Schizophrenia Research Forum - A Catalyst for Creative Thinking

Altered Connections Between Cortex and Striatum Mark Risk for Psychosis

September 17, 2013. Different signaling patterns between the cortex and striatum distinguish the brains of people with first-episode psychosis from those of control subjects, reports a study published online September 4 in JAMA Psychiatry. Moreover, the same circuit alterations are present in their healthy relatives.

Using resting-state functional magnetic resonance imaging (fMRI), Alex Fornito of Monash University in Melbourne, Australia, and colleagues at the University of Cambridge, U.K., examined patterns of co-activation between the striatum’s subregions and the cortical regions that target them. This revealed patterns of over- and under-coordination of activity in people with psychosis and those at genetic risk for it: In particular, dorsal regions of the striatum seemed decoupled from their cortical partners, whereas ventral regions showed overly tight connectivity with them, compared to healthy controls.

The striatum in focus
The findings suggest a vulnerability that may allow psychosis entry to the brain. Early on, prior to the development of full-blown psychosis, signs of aberrant brain signals can be detected in people at risk for psychosis (e.g., see SRF related news story). But it has been unclear whether this really indicates genetic risk or whether these abnormalities reflect the mild, intermittent psychotic experiences that put these people into the clinical at-risk category. In the new study, the researchers study people newly ill with psychosis, as well as their unaffected first-degree relatives; these people share some of the genetic risk for the disorder, but none of the psychosis. Anomalies detected in both groups, then, could reflect the risk state.

A recent study with this design implicates numerous brain networks (Khadka et al., 2013), and the new study focuses on the complicated connections between the striatum and cortex. The striatum has been seen as a locus of psychosis because its rich dopaminergic innervation makes it a candidate target of antipsychotic action. Previous work focused on the ventral parts of the striatum in psychosis, which receive inputs from limbic parts of the cortex as does the orbitofrontal cortex (OFC) involved in emotion and motivation. Recent studies, however, have suggested a role for the dorsal parts, also called the “associative” striatum, which receive inputs from cortical regions involved in cognition, such as the dorsolateral prefrontal cortex (DLPFC): Aberrant dopamine signaling localizes to the dorsal striatum in people showing mild signs of psychosis (see SRF related news story) and in people with schizophrenia (Kegeles et al., 2010; see also SRF Q&A with Anissa Abi-Dargham).

The new study examines the state of these dorsal and ventral striatal-cortical networks in psychosis and in people at genetic risk for psychosis. Using resting-state fMRI, the researchers measured the extent to which any two brain regions were simultaneously active while the study participant rested in the scanner: Regions that turn on and off in lockstep likely work together and are deemed to be functionally connected.

Seeds of psychosis
First author Alex Fornito and colleagues scanned 19 people with their first episode of psychosis, 25 of their unaffected first-degree relatives, and 26 healthy controls. The diagnoses of those with psychosis mainly consisted of schizophrenia or bipolar disorder. With the resulting images, the researchers selected six regions within the caudate and the putamen—the two main components of the striatum—spanning the dorsal and ventral parts. They then asked how activity in these predefined seed regions varied with activity in the rest of the brain.

To get at disease-related changes in how information flowed along the dorsal and ventral networks, the researchers compared those individuals with psychosis to controls. This revealed abnormally low functional connectivity in the dorsal network and abnormally high functional connectivity in the ventral network. Specifically, those with psychosis showed reduced connectivity between the dorsal caudate and DLPFC, but unusually heavy connectivity between the superior ventral caudate (sVC) and its limbic cortical partner, the OFC. The sVC also showed strong connectivity with the left DLPFC, a connection not seen at all in controls. This crossed connection suggests that the boundary between the ventral and dorsal systems is blurred in the brains of people who develop psychosis. A similar pattern held true for the dorsal and ventral parts of the putamen, and connectivity findings did not seem to vary with antipsychotic medication.

Risk regions
The unaffected relatives shared the same pattern of under-connectivity in the dorsal network and over-connectivity in the ventral network, suggesting this pattern marks a brain vulnerable to psychosis. Specifically, compared to controls, unaffected relatives showed weak functional connectivity between the dorsal caudate and DLPFC, and heavy connectivity between the sVC and the OFC. A similar gradient of connectivity was seen for the putamen regions. The magnitude of this over- and under-connectivity did not differ from that found in people with psychosis. Yet one key difference emerged: The unaffected relatives did not show a crossed connection between the ventral striatum and the DLPFC, which suggests that transitioning from risk to actual illness involves problems in additional circuits.

The researchers suggest that diminished dorsal connectivity triggers the ventral network problems; for example, enhanced dopamine signaling detected in dorsal striatum could promote noisy signals there, resulting in disjointed signaling between cortex and dorsal striatum, which could compromise top-down cognitive processes. This, in turn, may allow for runaway crosstalk between the ventral striatum and its cortical partners, which could lead to aberrant salience, a process that assigns importance to irrelevant stimuli and so could provide the grist for psychosis (Kapur, 2003).

Consistent with this, among people with psychosis, the weaker the connections between dorsal caudate and frontal cortical regions, the worse their psychotic symptoms, as measured by the positive symptom scores on the Brief Psychiatric Rating Scale (R = -0.53, p = 0.2). Although a positive correlation was also found between ventral corticostriatal network connectivity strength and positive symptom severity, this was not significant. The researchers also reported that lower connectivity between the dorsal caudate and the left DLPFC in particular correlated with worse negative symptom scores, which suggests this connection may be particularly relevant to schizophrenia.

Despite the tangle of connections between striatum and cortex, the research suggests some organization to the aberrant connectivity that appears in mental illness. This may stem from deviations in early brain development, which seems to follow a fairly basic blueprint (see SRF related news story). Future research will have to probe how genetic or environmental factors prod a vulnerable brain to full-blown illness.—Michele Solis.

Fornito A, Harrison BJ, Goodby E, Dean A, Ooi C, Nathan PJ, Lennox BR, Jones PB, Suckling J, Bullmore ET. Functional Dysconnectivity of Corticostriatal Circuitry as a Risk Phenotype for Psychosis. JAMA Psychiatry. 2013 Sep 4. Abstract

Comments on Related News

Related News: High Dopamine Levels in People With Evidence of Prodromal Schizophrenia

Comment by:  Anissa Abi-Dargham, SRF Advisor
Submitted 28 January 2009
Posted 28 January 2009

This paper introduces new knowledge and at the same time replicates many of the themes that have emerged in the area of dopamine research and schizophrenia. The new knowledge is that DA dysregulation precedes onset, and is present in the same anatomical area and with a similar magnitude to that observed in schizophrenia. It shows once again that the dopamine dysregulation in schizophrenia is one of the most replicated and consistent findings in the field.

The area of pathology within the striatum is, as described in schizophrenia, the associative, rather than the limbic or sensorimotor striatum (Kegeles et al., 2006). This represents the main area of projection of the DLPFC. Furthermore, this part of the striatum receives input from other limbic cortical areas (Haber et al., 2006) and may play a role in integrating emotions and cognition. Alterations in this integrative function could lead to misattributions or mislabelings leading to paranoia or “inappropriate” affect, in addition to the cognitive deficits.

The relationship to baseline psychotic symptoms, present for the prodrome but not detected in schizophrenia, as we have previously shown, too (Laruelle et al., 1999), is worth commenting on. It suggests that dopamine may play an essential role in the genesis of psychotic symptoms. This role may be most predominant early on in the disease, rather than later in the course of the illness. Later, symptoms may perpetuate for other reasons. Alternatively, it may be that their partial improvement due to treatment prior to the scan may make the relationship more difficult to detect, and that relationship would be detected if studies included only drug naïve patients. Either way, the relationship to severity of symptoms in the prodromal stage suggests that dopamine dysregulation may be an early part of the pathogenetic pathway, and not a subsequent consequence of other events. This is strengthened by the relationship to deficits on a verbal fluency task, suggesting again a general role for dopamine in the genesis of all symptom domains.

The next questions that emerge from these data are, How early in life is the DA alteration present and what are its effects on the developing brain and the relevant circuitry? As D2 overexpression in the striatum during development resulted in long-lasting alterations in cortical dopamine and cortical function in mice (Kellendonk et al., 2006), it may be that early alterations of D2 signaling within the associative striatum have effects on other targets within the associative striatum and along the whole integrated cortico-striato-thalamocortical circuit that become obvious only at onset. Identifying and halting this process early on are needed in order to prevent long-lasting deficits. In this regard, the paper does not present evidence that increased dopamine predicts conversion, which would be needed if one should propose a test of striatal dopamine for therapeutic intervention. For this to happen, large-scale imaging studies showing sensitivity and specificity of [18F]-DOPA as a biomarker for conversion are needed, similarly to studies conducted now in the field of Alzheimer disease using PET and tracers to label β amyloid.


Kegeles L, Frankle W, Gil R, et al. Schizophrenia is associated with increased synaptic dopamine in associative rather than limbic regions of the striatum: implications for mechanisms of action of antipsychotic drugs. J Nucl Med. 2006(47):139P.

Haber SN, Kim KS, Mailly P, Calzavara R. Reward-related cortical inputs define a large striatal region in primates that interface with associative cortical connections, providing a substrate for incentive-based learning. J Neurosci. Aug 9 2006;26(32):8368-8376. Abstract

Laruelle M, Abi-Dargham A, Gil R, Kegeles L, Innis R. Increased dopamine transmission in schizophrenia: relationship to illness phases. Biol Psychiatry. 1999;46(1):56-72. Abstract

Kellendonk C, Simpson EH, Polan HJ, et al. Transient and selective overexpression of dopamine D2 receptors in the striatum causes persistent abnormalities in prefrontal cortex functioning. Neuron. Feb 16 2006;49(4):603-615. Abstract

View all comments by Anissa Abi-Dargham

Related News: Signs of Things to Come? Seeking Biomarkers for the Schizophrenia Prodrome

Comment by:  Thomas McGlashan
Submitted 21 January 2012
Posted 21 January 2012

Three very recent publications have detailed that biomarkers can identify and quantify high-risk or prodromally symptomatic subjects who subsequently undergo conversion to psychosis. McGuire and his group (Howes et al., 2011) used fluoro-dopa positron emission tomography scanning to measure dopamine synthesis. Koutsouleris et al. (Koutsouleris et al., 2011) used structural MRI data to develop a neuroanatomical classification system for predicting psychosis conversion, and Amminger et al. (Amminger et al., 2011) used capillary gas chromatography of erythrocyte membrane fatty acid levels to provide information about brain phospholipids. All measures were significantly successful in identifying high-risk or prodromally symptomatic subjects who went on to develop a first episode of psychosis.

These papers point to an exciting future in our efforts to elaborate easily identifiable risk factors that can pinpoint among clinically identified "prodromal" subjects those who are most likely to become psychotic. That such diverse measures proved to be successful in identifying "true positives" can be regarded as a milestone in this line of investigation. It represents a quantitative leap forward in our ability to reduce the uncertainty of predicting psychosis, and hints at the day when tragedies such as the one occurring in Tucson, Arizona, become a thing of the past.


Howes OD, Bose SK, Turkheimer F, Valli I, Egerton A, Valmaggia LR, Murray RM, McGuire P. Dopamine Synthesis Capacity Before Onset of Psychosis: A Prospective [18F]-DOPA PET Imaging Study. Am J Psychiatry. 2011 Dec 1; 168: 1311-1317. Abstract

Koutsouleris N, Borgwardt S, Meisenzahl EM, Bottlender R, Möller HJ, Riecher-Rössler A. Disease Prediction in the At-Risk Mental State for Psychosis Using Neuroanatomical Biomarkers: Results From the FePsy Study. Schizophr Bull. 2011 Nov 10. Abstract

Amminger GP, Schäfer MR, Klier CM, Slavik JM, Holzer I, Holub M, Goldstone S, Whitford TJ, McGorry PD, Berk M. Decreased nervonic acid levels in erythrocyte membranes predict psychosis in help-seeking ultra-high-risk individuals. Mol Psychiatry. 2011 Dec 20. Abstract

View all comments by Thomas McGlashan