Stable signatures of schizophrenia in the cortical–subcortical–cerebellar network using fMRI of verbal working memory
Introduction
Schizophrenia involves a variety of symptoms that exhibit heterogeneous manifestations across patients in neuroimaging research, which prevents tangible imaging diagnoses. Nevertheless, general imaging markers of the illness have been proposed. One of the most frequently referred features is “hypofrontality”, i.e., reduced involvement of the prefrontal cortex (Potkin et al., 2009, Thormodsen et al., 2011). Another marker, “failure of deactivation,” was uncovered during the recent advancement in neuroimaging, specifically functional magnetic resonance imaging (fMRI), and demonstrates less pronounced signal decreases during the task period in a set of brain regions known as the “default mode network (DMN)” in patients compared to controls (Pomarol-Clotet et al., 2008, Whitfield-Gabrieli et al., 2009, Pomarol-Clotet et al., 2010, Guerrero-Pedraza et al., 2012). Functional connectivity in the DMN is also altered in schizophrenia (Ongur et al., 2010). Furthermore, another marker is “failure of lateralization,” which demonstrates bilateral activation, rather than the typical leftward asymmetry during language processing (Gur et al., 1989, Crow, 1997, Sommer et al., 2001, Mitchell and Crow, 2005, Andreasen et al., 2008, Angrilli et al., 2009, Bleich-Cohen et al., 2009). In this study, we propose another signature that is closely associated with an intrinsic impairment in schizophrenia, i.e., dysfunction in working memory (WM). In addition to the previously listed markers, we investigated the effect of verbal working memory (VWM) on patients, as well as controls, and aimed to obtain stable imaging signatures of schizophrenia that will benefit clinical diagnoses.
A dysfunction in WM is a core cognitive impairment in schizophrenia (Goldman-Rakic, 1994, Conklin et al., 2000, Manoach, 2003, Lee and Park, 2005, Honey and Fletcher, 2006, Reichenberg and Harvey, 2007). WM is associated with the cortical–subcortical–cerebellar system that involves cortical areas including the inferior frontal gyrus (IFG), and subcortical regions, including the thalamus and pons, and cerebellum (Desmond et al., 1997, Chen and Desmond, 2005a, Chen and Desmond, 2005b, Passamonti et al., 2011, Strick et al., 2009); these regions were often altered in schizophrenia (Andreasen et al., 1996, Honey et al., 2005, Rusch et al., 2007, Andreasen and Pierson, 2008, Mouchet-Mages et al., 2011). We employed an established Sternberg VWM paradigm of fMRI (Desmond et al., 1997, Chen and Desmond, 2005a, Chen and Desmond, 2005b, Marvel and Desmond, 2010); a prolonged maintenance of WM in the paradigm was advantageous because it required intensive cognitive resources including parallel processing, which were altered in schizophrenia (Andreasen et al., 1998). The application of a verbal task was favorable because language includes timing processing that is deteriorated in schizophrenia (Andreasen et al., 1999, Waters and Jablensky, 2009), and it also enabled the examination of the failure of lateralization in patients. Another advantage of our paradigm was its efficiency in examining the failure of deactivation in the DMN; signal intensity changes during the baseline period contrasting to the task period were compared between patients and controls (see Supplementary materials for detailed explanation).
One clinical concern in seeking stable signatures of schizophrenia is the duration of illness (DoI) (Premkumar et al., 2008, Elsabagh et al., 2009, Tanskanen et al., 2010). A long-term DoI may include side effects of the prolongation, which are independent of the illness. However, it is less well known to what extent the DoI affects WM-related activation. We extracted short and long DoI groups from our patient subjects to compare their activation.
Using whole-brain fMRI, we expected to observe lowered activation in the cortical–subcortical–cerebellar network and other indicators of illness in an all-in-one manner using the VWM paradigm. If these alterations were specific to the illness, the DoI would not affect these alterations. Taken together, these observations contributed to the characterization of schizophrenia and the attainment of stable diagnostic signatures of the illness.
Section snippets
Participants
Forty-six outpatients diagnosed with schizophrenia (SZ, DoI ≥ 1 year, 23 females, 23 males) and 46 healthy volunteers (CT) matched for age, gender and performance accuracy were examined with obtaining written informed consent (Table 1). The protocol was approved by the University Hospital Institutional Review Board in adherence with the Declaration of Helsinki. All patients were diagnosed by the attending psychiatrists in the Department of Psychiatry at the National Taiwan University Hospital as
Behavioral data
As participants were matched on task performance, neither the RT nor the accuracy was significantly different between patients and controls (Table 1) or between the long and short DoIs (Table 2).
Image data
The VWM contrast, i.e., high > low WM loads, yielded activation typically observed during VWM, including the IFG, intraparietal cortex, thalamus and cerebellum (Fig. 2). These areas are key regions of the cortical–subcortical–cerebellar network. A group comparison (CT > SZ) revealed a greater activation in
Promising signatures of schizophrenia
We confirmed previously proposed signatures of schizophrenia, i.e., a reduced activation in the cortical–subcortical–cerebellar network, a “failure of deactivation” in the DMN and a “failure of lateralization” in the posterior IFG. We demonstrated that the DoI did not affect these changes in general. Thus, these changes were independent of the DoI and viable as stable WM-related markers of schizophrenia.
Reductions in the cortical–subcortical–cerebellar network were particularly pronounced in
Role of the funding source
- 1.
Schizophrenia Project undertaken at the National Taiwan University Hospital.
Grant numbers: NTU97HM00273: National Taiwan University
NSC100-3112-B-002-016, NSC99-3112-B-002-030 (98HD017): National Science Council, Taiwan
NHRI-EX101-10145NI: National Health Research Institute, Taiwan
- 2.
Contributions from SHA Chen were in part supported by Tier2 Academic Research Fund (MOE2011-T2-1-031), Ministry of Education, Singapore.
The funding above was used for the machine usage charge, reward to subjects,
Contributors
Kayako Matsuo participated in all the MRI data collections and performed all the analysis and main interpretation of the data and wrote the draft of the manuscript.
Shen-Hsing Annabel Chen conceptualized and designed the fMRI experimental design and substantially determined the directions of the experiment, interpretation and discussion.
Chih-Min Liu has a principal responsibility in the interpretation of the results in regard to the clinical aspect of schizophrenia as a representative of the
Conflict of interest
We have no conflict of interest in regard to this study.
Acknowledgments
This study was supported by the Schizophrenia Project undertaken at the National Taiwan University Hospital (NTU97HM00273). It was also supported by the National Science Council, Taiwan (NSC100-3112-B-002-016, NSC99-3112-B-002-030 (98HD017)) and the National Health Research Institute, Taiwan (NHRI-EX101-10145NI). Contributions from SHA Chen were in part supported by Tier2 Academic Research Fund (MOE2011-T2-1-031), Ministry of Education, Singapore. We would like to specifically express our
References (73)
- et al.
The role of the cerebellum in schizophrenia
Biol. Psychiatry
(2008) - et al.
Defining the phenotype of schizophrenia: cognitive dysmetria and its neural mechanisms
Biol. Psychiatry
(1999) - et al.
Antipsychotic dose equivalents and dose-years: a standardized method for comparing exposure to different drugs
Biol. Psychiatry
(2010) - et al.
Reduced language lateralization in first-episode schizophrenia: an fMRI index of functional asymmetry
Psychiatry Res.
(2009) - et al.
Default-mode brain dysfunction in mental disorders: a systematic review
Neurosci. Biobehav. Rev.
(2009) - et al.
Self-projection and the brain
Trends Cogn. Sci.
(2007) - et al.
Cerebrocerebellar networks during articulatory rehearsal and verbal working memory tasks
Neuroimage
(2005) - et al.
Temporal dynamics of cerebro-cerebellar network recruitment during a cognitive task
Neuropsychologia
(2005) - et al.
A longer duration of schizophrenic illness has sex-specific associations within the working memory neural network in schizophrenia
Behav. Brain Res.
(2009) - et al.
Laterality and frontality of cerebral blood flow and metabolism in schizophrenia: relationship to symptom specificity
Psychiatry Res.
(1989)
Investigating principles of human brain function underlying working memory: what insights from schizophrenia?
Neuroscience
The representation of temporal information in perception and motor control
Curr. Opin. Neurobiol.
Reduced phonological similarity effects in patients with damage to the cerebellum
Brain Lang.
An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets
Neuroimage
Prefrontal cortex dysfunction during working memory performance in schizophrenia: reconciling discrepant findings
Schizophr. Res.
AveLI: a robust lateralization index in functional magnetic resonance imaging using unbiased threshold-free computation
J. Neurosci. Methods
The assessment and analysis of handedness: the Edinburgh Inventory
Neuropsychologia
Default mode network abnormalities in bipolar disorder and schizophrenia
Psychiatry Res.
Medial prefrontal default-mode hypoactivity affecting trait physical anhedonia in schizophrenia
Psychiatry Res.
Association between a longer duration of illness, age and lower frontal lobe grey matter volume in schizophrenia
Behav. Brain Res.
The critical relationship between the timing of stimulus presentation and data acquisition in blocked designs with fMRI
NeuroImage
Time perception disorders are related to working memory impairment in schizophrenia
Psychiatry Res.
Prefrontal–thalamic–cerebellar gray matter networks and executive functioning in schizophrenia
Schizophr. Res.
Dysmetria of thought: clinical consequences of cerebellar dysfunction on cognition and affect
Trends Cogn. Sci.
Language lateralization in schizophrenia, an fMRI study
Schizophr. Res.
Evidence for topographic organization in the cerebellum of motor control versus cognitive and affective processing
Cortex
Functional topography of the cerebellum for motor and cognitive tasks: an fMRI study
Neuroimage
Prefrontal hyperactivation during a working memory task in early-onset schizophrenia spectrum disorders: an fMRI study
Psychiatry Res.
Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain
Neuroimage
Time discrimination deficits in schizophrenia patients with first-rank (passivity) symptoms
Psychiatry Res.
Structure and function of auditory cortex: music and speech
Trends Cogn. Sci.
Schizophrenia and cognitive dysmetria: a positron-emission tomography study of dysfunctional prefrontal–thalamic–cerebellar circuitry
Proc. Natl. Acad. Sci. U. S. A.
“Cognitive dysmetria” as an integrative theory of schizophrenia: a dysfunction in cortical–subcortical–cerebellar circuitry?
Schizophr. Bull.
Theory of mind and schizophrenia: a positron emission tomography study of medication-free patients
Schizophr. Bull.
Schizophrenia as failure of left hemispheric dominance for the phonological component of language
PLoS One
The brain's default network: anatomy, function, and relevance to disease
Ann. N. Y. Acad. Sci.
Cited by (17)
Abnormal functional connectivity of motor circuit in the schizophrenic patients with tardive dyskinesia: A resting-state fMRI study
2021, Neuroscience LettersCitation Excerpt :It may be caused by neurodevelopmental abnormalities present before disease onset and/or susceptibility of neurons in these two areas to antipsychotic-induced neurodegeneration in patients with TD. The postcentral and the inferior frontal gyri of the triangular part belong to the cortical-subcortical-cerebellar network [35]. The cerebellum assumes a vital role in motor processing, and it may be related to the pathogenesis of TD.
How do abnormalities in the brain's spontaneous activity translate into symptoms in schizophrenia? From an overview of resting state activity findings to a proposed spatiotemporal psychopathology
2016, Progress in NeurobiologyCitation Excerpt :Are these changes in the cerebellum and cortex related to cognitive deficits? Matsuo (Matsuo et al., 2013) conducted an imaging study during a verbal working memory task. They observed decreased activation in cerebellum (vermis), default-mode network, pons, thalamus and lateral prefrontal cortex in schizophrenia.
The use of dynamic susceptibility contrast (DSC) MRI to automatically classify patients with first episode psychosis
2015, Schizophrenia ResearchCitation Excerpt :We restricted the regions we considered to left and right frontal, parietal, temporal and occipital lobes, insula, caudate and cerebellum. We selected these ROIs on the basis of the knowledge that the cortical–subcortical–cerebellar network results impaired in schizophrenia (Andreasen and Pierson, 2008) for example in relation to dysfunctions in working memory (Matsuo et al., 2013). Dysregulation of corticostriatal circuitry has also been found in FEP patients, involving striatum, including caudate, and insula (Fornito et al., 2013), which has been found to play a role also in schizophrenia (Corradi-Dell'Acqua et al., 2012).
Hyper-coupling between working memory task-evoked activations and amplitude of spontaneous fluctuations in first-episode schizophrenia
2014, Schizophrenia ResearchCitation Excerpt :We also found decreased correlations between the task-evoked activation and sensitivity in the bilateral cerebellum and the left fusiform gyrus in the FES patients. Even though the role of the cerebellum in the WM deficit in schizophrenia has been suggested (Matsuo et al., 2013; Sapara et al., 2014), the altered task-evoked activation–WM performance needs to be investigated in the future using a parametric WM task with the 3-back condition. Fourthly, we did not replicate the correlation between the ALFF and WM task-evoked deactivations, which had been reported in a previous study, in the healthy population (Zou et al., 2012).