Journal of

J Neurol (1984) 231 : 20-25

Neurology
© Springer-Verlag 1984

Alexia without agraphia:
a study of a case of verbal alexia without accompanying colour-naming defect
G. Rosati 1, P. De Bastiani 2, I. Aiello 1, and V. Agnetti 1
i Neurological Clinic of the University of Sassari, Italy
2 Neurological Clinic of the University of Ferrara, Italy

Summary. A 47-year-old man developed a persistent form of
alexia without agraphia as the result of a haemorrhagic intracerebral lesion in the left inferior temporo-occipital region,
due to the rupture of an arteriovenous malformation. Surgical
evacuation of the haematoma and excision of the malformation did not produce any modification of the reading deficit,
which remained unchanged during a 4-year follow-up. The
reading deficit was restricted to words and the patient was able
to read only letter-by-letter, so that the whole words were
reconstructed from the auditory names of the letters. So far,
the disconnection explanation is the standard explanation of
alexia without agraphia and the present case of verbal alexia
may be regarded as being within this overall category. However, this explanation meets with unanswered questions that
suggest more flexible interpretations. Neurolinguistic studies
have questioned the unique character of alexia without agraphia as a clinical entity and, in contrast to the disconnection
hypothesis, support the notion that the different varieties of
alexia that are traditionally described represent distinctive
syndromes, each with its own clinical features and pathophysiological basis. In this context, the reading properties in
this case seem fully compatible with a deficit of the visual
word-form system postulated by Warrington and Shallice, that
is, the relatively early stage of the reading process through
which a word-form or equivalent unit is attained. This system
might be lateralized to the left hemisphere, as suggested by
the fact that this case, like other cases of verbal alexia, had
sustained damage to the left hemisphere and did not show any
differential preservation of the reading of concrete words.
Key words: Alexia without agraphia - Verbal alexia - Letter-byletter reading - Disconnection - Visual word-form system

Zusammenfassung. Bei einem 47j/ihrigen Mann trat eine intracerebrale Blutung infolge Ruptur eines arterioven6sen
Angioms der linken basalen Temporookzipitalregion auf. Dies
f'tihrte zu einer persistierenden Alexie ohne Agraphie. Auch
die chirurgische Ausr/iumung des H/imatoms und der Mil3bildung bewirkte keine Ver/inderung der Leseschwierigkeiten,
die w/ihrend vier Jahren konstant blieben. Die Lesest6rung
beschr/inkte sich auf W6rter, und der Patient war lediglich
f'dhig, Buchstabe um Buchstabe zu lesen, so dab das Gesamtwort lediglich rekonstruiert wurde. Bisher beruht die Standarderkliirung der Alexie ohne Agraphie auf der Diskonektionstheorie und der vorgelegte Fall verbaler Alexie kann
innerhalb dieser Kategorie gedeutet werden. Immerhin wirft

Offprint requests to: Prof. Giulio Rosati, Clinica Neurologica dell'Universit/l, Viale S. Pietro 10, 07100 Sassari, Italy

dieser Fall eine Reihe von offenen Fragen auf, welche eine
flexiblere Interpretation nahelegen. Neurolinguistische Studien haben den einmaligen Charakter der Alexie ohne Agraphie als klinische Entitiit in Frage gestellt und sprechen - im
Gegensatz zu der Diskonektionshypothese - zugunsten der
Annahme, dab unterchiedliche Arten von Alexie auch eigenst/indige Syndrome darstellen, jedes davon mit seinen eigenen
klinischen Eigenheiten und pathophysiologischen Grundlagen. In diesem Zusammenhang scheinen die Leseeigentiimlichkeiten im vorliegenden Fall kompatibel zu sein mit einem
Defizit des visuellen Wort-Formsystemes, wie es von Warrington und Shallice postuliert wurde: das ist ein relativ friihes
Stadium des Leseprozesses durch welches eine Wortform oder
~iquivalente Einheit erreicht wird. Dieses System k6nnte im
Bereiche der rechten Hemisph~ire lokalisiert sein, woftir ein
Argument im vorliegenden Fall, wie auch in anderen Fallen
von verbaler Alexie sprechen wtirde: hier war eine L/ision der
linken Hemisph/ire erzeugt worden, ohne dab irgend eine
selektive Erhaltung der F/ihigkeit zum Lesen konkreter Worte
nachweisbar gewesen ware.

Introduction
Alexia without agraphia is a well-defined syndrome in which a
literate person loses the ability to read but retains the capacity
to write both to dictation and spontaneously [7]. Autopsies of
cases of alexia without agraphia have demonstrated the universal presence of lesions of the left occipital lobe, often including
the splenium of the corpus callosum [9]. In the context of the
widely accepted disconnection hypothesis revived by Geschwind [17], the importance of this double (occipital and callosal) lesion in the genesis of the syndrome has been stressed.
The lesion would disconnect the left angular gyrus from its
bilateral visual input impeding verbalization of the graphic
symbols. In alexia without agraphia, however, there are aspects
that remain difficult to explain in terms of callosal disconnection. A major obstacle in accepting this theory is that while the
patients are unable to name letters or words, they can name
objects, number or colours [20,23]. Further objections have
arisen from demonstration of different varieties of the reading
disorder (literal, verbal, sentence, and global alexia) within the
main syndrome. Neurolinguistic analyses of alexia without
agraphia would seem to suggest that these varieties do not
represent different degrees of severity of the same basic disorder, but rather that they represent structurally different
forms reflecting significant variations in the site of causative
pathology [15, 20].

21
This report describes the case of a right-handed man who
developed a persistent form of alexia without agraphia, resulting from localized haemorrhagic damage to the ventrolateral
aspect of the left occipito-temporal junction, with sparing of
the corpus callosum. In this case, the reading of words was
seriously impaired, while the reading of individual letters was
preserved. Spelling was normal and words were deciphered
letter-by-letter. This case provides further evidence that the
lesions resulting in verbal alexia are localized in the left temporo-occipital region and reinforces the importance of the ventral inputs from the visual cortex in the process of reading.

Case report
A 47-year-old, right-handed man, with a grade school certificate, was admitted to the Neurological Clinic on 28 December
1978 complaining of not being able to read any longer. He had
been in excellent health until the day prior to admission, when
he experienced the sudden onset of a throbbing headache
during lunch. In the ensuing hours the headache improved and
he went out to buy a packet of cigarettes. While he was going to
the tobacconist's, he wanted to read an advertising sign but was
not able to do so. Then, he went into a bookshop, but every
attempt to read failed. Therefore, he consulted his family doctor, who referred him to our clinic.
On admission, he was alert and oriented to person, place
and time. The neurological examination showed no abnormalities, except for right homonymous upper quadrantanopsia.
Visual acuity was normal. The patient was given a formal
Language Examination Test (Association for the Development of Neuropsychological Research, Milan, 1974). He
showed a specific impairment in word reading, but his ability
to read individual letters was preserved, as was his ability to
spell (a more detailed reading examination is reported below).
His writing capacity was unimpaired. Handwriting was fluid
and legible without elisions or substitutions and the patient
had no difficulty in writing sentences from dictation, although
a few minutes later he could not read without errors what he
had written before. He was able to repeat fluently words
spelled out by the examiner and to decipher, when blindfolded, single letters and common four-to five-letter words
assembled with plastic letters, tracing each of them by both
the right or the left index finger. His spontaneous speech was
fluent, prosodic, articulate, without any kind of errors. Repetition of words, non-words, and sentences was excellent and
the verbal comprehension of even complex commands was immediate. He had no difficulty in naming colours, objects and
pictures of objects presented from an unusual view. Number
reading was normal and mathematical basic operations were
performed verbally without difficulty. Finger recognition as
well as right-left discrimination were normal. There was no
kind of apraxia. On discrimination of unfamiliar faces he
scored 49 out of S1. On the Token Test he scored 32 out of 36.
On the Wechsler Adult Intelligence Scale (WAIS), the patient
obtained a verbal IQ of 100 and a non-verbal IQ of 82, the latter
being affected mainly by slowness in time-dependent tasks;
the full-scale IQ was 92. On the picture completion and picture
arrangement subtests of the WAIS (the stimuli are complex
visual scenes) he scored 9 and 7, respectively. Poppelreuter's
figures were recognized within the normal average range and
the patient scored normally on a test of discrimination of
embedded figures. On the Wechsler Memory Scale (WMS) he
scored 82. Digit span was 6 forward and 4 backward in both the
WAIS and the WMS.

To clarify further the patient's selective impairment of
word reading, an extensive reading examination was devised.
In different sessions, the patient was given a large sample of
controlled stimuli to read. The considered dimensions of the
stimuli were taken from Cornoldi [12], Roncato [27], and Bartolini et al. [5]. All the stimuli (i.e. words and non-words) printed
on a white sheet in standard lower-case letters (pica, 5 m m
high) were presented one at a time, free field, and without time
pressure. On occasions words and non-words were presented
in blocks, otherwise randomly intermixed. A set of stimuli
was also presented in CaSe AlTernAtiOn (i.e. lower and upper
case intermixed) and a set of script-like stimuli (Letraset Juliet,
7.6 mm). All the stimuli were arranged for the following different tasks: (1) letter reading: reading aloud (or pointing at)
a marked letter confused in an array of different letters (e.g.
a n i p f); (2) word reading: reading aloud concrete-abstract,
high-low frequency words of 5-11 letters (total stimuli=280);
(3) non-word reading: reading aloud non-words of different
length, all easily pronounceable (total stimuli = 120); (4) "part
of speech" reading: reading aloud nouns, adjectives, verbs, and
functors, all rated for frequency (total s t i m u l i = 80 + 60 + 60 +
100); (5) lexical decision: when given a word-non-word pair,
pointing at the word or non-word (total stimuli = 60); (6) comprehension: matching words to corresponding pictures (48
words plus 48 pictures); (7) semantic correlation: matching
words semantically, related in a multiple-choice task (30 words
semantically related to 30, plus 120 distractors).
To summarize briefly the results of the reading examination, the patient made no errors on the letter reading, lexical
decision, comprehension, and semantic correlation tasks. He
always knew which was a word and which not. His ability to
match words with pictures was perfect as well as his appreciation of the semantic relationship. On the other hand, the
patient's word reading was very laborious, marked by hesitations and abnormal slowness. He tended to read silently in the
correct order single letters of a given word, to pronounce the
letter sequence with frequent pauses, and then to pronounce
the whole word as a final unit. As shown in Table 1, a clear
effect of the word length was present (longer words taking
more time to be pronounced than shorter words) and the reading of non-words took much more time than the reading of
words. The reading of stimuli presented with case alternation
took the same amount of time as reading of lower-case stimuli.
A slight difference was found only for script-like stimuli, which
took a little more time to read than printed words. The patient,
who had been reported to be a normal reader before the accident, also made a number of errors, most often resulting in
single phonemic substitutions (e.g. MARE --- MORE; FORBICE ~ FERBICE; SAPONE -~ SAPENE; ORSO -~ ORZO;
A L L E A N Z A -~ ALLEONTA). Apart from his paralexia, he
very rarely made other types of errors that seemed to be visual
in nature (e.g. QUANTITA'---VANITA'; SEZIONE ~ SELEZIONE; DEVIAZIONE --, DELUSIONE). The types and distribution of the errors according to the syntactic classes of the
stimuli are shown in Table 2. As can be seen, the errors (66
errors out of 620 stimuli) were randomly distributed across
the different classes of stimuli. The patient was aware of all the
mistakes and able to correct 65% of them immediately. No concrete-abstract or high-low frequency effect was present. No
"part of speech" effect was detected. No relationship was
observed between errors and the imagery value or meaningfulness of the stimuli. The errors rate in reading non-words
was not different from the errors rate in reading words, the only
difference in reading words and non-words being the different

22
Table 1. Time (in seconds) taken to read a complete list of words and a list of non-words of different length (number of letters). Half of the words
were high-frequency stimuli and half low-frequency stimuli
Number of stimuli (words and non-words)

10

20

20

20

10

Number of letters for each stimulus
Time to read the list of words (s)
Time to read the list of non-words (s)

4
189
248

5
279
380

6
470
710

7
580
930

8
610
-

a 10 words and 10 non-words

Table 2. Type and distribution of errors for different syntactic classes and for non-words. In brackets: errors immediately corrected. Total errors
out of 500 meaningful stimuli-- 48 (9.6%), of which 34 (70.8%) were immediately corrected. Total errors out of 120 meaningless stimuli= 18 (15%),
of which 9 (50%) were immediately corrected. Total errors out of 620 stimuli = 66 (10.6%), of which 43 (65.2%) were immediately corrected
Stimulus

Words (280)
Non-words (120)
Adjectives(60)
Verbs (60)
Functors(100)

Number and type of errors
One-letter
substitution

Two-letter
substitution

Other

Total

16 (16)
8 (4)
4 (4)
2 (1)
3 (1)

7 (5)
6 (4)
2 (1)
2 (-)
--

3 (1)
4 (1)
3 (1)
6 (4)

26 (22)
18 (9)
6 (5)
7 (2)
9 (5)

Fig. la-d. Four consecutive slices from inferior (a) to superior (d) from the pre-operative non-enhanced CT scan. The three slices of the low
ventricular series (a-c) show a high-density lesion (in the range of blood) in the left inferior temporo-occipital region. In the highest slice
of the low ventricular series (c) and in the lowest slice of the high ventricular series (d), the splenium of the corpus callosum appears to be
unaffected

23

Fig.2a-c. Three consecutive slices of the low ventricular series from
the enhanced CT scan obtained 6 months after surgery. In the region
previously occupied by the haematoma, there is evidence of a lowdensity area. The neighbouring areas and the splenium of the
corpus callosum, the latter imaged in the highest slice (c),
appear to be unimpaired

lengths of time needed to read them (much longer for meaningless stimuli). The patient was only able to read the shortest
sentences owing to his often being depressed as a result of fatigue to read, the continuous stream of attention and the effort
needed to retain what he had been reading.
The findings from routine laboratory tests were normal. A
non-enhanced CT scan was indicative ofa haemorrhagic intracerebral lesion, surrounded by an oedematous area, in the
region of the left occipitotemporaljunction (Fig. 1). According
to the "sequential CT anatomy" data proposed by Gado et al.
[16], the lesion was shown in the three slices of the low ventricular series (Fig. la, b, c). The lesion damaged the inferior
aspect of the left temporal lobe in the part contiguous to the
lateral border of the cerebellum (Fig. la) and extended into the
occipitotemporal white matter reaching, on the lateral aspect
of the hemisphere, the posterior part of the inferior temporal
gyrus (Fig. lb) and the contiguous part of the lateral occipital
gyrus (Fig. lb, c). The lesion was surrounded by an oedematous area, with evidence of mass-effect, resulting in obliteration of the homolateral occipital horn. In the highest slice of
the low ventricular series (Fig. lc) and in the lowest slice of the
high ventricular series (Fig. ld), the latter showing no evidence
of the lesion, the splenium of the corpus callosum appeared to
be unaffected. Intravenous contrast did not produce any modification.
Percutaneous arteriography or the left carotid and retrograde arteriography of the left brachial artery showed filling of
thin opacified angiomatous vessels and of a wide arteriovenous
shunt in the left inferior temporo-occipital region. The arteriovenous malformation was supplied from vessels arising from

both the inferior temporal branch of the cerebl posterior artery
and the posterior temporal artery of the Sylvian system.
Surgical treatment was then decided upon, and 7 days after
admission an osteoplastic craniotomy was carried out in the
left parieto-temporo-occipital region. A mild expansion of the
left posterior inferotemporal region extending to the bordering
occipital region was noted. After cortical incision, a haematoma cavity appeared that extended deep into the subcortical
white matter. The evacuation of the haematoma provided the
exposure of a small arteriovenous malformation in the inferior
wall of the haematoma. The feeder of the malformation was
clipped and the malformation was totally excised.
During the initial postoperative period, the patient showed
only a mild increase of his visual defect. Verbal alexia
remained unchanged and no additional neurological sign was
detected. Subsequently, the visual defect contracted, and
2 weeks after surgery it had returned to its preoperative level.
The patient was discharged on 3 February 1979.
In the course of the following 4 years the patient's visits to
the neurological clinic were frequent and regular. On these
occasions, no significant change was observed in either the
verbal alexia or the right upper quadrantanopsia.
Six months after leaving hospital, an enhanced CT scan was
carried out, which demonstrated a low density area in the
region previously occupied by the haematoma. As in the preoperative CT scan, the lesion was shown in the three slices of
the low ventricular series (Fig. 2), but was especially evident in
the lowest (Fig. 2a) and intermediate (Fig. 2b) slices. Thus, the
lesion was clearly localized in the region corresponding to the
left occipitotemporal junction. The residual postoperative

24
lesion appeared to be well defined, and it was not surrounded
by oedema. There was no evidence of pathology in the neighbouring areas, particularly in the splenium of the corpus callosum.
The most recent neuropsychological examination, in
October 1982, showed no significant modification of the alexia;
he had even developed a useful strategy for reading based on
spelling. When the patient was given a written polysyllabic
word, he could only read it after a pause. He repeated each
letter to himself, then joined them into a complete word. His
capacity to read bisyllabic words was consolidated, as was the
capacity to read and to understand very short phrases.

Discussion

The case reported in this study clearly corresponds to those
previously described under the terms "verbal alexia" [15, 20],
"verbal alexia-simultanagnosia" [24,26,34], and "word-form
dyslexia" [33]. The patient was unable to read or grasp words
globally, but his ability to read or identify single letters was
preserved. He was able to read only letter-by-letter, so that the
whole word was reconstructed from the auditory names of the
letters, presumably relying on his intact spelling skills. Reading was very slow and laborious and attempts to read more
quickly were vitiated by paralexic errors, most often resulting
in single phonemic substitutions. The time taken to read
words was clearly related to their length, and reading script
was slightly more laborious than reading print. His ability to
name visually presented objects, numbers or colours was intact
and the memory span did not account for any memory-deficitsustaining errors in reading. On CT scan, the patient had a unilateral left-sided lesion in the inferior temporo-occipital
region, leaving the splenium and the primary visual cortex
intact. The optic radiations were partly disrupted, as suggested
by the right superior quadrantanopsia.
So far, the disconnection explanation is the standard explanation of alexia without agraphia [9] and many view verbal
alexia as being within this overall category [8,19, 31]. In the
context of the disconnection hypothesis, the different forms of
alexia (literal, verbal, and global alexia) that are traditionally
described are interpreted as points in a continuum, depending
on variations in the intensity of the same basic disorder [8]. In
all cases, the underlying pathology should disconnect the
left angular gyrus from its bilateral visual input, impeding verbalization of graphic symbols. This hypothesis, however,
meets with a number of problems that have not yet been
understood satisfactorily. A persisting problem is how to
explain the fact that while the patients are unable to name
words, they can name objects, numbers or colours [3, 20, 23]. In
Geschwind's opinion [17], object naming and number reading
might be preserved because visually presented objects and
numbers arouse tactile associative activity in the brain, leading
to excitation of anterior areas in the right hemisphere, which is
then transmitted to the left angular gyrus through caUosal
fibres anterior to the splenium. This is regarded as an ad hoc
explanation [20], which is difficult to reconcile with recent
observations on the nature of the callosal code [10, 23]. On the
other hand, this explanation does not account for the preserved ability to name colours in several cases of alexia without
agraphia, since colours, like words, have verbal associations
but, unlike objects, no tactile ones [18]. Some authors [2,13, 19]
have related this occasional preservation of colour naming in
alexia without agraphia to the preservation of the dorsal aspect

of the splenium; as a consequence, the view has emerged that
while colour naming is served by both the dorsal and ventral
outflow from the visual areas to the left angular gyrus, reading
is served by the ventral outflow only [2,19]. This latter view
makes the anatomical location observed in our case of verbal
alexia without accompanying colour naming defect fully compatible with the disconnection hypothesis. In our case, as in
other published cases of verbal alexia [1, 20-22, 24, 30, 31], the
reading disorder resulted from localized damage to the left
inferior temporo-occipital region. Although the splenium was
thought to be intact, the lesion was in the precise location to
disrupt bot the inferior association tracts from the left visual
association areas to the left angular gyrus and the inferior
callosal fibres coming from the right visual areas as they pass
from the splenium to the dominant hemisphere, producing the
same effect of the classical splenio-occipital lesion. However,
even though this explanation appears plausible from an anatomical point of view, there remains the problem of why the
reading failure should be restricted to word identification only.
Within the disconnection hypothesis [8], letters might be
spared because they would represent less complex visual
stimuli and, therefore, would be handled better than words by
the damaged transmission pathways. This assumption has
been questioned by Warrington and Rabin [32], who showed
that short-term storage of letter strings in patients with unilateral cerebral lesions imposes a greater information load than a
word of the same length. More recently, describing two personal cases of verbal alexia caused by unilateral left-sided
lesions, Warrington and Shallice [33] concluded that the dissociation between letter span and word reading makes an
explanation of spelling alexia in terms of the classical disconnection hypothesis most implausible.
Throughout the history of studies on alexia, arguments
have been provided that verbal alexia is a component element
of the syndrome of simultanagnosia, i.e. the failure of the
capacity to comprehend the "wholes" though the capacity to
name or identify the single parts is preserved [24, 26, 34]. In our
case of verbal alexia no formal test of simultanagnosia was performed, but the results on both the picture completion and picture arrangement subtests of the WAIS, together with the
normal performance on Poppelreuter's figures and on the
embedded figures indicate that complex picture interpretation
was relatively unaffected. A similar dissociation between
letter-by-letter reading and piecemeal perception of complex
scenes has recently been observed by Warrington and Shallice
[33], who concluded that the status of simultanagnosia as a
syndrome must be questioned.
During the last two decades, new impetus for research on
alexia has been provided by application of neurolinguistic
models. In general, the neurolinguistic studies support the
notion that reading is a multistage process and that alexia
without agraphia, like alexia with agraphia and aphasic alexia,
may be dissociated according to linguistic characteristics
into several distinctive syndromes, each with its own clinical
features and pathophysiological basis [4,11,15, 20, 25, 28, 29].
Within this approach, the idea has emerged that reading involves at least two processes, which must be distinguished at
the perceptuovisual level as well as at the linguistic level [6].
The first process (phonological reading process) would be analytical at the visual level and phonological at the linguistic
level, while the second process (non-phonological or lexical
reading process) would be global at the visual level (the different letters of a word are grasped simultaneously and globally
as a unit) and lexical or semantic at the linguistic level [6].

25
In our patient, words were read analytically, letter-by-letter,
and non-words, which should be read by m e a n s of the phonological reading process only [14], were read almost as efficiently
as words. S e m a n t i c errors were not observed and variables
such as frequency, concreteness and "parts of speech" did not
influence the reading performance. This w o u l d suggest that
his phonological processing was intact and that he was suffering from a disturbance of the non-phonological reading
process, presumably at a level prior to the semantic stage. In
other words, the properties of the reading disorder in our case
s e e m compatible with a deficit of the visual word-form system
recently postulated by W a r r i n g t o n and Shallice [33], that is the
stage b e t w e e n early visual processing and semantic processing,
through which a visual word-form is attained. This system
might be lateralized to the left h e m i s p h e r e [33], as suggested by
the absence of any differential preservation of the reading of
concrete words and by the fact that m a n y patients with letterby-letter reading had sustained damage to the left hemisphere,
as did the patient described in this study.

References
1. Aiax ET (1964) Acquired dyslexia. A comparative study of two
cases. Arch Neurol 11 : 66-72
2. Aiax ET, Schenkenberg T, Kosteljanetz M (1977) Alexia without
agraphia and the inferior splenium. Neurology (Minneap) 27 :
685-688
3. Albert ML (1979) Alexia. In: Heilman KM, Valenstein E (eds)
Clinical neuropsychology. Oxford University Press, Oxford, pp
59-91
4. Albert ML, Yamadori A, Gardner H, Howes D (1973) Comprehension in alexia. Brain 96 : 317-328
5. Bartolini V, Tagliavini C, Zampolli A (1972) Lessico di frequenza
della lingua italiana contemporanea. Garzanti, Milano
6. Beauvois MF, D6rouesn6 J (1979) Phonological alexia: Three
dissociations: J Neurol Neurosurg Psychiatry 42 : 1115-1124
7. Benson DF (1979) Aphasia, alexia, and agraphia. Churchill
Livingstone, London
8. Benson DF, Brown J, Tomlinson EB (1971) Varieties of alexia.
Neurology (Minneap) 21 : 951-957
9. Benson DF, Geschwind N (1969) The alexias. In: Vinken PJ,
Bruyn GW (eds) Handbook of clinical neurology, vol 4. North
Holland, Amsterdam, pp 112-140
10. Bogen JE (1979) The callosal syndrome. In: Heilman KM, Valenstein E (eds) Clinical neuropsychology. Oxford University Press,
Oxford, pp 308-359
11. Bottcher R (1974) Role of graphic and semantic-syntactic factors
in reading: a neuro-psycholinguistic study. J Psychol 182 : 40-67
12. Cornoldi C (1974) Imagery values for 310 italian nouns. Ital J
Psychol 1 : 211-225
13. Cumming WJK, Hurwitz I_J, Perl NT (1970) A study of a patient
who had alexia without agraphia. J Neurol Neurosurg Psychiatry
33 : 34-39

14. D6rouesn6 J, Beauvois MF (1979) Phonological processing in
reading: data from alexia. J Neurol Neurosurg Psychiatry 42 :
1125-1132
15. Dubois-Charlier F (1972) A p r o p o s de l'alexie pure. Langages
25 : 76-94
16. Gado M, Hanaway J, Frank R (1979) Functional anatomy of the
cerebral cortex by computed tomography. J Comput Assist
Tomogr 3 : 1-9
17. Geschwind N (1965) Disconnexion syndromes in animals and
man. Brain 88 : 237-294, 584-644
18. Geschwind N, Fusillo M (1966) Color-naming defects in association with alexia. Arch Neurol 15 : 137-146
19. Greenblatt SH (1973) Alexia without agraphia or hemianopsia.
Brain 96 : 307-316
20. H6caen H, Kremin H (1976) Neurolinguistic research on reading
disorders resulting from left hemisphere lesions: aphasic and
"pure" alexias. In: Whitaker H, Whitaker HA (eds) Studies in
neurolinguistics, vol 2. Academic Press, New York, pp 269-329
21. Johansson T, Fahlgren T (1979) Alexia without agraphia: lateral
and medial infarction of left occipital lobe. Neurology (Minneap)
29 : 390-393
22. Kinsbourne M, Warrington EK (1963) The localizing significance
of limited simultaneous visual form perception. Brain 8 : 697-705
23. LeDoux JE (1979) Parietooccipital symptomology: the split-brain
perspective. In: Gazzaniga MS (ed) Handbook of behavioral
neurobiology, vol 2. Neuropsychology. Plenum Press, New York,
pp 61-74
24. Levine DN, Calvanio R (1978) A study of the visual defect in verbal alexia - simultanagnosia. Brain 101 : 65-81
25. Marshall JC, Newcombe F (1973) Patterns ofparalexia: a psycholinguistic approach. J Psycholinguist Res 2 : 175-199
26. Poetzl O (1928) Die Aphasielehre vom Standpunkte der klinischen Psychiatrie. In: Handbuch der Psychiatrie. Franz Deuticke,
Wien
27. Roncato S (1974) Concreteness, imagery, meaningfulness and a
new index: values for 100 italian nouns. Ital J Psychol 1 : 195-209
28. Saffran EM, Marin OSM (1977) Reading without phonology: evidence from aphasia. Q J Exp Psychol 29 : 515-525
29. Shallice T, Warrington EK (1975) Word recognition in a phonemic
dyslexic patient. Q J Exp Psychol 27 : 187-199
30. Turgman J, Goldhammer Y, Braham J (1979) Alexia without agraphia due to brain tumor. Ann Neurol 6 : 265-268
31. Vincent FM, Sadowsky CH, Saunders RL, Reeves AG (1977)
Alexia without agraphia, hemianopia, or color naming defect: a
disconnection syndrome. Neurology (Minneap) 27 : 689-691
32. Warrington EK, Rabin P (1971) Visual span of apprehension in
patients with unilateral cerebral lesions. Q J Exp Psycho123 : 423431
33. Warrington EK, Shallice T (1980) Word-form dyslexia. Brain
103 : 99-112
34. Wolpert I (1924) Die Simultanagnosie. Z Gesamte Neurol Psychiatr 93 : 397-415

Received June 13, 1983