Aspects of cognitive functioning in schizotypy and schizophrenia: Evidence for a

continuum model
Mary Cochrane
a,
, Ian Petch
b
, Alan D. Pickering
c, 1
a
Division of Psychological Medicine and Psychiatry, Institute of Psychiatry, 4 Windsor Walk, London SE5 8BB, United Kingdom
b
Early Intervention Team, Springeld University Hospital, 61 Glenburnie Road, London SW17 7DJ, United Kingdom
c
Department of Psychology, Goldsmiths College, University of London, SE14 6NW, United Kingdom
a b s t r a c t a r t i c l e i n f o
Article history:
Received 27 August 2011
Received in revised form 7 February 2012
Accepted 8 February 2012
Keywords:
Schizotypy
Schizophrenia
Cognitive functioning
Continuum approach
This research consisted of two studies, the fundamental aim of which was to delineate the pattern of relationships
between measures of cognitive task performance and both symptom subtypes in schizophrenia and their
corresponding schizotypal personality traits in healthy individuals. Study 1 compared these relationships in healthy
individuals using the Schizotypal Personality Questionnaire (SPQ) and Study 2 assessed the relationships between
symptomatology assessed using the Scale for the Assessment of Positive Symptoms and Scale for the Assessment of
Negative Symptoms (SAPS/SANS) and cognitive task performance in a group of patients with schizophrenia. The
contribution of uid intelligence to task performance was also examined. In Study 1 high levels of negative schizo-
typy were associated with reduced verbal uency, and high levels of disorganised schizotypy were associated with
reduced negative priming in the healthy participants. In Study 2, closely corresponding relationships between
symptom measures and these tasks were found in the patients with schizophrenia. The associations between
the symptom and cognitive measures were independent of the effects of uid IQ on performance.
2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
A growing body of research suggests that symptoms of schizo-
phrenia and corresponding elevated schizotypal personality scores
in non-clinical samples are associated with similar disruptions in cog-
nitive functioning. This accords with the fully dimensional approach
to schizophrenia, which proposes that the dimension of schizotypy
lies on a continuum that begins with normality and proceeds towards
the schizophrenia spectrum disorders, with schizophrenia at the
upper end (Claridge and Beech, 1995).
Verbal uency (VF) and negative priming (NP) are two cognitive phe-
notypes which are consistently found to be impaired in patients with
schizophrenia and in healthy individuals scoring high on measures of
schizotypal personality. However, relationships betweenaspects of symp-
tomatology and cognitive performance are not always consistent and, in
the case of negative priming, there is limited research assessing the way
in which differential aspects of symptomatology relate to performance.
Verbal uency is a commontest of response generation, in whichpar-
ticipants are asked to generate, in a one-minute interval, as many words
as possible that beginwitheachof three letters. Patients withschizophre-
nia typically generate fewer words thancontrols onVF tasks andnegative
symptoms are associated with these decits (Liddle and Morris, 1991;
Frith, 1992). Negative symptomratings are also correlated with impaired
VF in patients with schizotypal personality disorder (Diforio et al., 2000),
and negative schizotypy is positively associated with VF decits in the
healthy siblings of schizophrenic patients (Franke et al., 1993).
Reported relationships between schizotypal personality traits and VF
in healthy participants are somewhat inconsistent. Tsakanikos and
Claridge (2005) found that reduced VF was associated with increased
O-LIFE (Oxford-Liverpool Inventory of Feelings andExperiences) negative
schizotypy, but also found that greater VF was associated with high levels
of O-LIFE positive schizotypy. Relationships have been found between
reduced VF and high scores on both negative and positive schizotypy
(Barrantes-Vidal et al., 2003) and between reduced VF and high scores
on both negative and disorganized schizotypy, or positive schizotypy
(Ruiz et al., 2008). These latter studies employed a cluster-analytic rather
than a dimensional approach to schizotypal personality, which may (at
least partially) explain the differential ndings.
The NP paradigm is believed to assess selective attention and cog-
nitive inhibition (Tipper, 1985). Reduced negative priming has been
found in patients with schizophrenia (Beech et al., 1989; Park et al.,
1996; Williams, 1996; Peters et al., 2000), although some authors
have not observed NP decits in patients (Laplante et al., 1992;
Baving et al., 2001). A possible explanation for this discrepancy is pro-
vided by Moritz et al. (2001). There is a striking paucity of research
assessing relationships between aspects of symptomatology and neg-
ative priming. Williams (1996) reported an association between both
positive and disorganised symptomatology and reduced NP in schizo-
phrenia, although this nding is yet to be replicated. Impaired NP in
Psychiatry Research 196 (2012) 230234
Corresponding author. Tel.: +44 207 848 0656.
E-mail addresses: mary.cochrane@iop.kcl.ac.uk (M. Cochrane),
ian.petch@swlstg-tr.nhs.uk (I. Petch), a.pickering@gold.ac.uk (A.D. Pickering).
1
Tel.: +20 7919 7334.
0165-1781/$ see front matter 2012 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.psychres.2012.02.010
Contents lists available at SciVerse ScienceDirect
Psychiatry Research
j our nal homepage: www. el sevi er . com/ l ocat e/ psychr es
healthy participants appears to be primarily associated with positive
schizotypy (Peters et al., 1994; Park et al., 1996).
The current research consisted of two studies, which attempted to
ascertain whether relationships between aspects of symptomatology/
schizotypal personality and performance on two popular cognitive
tasks would be consistent across a group of healthy participants and
a group of patients with schizophrenia. A consistent pattern of relation-
ships would provide convincing evidence for the continuum approach
to schizophrenia and uniquely demonstrate that both clinical symp-
toms and non-clinical characteristics impact in a similar way on cogni-
tive functioning. A measure of uid IQ was included to investigate
whether any relationships between symptomatology and cognition
were maintained despite differences in a general cognitive ability.
In the healthy participants in Study 1, it was expected that higher
levels of SPQ(Schizotypal Personality Questionnaire) negative schizo-
typy would be associated with reduced verbal uency, and that higher
levels of SPQ positive or disorganised schizotypy would be associated
with reduced negative priming. In the patient group we expected the
same pattern of relationships between symptomatology and cognitive
task performance, using the Scale for the Assessment of Positive
Symptoms/Scale for the Assessment of Negative Symptoms (SAPS/
SANS) inventory (Andreasen, 1983, 1984). Finally, it was assumed
that these effects would be independent of the expected positive rela-
tionships between uid IQ and the cognitive measures.
2. Method
2.1. Ethical approval
Ethical approval for Studies 1 and 2 was obtained from the Psychology Department
Ethics Committee at Goldsmiths College. For Study 2, ethical approval was also
obtained from the South West London NHS Local Research Ethics Committee.
2.2. Participants
Study 1 consisted of a convenience sample of 100 participants (50 females and 50
males), recruited from the student population at Goldsmiths College, University of
London. Participants were recruited by means of advertisements posted around the
college and received either course credits (psychology students) or a small remunera-
tion for their participation. Age range was between 18 and 45 years (M=25.00,
S.D. =6.09). Seventy one percent of 99 participants were native English speakers (lan-
guage information was not available for one participant), although all participants
were uent in English. The mean number of years spent in education was
16.17 years (S.D. =2.85). Exclusion criteria, determined through self-report, were pre-
vious or current psychological or neurological disturbance, a serious physical medical
condition, or currently taking psychoactive medication.
Twenty patients with schizophrenia took part in Study 2: 10 were inpatients and 10
were living in the community. There were 18 males and 2 females. All patients had an
ICD-10 diagnosis of schizophrenia, were taking atypical neuroleptic medication, and
were uent in English. Age range was between 18 and 50 years (M=34.24,
S.D. =8.42). Patients with a diagnosed substance abuse disorder, concomitant neurolog-
ical disorder or serious physical medical condition were excluded from the study. The in-
patients were recruited from three wards at Springeld University Hospital, South West
London, and were tested in a quiet consulting room on the ward. The community pa-
tients were recruited from three Community Mental Health Teams in the South West
London area and were tested at an outpatient clinic that they regularly attended.
2.3. Schizotypal Personality Questionnaire (SPQ, Raine, 1991)Study 1
The SPQ is a forced choice (Yes/No) self-report inventory which contains 74 items
and divides into nine subscales based on the nine criteria for DSM-III-R schizotypal per-
sonality disorder (American Psychiatric Association, 1987). These subscales are Ideas of
Reference (IoR), Unusual Perceptual Experiences (UPE), Odd Beliefs or Magical Thinking
(OBMT), Paranoid Ideation (PI), Excessive Social Anxiety (ESA), No Close Friends (NCF),
Constricted Affect (CA), Odd or Eccentric Behaviour (OEB), and Odd Speech (OS). The
SPQ also sub-divides into three factors: CognitivePerceptual, Interpersonal and Disorga-
nised; which are regarded as non-clinical analogues of the positive, negative and disorga-
nised symptoms of schizophrenia respectively (Raine et al., 1991).
2.4. Clinical Interview Schedule (SAPS/SANS)Study 2
The SAPS/SANS consist of a total of 51 items, with each item scored between 0 (ab-
sence of a symptom) to 5 (severe/complex symptomatology). The items fall into nine
global subscales, and the scores on these subscales are summed to yield three factors;
positive, negative and disorganised. Two attentional items were excluded from the
factor score calculations as their scores do not load consistently on any of the three fac-
tors (Andreasen et al., 1995 and personal correspondence with Andreasen).
2.5. Measures of cognitive functioning and uid IQ
Verbal uency was assessed using the Controlled Oral Word Association Test
(COWAT, Benton and Hamsher, 1976), in which participants are asked to produce as
many words as possible that begin with each of the letters F, A and S (with one minute
for each letter); without using proper nouns, plural words, or the same word with dif-
ferent sufxes. The dependent variable was the total number of correct words generat-
ed in response to the three letters.
The computerised spatial negative priming task was based on the paradigm used by
Park et al. (1996). Displays contained a target (letter O) and a distractor (letter X), and
there were four possible locations on the screen in which the target or distractor could ap-
pear. Each trial consisted of a prime followed by a probe display. Participants were asked to
respond only to the letter O and to ignore the X. In control trials, the locations of the targets
and distractors in both prime and probe displays were different. In NP trials, the target in the
probe display appeared in the same position as the distractor in the prime display, and thus
participants were required to respond to the previously ignored location. The overlap be-
tween the target and distractor in the NP trials is believed to produce response retardation
since the active inhibition of ignored information during the prime display may decrease its
accessibility for later selection during the probe display (Tipper, 1985).
Participants focused on a central xation point and responded to the location of the
target O as quickly and as accurately as possible. Responses were made by pressing one of
four keys on a special response box attached to the computer. Trials were initiated by
pressing the spacebar and each trial began with the presentation of the central xation
point for 0.8 s. The prime display was then presented until participants located the target
and responded. After a further 0.5 s the central xation point appeared (again for 0.8 s)
followed by the probe display which remained on the screen until a response was
made. After participants responded to the probe display, a pattern mask appeared for
1 s and then the next trial began. There were four blocks of 18 pairs of experimental tri-
als, consisting of 9 NP and 9 control trials, which were presented in a xed randomorder.
A set of 10 practice trials preceded the experimental trials.
The main dependent measure was the NP reaction time score (NPrt), which was
calculated by subtracting the mean reaction time for control probe trials from the
mean reaction time for NP probe trials. Separate mean reaction time scores for the
rst and second blocks of the task (n=36 trials in each block) were also computed
to ascertain whether any decits in performance and their corresponding schizotypal
features were specic to the initial or the latter phase of the task.
The matrices from the Wechsler Adult Intelligence Scale III (Wechsler, 1997) were
used as a brief measure of uid intelligence. In this untimed test participants were
shown a series of 26 matrices, each of which had a missing cell. For each item partici-
pants were asked to choose the correct completion from 5 options, based on the rela-
tionships (across rows and columns, etc.) for the other cells in the matrix. An initial
raw score was calculated by summing the number of correct items completed by par-
ticipants (with a maximum score of 26) and the raw scores were then converted to
age-adjusted scaled scores as described in the manual.
2.6. Procedure
All participants were asked to read and sign a consent form and to provide the de-
mographic information: age, gender, number of years in education, and whether En-
glish was their native language (a binary variable coded 1=native English,
0=English not native language). In Study 1 the healthy participants were then
asked to complete the SPQ and cognitive and uid IQ tasks. One participant failed to
complete both of the cognitive tasks and two failed to complete the negative priming
task; therefore, these participants were excluded from the analyses.
In Study 2 the patients records were accessed to obtain information on diagnosis,
medication and general medical history. Patients completed the SAPS/SANS interview,
followed by the cognitive tasks and WAIS-III matrices. In each of the two studies half of
the participants performed the uid IQ test before the cognitive tests and the remain-
ing participants performed the IQ test afterwards. Systematic order effects were mini-
mised by also counterbalancing the order of presentation of the verbal uency and NP
tasks. On completion of the tasks all participants were debriefed, provided with a small
remuneration, and thanked for their participation.
3. Results
3.1. Study 1healthy participants
3.1.1. Data distributions and outlier removal
Scores on the SPQ and verbal uency measures were normally dis-
tributed. One outlier score was found in the overall NP index and one
in the Block 1 NP index, and two outliers were present in the Block 2
NP index. These scores were removed in order to normalise the data
distributions.
231 M. Cochrane et al. / Psychiatry Research 196 (2012) 230234
3.1.2. Descriptive statistics
Descriptives for the SPQ, cognitive and uid IQ measures are pre-
sented in Table 1. The SPQ positive factor was composed of the Unusual
Perceptual Experiences, Ideas of Reference, OddBeliefs/Magical Thinking
and Paranoid Ideation scores, and the disorganised factor comprised the
summation of the Odd Speech and Odd/Eccentric Behaviour scores. The
negative factor was composed of the Excessive Social Anxiety, No Close
Friends and Constricted Affect scores. The internal consistency of the
SPQsubscales was assessed by coefcient alpha and the values accorded
closely with Raine's (1991) coefcients, which ranged from0.71 to 0.78.
The mean SPQ total score of 25.0 (S.D. =13.4) was consistent with the
mean SPQ total scores of two groups of healthy participants reported
by Raine et al. (1991): 26.9 (S.D. =11.0) and 26.3 (S.D. =11.4) in
Samples 1 and 2 respectively.
3.1.3. Correlational analyses
Bonferroni adjustments were made for multiple correlations be-
tween the personality measures, whereas the exploratory correla-
tions between uid IQ and the schizotypal personality measures,
and between the demographic and cognitive measures, were tested
at an unadjusted alpha of P0.05. Correlations between uid IQ and
the cognitive measures were tested one-tailed, as we expected that
higher levels of uid IQ would be associated with enhanced perfor-
mance (higher VF, increased NP) on the cognitive tasks.
3.1.3.1. Intercorrelations among the SPQ factors. The intercorrelations of
the main personality variables were assessed using 3 Pearson's
correlations and were therefore tested at an adjusted alpha of
P0.017 (~0.05/3). There was a signicant correlation between the
CognitivePerceptual factor and the Interpersonal factor (r=0.53,
Pb0.001) and between the CognitivePerceptual and Disorganised
factors (r=0.59, Pb0.001). There was also a signicant correlation
between the Interpersonal and Disorganised factors (r=0.35, Pb0.001).
3.1.3.2. Fluid IQ and schizotypy. There were no signicant relation-
ships between uid IQ and the SPQ positive (r =0.11), negative
(r =0.10) and disorganised (r =0.08) factors (P>0.05).
3.1.3.3. Fluid IQ and cognitive task performance. Fluid IQ was signi-
cantly positively associated with verbal uency (r =0.42, Pb0.001)
and with performance in Block 1 of the NP task (r =0.18, P=0.04,
1-tailed). However, there was no signicant correlation between
uid IQ and the overall NP index (r =0.06), or the Block 1 NP
index (r=0.03) (P>0.05).
3.1.3.4. Demographic variables and cognitive task performance. There
were no signicant relationships between the demographic and
cognitive measures (P>0.05).
3.2. Multiple regression analyses
Four standard multiple regressions were conducted to assess the
independent contribution of the personality measures to the predic-
tion of each cognitive (dependent) measure. Fluid IQ was added to
the regressions assessing VF and Block 1 NP, as was the language var-
iable, which was signicantly positively correlated with VF (r =0.37,
Pb0.001) and Block 1 NP (r=0.21, Pb0.05). Negative and disorga-
nised schizotypal features as predictors of VF and NP respectively
were tested one-tailed, given the direction of the effects reported in
previous research. The independent contribution of uid IQ to the
prediction of cognitive task performance was also tested one-tailed.
Positive schizotypal features were tested as predictors two-tailed
given the occasional report of an effect in the opposite direction to
the majority of effects (Tsakanikos and Claridge, 2005). The results
of the regression analyses are shown in Table 2.
3.2.1. Verbal uency
The SPQ factors, uid IQ and language were used as predictors,
with the number of words generated in the VF task as the dependent
variable. There was a near-signicant trend for a negative effect of
SPQ negative schizotypy (=0.17, Pb0.06, one-tailed), over and
above the robustly signicant independent contributions of uid IQ
(=0.33, Pb0.001) and English as rst language (=0.34, Pb0.001).
3.2.2. Negative priming (NP)
Regression analysis with the SPQ factors as predictors of the over-
all NP reaction time index revealed that higher levels of disorganised
Table 1
Descriptives for the personality and cognitive measuresStudy 1.
Mean (S.D.) Min Max N Alpha
SPQ factors
Positive (cognitiveperceptual) 11.14 (7.15) 0.00 30.00 99 0.89
Negative (interpersonal) 6.98 (5.08) 0.00 22.00 100 0.86
Disorganisation 7.00 (4.07) 0.00 14.00 99 0.85
SPQ subscales
Ideas of reference 3.94 (2.60) 0.00 9.00 100 0.77
Odd beliefs/magical thinking 2.26 (2.18) 0.00 7.00 100 0.81
Unusual perceptual experiences 2.49 (2.36) 0.00 9.00 99 0.77
Paranoid ideation 2.55 (2.26) 0.00 8.00 100 0.78
Excessive social anxiety 3.00 (2.29) 0.00 8.00 100 0.77
No close friends 2.06 (2.03) 0.00 9.00 100 0.72
Constricted affect 1.92 (1.65) 0.00 6.00 100 0.61
Odd or eccentric behavior 2.60 (2.19) 0.00 7.00 99 0.82
Odd speech 4.141 (2.49) 0.00 8.00 100 0.76
Cognitive measures
Verbal uency 40.78 (11.29) 13.00 73.00 99
Negative priming (NP) 0.02 (0.05) 0.27 0.08 98
Block 1 NP 0.00 (0.05) 0.19 0.13 97
Block 2 NP 0.01 (0.04) 0.12 0.11 96
Fluid IQ 14.10 (2.15) 9.00 19.00 99
Table 2
Predictors of executive functioningStudy 1.
Dependent measures
Verbal uency Negative priming Block 1 negative priming Block 2 negative priming
Predictors B SEB B SEB B SEB B SEB
Fluid IQ 1.752 0.473 0.334*** 0.004 0.002 0.173*
Language 8.405 2.238 0.341*** 0.020 0.011 0.182
SPQ pos 0.069 0.197 0.043 0.001 0.001 0.247 0.001 0.001 0.162 0.002 0.001 0.251
SPQ neg 0.373 0.233 0.167 0.001 0.001 0.107 0.000 0.001 0.057 0.001 0.001 0.069
SPQ disorg 0.021 0.308 0.008 0.002 0.001 0.244* 0.000 0.002 0.031 0.005 0.001 0.428***
R
2
=0.296, adj R
2
=0.258
(F (5, 92)=7.728, Pb0.001)
R
2
=0.050, adj R
2
=0.020
(F (3, 93)=1.647, ns)
R
2
=0.084, adj R
2
=0.033
(F (5, 90)=1.658, ns)
R
2
=0.121, adj R
2
=0.093
(F (3, 92)=4.236, Pb0.01)
Note: SPQ Pos = SPQ CognitivePerceptual (positive) factor; SPQ Neg = SPQ Interpersonal (negative) factor; SPQ Disorg = SPQ Disorganisation factor.
* P0.05, *** P0.001, trend effect (Pb0.06).
232 M. Cochrane et al. / Psychiatry Research 196 (2012) 230234
schizotypy signicantly predicted reduced negative priming (=
0.24, Pb0.03, 1-tailed). There was weak trend (=0.25, Pb0.08,
2-tailed) for a positive association between negative priming per-
formance and positive schizotypy, while the negative schizotypal
factor did not signicantly contribute to the variance in NP.
When the SPQ factors, uid IQ and language were used as predic-
tors of the NP reaction time index in the rst block of the task, uid IQ
was the only signicant predictor of (greater) NP (=0.17, P=0.05,
1-tailed). When the SPQ variables were used as predictors of NP per-
formance in the second block of the task, positive schizotypy again
tended to predict greater negative priming (=0.25, Pb0.07, 2-
tailed); however, disorganised schizotypy independently and signi-
cantly predicted reduced negative priming (=0.43, Pb0.001, 1-
tailed).
3.3. Study 2patients with schizophrenia
3.3.1. Descriptive statistics
Scores on the symptomand cognitive task measures were normal-
ly distributed. The descriptives are shown in Table 3.
3.3.2. Correlational analyses
The analyses were conducted in accordance with Study 1. Pear-
son's correlations used to assess the relationships between the symp-
tom and demographic measures were tested two-tailed and were
non-signicant (P>0.05). The correlations between uid IQ and the
cognitive measures were tested one-tailed. There were no signicant
correlations between the uid IQ measure and the negative priming
indices; however, there was a trend for a positive relationship be-
tween uid IQ and VF (r=0.35, P=0.06).
The relationships between the SAPS/SANS scores and cognitive
task performance were tested one-tailed for negative and disorga-
nised schizotypy and two-tailed for positive schizotypy. The negative
factor scores were signicantly negatively associated with VF (r=0.43,
P=0.03, 1-tailed) and the disorganised factor scores were signicantly
negatively associated with NP (r=0.53, Pb0.01, 1-tailed). The signi-
cant association between disorganised symptomatology and negative
priming was specic to performance in the latter phase of the task
(r =0.51, P=0.01, 1-tailed) as opposed to the initial phase (r =
0.27, P=0.13, 1-tailed); the same pattern that was found in the
healthy participants with SPQ disorganised schizotypy in Study 1.
There were no signicant correlations between the positive factor
scores and the cognitive measures.
3.3.3. Multiple regression analysis
Given the trend for a positive association between VF and uid IQ,
and the robust and signicant positive correlation between these var-
iables in the larger sample of healthy participants, VF was reanalysed
in a regression using both SANS negative factor score and uid IQ as
predictors. This reduced the independent prediction of VF by SANS
negative factor score to a near-signicant trend (=0.36, Pb0.06,
1-tailed), while the independent prediction of VF by uid IQ was
non-signicant (=0.27, P=0.12, 1-tailed). (R
2
=0.25, Adj. R
2
=
0.16, F(2, 17)=2.80, P>0.05).
4. Discussion
In accordance with expectations, high levels of negative schizo-
typy predicted reduced verbal uency and high levels of SPQ disorga-
nised schizotypy predicted reduced negative priming in the healthy
participants in Study 1. This latter effect was particularly evident in
the second block of the NP task. It seems that for participants high
in disorganisation, inhibitory processing was initially intact, but a dif-
culty in sustaining this ability was evident as the task progressed.
Higher levels of SPQ positive schizotypy tended to predict enhanced
inhibition in the latter part of the NP task; a nding which is at
odds with other published NP results showing a negative relationship
(Peters et al., 1994; Park et al., 1996) and therefore may require fur-
ther investigation to determine if it is replicable. This direction of ef-
fect has been previously reported for the inuence of positive
schizotypy on VF (Tsakanikos and Claridge, 2005); however, no
such effect was observed in the present study.
Study 2 examined the pattern of relationships between clinical
symptomatology and cognitive task performance in patients with
schizophrenia and revealed the same pattern of relationships. Higher
levels of disorganised symptoms were signicantly associated with
reduced negative priming, and this relationship was specic to the
second phase of the task. Higher levels of negative symptoms were
signicantly correlated with reduced verbal uency; however, the
prediction of VF by negative symptomatology was slightly reduced
in a regression controlling for uid IQ. Nevertheless, on both tasks,
continuity was observed across the groups of patients and healthy
participants.
The association between negative symptomatology and verbal u-
ency is consistent with previous research in patient groups (Liddle
and Morris, 1991; Frith, 1992) and in schizotypal participants
(Tsakanikos and Claridge, 2005). The relationship between SAPS dis-
organisation and reduced NP in patients with schizophrenia corre-
sponds with that reported by Williams (1996). However, the
association between disorganisation and reduced NP in the schizoty-
pal individuals does not accord with other reported relationships
which have emphasised the link between positive schizotypy and re-
duced NP. This discrepancy cannot be fully explained by the NP tasks
used in the different studies: Peters et al. (1994) used a Stroop-based
task but Park et al. (1996) used a spatial NP task, on which the current
task was closely based. The inconsistency across studies could arise
from the use of different schizotypy inventories: while Park et al.
and Peters et al. used the Perceptual Aberration Scale (PAS,
Chapman et al., 1978) and Combined Schizotypal Traits Questionnaire
(CSTQ, Claridge et al., 1996) respectively, the current schizotypy/NP
relationship was found using the SPQ. Interestingly, the relationship
between disorganised schizophrenia and NP found in this study and
by Williams (1996) was obtained using the same instrument (SAPS/
SANS). It may be difcult to clearly delineate the pattern of relation-
ships between symptomatology and cognition when methods of
assessing these phenomena are so varied.
The inclusion of the uid IQ measure in these studies was impor-
tant in highlighting the contribution of this process to cognitive task
performance, and in demonstrating that uid IQ did not fully account
for variations in performance. Fluid IQ may be regarded as a key com-
ponent of cognitive functioning (Duncan, 1995) and yet the ability of
schizophrenic symptomatology and schizotypal personality to predict
residual variance in verbal uency, after accounting for the contribu-
tion of uid IQ, was established. Fluid IQ made no signicant contri-
bution to the prediction of negative priming in the patients with
schizophrenia, but did contribute to the variance in Block 1 NP in
the healthy participants.
Table 3
Descriptives for the cognitive measures and uid IQStudy 2.
Mean (S.D.) Min Max N
Symptom measures (SAPS/SANS)
Positive factor 5.05 (2.91) 0.00 9.00 20
Negative factor 9.05 (3.58) 1.00 15.00 20
Disorganised factor 1.90 (1.86) 0.00 6.00 20
Cognitive measures
Verbal uency 34.80 (8.57) 17.00 46.00 20
Negative priming 0.01 (0.13) 0.47 0.26 20
Block 1 NP 0.03 (0.12) 0.11 0.47 20
Block 2 NP 0.04 (0.21) 0.91 0.76 20
Fluid IQ 10.85 (2.28) 6.00 16.00 20
233 M. Cochrane et al. / Psychiatry Research 196 (2012) 230234
Our ndings are interpreted with some caution since we did not
provide a formal assessment of some factors which may have impacted
on the relationships observed between symptomatology and cognitive
task performance. In Study 1 we did not have a formalised list of cri-
teria for establishing a serious physical medical condition, nor did
we perform any diagnostic assessment of psychopathology in the
healthy participants; rather, in both instances, the data were obtained
by means of self-report to general questions regarding these criteria.
In the patient group in Study 2 we did not include an assessment of
the class or dosage of atypical antipsychotic medication and therefore
could not statistically assess potential effects of medication. Further-
more, multiple regression analysis with SANS negative symptomatolo-
gy and uid IQ as predictors of verbal uency was conducted with low
power in such a small sample (n=20).
It could be argued that the lack of a control group is also a limita-
tion of Study 2. However, we do not regard this as a serious issue as
the research here is concerned with the relationships between symp-
toms and cognitive measures within a patient sample. The current re-
search questions are largely unaffected by whether the schizophrenic
group is signicantly impaired on a particular cognitive measure. If VF
(or NP) relate to a specic class of symptoms, then whether the par-
ticular schizophrenic sample, as a whole, is impaired with respect to
an appropriate matched control group will depend on the level of
the relevant symptom type: if it is high then the group will be im-
paired as a whole, if it is low, then the group might well not show
an impairment.
In conclusion, this is the rst report of a consistent pattern of
symptom-cognition relationships across a group of patients with
schizophrenia and a large group of healthy participants. For these
two specic relationships we thus provide convincing evidence for
the continuum model; demonstrating not only that schizophrenia-
like traits are present in the healthy population but also showing
that these traits may lead to disruptions in cognitive performance in
patients and schizotypal individuals in a similar manner. Future stud-
ies may be further informed by the inclusion of multiple measures of
cognition and additional culture-fair measures of general/uid
intelligence.
References
American Psychiatric Association, 1987. Diagnostic and Statistical Manual of Mental
Disorders, 3rd edition revised. APA Press, Washington.
Andreasen, N.C., 1983. The Scale for the Assessment of Negative Symptoms (SANS). The
University of Iowa, Iowa City.
Andreasen, N.C., 1984. The Scale for the Assessment of Positive Symptoms (SAPS). The
University of Iowa, Iowa City.
Andreasen, N.C., Arndt, S., Alliger, R., Miller, D., Flaum, M., 1995. Symptoms of schizophrenia:
methods, meanings and mechanisms. Archives of General Psychiatry 52, 341351.
Barrantes-Vidal, N., Fananas, L., Rosa, A., Caparros, B., Riba, M.D., Obiols, J.E., 2003.
Neurocognitive, behavioural and neurodevelopmental correlates of schizotypy
clusters in adolescents from the general population. Schizophrenia Research 61,
293302.
Baving, L., Wagner, M., Cohen, R., Rockstroh, B., 2001. Increased repetition and semantic
priming in schizophrenic patients. Journal of Abnormal Psychology 110, 6775.
Beech, A., Powell, T., McWilliam, J., Claridge, G., 1989. Evidence of reduced cognitive
inhibition in schizophrenia. British Journal of Clinical Psychology 28, 109116.
Benton, A.L., Hamsher, K.deS, 1976. Multilingual Aphasia Examination. University of
Iowa, Iowa City.
Chapman, L.J., Chapman, J.P., Raulin, M.L., 1978. Body-image aberration in schizophrenia.
Journal of Abnormal Psychology 86, 399407.
Claridge, G., Beech, T., 1995. Fully and quasi-dimensional constructions of schizotypy.
In: Raine, A., Lencz, T., Mednick, S.A. (Eds.), Schizotypal Personality. Cambridge
University Press, Cambridge, pp. 192216.
Claridge, G., McCreery, C., Mason, O., Bentall, R., Boyle, G., Slade, P., Popplewell, D.,
1996. The factor structure of schizotypal traits: a large replication study. British
Journal of Clinical Psychology 35, 103115.
Diforio, D., Walker, E.F., Kestler, L.P., 2000. Executive functions in adolescents with
schizotypal personality disorder. Schizophrenia Research 42, 125134.
Duncan, J., 1995. Attention, intelligence, and the frontal lobes. In: Gazzaniga, M.S. (Ed.),
The Cognitive Neurosciences. MIT Press, Cambridge, MA, pp. 721733.
Franke, P., Maier, W., Hardt, J., Hain, C., 1993. Cognitive functioning and anhedonia in
subjects at risk for schizophrenia. Schizophrenia Research 10, 7784.
Frith, C.D., 1992. The Cognitive Neuropsychology of Schizophrenia. Psychology Press,
East Sussex.
Laplante, L., Everett, J., Thomas, J., 1992. Inhibition through negative priming with
Stroop stimuli in schizophrenia. British Journal of Clinical Psychology 31, 307326.
Liddle, P.F., Morris, D.A., 1991. Schizophrenic syndromes and frontal lobe performance.
The British Journal of Psychiatry 158, 340345.
Moritz, S., Ruff, C., Wilke, U., Andresen, B., Krausz, M., Naber, D., 2001. Negative priming
in schizophrenia: effects of masking and prime presentation time. Schizophrenia
Research 48, 291299.
Park, S., Lenzenweger, M.F., Puschel, J., Holzman, P.S., 1996. Attentional inhibition in
schizophrenia and schizotypy: a spatial negative priming study. Cognitive
Neuropsychiatry 1 (2), 125149.
Peters, E.R., Pickering, A.D., Hemsley, D.R., 1994. Cognitive inhibition and positive
symptomatology in schizotypy. British Journal of Clinical Psychology 33, 3348.
Peters, E., Pickering, A., Kent, A., Glasper, A., Irani, M., Day, S., David, A., Hemsley, D.,
2000. The relationship between cognitive inhibition and psychotic symptoms.
Journal of Abnormal Psychology 109, 386395.
Raine, A., 1991. The SPQ: a scale for the assessment of schizotypal personality based on
DSM-III-R criteria. Schizophrenia Bulletin 17, 555564.
Raine, A., Reynolds, C., Lencz, T., Scerbo, A., Triphon, N., Kim, D., 1991. Cognitiveperceptual,
interpersonal, and disorganised features of schizotypal personality. Schizophrenia
Bulletin 20 (1), 191201.
Ruiz, M.C.A., Barrantes-Vidal, N., Guitart, M., Fananas, L., 2008. Study of neurocognitive
correlates of schizotypy personality clusters in healthy individuals. European Journal
of Psychiatry 22 (1), 1728.
Tipper, S., 1985. The negative priming effect: inhibitory priming by ignored objects.
Quarterly Journal of Experimental Psychology 37A, 571590.
Tsakanikos, E., Claridge, G., 2005. Less words, more words: verbal uency as a function
of positive and negative schizotypy. Personality and Individual Differences 39,
705713.
Wechsler, D., 1997. Wechsler Adult Intelligence Scale III. Manual for Administration
and Scoring. David Wechsler Corporation. Harcourt Brace & Co, San Antonio.
Williams, L.M., 1996. Cognitive inhibition and schizophrenic symptom subgroups.
Schizophrenia Bulletin 22, 139151.
234 M. Cochrane et al. / Psychiatry Research 196 (2012) 230234