two consecutive correct repetitions of the
stimulus series) in 20 trials, and did not reproduce more than two digits
in the correct order in any response.
(b2) (Free) verbal learning of 10 two-syllable words in 20 trials. The
patient did not reach the criterion (two consecutive correct repetitions
of all the 10 words). The maximum number of words correctly repeated
was 5.
(c) Description of places. The patient was asked to describe places
well known to him. The descriptions were incomplete and confused. On
more than one occasion the patient described architectural elements that
had long since disappeared (10 normal controls that live in the same
place performed this task without any difficulty).
(3) Language Examination (Table 4)
(a) Spontaneous speech: jluent, without paraphasia or anomia.
(6) Description of everyday objects (Basso, Capitani, & Vignolo, 1979).

Score 9110.
(c) Naming everyday objects (Basso et al., 1979). Score 1l/15.
(d) Understanding. Token test. Score 28136.
(4) Writing

No disturbance evident in either spontaneous or dictated writing (Table
5). The patient was not able to read the words written by himself, He
copied written words as they were spatial constructions.
(5) Phonemic spelling (Table 6)

The patient was asked to pronounce the phoneme components of presented words. We called this kind of spelling phonemic because the
patient had to avoid the naming of single letters but sound them out
(e.g., if a word contains the letter “b” the patient had to sound out “b”
and not to name “bi”). The presented words had a low frequency, 10
of them were four letters long, 5 were five letters long, and 5 were six
letters long. Each word had to be spelled correctly in its entirety.
TABLE 4
LANGUAGE EXAMINATION
Task
Spontaneous speech
Description
Naming
Token test

Score
Fluent without paraphasia or anomia
9110
11/15
28/36

READING

CAPACITY
TABLE

343

IN ALEXIA
5

WFUTING

N
Spontaneous
writing
Letters
Words
Sentences

N

Correct

Correct

10
28
10

Correct

20
20
10

20
20
3

Copying

Dictated
32
30
10

N

Letters
Words
Sentences
Numbers

32
20
10
20

28
20
10
20

Letters
Words
Sentences

(a) Orally presented words. Score 20120.
(b) Words in printed form. The printed words were not removed before

the patient

spelled them. Score M/20.

(bl) Reading of words spelled correctly. Score O/20.
(b2) Recalling letters of the words spelled correctly. Score: only one

letter per word.
The tests bl and b2 were given to demonstrate the patient’s inability
to combine and remember recognized graphemes. It is useful to mention
that letter-by-letter
reading in Italian is different from letter-by-letter
reading in English; in fact there are intrinsic differences between the two
orthographies.
(6) Reading
(a) Reading. The patient was asked to read letters, words, nonwords,
and sentences in printed and script form (Table 7); he was also presented
with overturned letters (top-bottom
reversal) and numbers (Table 7).
Reading of the letters was slightly deficient and there was not significant
difference between the type of writing presented. In most cases the
patient took a very long time to identify each single letter; on several
occasions he spoke a string of letters aloud until he came to the one
presented. He could read only two of the words printed and none of the
sentences or nonwords. He was able to match letters in printed capitals
and script form. Any type of paralexia was not found.
(b) Proprioceptive reading. The patient was blindfolded and his finger
was guided around the outline of letters, so every letter was transformed
TABLE
PHONEMIC

Task
Orally presented words
Words in printed form
Reading words spelled correctly
Recalling letters of correctly spelled words

6

SPELLING

Score
20120
18/20
o/20
1 letter per word

GROSS1 ET AL.

344

TABLE

7

READING

No. of tests

Correct responses

42
42
50
50
10
42
30
20
20

38
36
2
3
0
0
32
30
16
0

15
15
20
20
10
10

13
15
19
18
0
0

40

36

Reading
Printed capital letters
Letters in script form
Words in printed capitals
Words in script form
Sentences in printed capitals
Sentences in script form
Overturned printed capital letters
Simple numbers
Complex numbers
Nonwords of four letters
Proprioceptive reading
Letters recognized blindfolded
Letters recognized without blindfold
Words recognized blindfolded
Words recognized without blindfold
Sentences recognized blindfolded
Sentences recognized without blindfold
Letters in printed form and in script
form matching

10

into proprioceptive
information.
Letters and words were almost always
read, while reading of a sentence was impossible. When the blindfold
was removed, the patient traced the outline of the letters himself, with
the same result (Table 7).
(c) Matching test (Table 8). The aim of this test was to verify II tne
word form (the visual internal representation of the written word) was
intact. Lexical items are characterized in a visual code in word form
system; we attempted to reach the word forms usin‘g other codes. Word
stimuli were presented with various perceptive means (visual, auditory,
proprioceptive),
and the patient had to select the written equivalent from
TABLE
MATCHING

8
TEST

Delayed
Word stimulus
Auditory word
Proprioceptive word
Written word
Image
Object

No. of
tests

Correct
responses

No. of
tests

Correct
responses

12
12
12
12
12

10
9
0
9
9

12
12

9
8

12
12

8
8

READING

CAPACITY

IN ALEXIA

345

a list of 10 two-syllable words all beginning with the same letter. Five
words of the list were similar for form. For each of the five matching
tests the patient was given 12 trials and for each trial there was a different
list of IO two-syllable written words.
(cl) Auditory word-written
word. The examiner said a two-syllable
word which the patient had to select from a list of 10 written words.
Score 10/12.
(~2) Proprioceptive
word-written
word. The patient was blindfolded,
his finger was traced around the outline of a written word so every
written letter was transformed into proprioceptive information (we called
this type of stimulus “proprioceptive
word”), and then the patient wa:
requested to select the word from a list of 10 written words after removing
the stimulus written word. Score 9/12.
(~3) Picture-written
word. The patient was shown a picture of an
everyday object and had to identify it from a list of 10 written words
(the stimulus picture was not removed during this test). Score 9112.
(~4) Object-written
word. The patient was shown an object and had
to find the corresponding word in a list of 10 words (the stimulus object
was not removed during this test). Score 9/12.
(~5) Written word-written
word. The patient was shown a written
word and had to find the equivalent in a list of 10 words (the stimulus
word was not removed during this test). Score O/12.
(d) Delayed matching test (Table 8). The above mentioned tests cl,
c2, c3 and c4, were also presented with a delay of 30 set between the
presentation of the stimulus and the task performance. In c3 and c4
delayed performance tests the stimulus was removed during the tests.
COMMENT
Our patient is clearly affected by alexia without agraphia. He performed
well on writing tests, while he was never able to read a word or a
sentence. He had no difficulty in identifying letters regardless of the form
of writing presented. Spelling of written words was nearly always correct
even though it was slow and tedious: the patient compared each letter,
step by step, with a series of letters that he pronounced aloud until he
came to the one presented. He was never able to read a word even
though he spelled it correctly, neither was he able to remember the
letters he spelled correctly.
In accordance with other reports of this disorder, our patient also had
an impairment
of memory (Mohr, Leicester, Stoddard, & Sidman, 1971;
Caplan & Hedley-White,
1974). This deficit is due to lesions of the deep
structures of the temporal lobe and the thalamus of the left hemisphere
(Geshwind & Fusillo, 1966), which were clearly damaged in our patient.
The patient’s selective deficit of immediate memory for graphemes is
particularly
interesting, especially when compared to the good score

346

GROSS1

ET AL.

obtained at the span of phonemes (in our experience 60-year-old normal
subjects under the same conditions of testing have a grapheme span
between 5 and 7 well-remembered
units). There is a strong memory
interference for graphemes which arises from the slow, tedious device
adopted by the patient to recognize letters: the test of immediate memory
span was transformed into a long term memory test. The latter was
impaired particularly when stimuli were presented with graphemes. The
patient was unable to find a strategy which enabled him to retain written
letters for several seconds, even those which were correctly recognized
and named.
The most relevant finding is that the patient was able to match nonwritten
stimuli to written words, as has been found in other patients with this
disorder (Kreindler & Ionasescu, 1961). The internal representation of
the written words, the so-called word forms, was clearly preserved. Word
forms could be evoked through nonwritten stimuli and could be compared
with the written word. The comparison between written word and written
word failed in that the patient was unable to manipulate the stimulus as
a spatial pattern.
It should be stressed that no marked differences were found between
the performance on delayed tests and that on nondelayed tests. This
suggests that delay tasks require different routes to those required to
perform other memory tests. Similarly, comparison between word form
and written word must use nerve routes other than those used to compare
written word with word form, the former being possible and the latter
not in this condition.
The matching tests show that to evoke the internal representation of
a specific written word, a nonwritten stimulus must undergo a series of
transformations
(comparison with the internal representation relative to
the perceptive strategy used, naming and/or connection with a certain
meaning). All these transformations,
which are strictly concerned with
verbal processes, we call “internal language.”
The tests excluded the possibility that our patient had aphasic or
agnosic disturbances. His good performance on color tests contrasts with
observations that color agnosia is associated with pure alexia, especially
in forms due to lesions of the medial lesions of the left occipital lobe
(Mohr et al., 1971). The patient’s occupation as a decorator may have
some bearing on the results of color tests.
CONCLUSIONS

Andreewsky (1974) proposed a model of reading based on cognitivist
psychology. The model foresees a first reading process which consists
of an analytical-type perceptive level (letter-by-letter, syllable-by-syllable)
and a linguistic level which, depending on the phonological stage (graphemephoneme correspondence),
precedes and conditions the semantic stage.

READING

CAPACITY

IN ALEXIA

341

FIG. 3. “Acustic word,” ” proprioceptive word,” “picture,” and “object” are transformed
in “internal language” (meaning). From this stage access to the “word form” system is
possible; an evoked specific “word form” allows written word recognition. “Written word”
cannot access to the word form system directly because of splenium damage. The “letterby-letter reading” process works but fails because of memory defect (“interference”)
depending on “thalamus and hippocampus damage.”

The second reading process is global at the visual level, thus words are
simultaneously
taken as a whole, and the unit is made to coincide with
a specific internal representation. At the linguistic level, reading depends
on the lexical or semantic stage (nonphonological reading process). Clinical
evidence has shown the independence of the two systems (Beauvois &
Derouesne, 1979).
Our patient’s disturbance seems to be at a perceptive level of the
second reading process, that is of the global reading of a word. The
semantic stage is not impaired. An inverse process at the perceptive
level is possible (see Fig. 3).
(1) A specific meaning is given to a nonwritten stimulus (internal
language).
(2) The meaning evokes a specific word form.
(3) The word form allows the written word to be recognized.
This process shows that the meaning of a word is linked to all its
mental representations
(different depending on the different stimuli);
moreover it shows that the meaning evokes a specific mental representation
and vice versa. The meaning itself is a link between several representational
events, that thus become a whole.
Regarding the first reading process, the perceptive level in our patient
is intact. The phonological level works but reading is not possible because
memory disturbances hinder the performance of the correct strategy.
It is likely that the splenium of the corpus callosum is indispensable

348

GROSS1 ET AL.

for the perceptive level of the “nonphonological”
reading process. The
process of word recognition could occur through other routes. Our interpretation of the selective deficit of memory span for graphemes supports
the hypothesis that the left hyppocampal structure is necessary for phonological reading.
We have demonstrated a residual reading capacity only when the patient
was requested to read using word forms in simple recognizing tests. The
patient had lost the “key” that gives access to the internal representations,
and therefore, according to our model, to the meanings. The “key” is
the correct combination
of “grapheme clusters” (written words) made
reading propossible by both the phonological and “nonphonological”
cesses. In our patient the former was impaired by a disorder at perceptive
level, and the latter by a disturbance of selective memory for graphemes.
REFERENCES
Albert, M. L., Yamadori, A., Gardner, H., & Howes, D. 1973. Comprehension in alexia.
Brain, 96, 317-328.
Andreewsky, E. 1974. Un modele semantique. Application a la pathologie du language:
Alexia aphasique. Traitemente
Automatique
Informations,
2, 3-27.
Basso, A., Capitani, E., & Vignolo. L. A. 1979. Influence of rehabilitation on language
skills in aphasic patients. A controlled study. Archives of Neurology,
36, 190-196.
Beauvois, M. E., & Derouesne, J. 1979. Phonological alexia: Three dissociations. Journal
of Neurology,
Neurosurgery,
and Psychiatry,
42, 1115-l 124.
Benson, D. F., Brown, J., & Tomlinson, E. B. 1971. Varieties of alexia. Word and letter
blindness. Neurology
(Minn.),
21, 951-957.
Benson, D. F., & Geshwind, N. 1969. The alexias. In P. J. Winken &G. W. Bruyn (Eds.),
Handbook
of clinical
neurology.
Amsterdam: North-Holland. Pp. 112-140.
Bortolini, U., Tagliavini, C., & Zampolli, A. 1972. Lessico di Frequenza
della Lingua
Italiana
Contemporanea.
Milanno: Garzanti.
Caplan, L. R., & Hedley-White, T. 1974. Cueing and memory dysfunction in alexia without
agraphia. A case report. Brain, 97, 251-262.
Geschwind, N., & Fusillo, M. 1966. Color naming defects in association with alexia.
Archives
of Neurology,
15, 137-146.
Gloning, I., Gloning, K., & Hoff, H. 1968. Neuropsychological
symptoms
and syndromes
in lesions of the occipital
lobe and the adjacent
areas. Paris: Gautlier-Villars.
Pp. 75
81.
Grossi, D., Matarese, V., & Orsini, A. 1980. Sex differences in adults’ spatial and verbal
memory span. Cortex, 16, 339-340.
Kreindler, A., & Ionagescu, Y. 1961. A case of “pure” word blindness. Journal of Neurology,
Neurosurgery,

and Psychiatry,

24, 275-280.

Mohr, J. P., Leicester, J., Stoddard, L. T., & Sidman, M. 1971. Right hemianopia with
memory and color deficits in circumscribed left posterior cerebral artery territory
infarction. Neurology
(Minn.),
21, 1104-l 113.
Orsini, A., Schiappa, 0.. Chiacchio, L., & Grossi, D. 1982. Sex differences in children’s
spatial serial-learning task. The Journal of Psychology,
111, 67-71.
Rothi, L. J., L Heilman, K. M. 1981. Alexia and agraphia with spared spelling and letter
recognition abilities. Brain and Language,
12, l-13.