CONNECTING THE WRONG DOTS:
 CAN THALAMO
-CORTICAL DYSCONNECTIVITY 
EXPLAIN ALTERED COROLLARY DISCHARGE IN SCHIZOPHRENIA?
  By Beier Yao
        A THESIS
 Submitted to
 Michigan State 
University
 in partial fulfillment of the requirements
 for the degree of
  Psychology
 Ð Master
 of Arts
 2018   ABSTRACT
 CONNECTING THE WRONG DOTS:
 CAN THALAMO
-CORTICAL DYSCONNECTIVITY 
EXPLAIN 
ALTERED COROLLARY DISCHARGE IN SCHIZOPHRENIA?
  By Beier Yao
 Corollary discharge (CD)
 signals
 are
 Òcop
ies
Ó of motor signals sent to sensory areas to 
predict the 
impending
 input. Because 
they are
 used to distinguish actions generated by oneself 
versus external forces, altered CD has been hypothesized to result in 
the commonly
-observed 
agency disturbances in schizophrenia patients (SZP).
 Behavioral evidence for altered CD in SZP 
has been observed i
n multiple sensorimotor domains, including the 
oculomotor 
system; however
, its exact neural underpinning is unknown. One oculomotor CD pathway identified in primates 
projects from motor neurons in 
the superior colliculus (SC) 
to visual neurons in the 
front
al eye 
fields (FEF) 
via the 
mediodorsal thalamus (MD
T). The current study aimed to examine the 
structural connectivity of MD
T-FEF pathway in SZP and whether it relates to 
oculomotor 
CD 
abnormalities.
 Twenty
-four
 SZP and 
22 healthy controls (
HC) underwent d
iffusion tensor 
imaging (DTI)
, and 
a large subset of those individuals also
 performed the blanking task, an eye 
movement task that measures the influence of CD on visual perception
. Probabilistic 
tractography was used to identify white matter tracts connec
ting FEF and MD
T. Microstructural 
integrity of these tracts 
was
 compared across groups and correlated with 
behavioral indices of 
oculomotor CD from the blanking task
 and symptom severity.
 We found that SZP had 
compromised microstructural integrity in 
MDT
-FEF
 pathway
. This hypoconnectivity was 
correlate
d with 
both 
impaired oculomotor CD signals and more severe positive symptoms in 
SZP.
 These data suggest that the MDT
-FEF pathway may serve an important role in transmitting 
oculomotor CD signals
, which in
 turn may relate to 
positive symptom manifestation in SZP.
  !iii ACKNOWLEDGEMENTS
  The author would like to
 thank 
Dr. 
Katy
 Thakkar for her guidance and support, 
Dr. Jason 
Moser and Dr. Mark Becker for their helpful feedback, 
Lara Rısler
 for sharing her analyses 
results, 
Jingtai Liu for his assist
ance o
n 3D tract visualization,
 Livon Ghermezi for visually 
inspecting 
part of 
the imaging data, 
Xiaochen Luo for providing constructive distractions and 
meaningful companionship, 
and 
all the participants for their partic
ipation
.    !iv TABLE OF CONTENTS
  LIST OF TABLES
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ..v
 LIST OF FIGURES
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
...
Évi INTRODUCTION ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ..1
  Corollary discharge abnormalities in psychosis
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ.3  Altered 
thalamo
-cortical connectivity in schizophrenia
 ÉÉÉÉÉÉÉÉÉÉÉÉÉ.5
  METHODS ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É.7
  Overview
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É.7
  Participants ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É..7
  Blanking task ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É..8
  Image a
cquisition ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É..10
  Regions of interest ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
ÉÉ11  Preprocessing ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É11 Probabilistic tractography ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
ÉÉ.12
  Statistical analyses ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
ÉÉ14  RESULTS
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ.16
  Participant characteristics ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É.16
  Hypoconnectivity in MDT
-FEF pathway
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É.16
  Relationship with oculomotor CD
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
ÉÉ17  Relationship with 
clinical symptoms 
ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É..
É..
18   DISCUSSION ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
...19
  APPENDICES ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É..24
  APPENDIX A: Tables ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
ÉÉ..25
  APPENDIX B: Figures ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
ÉÉ.27
  APPENDIX C: Supplementary 
Results ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
É37   REFERENCES ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
ÉÉ.41
    !v LIST OF TABLES
  Table 1
. Demographic information
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
...25
 Table 2
. Mean microstructural integrity measures (FA, MD, RD, AD) in the MDT
-FEF tract by 
hemisphe
re 
ÉÉÉÉÉÉÉÉÉÉÉÉÉÉ.ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
...26
  Table S1
. Demographic information
 of the subset sample with behavioral data ÉÉÉÉÉÉ
.39
  Table S2
. Mean microstructural integrity
 measures
 (FA, MD, RD, AD) in the MDT
-FEF tract by 
hemisphere in the subset
!sample 
with behavioral data
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉ.ÉÉ40    !vi LIST OF FIGURES
  Figure 1
. Blanking task ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ.
27  Figure 2
. Blanking task perceptual judgment
 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
...28
  Figure 3
. Blanking task pe
rformance in chronic SZP and HC ÉÉÉÉÉÉÉÉÉÉÉ
ÉÉ.29
  Figure 4
. Blanking task performance and symptom severity in chronic SZP
 ÉÉÉÉÉÉÉ..30
  Figure 5
. Group
-averaged blanking task per
formance in chronic SZP and HC ÉÉÉÉÉÉ
...31
  Figure 6
. Probabilistic tractography results of the
 MDT
-FEF pathway in both groups ÉÉÉ...
32  Figure 7
. 3D Probabilistic tractography resul
ts of the MDT
-FEF pathway in HC ÉÉÉÉÉ
...33
  Figure 8
. Group differences in the MDT
-FEF tract microstr
uctural integrity by hemisphere É
.34
  Figure 9
. Scatterplots of JND 
against MDT
-FEF tract
-specific indic
es of microstructural 
integrity ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ................................
.............................35
  Figure 10
. Scatterplots of PANSS positive symptoms scores against MDT
-FEF trac
t-specific 
indic
es of microstructural integrity .ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ
..36
 !1 INTRODUCTION
 It has always been a challenge to come up with a mechanistic model 
of schizophrenia that 
explain
s its wide range of symptoms on both phenomenological and neuro
bio
logical levels. 
One 
recent theory 
that 
attempted to address this problem was the prediction error 
(PE) 
model of 
psychosis 
(Corlett, Honey, & Fletcher, 2016; Fletcher & Frith, 2009; Gray, Feldon, Rawlins, 
Hemsley, & Smith, 1991)
. Simply put, 
we 
form a model of the world that is based 
on sensory 
inputs and expectancies, and 
we 
update our model of 
the world whenever we encounter a PE 
Ð a mismatch between our prediction and actual 
sensory 
input
, to make the model closer to the ever
-changing reality
. However, individuals with schizophrenia 
(SZP
) may have excessive false PEs 
that constantly prompt them to update a perfectly working model. Because these abnormal PEs 
were 
due to systematically altered predictions rather than real changes in the environment
, the 
model can never be updated satisfacto
rily, hence leading to unusual sensory experiences 
commonly seen in SZP
.  To break 
the
 PE 
model down, 
basic 
predictions 
rely on simple correlations between 
commonly co
-occurring stimulus
-stimulus pairs (e.g. 
police cars by the roadside predict
 accidents
) an
d action
-stimulus pairs (e.g. 
speaking predicts hearing your own voice
). This
 study 
focus
ed on the latter. Predictions
 about our own actions
 are established 
via
 corollary discharge 
(CD), or efference copy, signals
 in our brains
. CD signals are
, in essence
, ÒcopiesÓ of motor 
signals
 containing information about the intended movement (e.g. direction, amplitude, velocity, 
etc.)
. They originate from the same brain areas as the motor signals do but are then sent to 
sensory areas
, where a prediction of the 
sensor
y consequences of the intended movement
 can be 
computed
. This match
 between 
predict
ed and 
actual 
sensory inputs 
is argued
 to be
 the basis of 
our sense of agency 
- the subjective sense of being in control of the actions we produce.
  !2 However
, an altered
 CD si
gnal could lead directly to a wrong 
(or absent) 
prediction and hence a 
PE.
 One direct consequence 
then 
could be misattributing self
-generated actions to external 
forces
. Interestingly, almost all first rank symptoms of schizophrenia 
(Schneider, 1959)
Ñsymptoms which were argued to be pathognomonic to schizophrenia
Ñcould be conceptualized 
as agency disruptions, e.g.
 alien control delusions
, auditory hallucinations (misattributing oneÕs 
sub vocal 
voices as someone else speaking;
 Green & Kinsbourne, 1990)
. Therefore, altered CD 
has been theorized as the basis for these agency disruptions in 
schizophrenia
 (Feinberg, 1978)
. There is 
indeed 
a growing evidence base that CD is impaired in 
SZP
 in multiple sensory 
modalities 
(Blakemore, Smith, Steel, Johnstone, & Frith, 2000; Ford & Mathalon, 2012; Shergill, 
Samson, Bays, Frith, & Wolpert, 2005)
. A
 thoro
ugh understanding of how CD signals go awry in 
the brains of 
SZP
s can shed light on key mechanism
s of the disorder and help develop more 
targeted interventions.
 One potential neural pathway transmitting CD signals associated with saccadic eye 
movements has been identified in recent primate studies: 
neurons in the
 mediodorsal thalamus 
(MDT) seem to play a crucial role in relaying 
CD signals 
originating 
from
 movement neurons in
 the superior colliculus (SC) to the
 visual neurons in the
 frontal eye fields (FEF;
 Sommer & 
Wurtz, 2002, 2004, 2008)
. Results from humans with specific MDT lesions support 
this
 finding 
(Gaymard, Rivaud, & Pierrot
-Deseilligny, 1994; Ostendorf, Liebermann, & Ploner, 2010)
. A large body of animal physiology work 
support
s the notion 
that 
the 
thalamus may play a 
more 
general key
 role in transmitting CD sig
nals. The branching axons innervating thalamus enable 
the 
distribution
 of an identical message to both motor centers and sensory areas, making them a 
strong candidate for CD circuits 
that permit
 sensory predictions of almost all motor actions
 (Sherman, 2016)
. If this is the case, then thalamus essentially helps modulate all cortical 
 !3 processing of motor commands. In other words, thalamo
-cortical connectivity could play a 
crucial role in abnormal transmission of CD signals. Interestingly, there has long been a model of 
schizophrenia as a 
Òmisconnection syndrome
Ó, in which thalamus 
is a 
key link
 (Andreasen, 
1999). Altered CD could be a direct consequence o
f misconnection in 
neural
 circuit
s and could 
be a more proximal mechanism of schizophrenia symptoms.
 Though there is mounting 
behavioral evidence for altered oculomotor CD in SZP 
(See 
Thakkar
, Diwadkar, & Rolfs
, 2017 for review
), no study to date has studied whether abnormalities in the MDT
-FEF pathway 
contribute to this.
 Corollary discharge abnormalities in psychosis
 To date
, most 
of our understanding
 of 
CD 
in human and non
-human primates 
come
s from the
 oculomotor system. Current evidence 
indicates 
that we rely on CD signals 
for different 
oculomotor
 processes such as
 maintain
ing
 visual stability 
(i.e. perceiving a 
stable
 and continuous
 world de
spite constant eye movements;
 Cavanagh, Hunt
, Afraz, & Rolfs, 2010)
 and generat
ing
 rapid successive
 goal
-directed
 saccades
 (i.e. 
preparing 
a second saccade command while 
executing
 the first saccade
; Hallett & Lightstone, 1976a, 1976b)
. Likewise
, a large body of
 evidence for CD abnormalities in psychosis 
also 
comes from the oculomotor system 
(see 
Thakkar et al.
, 2017
 for review
).  One commonly used task to assess visual stability is the blanking task, which measures 
visual perception shortly following an eye movement 
(Figure 1; 
Deubel, Schneider, & 
Bridgeman, 1996)
. In this task, following a period of fixation, a visual stimulus appear
s at a 
peripheral location 
on 
the screen 
(at the 
pre
-saccadic location
), and the participant is instructed 
to 
look at
 the stimulus. Once the participant initiates a saccade, the stimulus disappear
s for a 
brief moment and reappear
s at a new location (
post
-saccadic location
) slightly bac
kward or 
 !4 forward
 relative to the pre
-saccadic location
. The participant is then asked to judge 
in which 
direction the stimulus jumped relative to its pre
-saccadic location. Because saccadic eye 
movements are often 
imprecise, the saccade 
typically
 fall
s short or long of 
the
 target. Therefore, 
the participant cannot rely on the saccade landing site
 to inform them of the pre
-saccadic location
 to make an accurate judgment on the post
-saccadic location. 
For example, if the stimulus jumps 
backward by 0.5
¡ and
 the saccade falls short of the target by 1
¡ (i.e. 0.5
¡ short of the post
-saccadic location), it will appear that the stimulus jumps 
forward
 relative to the
 saccade
 landing 
site
 (see
 Figure 2 for an illustration).
 The participant has to ÒknowÓ 
by 
exactly h
ow much the 
saccade f
ell short
 or long 
of target
 in order to remap the actual pre
-saccadic location correctly, 
which requires access to
 information stored in a CD signal. In other words, if CD signals are 
intact, there should be no correlation between the 
actual saccade landing site and the 
participant
Õs perceptual judgment 
Ðsaccades fall
ing
 short or long of the target are random errors 
that have no relationship with the pre
-saccadic location, and hence should have no influence on 
the judgment of post
-sacca
dic location either.
 Indeed
, this is exactly how healthy people perform 
on the task 
(Collins, 
Rolfs, Deubel, & Cavanagh, 2009)
. However, if CD signals are 
compromised, the participant may not have access to enough information 
about the spatial
 accuracy of the saccade. In such
 a case, the participant can only rely on the actual saccade 
landing si
te to make the perceptual judgment
 - there is no information that would prompt the 
participant to think that the saccade was not precise. Therefore, when CD is impaired, one would 
expect to see a correlation between the saccade landing site and the partici
pant
Õs perceptual 
judgment
 because
 as 
the saccade falls increasingly short of the target, there is a greater chance 
that the post
-saccadic location will be forward of the saccade landing site. 
This has indeed been 
observed in humans with 
MDT lesions 
(Ostendorf et al., 2010)
 and 
in 
primates whose MD
T was 
 !5 temporarily
 inactivated 
(Cavanaugh, Berman, Joiner, & Wurtz, 2016)
, providing
 more evidence 
to suggest that
 MDT is a key relay station 
for
 CD signals
 related to saccadic eye movements
. Recently, 
Rısler
 and colleagues (2015) examined CD in 
SZP
 using the blanking task. 
SZP 
exhibited reduced precision of perceptual judgments, a
s evidenced by flatter psychometric curves 
plotting performance against target 
displacement
. That is, their perceptual judgments were less 
sensitive to the actual target displacement. 
On a group level, t
hey found 
a correlation between
 saccade landing site 
and
 perceptual 
judgment 
in SZP, but not in HC
 (Figure 3)
, consistent with 
previ
ous literature of impaired oculomotor CD in
 SZP. That is, SZP with shorter mean saccade 
amplitudes made more forward judgements. Although this relationship did not manifest on the 
single
-subject level for most participants, 
SZP with more severe pos
itive sy
mptoms showed a 
greater reliance on saccade landing sites
 (Figure 4)
, supporting the theory that abnormal CD is a 
key mechanism of 
psychosis
 (Feinberg & Guazzelli, 1999)
.  Altered thalamo
-cortical connectivity in schizophrenia
 There is 
growing evidence that thalamo
-cortical connectivity is impaired in schizophrenia 
and is 
closely 
related to symptomatology
 and functioning. 
Recent 
structural 
imaging studies 
found white matter alterations in thalamo
-cortical circuits 
across differ
ent illness
 stage
s (see 
Canu,
 Agosta, & Filippi, 2015
 for review
), which are 
consistent with 
altered 
functional 
connectivity 
in thalamo
-cortical circuits
 (see 
Pergola, Selvaggi, Trizio, Bertolino, & Blasi, 2015
 for review
). More specifically, pathways involving 
MDT and its project
ion
 sites in frontal cortex 
seem to have the strongest support for being crucial to the pathophysiology of schizophrenia, 
though not many studies 
have 
examined individual thalamic nuclei
 (Anticevic, Cole, et al., 2014; 
Giraldo
-Chica, Rogers, Damon, Landman, & Woodward, 2018; Shepherd, Laurens, Matheson, 
Carr, & Green, 2012; Woodward, Karbasforoushan, & Heckers,
 2012). Importantly, thalamo
- !6 frontal hypoconnectivity has also been found in individuals at clinical high risk for psychosis
 and 
bipolar patients with a history of psychosis 
(Anticevic et al., 2015; Anticevic, Yan
g, et al., 2014)
, ruling out medication effects or illness chronicity as main contributors
 and suggesting that 
thalamo
-frontal hypoconnectivity may relate to psychosis in a trans
-diagnostic manner
. Attesting 
to their clinical relevance, these 
white matt
er 
alterations are further correlated with 
clinical 
symptoms
, cognitive impairments, and treatment outcomes
 (see 
Canu
 et al.
, 2015
 for review
). Relevant to this study, there is evidence that SZP with passivity symptoms (i.e.
 symptoms related 
to a sense of losing agency over oneÕs body/thoughts, directly reflecting altered CD) have more 
compromised white matter integrity in thalamus, relative to SZP without passivity symptoms 
(Sim et al., 2009)
. This 
provides indirect evidence for
 our hypothesis that impaired thalamo
-cortical structural connectivity 
may 
underlie the CD
 abnormalities
 in schizophrenia, which we 
propose could be 
a key mechanism of 
symptom expression 
of 
the disease.
  In this study, we used diffusion tensor imaging (DTI) and the blanking task to study white 
matter connectivity and oculomotor CD in SZP
, respectively
. DTI i
s an imaging method 
commonly used to assess structural connectivity in human brains. 
Specifically, we tested the 
following three hypotheses: (1)
 there
 will be
 abnormal 
structural connectivity in the MD
T-FEF 
pathway in 
SZP
 relative to HC
; (2)
 connectivity 
of the MD
T-FEF pathway will be positively 
correlated with 
oculomotor indices of CD 
signal 
integrity; (3) the integrity of MD
T-FEF white 
matter pathway will also be correlated inversely with
 positive 
symptom severity in 
SZP
.    !7 METHODS
 Overview
 Participants 
performed 
the
 blanking task
 and underwen
t DTI
. Putative white matter tracts 
connecting our regions of interest (ROIs) were computed using p
robabilistic tractography
, an 
analysis technique that 
reconstructs anatomical pathways
 (presumably white matter tracts)
 between 
any given 
brain regions 
based on a distribution profile of probable fiber orientations in 
each voxel 
(Behrens et al., 2003; Behrens, Berg, Jbabdi, Rushworth, & Woolrich, 2007)
. The 
FEF was defined based on 
group
-level functional MRI (
fMRI
) activation maps 
from the same 
subjects performing a different saccadic 
eye movement task in the scanner 
(Thakkar, van den 
Heiligenberg, Kahn, & Neggers, 2014)
. The 
MDT
 was
 localized using the 
Harvard
-Oxford 
thalamic connectivity atlas, a probabilistic
 atlas of 7 sub
-thalamic regions segmented according 
to their white matter connectivity to cortical areas
 (Desikan et al., 2006)
. ParticipantsÕ indiv
idual 
probabilistic tracts in native space were normalized into a common space and averaged within 
groups. Then a threshold was applied to obtain final group tracts. Within each group
-averaged
 tract, 
mean fractional anisotropy
 (FA)
, mean diffusivity
 (MD)
, axial diffusivity
 (AD), and radial 
diffusivity
 (RD) were then calculated for
 each 
participant. These indices of microstructural 
integrity of white matter tracts 
were used 
for 
further
 group comparisons and correlation analyses 
with 
behavioral indices of CD 
obtained from the blanking task, as well as symptom scores within 
the patient group.
 Participants
 Twenty
-two
 antipsychotic
-medicated 
SZP
 were recruited 
from a longitudinal study
 (Genetic 
Risk and Outcome in Psychosis (GROUP) Investigators, 2011)
 and an out
patient psychi
atric 
facility in The Netherlands. Schizophrenia or schizoaffective disorder diagnoses were based on 
 !8 Diagnostic and 
Statistical Manual of Mental Disorders, fourth edition (DSM
-IV) criteria and 
verified with the Comprehensive Assessment of S
ymptoms and History inter
view
 (Andr
easen, 
Flaum, & Arndt, 1992)
 or Schedules for Clinical Assessment for Neuropsychiatry, version 2.1
 (Wing et al., 1990)
. Chlorpromazine (CPZ) equivalent antipsychotic dosages were calcula
ted for 
each patient 
(Woods, 2003)
. Clinical symptoms were assessed with the Positive and Negative 
Syndrome Scale (PANSS
; Kay, Fiszbein, & Opler, 1987)
. Twenty
-four
 HC without a personal or 
family history of DSM
-IV Axis I diagnosis were rec
ruited via community adve
rtise
ments. 
Criteria for participant exclusion were a history of head trauma or neurologic
al illness and recent 
substance 
abuse or dependence. All subjects gave written informed consent and were reimbursed 
for participation. The study was approved by the H
uman Ethics Committee of the University 
Medical Center, Utrecht.
 Blanking
 task
 This task measure
s the degree to which CD influences visual perception
 immediately 
following
 a saccade
 (Figure 
1). Each trial start
ed with the participant fixating on a
 red circle 
on a grey
 background
 in 
a dimly lit
 room. To reduce anticipation effect or stereotypical behavior 
(Collins et al., 2009)
, the circle randomly 
appear
ed at one of six locations (a combination of a 1¡ 
or 
!1¡ of visual angle 
displacement horizontally and a 
0¡, 1¡, or 
!1¡ displacement vertically 
relative to the center of the screen) with equal 
probability
. Once the eye tracker detect
ed that the 
parti
cipant ha
d maintained fixation for 200 ms, 
the 
circle
 would
 turn
 black. After a random delay 
of 
500-1000 ms,
 the circle reappear
ed at a new location 10
¡ to the left or right of 
the fixation
 position
 (pre
-saccadic 
location
). 
The
 participant 
was instructed t
o look at
 the stimulus
 as quickly 
as possible
. Once the participant initiate
d a saccade, the stimulus 
would 
disappear
 (i.e. blank)
 for 
250 ms
 and reappear at a location 
(post
-saccadic location
) somewhere between 
!3¡ to 3¡
 (in 
 !9 increments of 0.5
¡) to the left or right of the pre
-saccadic location
 and 
would 
stay on the screen
 for the remainder of the
 trial
. The participant 
was
 then asked to 
report
 in which direction the 
stimulus jump
ed relative to its pre
-saccadic location
 by key pressing
. Although 
participants 
indicated a left or right response, for 
analysis purpose
s we recoded these responses: 
Òforward
Ó means
 jumping away from the fixation position, and 
Òbackward
Ó means
 jumping towards the 
fixation. 
The combination of 6 fixation positions 
"
 13 
post
-saccadic locations (including 0
¡ displacement) 
"
 2 directions (left, right) result
ed in 
156 total trials. 
Duration
 of the task
 varied 
across participants and typically ranged from 15 to 30 min
utes
. See 
Rısler et al., 2015
 for more 
details on the eye
-tracking apparatus and task stimuli.
 Eighteen SZP and
 17 HC reported in 
Rısler et al., 2015
 also participated in this study
 and their data was used in the analyses 
involving blanking task 
performance.
 ParticipantsÕ performance on this paradigm was already analyzed and reported elsewhere 
(Rısler et al., 2015)
. For
 this study
, two 
measu
res of 
CD 
were
 used in relevant statistical 
analyses: 
just noticeable difference (
JND
) and the 
loss
 slope. Briefly, 
for each individual, trials 
were collapsed across 
fixation positions and saccade directions. 
A logistic function was 
fit 
to the 
data 
plotting
 percentage of forward responses 
as 
a function of target displacement
 (i.e. the 
position of post
-saccadic location relative to the pre
-saccadic location
; see Figure 5 for 
examples
). A JND
 was derived from the function as the difference in target di
splacements 
between the points where the function reached 50% 
and 75% of its full growth. This measure 
indicates individualsÕ sen
sitivity to target displacement
, which presumably relies on CD input to 
infer the pre
-saccadic location. Therefore, smaller JND
 indicates greater CD integrity. Next, for 
each individual, trials were collapsed across target displacements
. Landing site errors (i.e. 
distance between the actual saccade landing site and the pre
-saccadic location) were divided into 
 !10 multiple 
bins. Mean s
accade landing site was then calculated for each bin. The 
loss
 slope
 was 
derived from
 a weighted linear regression where percentage of forward responses corresponding 
to each bin was fit against mean saccade landing site
 per bin
, weighted 
by 
the number of 
trials 
per bin. Because individuals with intact CD signals should not rely on saccade landing sites to 
make perceptual judgments about target displacements, the slope should be almost zero in HC. 
More negative 
slope
 values 
indicate
 more reliance on 
landing
 site and therefore a larger loss of 
CD signals, 
hence the name Ò
loss slope.Ó
 Image
 acquisition
 All 
DTI
 data were acquired at the University Medical Center Utrecht on a Philips Achieva 
3T scanner (Philips Medical Systems, Best, The Netherlands) equipped with an eight
-channel 
head coil allo
wing parallel imaging. Two dif
fusion images were acquired using singl
e-shot 
echoplanar imaging sequences, consisting of 30 diffusion
-weighted scans (
b = 1,000 s/mm
2) wi
th 
noncollinear gradient direc
tions and one im
age without diffusion weighting
 (b = 0 s/ mm
2), 
covering the entire brain (Repetition Time (TR)
 = 7,057 ms; Ech
o Time (TE)
 = 68 ms; field of 
view
 = 240 mm
 " 240 mm 
"
 150 mm; in plane resolution = 
1.875
 mm 
"
 1.875 mm; slice 
thickness = 
2 mm; no slice gap; 75 axial slices; matrix size 128 mm 
" 99 mm). The diffusion 
weighted scans were measured twice, once with phase 
encoding direction r
eversed (first scan 
posterior
-anterior, second scan anterior
-posterior), in order to correct for susceptibility induced 
spatial distortions
 (Andersson & Skare, 2002)
. For registration purposes, a whole
-brain three
-dimensional T
1-weighted scan (200 
slices; TR = 10 ms; TE = 4.6 ms; flip angle = 8¡
; field of 
view, 240 mm 
" 240 mm 
" 160 mm; voxel size: 0.75 mm 
" 0.8mm
 " 0.75 mm) was acquired.
    !11 Regions of interest
 We included
 two regions of interest (ROI
s) in each hemisphere
: FEF and MD
T. As FEF
 does not have clear anatomical boundaries, 
the group
-level 
functional MRI (
fMRI
) activation maps 
from the same subjects performing a 
different 
saccadic
 eye movement
 task in the scanner 
(Thakkar et al., 2014)
 were
 used to define th
is region
. More specifically, 
FEF
 was
 cre
ated based 
on areas that showed greater a
ctivation on trials
 during which 
participants 
were making saccades 
versus where they were 
fixating 
(th
reshold 
p < 0.001 uncorrected) 
across both groups
. The thalamus
 was
 localized using the 
Harvard
-Oxford thalamic connectivity atlas, which is a 
probabilistic atlas of 7 sub
-thalamic regions segmented according to their white matter 
connectivity to cortical areas
 (Desikan et al., 2006)
. MD
T was operationally defined as the 
region of the atlas with highest probability 
(
#
 25%) 
of connectivity to the prefrontal cortex.
 Both 
ROIs were
 limited to whi
te matter structure only and did not
 include grey matter.
 Individual 
white matter masks were created 
by first 
segmenting the T1 image into grey matter, white 
matter, and cerebrospinal fluid using
 SPM, and then 
extracting 
those 
voxels 
that had the 
highest 
probability 
of being identified as white matter.
 Preprocessing
 The diffusion
-weighted scans 
were
 prepr
ocessed and analyzed using FSL 5.0 
(FMRIBÕs 
Software Library, www.
fmrib.ox.ac.uk/fsl). As DTI
 scans suffer from spatial dis
tortions along 
the phase enc
oding direction, two diffusion
-weighted scans were acqu
ired with reversed phase 
encod
ing blips, resulting
 in pairs of images with distortions going in opposite directions. From 
these two images, the off
-resonance field 
were
 estimated using a method similar to that de
scribed 
by Andersson and Skare (2002) as imple
mented in FSL
 (Smith et al., 2004)
. Next, the 30 
diffusion
-weighted images from each phase
-encoding direction 
were
 realigned to the b0 image 
 !12 using affine registration, and eddy current correction 
was
 applied. The eddy
-corrected scans with 
opposite phase encoding blips 
were 
then
 com
bined into a single corrected image using the 
previously estimated off
-resonance field. A brain mask 
was
 created for the mean b0 image and 
applied to all di
ffusion
-weighted 
images. DTI analyses must be performed in native space as 
diffusion gradients are specified in this space; however, ROIs 
were
 created in standard (Montreal 
neuroimaging; MNI) space. To transform the ROIs into each subjectÕs native space, the 
anatomical T1
-weighted volume 
was
 realigned to the mean b0
-weighted
 image and subsequently 
normal
ized to MNI
-space using
 the unified segmentation algo
rithm as implemented in SPM8
 (Ashburner & Friston, 2005)
. The inverse warping parameters from this step 
were
 used to 
transform ROIs from MNI space to native space
 (Neggers, Za
ndbelt, Schall, & Schall, 2015)
. Probabilistic tractography
 We performed 
probabilistic tractography between MD
T and FEF in both hemispheres
 from 
both directions (i.e. MD
T to FEF and FEF to MD
T), each serving
 as both a seed (start point) and 
a target (end point), 
since
 diffusion MRI cannot distinguish between forward and backward 
projections. 
The distribution profile
 of probabilistic connectivity i
s computed by iteratively 
sending out 
5000 streamlines from the seed area, going 
through
 all 
probab
le principal diffusion 
directions in each voxel
 until it was determined impossible to continue
. Only streamlines that 
reach
ed the target 
successfully 
were
 included in the analysis
, and streamlines were not
 allowed 
to 
continue after reaching the target. 
Two crossing fibers per vo
xel were allowed. 
In addition
, the 
mid-sagittal plane was used as 
exclusion mask to avoid streamlines travelling 
falsely 
into the 
other hemisphere
 (Landman et al., 2012; Mori & van Zijl, 2002; Mori & Zhang, 2006)
. Upon 
examination of preliminary tractography results, an axial plane located one voxel beneath MDT 
 !13 was added as another exclusion mask to avoid streamlines travelling in the opposite direction of 
FEF and looping back on themselves.
 For each subject
, a Òcon
nectivity valueÓ was
 generated for each voxel 
at the end of
 the 
tractography 
process
 by computing
 the number of streamlines going through the specific voxel. 
Therefore, 
higher values indicate 
a higher probabilit
y that the voxel belongs to the tract of 
inte
rest
. In order to perform group
 analyses, 
individual
 tractography 
results were
 transformed to 
MNI spac
e. To make sure only white matter structures were included in the analyses, each 
individualÕs whole brain white matter map was used as a mask to preserve 
connectivity values in 
these structures only.
 After that,
 these values were
 averaged across 
participants
 within each 
group
. Connectivity values are not only determined by
 actual structural connectivity, but also
 influenced
 by other factors 
like size of the
 seed ROI (i.e. bigger seeds send out more 
streamlines)
. Thus, 
higher
 connectivity values 
in a given tract 
do not necessar
ily indicate a larger
 probability
 of a
n actual
 structural connection. 
That is, the number of streamlines passing through 
a given voxel
 cannot be 
directly 
compared across seed/target combinations. 
Therefore, to avoid 
potential biases, we 
did
 not use the same 
absolute 
value 
for threshol
ding each group
-averaged 
tract but
 use
d the top 0.27% (
3$
 above mean in normal distribution
) of all voxels 
as the threshold 
instead.
 After applying the thres
hold, these group tracts were
 binarized into masks per tract. 
Masks from both directions (
i.e
. FEF to MD
T and MD
T to FEF) were
 considered equally 
accurate
 and combined
 to 
achieve
 a single mas
k for 
the MD
T-FEF pathway
 per each group
.  For each 
participant
, mean 
FA, MD, AD, and 
RD within each of the 
group
-thresholded tracts 
were
 extracted. 
FA is 
currently 
the most 
widely
 used
 measure of anisotropy, highly sensitive to 
microstructural
 integrity of white matter tracts
. Higher FA value means higher anisotropy and 
 !14 lower diffusivity
 (e.g. better microstructural integrity)
. However, 
FA value
 contains no 
information on the orientation of anisotropy and thus can be 
difficult 
to interpret 
in i
solation
. A 
decrease in FA could be caused by multiple factors: white matter neuropathology, fiber crossing, 
normal aging, etc
. (Alexander, Lee, Lazar, & Field, 2007)
. Therefore, it is recommended to u
se 
multiple 
measures 
for better understanding of the microstructure of white matter
. Specifically, 
RD appears to be a more sensitive indicator 
of myelination, while 
AD seems to 
indicate
 axonal 
degeneration
 (Song et al., 2002)
. MD seems to be best at assessing white matter maturation 
and
 aging 
(Abe et al., 2002; Snook, Plewes, & Beaulieu, 2007)
. Higher values 
for RD, AD, and MD
 mean higher diffusivity
 (e.g. poorer mic
rostructural integrity)
. Statistical analyses
 Statistical analyses 
were
 performed using IBM SPSS Statistics 20.0 (IBM, Armonk, NY)
. First, to
 examine group difference
s of tract locations
, a probability value 
was
 calculated for each 
voxel by dividing the 
number of tracts going through the voxel
 by the
 total number of tracts
 in 
the MD
T-FEF pathway
 in each hemisphere
. This value indicates the probability of the specific 
voxel belonging to the actual MD
T-FEF structural connection.
 Then between
-group t
-tests 
were
 performed
 on probability values
 at each voxel
 on the whole brain level 
with family
-wise error 
correction 
using SPM
. Next, to compare 
microstructural
 integrity
 of MD
T-FEF 
white matter 
tracts
, repeated measures AN
OVAs 
were
 conducted 
on each of the four 
microstructural integrity
 measures
 (FA, MD, RD, 
and 
AD), including diagnostic group as 
a between
-subjects variable and
 hemisphere as 
a within
-subject variable
. Since we we
re interested in the specific infl
uence 
of 
structural connections between MD
T and FEF 
on behavior, mean 
integrity measures
 over the whole brain 
were
 regressed out of the 
tract 
measure
s that did not differ between hemispheres
. For measures with a significant 
 !15 hemisphere effect or hemisphere by group interaction, mean integrity measur
es over each 
hemisphere were regressed out of the corresponding tract measures.
 Thes
e standardized residuals 
were
 used for 
all the correlation
 analyses. 
All
 standardized
 values
 are 0 when equal to the whole 
brain average
 of the corresponding measure. For s
tandardized residuals of FA
, higher 
values 
indicate
 more anisotropic voxels and lower more isotropic voxels
 (i.e. more impaired integrity)
. For standardized residuals of MD, RD, and AD, higher values indicate more diffusive voxels, 
and lower values less di
ffusive
. We perform
ed Pearson 
correlation analyses
 between
 two
 behavioral measures (
loss slope and JND
) and 
the 
standardized
 residuals
. Using
 SZP data only, 
we 
conducted
 SpearmanÕs rank 
correlation analyses
 between 
PANSS
 subscale scores (positive 
and 
negative symptom
s) and 
standardized 
residuals
. To 
examine
 pote
ntial confounding effects o
f antipsychotic
 use, 
standardized medication dose was correlated with the behavioral measures and 
the microstructural integrity
 measures
.    !16 RESULTS
 Participant 
characteristics
  The 
SZP
 and HC were ma
tched for sex, age, 
IQ, 
and 
handedness
. See
 Table 1 for detailed 
participant characteristics.
 Only a subset of the sample (18 SZP, 17 
HC) has both DTI and 
behavioral data. The two groups in this subset 
also 
did no
t differ on sex, age, 
IQ, 
and handedness
 (Table S1)
. Hypoconnectivity in MDT
-FEF pathway
  The spatially normalized, group
-averaged, and statistically thresholded 
MDT
-FEF 
probabilistic tracts 
in both the left and right hemisphere 
are shown in Figure 
6 and Figure 
7. We performed a
 between
-group t
-test
 on whole
 brain
 probability values at each voxel. 
No significant 
differences were found between the SZP and HC
 at the family
-wise error rate of 0.05
, indicating 
that the MDT
-FEF pathway occupied the same
 spatial location in the brain.
 Next, 
we conducted four 
repeated measures AN
OVAs 
where
 one of the
 microstructural 
integrity
 measures
 (FA, MD, 
RD, AD) was the dependent variable each time
, using
 diagnostic 
group as 
a between
-subjects variable and 
hemisphere as 
a within
-subject variable
 (see Figure
 8). Means and standard deviations are presented in Table 2. 
Again, 
higher
 microstructural integrity 
is indicated by higher FA values and lower MD, RD, and AD values. 
For FA,
 MD, and RD, 
there 
was a signif
icant effect of group such that 
SZP
 showed evidence of reduced microstructural 
integrity of the MDT
-FEF pathway 
(FA:
 F(1,44)
 = 11.05
, p = 0.002
, partial 
!2 = 0.20
; MD: 
F(1,44) = 4.69
, p = 0.036, 
partial 
!2 = 0.10; RD:
 F(1,44) = 10.88
, p = 0.002, 
partial 
!2 = 0.20)
. There was no group difference in tract AD, 
F(1,44) = 0.001
, p = 0.98
, partial 
!2 = 0.00
. For FA 
and RD, t
here was 
an additional
 group 
" hemisphere interaction
 effect (FA:
 F(1, 44) = 31.98, 
p < 0.001, 
partial 
!2 = 0.42
; RD: 
F(1, 44) = 13.16, 
p = 0.001, 
partial 
!2 = 0.23), such that these 
 !17 measures only differed significantly between groups in the right hemisphere
 (FA: 
t(44) = 5.86, 
p < 0.001; RD: 
t(44) = 4.63, 
p < 0.001; 
Table 2)
. This group 
" hemisphere interaction effect was 
not significant for MD, 
F(1,44) = 0.42
, p = 0.52
, partial 
!2 = 0.01. Finally, there was no main 
effect of hemisphere for any of the measures. 
There were
 no correlation
s between CPZ 
equivalent dose and any of the microstructural integrity
 measures
 (-0.44 < 
rs < 0.25, 0.07 <
 p < 0.98)
. In 
the
 subsample of participants
 with behavioral data
, there was still a main effect of group 
on FA, MD, and RD
, such that patients 
showed 
reduced microstructural integrity
. Complete 
results from this subsample are presented in Supplementary Results
. Relationship with oculomotor CD
 Because we only found significant group differences in 
FA, MD, and 
RD, we
 performed 
Pearson 
correlation analyses
 between
 the 
standardized
 residuals of these three measures
 (after 
controlling for variance 
due to whole brain structural integrity)
 and 
two
 behavioral measures 
(loss slope and JND
). In SZP only, w
e found a significant correlatio
n between standardized 
resi
dual
 of MD and JND
 (r = 0.57, 
p = 0.013
) and a significant correlation between standardized 
residual of 
RD in the left tract and JND
 (r = 0.47, 
p = 0.049
; see Figure 
9). 
Combined, these 
results are cons
istent with our predictions: 
higher JND, which is indicative of poorer CD 
signaling, is related 
to compromised white matter integrity in the MDT
-FEF pathway in SZP. 
No 
correlations were found between 
the 
loss slope
 and the standardized residuals. No correlations 
were found between the
 blanking task 
measures and standardized residuals in HC.
 There were no 
correlations between CPZ equivalent dose and the behavioral measures in SZP (JND: 
rs = -0.31, 
p = 0.24; loss slope: 
rs = 0.42, 
p = 0.11).
    !18 Relationship with clinical symptoms
 Lastly
, we 
conducted 
SpearmanÕs rank 
correlation analyses
 between 
PANSS
 subscale 
scores (positive and negative symptom
s) and 
the three 
residual
 tract measures (FA, MD, RD)
 that were altered in 
SZP
. We found a significant correlatio
n between MD 
residual values 
and PANSS positive symptom score
 (rs = 0.43, 
p = 0.047
) and between standardized residual of 
RD in the left tract and PANSS positive symptom score
 (rs = 0.48, 
p = 0.026
; see Figure 
10). No 
correlations were found between PANSS negative symptom score and the 
standardized residuals
 (MD: 
rs = 0.12, 
p = 0.58; left RD: 
rs = 0.08, 
p = 0.74)
. Taken together, these correlations suggest 
that more severe positive symptoms in SZP, but not negative symptoms, 
were associated 
with 
compromised white matter integrity in the 
MDT
-FEF pathway.
    !19 DISCUSSION
  In t
his
 study, 
we examined 
whether
 SZP differed from HC on the microstructural 
integrity of the MDT
-FEF pathway, and whether 
putative abnormalit
ies
 were 
correlated with 
behavioral indices of oculomotor CD and psychotic symptoms. 
Using DTI and probabilistic 
tractography, we found that SZP and HC shared the same spatial location of the tracts connecting 
MDT and FEF. However, SZP had compromised microstructur
al integrity 
of 
these tracts, such 
that they had lower FA and higher MD and RD than HC. This hypoconnectivity was found to 
further 
correlate with 
behavioral indices of reduced 
oculomotor CD 
signaling
 and more severe 
positive symptoms in SZP, indicating a p
oten
tial disease mechanism that has 
specific 
behavioral 
correlates and symptom implications.
  Our finding that 
SZP had 
reduced MDT
-FEF 
structural 
connectivity is consistent with 
recent
 findings of 
white matter alterations in 
prefrontal
-thalam
ic pathways 
in SZP 
(Canu et al., 
2015; Gi
raldo
-Chica et al., 2018)
. This was also consistent with previous functional connectivity 
findings of reduced prefrontal
-MDT connectivity
 in SZP 
(Anticevic, Cole, et al., 2014; Welsh, 
Chen, & Taylor, 2010; Woodward et al., 2012)
 and thalamo
-prefrontal hyp
oconnectivity in 
clinical high risk populations 
(Anticevic et al., 2015)
. Notably
, there have been consistent case 
reports on patients with MDT lesions experiencing 
psychotic
-like symptoms such as 
hallucinations and delusions
 (Carrera & Bogousslavsky, 2006; Crail
-Melendez, Atriano
-Mendieta, Carrillo
-Meza, & Ramirez
-Bermudez, 2013; Zhou e
t al., 2015)
, supporting the link 
between thalamo
-cortical communication and positive symptom formation.
 Besides decreased FA, SZP also exhibited increased MD and RD in the MDT
-FEF 
pathway in current study. 
As a marker of normal aging 
(Abe et al., 2002)
, our finding of 
increased MD
 in SZP
 provides 
indirect evidence 
supporting 
the accelerated aging theory of 
 !20 schizophrenia 
(see 
Islam, Mulsant, Voineskos, & Rajji, 2017
 for review
). In fact, a recen
t study 
found
 that the rate of white matter decline was 60% steeper in SZP than in HC
 and the difference 
was present throughout the life span
 (Cropley et al., 2017)
. On the neurobiological level, 
incre
ased RD in SZP indicated that the hypoconnectivity might be related to myelin and 
oligodendroglia dysfunctions 
(Davis et 
al., 2003)
.  Somewhat surprisingly
, we also found
 a significant
 group 
by 
hemisphere interaction 
effect in FA and RD.
 To date, 
very few studies have assessed 
the
 connectivity between specific 
thalamic nuclei and 
cortical areas in SZP. 
In the few studies that 
examined
 MDT
-cortical 
functional connectivity, 
a group 
by hemisphere interaction
 was either not mo
deled, 
not reported
, or not found
 (Anticevic, Cole, et al., 2014; Welsh et al., 2010; Woodward et al., 2012)
. One 
recent
 structural imaging
 study 
that found
 decrea
sed 
prefrontal
-MDT connectivity in SZP 
reported a more prominent difference in the left hemisphere 
(Giraldo
-Chica et al., 2018)
, contrary to our finding of a more prominent difference in the right hemisphere. However, there 
were a few ke
y differences between their study and the current one.
 First, the SZP group in their 
study was 
approximately 
10 years younge
r and 
was
 earlier in the course of illness
. Second
, and
 perhaps
 more importantly
, their main connectivity analyses used Òtotal 
connectivityÓ
 values 
(calculated by
 dividing
 the number of streamlines reaching a certain cortical area by the number 
of streamlines
 reaching
 all cortical regions
 from thalamus), rather than indices of microstructural 
integrity such as FA and MD
. In other 
words, their group comparison revealed the differences in 
whole brain thalamo
-cortical connectivity patterns
 (i.e. how thalamo
-cortical connectivity was 
distributed among different cortical regions)
, rather than the absolute difference on 
microstructural l
evel within a specific 
thalamo
-cortical 
pathway
 such as prefrontal
-thalamic 
 !21 connection
. Clearly, more research is needed to address whether there is a hemispheric effect to 
the altered thalamo
-prefrontal connectivity, or thalamo
-cortical connectivity in ge
neral, in SZP.
  The most novel and arguably most important 
finding of 
current
 study 
is that we found 
a relationship
 between 
the 
MDT
-FEF 
structural 
connectivity and oculomotor CD signal integrity
, mirroring findings from 
primate 
neurophysiology 
studies 
(Sommer & Wurtz, 2008)
. In other 
words, 
one interpretation of the current findings is 
that due to reduced 
connectivity between 
MDT and FEF, oculomotor CD
 signals 
may have been lost or disrupted
, resulting in 
poorer 
behavioral performance on the 
blanking 
task. Presumably, the other consequence of these 
compromised CD signals were frequent inaccurate predictions
 about visual inputs, which then 
led to constant false PEs that were ultimately 
Òresolved
Ó by abnormal perceptual experiences. 
This could potentially explain the correlation between 
the 
MDT
-FEF hypoconnectivity and 
positive symptom severity
 and the lack of
 correlation with negative symptom severity
. Because 
whole brain white matter connectivity differences were already regressed out in the 
correlation 
analyses, 
this relationship between white matter integrity and positive symptom w
as specific to 
the MDT
-FEF
 path
way
 rather than
 a result of general 
reduced structural 
connectivity
. Taken 
together, the MDT
-FEF white matter tracts seem to play a key role in transmitting oculomotor 
CD signals that may also explain some of the symptom manifestation in SZP.
 This hig
hlights the 
importance of thalamo
-cortical connectivity as a potential biomarker of and/or a treatment target 
for schizophrenia 
(Ramsay & MacDonald, 2018)
.  The in
terpretation
 of current findings is limited
 by several factors. 
First
, JND 
may 
not 
be 
the best indicator of CD 
signaling
 in the blanking task because it is also
 affected by
 general 
sources of noise (e.g. in 
visual input 
or in effector output)
. Indeed, it 
has been found that
 SZP 
have a wide range of
 visual 
processing
 deficits
 (reviewed in 
Yoon, Sheremata, Rokem, & Silver, 
 !22 2013). A more sensitive index of CD 
signaling
 would be the loss slope obtained from the 
blanking task. However,
 we did not find any correlations between the loss slope and the 
microstructural integrity indices.
 One potential explanation is that SZP may fail to encode the 
pre
-saccadic target locations properly due to impaired spatial working memory 
(Lee & Park, 
2005), resulting in noisier perceptual judgments on 
the 
single
-trial level.
 Consequently, due to a 
relatively small trial number
, the regression analyses 
may not be powerful enough 
to obtain the 
most accurate estimates of loss slopes.
 Future studies utilizing 
more trials 
may help 
resolve this 
issue
. Second
, we 
would expect that a
ltered CD should most directly relate to symptoms 
explained by agency loss; however, we 
did not 
include
 measures that assessed agency 
disturbances directly. Future studies should employ a more comprehensive and detailed measure 
on agency disruptions to tes
t a more fine
-grained hypothesis of the role 
that 
the MDT
-FEF 
pathway plays in the 
PE model of psychosis.
 Nevertheless, altered CD has been 
posited as a 
general mechanism underlying hallucinations and delusions 
(Corlett et al., 2016)
, which were 
both captured by the PANSS positive symptom score.
 Third, SZP in 
the 
current study 
were 
relatively high
-functioning and exhibit
ed a limited range of symptom severity. Future studies 
using a more clinically heterogeneous sample 
(e.g. recent
-onset and unmedicated patients) 
could 
further inform the relationships between thalamo
-cortical structural connectivity and clinical 
status in SZP. Lastly, we were not able to rule out potential con
founding effects of antipsychotic 
treatment because current study used a medicated sample. Ho
wever, the CPZ equivalent dose
 did 
not correlate with any of the behavioral or microstructural measures.
 We anticipate that findings 
from current study will pave t
he way for more in
-depth 
mechanistic questions on CD and psychosis
. In terms of future directions, studies utilizing 
larger
 sample size 
are necessary to 
obtain adequate
 power
 for regression analyses 
in which 
white matter 
 !23 connectivity indices, behavioral in
dices of oculomotor CD, and symptom
 severity 
could be 
entered into 
one 
model, allowing 
us to test the hypothesis that 
structural
 connectivity
 in the 
MDT
-FEF pathway directly mediates the relationship between oculomotor CD signaling 
abnormalities and positi
ve symptom severity
. Second, to 
better inform treatment development, a 
causal or directional relationship between white matter integrity, oculomotor CD signals, and 
psychotic symptoms is necessary. F
uture studies could try to establish directions in 
the br
ain
-behavior correlations observed in current study
 by following clinical high risk populations over 
time and measuring their white matter integrity and behavioral indices of CD 
signaling
 before 
and after the onset of illness.
 Third, it is important to know whether 
altered CD 
is 
a ph
enotype of 
psychosis in general or a specific biomarker of schizophrenia spectrum disorders.
 To this date
, only
 one study 
has found
 that CD 
in the auditory domain 
was similarly impaired in 
SZP
 and 
bipolar patients with a history of psychotic features
 (Ford et al., 2013)
. More replications in 
recent
-onset affective and non
-affe
ctive psychotic disorders, clinical high risk population
s, and 
unaffected first
-degree relatives are needed to address
 this question. 
The p
resence 
of altered CD 
in healthy relatives would provide ev
idence that CD abnormalities represent vulnerability 
towards psychosis, but 
that 
other factors are involved in the expression of full
-blown psychosis.
 In conclusion, we identified significant decreases in white matter connectivity in SZP 
relative to HC
 in 
the MDT
-FEF pathway, which belongs to a key neur
al 
circuit transmitting 
oculomotor CD signals previously established in
 the
 animal literature. We demonstrated that 
hypoconnectivity in the MDT
-FEF pathway was correlated with impaired oculomotor CD 
signals a
nd more severe positive symptoms in SZP.
 These findings have important disease 
mechanism and treatment implications.
    !24           APPENDICES
 !! !25 APPENDIX A: Tables
 Table 1. 
Demographic information
.   HC (
n = 24)
 SZP (
n = 22)
      Mean (SD)
 Mean (SD)
 Statistic
 p Age (Years)
 34.9 (8.1)
 37.4 (7.8)
 t = -1.1
 0.29
 Sex (Female/Male)
 11/13
 5/17
 !2 = 2.7
 0.13
 IQa 100.4 (13.8)
 96.0 (12.8)
 t = 1.1
 0.29
 Education
b 7.2 (1.3)
 4.8 (1.7)
 t = 5.2
 < .001
 Handedness
c 0.89 (0.41)
 0.85 (0.45)
 t = 0.35
 0.73
 Illness 
Duration (Years)
 14.2 (5.2)
     PANSS Positive
   11.6 (5.2)
     PANSS Negative
   13.1 (6.3)
     PANSS General
   25.0 (7.9)
     CPZ Equivalent (mg)
 281.3 (249.6)
      Notes: CPZ, chlorpromazine; HC, healthy control subjects; PANSS, Positive and 
Negative 
Syndrome Scale; SZP, patients with schizophrenia.
 aBased on the Nederlandse Leestest voor Volwassenen.
 bEducation category: 
0 = 
<6 years of primary education; 1 =
 finished 6
 years of primary 
education; 2 =
 6 years of primary education and low
-leve
l secondary education; 3 =
 4 years of 
low-level secondary education; 4 =
 4 years of averag
e-level secondary education; 5 =
 5 years of 
averag
e-level secondary education; 6 =
 4 years of se
condary vocational training; 7 =
 4 years of 
high
-level professional ed
ucation; 8 =
 university degree.
 cBased on the Edinburgh Handedness Inventory; scores range from 0 indicating complete left
-handedness to 1 indicating complete right
-handedness.
    !26 Table 
2. Mean microstructural integrity
 measures
 (FA, MD, RD, AD) in the MDT
-FEF tract by 
hemisphere.
    HC (
n = 24)
 SZP (
n = 22)
   Mean (SD)
 Mean (SD)
 FA Left
 0.41 (0.02)
 0.41 (0.02)
 Right
 0.42 (0.02)
 0.39 (0.02)
 MD Left
 7.05 
"
 10
-4 (2.40 
"
 10
-5) 7.18 
"
 10
-4 (2.65 
"
 10
-5) Right
 7.06 
"
 10
-4 (1.84 
"
 10
-5) 7.21 
"
 10
-4 (2.51 
"
 10
-5) RD Left
 5.43 
"
 10
-4 (2.57 
"
 10
-5) 5.56 
"
 10
-4 (2.46 
"
 10
-5) Right
 5.36 
"
 10
-4 (1.82 
"
 10
-5) 5.66 
"
 10
-4 (2.59 
"
 10
-5) AD Left
 1.03 
"
 10
-3 (3.72 
"
 10
-5) 1.04 
"
 10
-3 (3.70 
"
 10
-5) Right
 1.05 
"
 10
-3 (3.12 
"
 10
-5) 1.03 
"
 10
-3 (2.92 
"
 10
-5)  Notes: 
AD: 
axial diffusivity
; FA, 
fractional anisotropy
; FEF, 
frontal eye fields
; HC, healthy 
controls; MD, 
mean diffusivity
; MDT, 
mediodorsal thalamus
; RD,
 radial diffusivity
; SZP, 
schizophrenia patients.
    !27 APPENDIX B: Figures
  Figure 1. 
Blanking task.
 Dotted circles represent gaze position
s (Rısler et al., 2015)
.     !28  Figure 2. 
Blanking task perceptual judgment.
 Top:
 After the stimulus appears at the pre
-saccadic location, a corollary discharge (CD) vector can be computed accordingly when 
preparing for the saccade initiation. In this example, the predicted saccade landing site (grey 
cross) will fall short of the targe
t. Bottom: Upon saccade initiation, the stimulus will disappear 
and reappear to the left or right of the pre
-saccadic location. If participant can use CD to remap 
the pre
-saccadic location correctly, then vector 
a should match the actual stimulus displacem
ent. 
Here in this example, the post
-saccadic location is to the left of the pre
-saccadic location, so the 
participant should judge the displacement as backward. However, if participant does not have 
access to complete CD information, saccade landing site m
ay be used as a proxy of the pre
-saccadic location. In this case, participant will perceive vector 
b as the target displacement and 
give a forward response instead
 (Rısler et al., 2015)
.    !29  Figure 3
. Blanking task performance in chronic SZP and HC.
 Mean saccade amplitude was 
correlated with judgment on pre
-saccadic location of 
the
 stimulus in SZP, but not in HC 
(Rısler 
et al., 2015)
.    !30  Figure 4
. Blanking task performance
 and symptom severity
 in chronic SZP.
 When making 
perceptual judgments of target displacement, SZP with more severe p
ositive 
symptom 
showed 
greater reliance on actual saccade landing site, consistent with less influence of CD signals
 (Rısler et al., 2015)
.    !31  Figure 5
. Group
-average
d blanking task performance in chronic SZP and HC.
 Mean 
percentage of forward responses 
averaged over each group 
as a function of target displacement. 
On the x
-axis, negative values mean backward jumps (towards the fixation) and positive value
s mean forward jumps (away from the fixation) relative to the pre
-saccadic location 
(Rısler et al., 
2015).    !32  Figure 
6. Probabilistic tractography 
results of the MDT
-FEF pathway in both groups.
 Top 
panel: ROI masks
 projected onto a standard
-space brain in multi
-slice coronal view; bottom 
panel: group
-averaged and thresholded white matter tracts in standard space. ROIs are color
-coded as follows: FEF 
Ð green, MDT 
Ð yellow. Group tracts are color
-coded as follows: HC 
tracts 
Ð red, SZP tracts 
Ð blue, overlapping areas 
Ð pink. Coordinates of each slice in standard 
space were presented in the middle.
    !33  Figure
 7.  3D 
Probabilistic tractography results of 
the MDT
-FEF pathway in HC.
 3D-rendered representations of the group
-averaged and thresholded white matter tracts reverse 
normalized and overlaid on an HC participantÕs white matter skeleton, then projected onto the 
same participantÕs structural image using
 AFNI FATCAT Visualization. Results are essentially 
indistinguishable in SZP. ROIs are color
-coded as follows: FEF 
Ð red; MDT 
Ð yellow.
 Fiber tract 
reconstructions are colored according to the DTI visualization convention (blue: superior
-inferior, green: a
nterior
-posterior, red: left
-right).
  !34  Figure 
8. Group differences in the MDT
-FEF tract 
microstructural integrity by 
hemisphere.
 Top left: FA; 
top 
right:
 MD; bottom left: RD;
 bottom right AD
. AD: 
axial 
diffusivity
; FA, 
fractional anisotropy
; FEF, 
frontal 
eye fields
; HC, healthy controls; MD, 
mean 
diffusivity
; MDT, 
mediodorsal thalamus
; RD,
 radial diffusivity
; SZP, schizophrenia patients.
   !"#$!"#%!"&!!"&'!"&&!"&$!"&%()*+,,,,,,,,,,,,,,,,,,,,,,,,,,-./0+
123!3&3%()*+,,,,,,,,,,,,,,,,,,,,,,,,,,-./0+453$3'$%6,7!899'9&9%$'6,7!89()*+,,,,,,,,,,,,,,,,,,,,,,,,,,-./0+
-59$$!7!!7!977!25()*+,,,,,,,,,,,,,,,,,,,,,,,,,,-./0+
6,7!89 !35  Figure 
9. Scatterplots of JND against 
MDT
-FEF tract
-specific indices of microstructural 
integrity. 
Left: 
standardized residuals of MD acr
oss hemispheres
; right:
 standardized residuals 
of 
RD of
 the left tract
. These standardized
 values
 are 0 when equal to the whole brain 
MD or RD
 average, higher for more 
diffusive
 voxels and lower for 
less
 diffusive
 voxels.
 Smaller JND 
indicates more intact oculomotor CD signals. More compromised oculomotor CD signals 
correlated with more impaired white matter microstructure in SZP only. HC, healthy controls; 
JND, just noticeable difference; MD, 
mean diffusivity
; RD,
 radial 
diffusivity
; SZP, 
schizophrenia patients.
   !"!#!$%$#"%&%%&#%&'%&(%&)$&%*+,-.,/.012.3/240.5,63783+/,9+3:;<=;>?*@A!"!#!$%$#%&%%&#%&'%&(%&)$&%*+,-.,/.012.3/240.5,63783628+3+/,9+3B;<=; !36  Figure 
10. Scatterplots of PANSS positive symptoms scores against 
MDT
-FEF tract
-specific indices of microstructural integrity. 
Left: standardized residuals of MD across 
hemispheres; right: standardized residual
s of RD of the left tract.
 These standardized
 values
 are 0 
when equal to the whole brain 
MD or RD
 average, higher for more 
diffusive
 voxels and lower 
for 
less
 diffusive
 voxels.
 Larger PANSS scores indicate more severe symptoms. More severe 
positive symptom
s correlated with more impaired white matter microstructure. MD, 
mean 
diffusivity
; PANSS, 
Positive and Negative Syndrome Scale
; RD,
 radial diffusivity
.   !"!#$#"%#$#&"$"&'()*+),+-./+0,/1-+2)30450(,)6(0789:;''0<41-(-=/01>?<(4?10164,/!"!#$#"&#$#&"$"&'()*+),+-./+0,/1-+2)304503/5(0(,)6(0@89:;''0<41-(-=/01>?<(4?10164,/ !37 APPENDIX C: 
Supplementary Results
  Using the smaller subsample with behavioral data (see Table S1 for demographic 
information), the spatially normalized, group
-averaged, and statistically thresholded MDT
-FEF 
probabilistic tracts were very similar to those shown in Figure 6 and Figure 7. We 
performed a 
between
-group t
-test on 
whole
 brain probability values at each voxel. Again, no significant 
differences were found between the SZP and HC at the family
-wise error rate of 0.05, indicating 
that the MDT
-FEF pathway occupied the same spatial locat
ion in the brain.
 Next, 
we conducted four repeated measures AN
OVAs 
where
 one of the
 microstructural 
integrity
 measures
 (FA, MD, RD, AD) was the dependent variable each time
, using
 diagnostic 
group as 
a between
-subjects variable and 
hemisphere as 
a within
-subject variable. Means and 
standard deviations are presented in Table S2. Again, higher microstructural integrity is indicated 
by higher FA values and lower MD, RD, and AD values. For FA, MD, and RD, there was a 
significant effect of group such that 
SZP
 showed evidence of reduced microstructural integrity of 
the MDT
-FEF pathway (FA: 
F(1,33) = 23.03
, p < 0.001, 
partial 
!2 = 0.41
; MD: 
F(1,33) = 6.61
, p = 0.015, 
partial 
!2 = 0.17; RD:
 F(1,33) = 19.61
, p < 0.001, 
partial 
!2 = 0.37)
. There was no 
group difference in tract AD, 
F(1,33) = 0.34
, p = 0.57, 
partial 
!2 = 0.01
. For FA and RD, there 
was an additional hemisphere effect such that the MDT
-FEF pathway in the right hemisphere 
showed lower microstructural integrity than that in the 
left hemisphere across groups (FA: 
F(1,33) = 9.56
, p = 0.004, 
partial 
!2 = 0.23
; RD:
 F(1,33) = 4.26
, p = 0.047, 
partial 
!2 = 0.11). 
This hemisphere effect was not significant for MD, 
F(1,33) = 2.98
, p = 0.09, 
partial 
!2 = 0.08. 
For FA only, there was a group 
" hemisphere interaction effect, 
F(1, 33) = 7.97, 
p = 0.008, 
partial 
!2 = 0.19, such that the hemisphere effect was only significant in SZP, 
t(17) = 4.01, 
p = 0.001, but not in HC, 
t(16) = 0.20, 
p = 0.84
. This group 
" hemisphere interaction effect was not 
 !38 significant for MD or RD (MD: 
F(1,33) = 0.00
, p = 1.0, 
partial 
!2 = 0.00; RD: 
F(1,33) = 3.58
, p = 0.07, 
partial 
!2 = 0.10). There were no correlations between CPZ equivalent dose and any of 
the microstructural integri
ty measures
 (-0.49 < 
rs < 0.15, 0.06 <
 p < 0.77)
.    !39 Table 
S1. 
Demographic information
 of the subset sample with behavioral data
    HC (
n = 17)
 SZP (
n = 18)
      Mean (SD)
 Mean (SD)
 Statistic
 p Age (Years)
 34.1 (8.9)
 36.7 (8.2)
 t = -0.9
 0.37
 Sex 
(Female/Male)
 7/10
 2/16
 !2 = 4.1
 0.06
 IQa 99.2 (14.6)
 95.8 (12.2)
 t = 0.7
 0.48
 Education
b 7.2 (1.1)
 4.9 (1.8)
 t = 4.6
 < .001
 Handedness
c 0.85 (0.49)
 0.93 (0.20)
 t = -0.7
 0.49
 Illness Duration (Years)
   14.6 (5.4)
     PANSS Positive
   11.4 (5.2)
     PANSS Negative
   12.6 (6.1)
     PANSS General
   24.6 (7.9)
     CPZ Equivalent (mg)
   302.0 (257.7)
     !Notes: CPZ, chlorpromazine; HC, healthy control subjects; PANSS, Positive and Negative 
Syndrome Scale; SZP, patients with schizophrenia.
 aBased on 
the Nederlandse Leestest voor Volwassenen.
 bEducation category: 
0 = 
<6 years of primary education; 1 =
 finished 6
 years of primary 
education; 2 =
 6 years of primary education and low
-level secondary education; 3 =
 4 years of 
low-level secondary education; 
4 =
 4 years of averag
e-level secondary education; 5 =
 5 years of 
averag
e-level secondary education; 6 =
 4 years of se
condary vocational training; 7 =
 4 years of 
high
-level professional education; 8 =
 university degree.
 cBased on the Edinburgh Handedness In
ventory; scores range from 0 indicating complete left
-handedness to 1 indicating complete right
-handedness.
 !! !40 Table 
S2. Mean microstructural integrity
 measures
 (FA, MD, RD, AD) in the MDT
-FEF tract by 
hemisphere in the subset
!sample with behavioral data
.     HC (
n = 17)
 SZP (
n = 18)
   Mean (SD)
 Mean (SD)
 FA Left
 0.43 (0.02)
 0.41 (0.02)
 Right
 0.43 (0.02)
 0.39 (0.02)
 MD Left
 6.98 
"
 10
-4 (2.15 
"
 10
-5) 7.16 
"
 10
-4 (2.42 
"
 10
-5) Right
 7.01 
"
 10
-4 (1.92 
"
 10
-5) 7.20 
"
 10
-4 (2.30 
"
 10
-5) RD Left
 5.27 
"
 10
-4 (2.22 
"
 10
-5) 5.54 
"
 10
-4 (2.24 
"
 10
-5) Right
 5.28 
"
 10
-4 (1.87 
"
 10
-5) 5.63 
"
 10
-4 (2.31 
"
 10
-5) AD Left
 1.04 
"
 10
-3 (3.65 
"
 10
-5) 1.04 
"
 10
-3 (3.50 
"
 10
-5) Right
 1.05 
"
 10
-3 (3.31 
"
 10
-5) 1.03 
"
 10
-3 (2.95 
"
 10
-5) !Notes: 
AD: 
axial diffusivity
; FA, 
fractional anisotropy
; FEF, 
frontal eye fields
; HC, healthy 
controls; MD, 
mean diffusivity
; MDT, 
mediodorsal thalamus
; RD,
 radial diffusivity
; SZP, 
schizophrenia patients.
    !41            REFERENCES
    !42 REFERENCES
  Abe, O., Aoki, S., Hayashi, N., Yamada, H., Kunimatsu, A., Mori, H., É Ohtomo, K. (2002). 
Normal aging in the central nervous system: Quantitative MR diffusion
-tensor analysis. 
Neurobiology of Aging, 23(3),
 433
Ð441.  Alexander, A. L., Lee, J. E., Lazar, M., & Field, A. S. (2007). Diffusion tensor imaging of the 
brain. 
Neurotherapeutics, 4(3),
 316Ð329.  Andersson, J
. L. R., & Skare, S. (2002). A model
-based method for retrospective correction of 
geometric distortions in
 diffusion
-weighted EPI. 
NeuroImage, 16(1),
 177Ð199.
  Andreasen, N. C. (1999). A unitary model of s
chizophrenia. 
Archives of General Psychiatry, 
56(9),
 781Ð787.  Andreasen, N. C., Flaum, M., & Arndt, S. (1992). The Comprehensive Assessment of Symptoms 
and 
History (CASH). An instrument for assessing diagnosis and psychopathology. 
Archives 
of Gene
ral Psychiatry, 49(8),
 615Ð623.
  Anticevic, A., Cole, M. W., Repovs, G., Murray, J. D., Brumbaugh, M. S., Winkler, A. M., É 
Glahn
, D. C. (2014). Characterizing thala
mo-cortical disturbances in schizophrenia and 
bipolar i
llness. 
Cere
bral Cortex, 24(12),
 3116Ð3130.
  Anticevic, A., Haut, K., Murray, J. D., Repovs, G., Yang, G. J., Diehl, C., É Cannon
, T. D. 
(2015). Association of thalamic dysconnectivity and conversion t
o psychosis in youth and 
young adults at elevated clinical r
isk. 
JAMA Psychiatry, 72(9),
 882Ð891.
  Anticevic, A., Yang, G., Savic, A., Murray, J. D., Cole, M. W., Repovs, G., É Glahn,
 D. C. 
(2014). Mediodorsal and visual thalamic connectivity differ in sch
izophrenia and bipolar 
disorder with and without psychosis h
istory. 
Schizophrenia Bulletin, 40(6),
 1227Ð1243. 
  Ashburner, J., & Friston, K. J. (2005). Unified segmentation. 
NeuroImage, 26(3),
 839Ð851.  Behrens, T. E. J., Berg, H. J., Jbabdi, S., Rushworth, M. F. S., & Woolrich, M. W. (2007). 
Probabilistic diffusion tractography with multiple fibre orientations: What can we gain? 
NeuroImage, 34(1),
 144Ð155.  Behrens, T. E. J., Woolrich, M. W., Jenkinson, 
M., Johansen
-Berg, H., Nunes, R. G., Clare, 
S., É Smith, S. M. (2003). Characterization and propagation of uncertainty in diffusion
-weighted MR imaging. 
Magnetic Resonance
 in Medicine, 50(5),
 1077Ð1088.  Blakemore, S.
-J., Smith, J., Steel, R., Johnstone, E. C., & Frith, C. D. (2000). The perception of 
self
-produced sensory stimuli in patients with auditory hallucinations and passivity 
experiences: evidence for a breakdown in self
-monitoring. 
Psychological 
Medicine, 30(5),
  !43 1131Ð1139.  Canu, E., Agosta, F., & Filippi, M. (2015). A selective review of structural connectivity 
abnormalities of schizophrenic patients at different stages of the disease. 
Schizop
hrenia 
Research, 161(1),
 19Ð28.  Carrera, E., & Bogous
slavsky, J. (2006). The thalamus and behavior: Effects of anatomically 
distinct strokes
. Neurology, 66(12),
 1817Ð1823.
  Cavanagh, P., Hunt, A. R., Afraz, A., & Rolfs, M. (2010). Visual stability based on remapping of 
attention pointers. 
Trends in Cogni
tive
 Sciences, 14(4),
 147Ð153.  Cavanaugh, J., Berman, R. A., Joiner, W. M., &
 Wurtz, R. H. (2016). Saccadic corollary 
discharge underlies stable visual p
erception. 
Journal
 of Neuroscience, 36(1),
 31Ð42.  Collins, T., Rolfs, M., Deubel, H., & Cavanagh, P. (200
9). Post
-saccadic location judgments 
reveal remapping of saccade targets to non
-foveal locations
. Journal of Vision, 9(5),
 1Ð9.  Corlett, P. R., Honey, G. D., & Fletcher, P. C. (2016). Prediction error, ketamine and psychosis: 
An updated model. 
Journal of 
Psychopharmacology, 30(11),
 1Ð11.  Crail
-Melendez, D., Atriano
-Mendieta, C., Carrillo
-Meza, R., & Ramirez
-Bermudez, J. (2013). 
Schizophrenia
-like psychosis associated with right lacunar thalamic in
farct. 
Neurocase, 
19(1),
 22Ð26.  Cropley, V. L., Klauser, P
., Lenroot, R. K., Bruggemann, J., Sundram, S., Bousman, C., É 
Zalesky, A. (2017). Accelerated gray and white matter deterioration with age in 
schizophrenia. 
American Journal 
of Psychiatry, 174(3),
 286Ð295.  Davis, K. L., Stewart, D. G., Friedman, J. I., B
uchsbaum, M., Harvey, P. D., Hof, P. R., É
 Haroutunian, V. (2003). White matter changes in s
chizophrenia. 
Archives of General 
Psychiatry, 60(5),
 443-456.  Desikan, R. S., S”gonne, F., Fischl, B., Quinn, B. T., Dickerson, B. C., Blacker, D., É Killiany, 
R. 
J. (2006). An automated labeling system for subdividing the human cerebral cortex on 
MRI scans into gyral based regions of intere
st. 
NeuroImage, 31(3),
 968Ð980.
  Deubel, H., Schneider, W. X., & Bridgeman, B. (1996). Postsaccadic target blanking prevents 
saccadic suppression of image displacement. 
Vision Research, 36(7),
 985Ð996.
  Feinberg, I. (1978). Efference copy and corollary discharge: implications for thinking
 and its 
disorders. 
Schizophrenia Bulletin, 4(4),
 636Ð640.  Feinberg, I., & Guazzelli, M. (1999). Schizophrenia
--a disorder of the corollary discharge 
systems that integrate the motor systems of thought with the sensory systems of 
consciousness. 
The Britis
h Journal of Psychiatry, 17
4(3),
 196Ð204.  !44 Fletcher, P. C., & Frith, C. D. (2009). Perceiving is believing: a Bayesian approach to explaining 
the positive symptoms of schizophrenia. Nature Revie
ws. 
Neuroscience, 10(1),
 48Ð58.
  Ford, J. M., & Mathalon, D. H.
 (2012). Anticipating the future: Automatic prediction failures in 
schizophrenia. 
International Journal of Ps
ychophysiology, 83(2),
 232Ð239.
  Ford, J. M., Mathalon, D. H., Roach, B. J., Keedy, S. K., Reilly, J. L., Gershon, E. S., & 
Sweeney, J.
 A. (2013). 
Neurophysiological evidence of corollary discharge function during 
vocalization in psychotic patients and their nonpsychotic first
-degree r
elatives. 
Schizophrenia Bulletin, 39(6),
 1272Ð1280.  Gaymard, B., Rivaud, S., & Pierrot
-Deseilligny, C. (1994). Impai
rment of extraretinal eye 
position signals after central thalamic lesions in humans. 
Experimental Brain Research, 
102, 1Ð9.  Genetic Risk and Outcome in Psychosis (GROUP) I
nvestigators. (2011). Evidence that familial 
liability for psychosis is expressed as
 differential sensitivity to c
annabis: an analysis of 
patient
-sibling and sibling
-control pairs. 
Archives of Gene
ral Psychiatry, 68(2),
 138Ð147.
  Giraldo
-Chica, M., Rogers, B. P., Damon, S. M., Landman, B. A., & Wood
ward, N. D. (2018). 
Prefrontal
-thalamic 
anatomical connectivity and executive c
ognitive
 function in 
schizophrenia. 
Biologi
cal Psychiatry, 83(6),
 509Ð517.
  Gray, J. A., Feldon, J., Rawlins, J. N. P., Hemsley, D. R., & Smith, A. D. (1991). The 
neuropsychology of schizophrenia. 
Behavioral and Brain
 Sciences, 14(1),
 1Ð20.  Green, M. F., & K
insbourne, M. (1990). Subvocal activity and auditory hallucinations: clues for 
behavioral t
reatments? 
Schizoph
renia Bulletin, 16(4),
 617Ð625.
  Hallett, P. E., & Lightstone, A. D. (1976a). Saccadic eye movements to flashed targets. 
Vision 
Research, 16(1),
 107Ð114.  Hallett, P. E., & Lightstone, A. D. (1976b). Saccadic eye movements towards stimuli triggered 
by prior saccades. 
Vision Research, 16(1
), 99Ð106.
  Islam, F., Mulsant, B. H., Voineskos, A. N.
, & Rajji, T. K. (2017). Brain
-derived neurotrophic 
factor expression in individuals with s
chizophrenia a
nd healthy aging: Testing the 
accelerated aging hypothesis of s
chizophrenia. 
Current
 Psychiatry 
Reports, 19(7),
 36.  Kay, S. R., Fiszbein, A., & Opler, L. A. (1987). The Positive and Negative Syndrome Scale 
(PANSS) 
for s
chizophrenia. 
Schizophrenia Bulletin, 13(2),
 261Ð276.  Landman, B. A., Bogovic, J. A., Wan, H., ElShahaby, F. E. Z., Bazin, P.
-L., & Prince, J. L. 
(2012). Resolution of crossing fibers with constrained compressed sensing using diffusion 
tensor MRI. 
Neu
roImage, 59(3),
 2175Ð2186.   !45 Lee, J., & Park, S. (2005). Wor
king memory impairments in schizophrenia: A meta
-analysis. 
Journal of Abnormal Psychology, 114(4),
 599Ð611.
  Mori, S., & van Zijl, P. C. M. (2002). Fiber tracking: principles and strategies 
- a technical 
review. 
NMR in Biomedicine, 15(7
Ð8),
 468Ð480.
  Mori,
 S., & Z
hang, J. (2006). Principles of diffusion tensor imaging and its applications to basic 
neuroscience r
esearch. 
Neuron, 51(5),
 527Ð539.  Neggers, S. F. W., Zandbelt, B. B., Schall, M. S., & Schall, J. D. (2015). Comparative diffusion 
tractography of c
orticostriatal motor pathways reveals differences between humans and 
macaques. 
Journal of Neurophysiology, 
113(7),
 2164
Ð2172.  Ostendorf, F., Liebermann, D., & Ploner, C. J. (2010). Human thalamus contributes to perceptual 
stability across eye movements. 
Proceedings of the National Academy 
of Sciences, 107(3),
 1229Ð1234.  Pergola, G., Selvaggi, P., Trizio, S., Bertolino, A., & Blasi, G. (2015). The role of the thalamus 
in schizophrenia from a neuroimaging perspective. 
Neuroscience & Biobehavioral Reviews, 
54, 57Ð75.
  Ramsay, I. S., 
& MacDonald, A. W. (2018). The ups and downs of thalamocortical connectivity 
in s
chizophrenia. 
Biological Psychiatry, 83(6),
 473Ð474. 
  Rısler, L., Rolfs, M., van der Stigchel, S., Neggers, S. F. W., Cahn, W., Kahn, R. S., & Thak
kar, 
K. N. (2015). Failure to use corollary discharge to remap visual target locations is 
associated with psychotic symptom severity in schizophrenia. 
Journal of Neurophysiology, 
114(2),
 1129Ð1136.  Schneider, K. (1959). 
Clinical Psychopathology.
 New York,
 NY: Grune &Stratton.
  Shepherd, A. M., Laurens, K. R., Matheson, S. L., Carr, V. J., & Green, M. J. (2012). Systematic 
meta
-review and quality assessment of the structural brain alterations in schizophrenia. 
Neuroscience & Biobehavioral Reviews, 36(4),
 1342Ð1356. 
  Shergill, S. S., Samson, G., Bays, P. M., Frith, C. D., & Wolpe
rt, D. M. (2005). Evidence for 
sensory prediction deficits in s
chizophrenia. 
American Journal of Psychiatry, 162(12),
 2384Ð2386. 
  Sherman, S. M. (2016). Thalamus plays a central rol
e in ongoing cortical func
tioning. 
Nature 
Neuroscience, 19
(4),
 533Ð541.  Sim, K., Yang, G. L., Loh, D., Poon, L. Y., Sitoh, Y. Y., Verma, S., É Pantelis, C. (2009). 
White matter abnormalities and neurocognitive deficits associated with the passivity 
phenomenon in schizophrenia: A diffusion tensor imaging study. 
Psychiatry Resea
rch 
- Neuroimaging, 172(2),
 121Ð127.  !46 Smith, S. M., Jenkinson, M., Woolrich, M. W., Beckmann, C. F., Behrens, T. E. J., Johansen
-Berg, H., É Matthews, P. M. (2004). Advances in functional and structural MR image 
analysis and implementation as FSL. 
Neuro
Imag
e, 23(SUPPL. 1),
 S208
ÐS219.
  Snook, L., Plewes, C., & Beaulieu, C. (2007). Voxel based versus region of interest analysis in 
diffusion tensor imaging of neurodevelopme
nt. 
NeuroImage, 34(1),
 243Ð252.
  Sommer, M
. A., & Wurtz, R. H. (2002). A pathway in prima
te brain for internal monitoring of 
movements. 
Science, 296(5572),
 1480Ð1482.  Sommer, M. A., &
 Wurtz, R. H. (2004). What the brain stem tells the frontal c
ortex. II. Role of 
the SC
-MD-FEF pathway in corollary d
ischarge. 
Journal of Neurophysiology, 91(3),
 1403Ð1423.  Sommer, M. A., & Wurtz, R. H. (2008). Brain circuits for the internal monitoring of movements. 
Annual Review of Neuroscience, 31,
 317Ð338.  Song, S.
-K., Sun, S.
-W., Ramsbottom, M. J., Chang, C., Russell, J., & Cross
, A. H. (2002). 
Dysmyelinatio
n revealed through MRI as increased radial (but unchanged axial) diffusion 
of w
ater. 
NeuroImage, 17(3),
 1429Ð1436. 
  Thakkar, K. N., Diwadkar, V. A., & Rolfs, M. (2017)
. Oculomotor prediction: A window into 
the psychotic m
ind. 
Trends in Cognitive Sciences,
 21(5),
 344Ð356. 
  Thakkar, K. N., van den Heiligenberg, F. M. Z., Kahn, R. S., & Neg
gers, S. F. W. (2014). 
Frontal
-subcortical
 circuits involved in reactive control and monitoring of g
aze. 
Journal of 
Neuroscience, 34(26),
 8918Ð8929.  Welsh, R. C., Chen, A. C., & Taylor, S. F. (2010). Low
-frequency BOLD fluctuations 
demonstrate altered thalamocortical connectivity in schizophrenia. 
Schizophrenia Bulletin, 
36(4),
 713Ð722. 
  Wing, J. K., Babor, T., Brugha, T., Burke, J., Cooper, J. E., Gi
el, R., É Sartorius, N. (1990). 
SCAN: Schedules for Clinical Assessment in Neuropsychiatry. 
Archives of Gene
ral 
Psychiatry, 47(6),
 589Ð593.  Woods
, S. W. (2003). Chlorpromazine equivalent doses for the newer atypical a
ntipsychotics. 
The Journal of Clinical
 Psychiatry, 64(6),
 663Ð667.
  Woodward, N. D., Karbasforoushan, H., & Heckers, S. (2012). Thalamocortical dysconnectivity 
in schizophrenia. 
American Journal of 
Psychiatry, 169(10),
 1092Ð1099.  Yoon, J. H., Sheremata, S. L., Rokem, A., & Silver, M. A. (2013
). Windows to the soul: vision 
science as a tool for studying biological mechanisms of information processing deficits in 
schizophrenia. 
Frontiers in Psychology, 4
, 681
.   !47 Zhou, Y., Fox, D., Anand, A., Elhaj, A., Kapoor, A., Najibi, F., É Jaya
m-Trouth, A. 
(2015). 
Artery of percheron infarction as an unusual cause of KorsakoffÕs s
yndrome. 
Case Reports 
in Ne
urological Medicine, 2015,
 1Ð6.