King’s Research Portal
DOI:
10.1017/S0033291705005738
Document Version
Publisher's PDF, also known as Version of record
Link to publication record in King's Research Portal
Citation for published version (APA):
Maccabe, J. H., Simon, H., Zanelli, J. W., Walwyn, R., McDonald, C. D., & Murray, R. M. (2005). Saccadic
distractibility is elevated in schizophrenia patients, but not in their unaffected relatives. Psychological Medicine,
35(12), 1727 - 1736. https://doi.org/10.1017/S0033291705005738
Citing this paper
Please note that where the full-text provided on King's Research Portal is the Author Accepted Manuscript or Post-Print version this may
differ from the final Published version. If citing, it is advised that you check and use the publisher's definitive version for pagination,
volume/issue, and date of publication details. And where the final published version is provided on the Research Portal, if citing you are
again advised to check the publisher's website for any subsequent corrections.
General rights
Copyright and moral rights for the publications made accessible in the Research Portal are retained by the authors and/or other copyright
owners and it is a condition of accessing publications that users recognize and abide by the legal requirements associated with these rights.
•Users may download and print one copy of any publication from the Research Portal for the purpose of private study or research.
•You may not further distribute the material or use it for any profit-making activity or commercial gain
•You may freely distribute the URL identifying the publication in the Research Portal
Take down policy
If you believe that this document breaches copyright please contact librarypure@kcl.ac.uk providing details, and we will remove access to
the work immediately and investigate your claim.
Download date: 10. Aug. 2022
Saccadic distractibility is elevated in schizophrenia
patients, but not in their unaffected relatives
JAMES H. MACCABE
1
*, H E L E N S I M O N
1
, JOLANTA W. ZANELLI
1
,
REBECCA WALWYN
2
,COLMD.MC DONALD
1
AND ROBIN M. MURRAY
1
1
Division of Psychological Medicine,
2
Department of Biostatistics and Computing, Institute of Psychiatry,
London, UK
ABSTRACT
Background. Saccadic distractibility, as measured by the antisaccade task, has attracted attention
as a putative endophenotypic marker for schizophr enia. Some studies have suggested that this
measure is elevated in the unaffected relatives of schizophrenia patients. However, recent studies
have called this into question and the topic remains controversial.
Method. Saccadic distractibility was measured in 53 patients with DSM-I V schizophrenia, 80
unaffected first-degree relatives and 41 unaffected controls.
Results. Schizophrenia patients performed worse than relatives and controls combined (p<
0
.
00001), but relatives did not differ significantly from controls. Performance in multiply affected
families was no worse than that in singly affected families. Relatives with a high presumed genetic
risk for schizophrenia performed no worse than other relatives. The performance of the patients
did not predict that of their relatives.
Conclusions. These results demonstrate that saccadic distractibility is strongly associated with
disease status but not with genetic loading for schizophrenia. We conclude that saccadic distract-
ibility is unlikely to be useful as an endophenotypic marker in schizophrenia.
INTRODUCTION
Family, twin and adoption studies have indi-
cated that operationally defined schizophrenia
has a heritability of over 0
.
8 (Cardno et al.
1999). However, the non-Mendelian segregation
within families and the disparate linkag e find-
ings suggest that schizophrenia is a complex,
polygenic disorder (Gottesman & Shields, 1982).
Genetic transmission in schizophrenia is likely
to be further complicated by epistasis (Wade,
2001) (non-additive interactions between genes),
pleiotropy (Hodgkin, 1998) (a single gene deter-
mining two or more characteristics), incomplete
penetrance (Levinson et al. 1996), interactions
with en vironmental factors (Van Os & Sham,
2002), and the probable aetiological hetero-
geneity of the clinically defined disorder
(Cardno & Gottesman, 2000).
One approach to overcoming these problems
is the use of intermediate phenotypes, or bio-
logical markers. These are anatomical, physio-
logical or biochemical variables that segreg ate
with genetic risk for the disorder, and which
are assumed to have a simpler genetic architec-
ture than the disorder itself (Weinberger, 2002),
with a more proximal relationship to the
underlying genes. It is hoped that the use of
biological markers will identify more homo-
geneous groups of subjects within the broader
clinically defined phenotype, and thus lead to
improved success in the search for susceptibility
genes.
The necessary criteria for a biological marker
were set out by Wickham & Murray (1997).
* Address for correspondence: Dr J. H. MacCabe, PO Box 63,
Institute of Psychiatry, London SE5 8AF, UK.
(Email: j.maccabe@iop.kcl.ac.uk)
Psychological Medicine, 2005, 35, 1–10. f 2005 Cambridge University Press
doi:10.1017/S0033291705005738 Printed in the United Kingdom
1
The most important criterion, and the one that
has received the most research attention, is an
increased prevalence of the marker in the
unaffected relatives of affected subjects. Several
putative biological markers have been studied,
including structural brain changes (McDonald
et al. 2004), endogenous event-related poten-
tials (Bramon et al. 2004) and performance on
a variety of neuropsychological tasks (Toulo-
poulou et al. 2003).
The study of eye-movement abnormalities as
potential biological markers for schizophrenia
dates back to the 1970s (Holzman et al. 1974).
Initially research focused on abnormalities of
smooth pursuit eye movements, but recently,
the study of saccadic eye movements has
attracted attention, particularly the antisaccade
task.
The antisaccade task
The standard antisaccade task (Hallett, 1978)
begins with the subject fixating on a central
illuminated target. The target then moves
rapidly to a peripheral location, and the subject
is required to inhibit his/her reflexive saccade
(in pursuit of the target) and to generate, in-
stead, a saccade in the opposite direction, which
is termed an antisaccade. A saccadic distract-
ibility error is recorded if a subject fails to
inhibit the reflexive saccade in the direction
of the target. Saccadic distractibility is usually
expressed as the percentage of trials in which the
initial eye movement was towards the stimulus.
Variations of the task include a gap or overlap
(McDowell & Clementz, 1997) between the off-
set of the central stimulus and the onset of the
peripheral stimulus.
Every published study to date examining
the antisaccade task in schizophrenia and its
variants has demonstrated that schizophrenia
patients perform more poorly than control
subjects in suppressing reflexive saccades,
regardless of variations in the paradigm (Levy
et al. 2004). By contrast, studies that have
examined the ability of schizophrenia patients
to perform simple reflexive saccades have found
that their performance is within the normal
range (Crawford et al. 1995a).
Several studies have examined the prevalence
of saccadic distractibility errors in unaffected
relatives of patients with schizophrenia, in-
cluding one study by our research group
(Crawford et al. 1998). Some studies have
shown evidence of elevated rates of saccadic
distractibility errors in well relatives compared
to control populations (Curtis et al. 2001;
Karoumi et al. 2001) but others have either
found non-significant differences, or have not
presented a statistical test comparing well
relatives with controls (Thaker et al. 1996,
2000; Katsanis et al. 1997; Crawford et al.
1998; Brownstein et al. 2003). Two recent meta-
analyses have investigated the presence of
abnormalities in the unaffected relatives of
schizophrenia patients. Calkins and colleagues
(2004) demonstrated that when all studies are
combined, relatives perform worse than con-
trols. However, Levy et al. (2004) showed that
studies which applied more stringent exclusion
criteria to controls than to relatives reported
large effect sizes, whereas those with symmetri-
cal criteria showed small and non-significant
differences.
As well as being over-expressed in the well
relatives of patients, biological markers should
correlate with the likelihood of carrying suscep-
tibility genes, both between and within families.
Thus, members of multiply affected families are
hypothesized to have poorer eye tracking than
members of singly affected families (Lewis et al.
1987). This hypothesis has previously been
tested in a small study by Ross et al. (1998), with
negative results.
A further prediction is that within multiply
affected families, relatives who appear to be
‘carriers ’ of the genetic risk for schizophrenia
have worse performance than other members of
multiply affected families. An example of such
a ‘presumed obligate carrier’ would be a man
whose daughter and brother both had schizo-
phrenia, but who was not affected himself. Ross
et al. (1998) found that these ‘presumed obligate
carriers’ had worse performance than other
well relatives.
To summarize, previous research has demon-
strated that schizophrenia patients perform
worse than control subjects on the antisaccade
task, but the crucial question as to whether this
trait is related to genetic risk for schizophrenia,
thus warranting further investigation as a puta-
tive biological marker for schizophrenia, re-
mains controversial (Brownstein et al. 2003).
We therefore examined antisaccades in both
schizophrenia patients and their relatives from
2 J. H. MacCabe et al.
both singly and multiply affected families. Our
hypotheses were as follows :
(1) Schizophrenia patients will perform
worse than their unaffected relatives and normal
controls.
(2) Unaffected first-degree relatives of pa-
tients will perform worse than normal controls.
(3) Patients from multiply affected families
will perform worse than those from singly
affected families.
(4) Relatives from multiply affected families
will perform worse than those from singly
affected families.
(5) Among the relatives from multiply affec-
ted families, presumed obligate carriers will
perform worse than the remainder.
(6) Relatives of patients who perform poorly
will have worse performance than relatives
of patients who perform well.
METHOD
The study was conducted as part of the
Maudsley Family Study (McDonald et al . 2004),
a larger investigation of biological markers in
schizophrenia patients and their relatives. There
is no overlap between the data presented here
and the data previously reported by Crawford
et al. (1998).
Subjects
All su bjects were Caucasians, aged 16–69 years,
whose first language was English. Exclusion
criteria were substance or alcohol dependence in
the previous year, or a lifetime history of sig-
nificant head injury or organic brain disease.
Subjects gave informed written consent for their
participation. The study was approved by the
local Ethical Committee.
The study groups are represented diagram-
matically in Fig. 1. A total of 53 patients
fulfilling DSM-IV (APA, 1994) criteria for
schizophrenia (n=47), schizoaffective disorder
(n=5) or schizophreniform disorder (n=1) were
included in the study. Patients were divided into
two groups on the basis of family history : (a)
multiply affected, where the proband had at
least one first- or second-degree relative with
schizophrenia or other psychotic disorder, and
(b) ‘singly affected ’, where the proband had
no family history of schizophrenia or other
psychotic disorder as far as third-degree rela-
tives. Patients were recruited either in response
to advertisements through voluntary organiz-
ations or by direct referral from their treating
clinicians. Due to the comparative rarity of
multiply affected families, these were specifically
targeted when advertising the study.
All available first-degree relatives were
screened for mental health prob lems. Any who
were suspected of psychotic symptoms were
interviewed using the Schedule for Affective
Disorders and Schizophrenia – Lifetime version
(SADS-L; Endicott & Spitzer, 1978), and were
reassigned to the Patient category if they
fulfilled diagnostic criteri a for schizophrenia,
schizophreniform or schizoaffective disorder.
However, those with a lifetime history of a non-
psychotic DSM-IV disorder were not excluded.
The relatives from multiply affected families
included 10 parents who were classified as ‘pre-
sumed obligate carriers’, on the basis that (a)
they also had a sibling or parent affected and (b)
transmission of liability was apparently uni-
lineal within that family.
Forty-one control subjects were recruited
from the local community via newspaper ad-
vertisements and from hospital and university
staff. None of the control subjects had a per-
sonal or family history of schizophren ia or other
Total sample
(n =174)
Schizophrenia
(n =53)
Relatives
(n =80)
MA family
(n =25)
MA family
(n =31)
Obligate
carrier
(n =10)
Other
(n = 21)
SA family
(n =28)
SA family
(n = 49)
Controls
(n = 41)
Non-schizophrenia
(n =121)
Contrast 1
Contrast 2
Contrast 3
Contrast 4
Contrast 5
FIG. 1. Composition of subject groups and contrasts. MA,
multiply affected ; SA, singly affected.
Saccadic distractibility in schizophrenia families 3
psychotic disorde r. A lifetime personal or family
history of other psychiatric disorders was not an
exclusion criterion. Therefore, unlike many
previous studies, having a relative with schizo-
phrenia (or other psychotic disorder), was the
only criterion that distinguished between unaf-
fected relatives and controls. The diagnoses in
the six groups are shown in Table 1.
Clinical assessment
Patients, relatives, and controls were assessed
using the same clinical scales in face-to-face
interviews. Diagnoses were made using the
SADS-L and family history was assessed using
the Family Interview for Genetic Studies
(Nurnberger et al. 1994). Additional information
regarding the timing and nature of psychopath-
ology was collected for all participants allowing
DSM-IV diagnoses to be made. Clinical infor-
mation was always supplemented with collateral
histories with informants, and from medical
notes where available.
Eye-movement assessment
The antisaccade task was conducted using
the Amtech ET3 eye-tracking system (AmTech
GmbH, Weinheim, Germany). A Dell Optiplex
560/L computer (Del l Inc., Round Rock, TX,
USA) was used to control the apparatus, and
to record eye movements, using AmTech ET3
software. Eye movements were detected by
means of an infra-red reflection oculogra ph,
with eye position sampled at 200 Hz. The
stimuli consisted of a central LED, with per-
ipheral LEDs at 15x horizontally, mounted on
a board, 180 cm away from the subject’s eyes.
Eye-movement recordings were later analysed
using AmTech Eyemap version 2.0, by a trained
rater (H.S.) who was blind to diagnosis or sub-
ject group.
The tests were carried out as the last of
a battery of eye-movement tasks, including
smooth pursuit and prosaccade tasks. The sub-
jects were seated in a darkened room, and were
requested not to move their head, which rested
on an adjustable frame. At the start of each
trial, the central LED was illuminated. After
800 ms, the central LED was extinguished and
simultaneously, one of the peripheral targets
(¡15x eccentricity) was illuminated for 3 s
(accompanied by an audible signal ) and then
returned to the centre. The subjects were
instructed to fixate on the central target until it
moved to the peripher al position, and then to
direct their gaze as quickly and accurately as
possible to its mirror position (of equal distance
from, but in the opposite direction to, the
peripheral target). Blinks and other artefacts
were identified by inspection of the trace, and
removed from the analysis. A saccade was
defined as a deflection of o2
.
5x. Any initial
saccade (discounting the first 80 ms) towards
the peripheral target was scored as a distract-
ibility error. Where this was followed by a sac-
cade in the correct direct ion, the latter was
termed a corrective saccade. The number of
corrected distractibility errors divided by the
number of analysable trials gave the distract-
ibility error score. As well as the distractibility
error score, the mean latency of each type of
saccade (antisaccade, corrective saccade, and
distractibility error) was measured for each
subject.
Each pa rticipant completed a set of 12 prac-
tice trials to ensure that they understood the test
instructions. Subjects with an error rate of
Table 1. Lifetime DSM-IV diagnoses and demographic details in the six groups
Group
Schizo-
phrenia
Schizo-
phreniform
disorder
Schizo-
affective
disorder MD
Other
DSM-IV
diagnosis
No
disorder
Age (yr) Education (yr)
Gender
(% males)Mean
S.D. Mean S.D.
Sz, MA family 22 1 2 36
.
411
.
514
.
13
.
969
.
6
Sz, SA family 25 3 32
.
78
.
813
.
62
.
266
.
7
Obligate carrier 2 1 7 54
.
03
.
714
.
03
.
330
.
0
Relative, MA family 5 2 14 42
.
614
.
414
.
33
.
045
.
5
Relative, SA family 6 1 42 50
.
114
.
113
.
82
.
832
.
7
Control 4 37 41
.
914
.
713
.
73
.
541
.
5
F 9
.
90
.
2
x
2
18
.
8
p <0
.
001
N.S. <0
.
01
MD, Major depression; Sz, schizophrenia ; MA, multiply affected; SA, singly affected.
4 J. H. MacCabe et al.
