2011. Vol.2, No.5, 416-425
Copyright © 2011 SciRes. DOI:10.4236/psych.2011.25065
On the Relationship between Mind and Language:
Teaching English Vocabulary to Schizophrenics
Abdolmajid Hayati1, Khaled Shahlaee2
1Shahid Chamran University of Ahvaz, Ahvaz, Iran;
2University of Science and Research, Ahvaz Branch, Ahvaz, Iran.
Received March 25th, 2011; revised May 12th, 2011; accepted June 23rd, 2011.
Undoubtedly mind is the most complicated member of human being and respectively a small breakdown in
whatever processes undergone will cause great difficulties for the beholder. One of the most serious psycho-
logical impairments is known as schizophrenia which severely distorts thought and action (Hockenbury &
Hockenbury, 2004). The present study is set out to investigate the capability of patients with schizophrenia in
doing different vocabulary tasks on second language and retaining them for a period of time. Eight male patients
diagnosed with schizophrenia of different types were selected from two mental centers to participate in the study.
They were invited to attend a class for 16 sessions in 8 weeks and do vocabulary ta s ks i ncl udin g matching words
to the pictures, filling the blanks, word association, crossword puzzle, and word finding. A test was administered
right after the period and the retest was administered 16 days later. The study illustrated that there is no signifi-
cant relationship between age and learning vocabulary items by patients suffering from schizophrenia. More,
schizophrenics showed different capabilities in doing vocabulary tasks, and they also showed different impair-
ments in their attempt to learn foreign language words. They also found some of the tasks demanding as they
become more cognitively difficult.
Keywords: Psychological Im pairment, Schizophre nia , Vocabulary Task
One of the main branches of linguistics known as psycho-
linguistics has long been an area attracting researchers and
linguists to investigate and find any trace of language in human
mind. In this respect, this study seeks to shed light on a new
dimension of psycholinguistic viewpoint regarding second lan-
guage learning to those who suffer from schizophrenia, which
Hockenbury and Hockenbury (2004) refer to as one of the most
serious psychological disorders involving severely distorted
beliefs, perceptions, and thought processes (p. 525). Myers
(1996) defines psychological disorders a condition in which
behavior is judged atypical, disturbing, maladaptive, and unjus-
tifiable (p. 415). Due to the mental impairment, schizophrenics
produce a kind of language which might be senseless or irrele-
vant and, to some extent, all aspects of language will be dis-
turbed. Teaching a language other than their mother tongue to
schizophrenics will magnify the possibilities of opening a new
horizon to second language learning approaches. The acquisi-
tion of first language takes place spontaneously under an in-
herent ability given to human beings (Chomsky, 1972 as cited
in Aitchison, 2008), while learning a second language requires
awareness of the stages and processes during the course.
Therefore, the aim of this study is to investigate the relati onship
between mind and learning second language vocabulary items
in patients suffering from schizophrenia.
Review of Literature
Schizophrenia literally means split mind. It refers to both a
multiple personality split and a split from reality that shows
itself in disorganized thinking, disturbed perceptions, and inap-
propriate emotions and actions (Myers, 1996). Elsewhere,
Hockenbury and Hockenbury (2004) define schizophrenia as
one of the most serious psychological disorders which involves
severely distorted beliefs, perceptions, and thought processes (p.
525). Generally, schizophrenia is defined as a psychotic disor-
der marked by delusion, hallucination, apathy, thinking abnor-
malities, and a split between thought and emotion. One person
in 100 will become schizophrenics, and roughly half of all the
people admitted to mental hospitals are schizophrenics. Most
are young adults, but schizophrenia can occur at any age (Coon,
2001: p. 575). Cardwell (2003) defines schizophrenia as a seri-
ous mental disorder characterized by severe disruption in psy-
chological functioning.
Hockenbury and Hockenbury (2004) define four types of
schizophrenia. Paranoid type shows such characteristics as
well-organized delusional beliefs reflecting persecutory or
grandiose ideas, f requent auditory hallucinations, usually voices,
and little or no disorganized behavior, speech or flat effect.
Catatonic type having characteristics like highly disturbed
movements or actions, such as extreme excitement, bizarre
postures or grimaces, being completely immobile, and echoing
of words spoken by others or imitations of movements of others.
Undifferentiated type displays characteristic symptoms of
schizophrenia but not in a way that fits the pattern for paranoid,
catatonic, or disorganized type. Patients with disorganized type
show flat or inappropriate emotional expressions, severely dis-
organized speech and behavior, and fragmented delusional
ideas and hallu c i n a t io n s .
Schizophrenia and the Brain
The search for a brain abnormality that causes schizophrenia
began as early as the syndrome was identified. In the last two
decades, however, stimulated by a number of technological
advances, the field has reawakened and yielded some promising
evidence. Some patients with schizophrenia have been found to
have observable brain pathology. Postmortem analyses of the
patients with schizophrenia are one source of evidence. Such
studies constantly reveal abnormalities in some areas of the
brains of patients with schizophrenia, although the specific
problems reported vary from study to study and there are many
contradictory findings. The most consistent finding is of
enlarged ventricles, which implies a loss of subcortical brain
cells. Moderately consistent finding indicated structural prob-
lems in subcortical temporal-limbic areas, such as hippocampus
and the basal ganglia, in the prefrontal and temporal cortex
(Dwork, 1997; Heckers, 1997 as cited in Davison & Neale,
Even more impressive and surprising are the images obtained
in Computed Tomography (CT) scan and Magnetic Resonance
Imaging (MRI) studies. Researchers were quick to apply these
new tools to brains of living patients with schizophrenia. Thus
far these images of living brain tissue have most consistently
revealed that some patients, especially males have enlarged
ventricles (Nopoulos, Flaum, & Andreasen, 1997 as cited in
Davison & Neale, 2001). Research also shows a reduction in
cortical gray matter in both the temporal and frontal regions
(Goldstein et al., 1999 as cited in Davison & Neale, 2001) and
reduced volume in basal ganglia (e.g., the caudate nuleus) and
limbic structures (Chua & McKenna, 1995; Gur & Pearlson,
1993; Keshavan et al., 1998; Lim et al., 1998; Velakoulis et al.,
1999 as cited in Davison & Neale, 2001), suggesting deteriora-
tion or atrophy of brain tissue.
Further evidence concerning large ventricles comes from an
MRI study of fifteen pairs of Monozygotic (MZ) twins who
were discordant for schizophrenia (Suddath et al., 1990 as cited
in Davison & Neale, 2001). For twelve of the fifteen pairs, the
twin with schizophrenia could be identified by simple visual
inspection of the scan. Because the twins were genetically iden-
tical, these data also suggest that the origin of these brain ab-
normalities may not be genetic.
There are two distinct forms of impairment (Sadock & Sa-
dock, 2005). One, deficits that arise early in development and
the other is deficits that arise in the context of clinical illness.
These two types of deficits are conceptually, as well as tempo-
rally, distinct: developmentally based deficit limit the normal
acquisition of cognitive skills, as seen on measures of intelli-
gence and academic ability. This developmental comprise is
relatively subtle, requiring sensitive formal testing and com-
parison to well-matched control population. In contrast, ill-
ness-onset-related deficits appear to present an actual decline in
function: they reflect a compromise in the ability to access, to
deploy efficiently, and to coordinate the cognitive skills that
were successfully acquired during the developmental period.
Illness-onset deficits tend to be far more severe than develop-
mentally based deficits and are obvious in the context of clini-
cal interaction.
As far as the relationship between age and schizophrenia is
concerned, an attempt has been carried out to magnify the ef-
fects of aging in patients diagnosed with schizophrenia. Older
people with schizophrenia are an often neglected group. Since
there has been no retirements and barely government support,
they shuffle into poor houses, county farms, and nursing homes.
Cognitive decline is often seen in schizophrenia. This is a
common diagnostic problem once the person has become eld-
erly. Many older people have dementia. This can be very diffi-
cult to sort out in the elderly person with schizophrenia if the
history is unavailable. The cohabitatio n of dement ed elders and
cognitively impaired elders with psychosis is commonly seen in
nursing homes. In this setting, misdiagnosis and the subsequent
incorrect treatment of elderly patients causes a multitude of
Even before coming to old age, the elderly person with
schizophrenia has had a number of treatments not shared by
elderly people without chronic mental illness. Neuroleptic use
is common, and Electroconvulsive Therapy (ECT) and even
psychosurgery occur often in the histories of elderly patients
with schizophrenia. Many former treatments lead to complica-
tions in the assessment of cognitive fall. As these people come
into treatment settings, triage and care must be provided. (Har-
vey, Wilcox, & Boudjenah, 2005).
Teaching Vocabulary
Almost all EFL courses apply vocabulary teaching as a step
toward learning a foreign language and this is executed by us-
ing a number of techniques and tasks which best facilitate this
As Scrivener (1994) stated, activities on vocabulary should
contain tasks such as matching pictures to words, matching
parts of the words to other parts, matching words to other
words like collocation, synonyms, opposites, using prefixes and
suffixes to build new ones, classifying items into lists, using
given words to complete specific task, filling in crosswords,
filling in gaps in sentences, memory games (p. 83).
Elsewhere, Celce-Murcia (2001: p. 288 ) proposes that, es-
pecially at the beginning levels, teaching word lists through
word association techniques has proven to be a successful way
to learn a large number of vocabulary items in a short period of
time and also helps to retain them. Celce-Murcia (2001) also
suggests a number of strategies such as collocations, semantic
associations, syntactic collocation types, teaching activities and
idioms (pp. 292-294).
To Schmitt (2002), deliberate vocabulary teaching may take
a variety of forms including pre-teaching of vocabulary before
a language use activity, exercises that follow a listening or
reading text like matching words and definition, and creating
word families using word parts or semantic mapping, self con-
tained vocabulary activities, word detectives where learners
report on words they have found, collocation activities, quickly
dealing with words as they occur in a lesson (p. 43).
Studies on Language, and Schizophrenia
Much has been done to investigate schizophrenia as well as
the features and properties of the schizophrenic's language
(Covington, Brown, Chaika, Congzhou, Lebedinskij, Semple,
& Sirmon, 2005). The study of schizophrenic language disorder
by linguists began with Chaika (1974, as cited in Covington et
al. 2005: p. 86), who studied a single patient who spoke nor-
mally for weeks at a time, her deviant language coinciding with
what her psychiatrists term psychotic episodes. The abnormali-
ties that Chaika (1974, as cited in Covington et al, 2005) ob-
served were: 1) Failure to utter the intended lexical item, 2)
Distraction by the sounds or senses of words, so that a dis-
course becomes a string of word associations rather than a
presentation of previously intended information, 3) Breakdown
of syntax and/or discourse, 4) Lack of awareness that the utter-
ances are abnormal. Of these, number two is the most charac-
teristic of schizophrenia; one and three resemble ordinary
speech errors, and number four resembles some forms of apha-
The bizarre language of schizophrenics is nowadays gener-
ally believed to be a meaningless breakdown product of a
neurobiological disorder. As Graham, Gath, and Mayou (1999)
point out, in the early stages of schizophrenia, speech is vague
and incomprehensive and later on we see more formal disorders.
More, some have problems in thinking of concretes while oth-
ers are involved in semi-scientific or codified beliefs. In addi-
tion, there is no connection in their belief (loosing association).
Also, in some cases, the connection in their thought is illogical
and diverts from one subject to another, and in its most severe
case, loosing of association leads to complete unorganized
speech (word salad), and there might be thought impairments
such as hard speech, poverty of thought, and thought blocking
(p. 97). Lebedinskij (1959) argues that the interrelationships
between ideation and language are, to a large extent, distinc-
tively structured in schizophrenia. Even those schizophrenics
who are apparently not suffering from speech disturbances
exhibit a series of distinctive ways of expressing themselves
with different speech production and vocabulary, all of which
are connected with disturbances in thought. Structural imaging
studies taken together show that brain cortical regions specifi-
cally known to be responsible for components of language are
reduced in size and reduced in the normal left greater than right
asymmetries (i.e. prefrontal cortex, superior temporal gyrus,
planumtemorale). It is proposed that the underlying cerebral
basis for schizophrenia comes from anomalies in the neuronal
connections between these crucial structures for normal human
language functioning and that these anomalies are genetically
controlled and develop slowly over time. From late adolescence
to early adulthood as these connections reach maturity and a
peak level of myelination, a threshold is reached whereby nor-
mal language pathways are disrupted (DeLisi, 2004).
The standard account of schizophrenic language today is that
of Anderson (1979 as cited in Covington et al., 2005: p. 88).
This is a scale which comprises 16 symptoms: poverty of
speech, poverty of content (wordy vagueness), pressure of
speech (excessive speech or emphasis), distractibility (by stim-
uli in the environment), tangibility (partly irrelevant replies),
loss of goal, derailment (loss of goal in gradual steps), circum-
stanciality (numerous digressions on the way to the goal), il-
logicality, incoherence (word salad), neologisms (novel
made-up words), word approximations (coined substitutes for
existing words, stilted speech (pompous or overly formal style),
clanging perseveration, echolalia, blocking (sudden stoppage),
and self-reference (talking about oneself excessively).
A community sample including 10 males ranging from 25 to
44 years old was selected from a mental hospital and a charity
center in which mental patients were kept. Two participants
who had a good command of English were ignored after the
interview, and only eight could attend the class and fulfill the
tasks. They all were educated and had passed high school
courses. The method of selection was purposive sampling (ob-
taining certain participants with pre-determined characteris-
tics), due to the limitation we had in accessing schizophrenics.
Besides, schizophrenia is an impairment that causes the same
impairments in all sufferers worldwide and might not make
much difference in different settings, ethnic groups, or races
(Graham, Gath, & Mayou, 1999). All of the participants were
recognized as schizophrenics by the psychiatrists and had re-
cords in the hospital and had been hospitalized there for at
least five months at the time of the study. Based on the ar-
chived records, which were attested later during the class, two
of them were recognized as disorganized type with flat emo-
tional expression and severely disorganized speech, five para-
noid types and one undifferentiated type. Kurdish was their
native and Persian their second language and, based on the
interview administered to homogenize the samples, their Eng-
lish proficiency level was limited to just familiarity with Eng-
lish alpha bet.
The background information of the patients was the first to
be provided in order to select proper samples for the research
since patients with schizophrenia are the goal. This information
was provided in the archive room of both mental hospital and
the mental center with the permission of their managers. An
interview then was designed to check the patients’ vocabulary
knowledge making sure that they had no prior word power.
This interview took about 15 minutes for each as they were
asked to say English alphabet, then naming objects around and
recognizing some words chosen from those in the syllabi. Since
this study aims to investigate the ability if schizophrenics to
recall the thought vocabulary in a kind of achievement course,
those samples with no prior word power would best work. Each
syllabus contained three to four words. In this regard, Thorn-
bury (2002) suggested that the vocabulary presentation should
not extend at most about a dozen items which is based on a
number of considerations such as the level of the learners, the
difficulty of the items, teachability of them and even whether
the items are being learnt for production or recognition only.
Since more time will be needed for the former, the number of
items is likely to be fewer than when the aim is only recogni-
tion (pp. 75-76).
Since nouns are more teachable, that is, they can be ex-
plained or demonstrated (Thornbury, 2002: p. 76), 18 of the 38
items chosen in the syllabi were nouns. These 38 items were of
seven categories as objects, foods, sports, animal, colors, ac-
tions, and adjectives (see Appendix 1). Each syllabus was de-
signed to help the patients practice and memorize the items
every session with a memory game in the end.
The words in the syllabi were presented through five tasks:
matching words to the pictures, word association, filling the
gaps, crossword puzzle, and finding words.(see Appendix 2)
Two sets of cards were selected: let’s go flash cards by Tavak-
koli (2008) and the other opposite’s flashcards by Shirkhani
Since schizophrenics are not generally found in every hospi-
tal, and are only hospitalized in mental hospitals or centers
where they are taken care of, two of these centers out of three
located in Sanandaj were selected. Qods Mental Hospital lo-
cated in a remote part of the city was the first to visit and, with
the permission of the manager, it was possible to reach the ar-
chive, library, as well as the patients and a room in which the
classes could be held two sessions a week. Each session lasted
around 40 minutes in an eight week period. The above coordi-
nation was made in the other center which was a charity one
responsible for the chronic schizophrenics under the supervi-
sion of welfare organization. Based on the list provided, an oral
interview was designed and administered in order to select the
samples that were both schizophrenics and had very limited
number of English words.
The first session was spent on practicing the English alphabet
making sure that all participants were familiar with English
sounds and letters. Two methods were used to present the
meaning of the vocabulary presented. Three new words were
presented each session first through showing pictures on flash
cards as they were read aloud in the class. As Thornbury (2002)
points out, this method is meaning-first since the sequence of
presenting might be effective. There is an argument that pre-
senting the meaning first creates a need for the form, opening
the appropriate mental files and making the presentation both
more efficient and more memorable. On the other hand,
form-first presentation works best when the words are pre-
sented in some kind of context, so that the learners can work
out the meaning for themselves. Second, Since translation has
been the most widely used means of presenting the meaning of
a word in monolingual classes it was provided then because it
has the advantage of being the most direct way to a word’s
meaning assuming that there is a close match between the target
word and its L1 equivalent (Thornbury, 2002: pp. 76-77). After
the students did all the tasks, a memory game was set out. Four
pictures were shown to a patient letting him watch for a while,
and then one picture was removed asking the individual par-
ticipant to name the removed picture. The final task was to ask
them to draw the learnt words on the board. These two final
tasks, however; is not included for the evaluation of the patients
and were administered for the sake of practicing the taught
The class lasted for 16 sessions and the first test was admin-
istered two weeks later to find out how many of the vocabulary
items were retained. The retest, exactly the same test as the first,
was administered 16 days later. Twenty words were randomly
chosen from the syllabi and one picture was shown at a time to
the patient to elicit the English word. Since the aim was to find
out if they could remember or even recognize the words, a list
was shown in case they could not remember.
The result is based on data provided from three assessment
devices: tasks which have been done during the sessions, test,
and retest. Papers given from the sessions told us about pa-
tients’ capability in fulfilling different kinds of tasks such as
matching words to the pictures, word association, filling the
gaps, crossword puzzle, and finding words out of which the
final two require more mental ability on the part of the learners.
The effect of age on patients’ achievement is determined by
putting them into two groups; first, patients being more than 30
years and second, those with below 30 years of age. On two
different diagrams, the groups’ average percentages of task
done is shown and compared.
The data on matching words to the pictures shows a fluctua-
tion with the undifferentiated type patient doing all the tasks
correctly and two disorganized types scoring the worst with
lack of ability in doing any of the tasks in the syllabi. More,
only three out of eight patients showed a figure below average
whereas most of them could reach out to above average point of
53 percent (Figure 1).
The Figure 2 on word association task shows a marked de-
crease in the patients’ attendant to the tasks with the two disor-
ganized type with no correct answers and undifferentiated
scoring well. The average in this task is 42 percent meaning
that they have done 11 percent less compared with that of
matching words to picture.
Filling the gap tasks which required the patients to provide
written answers to the blank spaces based on the pictures show
almost similar results to the Figure 3 which was on word asso-
ciation. One paranoid type patient seemed to show poorer abil-
ity compared to the previous tasks and its performance is going
downward while others are relatively doing well.
One cognitive demanding task is crossword puzzle in which
patients are required to fill in the blank boxes based on small
clues available. The diagram below shows the percentage of
correctly responded items in the crossword puzzle tasks by the
Figure 1.
Percentage of tasks by the patie nt s on m at c hi n g w o r ds to pictures.
Figure 2.
Percentages of tasks by the patients on word associ ation.
Figure 3.
Percentages of tasks by the p atients on filling the gaps.
Figure 4.
Percentages of tasks by the pati en t s o n cr o ss w o rd puzzle.
patients. As can be seen from Figure 4, one paranoid ty pe hit the
trough doing exactly no single task and another paranoid pa-
tient had a significant decrease in its performance. The two
disorganized types were not still able to carry out all the ques-
Figure 5 indicates the percentage of the patients’ ability in
doing the final vocabulary task called finding words. The graph
shows a drop in the performance of patients in doing the task.
Compared with the other tasks, this graph has almost four pa-
tients hitting the trough and a marked decline in two of them.
While the average reached the bottom, just two patients could
overtake the average and six stayed lower. The undifferentiated
type still has the highest percentage with 100 percent perform-
The Figure 6 compares the patients’ performance on the tests.
The first test was administered a week after the class and the
retest was administered 16 days after the first test. Two para-
noid patients scored lower on the retest while two performed
better in it and two did the same. The two disorganized type
could not answer any of the questions asked while the undif-
ferentiated patient correctly answered all the questions.
The Figure 7 shows the percentage of the tasks performed by
patients in an age categorization. The highest percentage goes
to a 44 year old undifferentiated patient and the lowest per-
centage belongs to the two disorganized patients with 40 and 44
years old.
The above figures can be categorized into two groups; first,
patients being more than 30 years and those below 30 years of
age. The Figure 8 shows that there is a very slight difference
between the two groups of patients in their average tasks done.
Figure 5.
Percentages of tasks by the patients on fin di ng w ords.
Figure 6.
Percentage of patientsperf o r m a nc e o n t e s t a n d r e t e st .
Figure 7.
Average percentages of all the tasks done by the patients.
Figure 8.
Average percentages of tasks by two groups of patients; under 30 and
over 30.
The study yields significant results in schizophrenia’s foreign
language domain. Patients with schizophrenia showed different
impairments in their attempt to learn foreign language vocabu-
lary. These can be categorized into six abnormalities.
Breakdown in Doing the Tasks
Schizophrenics seem to have serious problem in thinking and
doing tasks which are cognitively demanding. They failed not
only in answering different vocabulary tasks but also in under-
standing the tasks. They seemed very far from understanding
them and their minds function as if there is just a simple logic
behind every task. As they started with matching word to the
pictures and drawing a line to match them, they simply did the
same for all other tasks. The following figures show some of
those tasks in which the patients showed abnormal perform-
ances. They were supposed to fill in the gaps and provide the
correct word but they simply drew a line and matched them to
the pictures unexpectedly.
Distorted thoughts resulted in a number of abnormal per-
formances which are clear in the following task done by
Figure 9.
Average percentage of tasks d o ne by the patients.
In addition, as the order of the tasks move from simple to
difficult and more cognitively demanding, the patients showed
poorer results. Figure 9 on the average percentage of tasks done
by schizophrenic shows the degree of tasks difficulty. As it is
clear from the figure, they found finding words task most diffi-
cult and it is the task in which the patients showed the least
performance with doing almost a fifth of it. Word association
also challenged their ability. The task with the highest percent-
age was matching words to the pictures and they did well rela-
Novel Made-up Words
Another abnormality observed was the creation of some non-
sense words and writing them in the blanks as for the answers.
This abnormality, known as neologisms, is clearly known and
exists among the Anderson’s standard account of schizo-
phrenic’s first language (1979 as cited in Covington et al.,
2005). These made-up words and letters are so illegible and
sometimes have no similarity with English letters. The following
picture illustrates some of these abnormal language utterances.
Schizophrenics tend to switch back to native language in the
middle of the tasks while they were fully unaware of this. One
of the patients diagnosed with disorganized type who suffers
from speech disorders, when asked to read the English alphabet
from the board, preceded as follows:
/gi/ /pi/ /ye/ /gi/ /es/ /ge/ /dal/ /zal/
He started with /g/ a letter from the middle of the alphabet
list, saying some random letters; he switched to Persian alpha-
bet all of a sudden and stopped soon after.
Failure to Utter Lexical Items
Schizophrenic patients, particularly those with language dis-
order who failed to learn even a single English word, had seri-
ous problems in naming objects and, as Chaika (1974 as cited in
Covington et al.) observed, they fail to utter an intended word
and their thought goes around word associations. These patien t s
had difficulty in even repeating English words and tended to
utter associative words in native language. When I asked an-
other patient to say the sessions taught vocabulary, he said the
following native words with confidence: edaray kargozini
(personnel department), emarat (mansion), sakhteman (build-
ing), bimarestan (hospital), hotel (hotel), rasturansh (restau-
rant), kabara (bar). These irrelevant replies, named tangibility,
are among the standard account of schizophrenic’s language of
Anderson (1979 as cited in Covington et al., 2005: p. 88).
When I showed him the following picture and asked him
‘’what are they doing’’(he was supposed to say ‘watching tele-
vision’), he uttered these words; madrese (school), hayat
(yard), baziye bacheha (children game).
In the following pictures, notice the gap for the picture sand-
wich in which one of the patients devises a blend word very
similar to both sandwich and dog, while he could not simply
name the dog.
Another case is the repetition of words ‘tree’ and ‘door’ by
one of the patient. When I asked him to pronounce these words,
he followed as ‘’ tere’’, ‘’’’, ‘’ter’’, ‘’der’’, ‘’dar’’ and he
found it very difficult to pronounce the consonant cluster in the
word ‘tree’ and finally he could not.
This kind of abnormality also exists in schizophrenics’ L1
(first language) production and has been observed by Chaika
known as distraction by the sound or senses of words, so that a
discourse becomes a string of word association rather than a
presentation of previously intended information. (1974 as cite
in Covington et al.)
The last exercise in the following graph indicates how the
girl’s clothing in the picture distracts the patient and led him to
a wrong answer [Zendani (prisoner) instead of reading a book].
In the second figure the word food reminds him of the word
After teaching the word ‘chair’, I asked one of them what
‘chair’ means and he answered ‘’yani chikar mikoni’’(what are
you doing?), the sound /ch/ distracted him and reminded him of
the expression in Persian beginning with the same sound. The
following conversation happened between me and one of the
Me What is this in English (in native language)?
[Pointing to the chair]
The patient Sanandaj
Me What is it again?
The patient Sandali (chair)
Me in English please!
The patient Ja (seat)
The study has shed light on second language learning from a
very different point of view and has provided some implica-
tions regarding the similarities of first and second language
learning from a psycholinguistic perspective.
Since schizophrenics already speak a language and due to
mental impairments they suffer from language disorder, conse-
quently they show the same difficulty in an attempt to learn a
foreign language. Due to mental problems, all schizophrenics
show a kind of language disorder in their first language and this
is more complicated when it comes to those who are diagnosed
with language disorders. As the study figured out, those with
such a diagnosis could not perceive, understand, or produce a
single vocabulary item of a foreign language. They tend to
reveal the same impairments on their way toward learning a
foreign language. Other types of schizophrenics who attempted
to learn English vocabulary items also had difficulty in doing
the activities; the more cognitively demanding the task, the
harder they found it to do.
The results on patients’ vocabulary retention also reveal that
schizophrenics are able to recall English vocabulary items.
They either are quick to produce the English word as soon as
the picture is shown or they soon recognize the word in a list
and point it if they are not able to pronounce it. The study also
magnifies that age has no significant effect on schizophrenics to
learn English vocabulary items. Both groups of patients,
younger as well as older than 30, showed the same ability and
had very similar results in doing all the vocabulary tasks.
Aitchison, J. (2008). The articulate mammal: An introduction to psy-
cholinguistics (5th ed.). Canada: Routledge.
Cardwell, M. (2003). Complete A-Z psychology: Handbook (3rd ed.).
London: Bath Press Ltd.
Celce-Murcia, M. (2001). Teaching English as a second or foreign
language (3rd ed.). Boston: Heinle & Heinle.
Coon, D. (2001). Introduction to psychology: Gateways to mind and
behavior (9th ed.). Belmont: Wadswor t h.
Covington, M. A., He, C., Brown, C., Naci, L., McClain, J. T., &
Fjordbak, B. S. et al. (2005). Schizophrenia and the Structure of Lan-
guage: The Linguist’s View. URL (last checked 7 December 2008)
Davison, G. C., & Neale J. M. (2001). Abnormal psychology (8th ed.).
New York, NY: John Wiley & Sons, Inc.
DeLisi, L. E. (2004). Anomalous Language Perception and Speech of
Schizophrenia. URL (last checked 21 June 200 9) evolang/2004/ABSTRACTS/.../delisi.rtf
Harvey, D. P., Wilcox, J. A., & Boudjenah, D. (2005). Schizophrenia in
late life: Aging effects on symptoms and course of illness. The Ame-
rican Journal of Psychiatry, 162, 2204.
Graham, G. M., Gath, D., & Mayou, R. (1999). Concise oxford text-
book of psychiatry (2nd ed.). Oxford: Oxford University Press.
Hockenbury, D. H., & Hockenbury, S. E. (2004). Discovering psy-
chology (3rd ed.). New York, NY: Worth Publisher.
Lebedinskij, M. S. (1959). On Certain Features of Thought and Lan-
guage of Schizophrenic s. URL (last checked 6 August 2008)
Myers, D. G. (1996). Exploring psychology (3rd ed.). New York, NY:
Worth Publisher.
Sadock, B. J., & Sadock, V. A. (2005). Kaplan & Sadock’s comprehen-
sive textbook of psychiatry (8th ed.). Philadelphia: Loppincott Wil-
liams & Wilkins.
Schmitt, N. (2002 ). Applied lin g ui st ics. London: Arnold.
Scrivener, J. (1994). Learning teaching. Oxford: Macmillan Heinemann
English Language Teaching.
Shirkhani, E. (2008). Opposites flashcards. Esfahan: Jungle Publica-
Tavakoli, S. (2008). Let’s go flashcards. Tehran: Koshamehr.
Thornbury, S. (2002). How to teach vocabulary. Essex: Pearson Educa-
tion Limited.
Appendix 1
Different Word Categories in the Syllabi
Adjectives Actions ColorsAnimalsSports Foods Objects
/slow Writing Red elephantSwimmingPasta Comb
Strong/weak Watching GreenDog BasketballSandwichKey
Hot/cold Driving Blue Mouse Skating Cake Clock
Young/old Playing BrownCow Chess Hat
Big/small Listening Duck Boat
Reading Chicken
10 6 4 6 4 3 5
Appendix 2
Sample Tasks