Fronto-striatal Dysfunction During Reward Processing in Unaffected Siblings of Schizophrenia Patients

Abstract

Schizophrenia is a psychiatric disorder that is associated with impaired functioning of the fronto-striatal network, in particular during reward processing. However, it is unclear whether this dysfunction is related to the illness itself or whether it reflects a genetic vulnerability to develop schizophrenia. Here, we examined reward processing in unaffected siblings of schizophrenia patients using functional magnetic resonance imaging. Brain activity was measured during reward anticipation and reward outcome in 27 unaffected siblings of schizophrenia patients and 29 healthy volunteers using a modified monetary incentive delay task. Task performance was manipulated online so that all subjects won the same amount of money. Despite equal performance, siblings showed reduced activation in the ventral striatum, insula, and supplementary motor area (SMA) during reward anticipation compared to controls. Decreased ventral striatal activation in siblings was correlated with sub-clinical negative symptoms. During the outcome of reward, siblings showed increased activation in the ventral striatum and orbitofrontal cortex compared to controls. Our finding of decreased activity in the ventral striatum during reward anticipation and increased activity in this region during receiving reward may indicate impaired cue processing in siblings. This is consistent with the notion of dopamine dysfunction typically associated with schizophrenia. Since unaffected siblings share on average 50% of their genes with their ill relatives, these deficits may be related to the genetic vulnerability for schizophrenia.

Introduction

Schizophrenia is a highly heritable psychiatric disorder characterized by positive symptoms such as delusions and hallucinations, negative symptoms including affective flattening, as well as cognitive impairments.¹ Underlying these symptoms may be dysfunctions in the frontal lobe and the striatum.^2–8 Indeed, functional magnetic resonance imaging (fMRI) studies have demonstrated abnormal fronto-striatal activity in the context of various cognitive tasks,^2–4,9–15 in particular those that require the processing of rewards.^16–21

Reward processing can be divided into at least 2 sub-processes: anticipation of reward and receipt of reward,²² both of which are consistently reported to be abnormal in schizophrenia.^16–21,23 Functional MRI studies have revealed that in the healthy human brain the striatum is most active during reward anticipation, while during receipt of reward the frontal cortex, particularly the orbitofrontal cortex, becomes most activated.^22,24–26 However, in schizophrenia patients, several studies reported blunted ventral striatum activation during reward anticipation,^16–21 whereas 1 study showed increased activation in the orbitofrontal cortex during receipt of reward.²⁰ Moreover, these impairments, in particular decreased ventral striatal activity, have been associated with symptom severity in schizophrenia patients.^{16–18,21,27} However, as of yet it remains unclear whether these abnormal activations are related to the illness itself or to a genetic vulnerability for the disorder.

Therefore, we investigated reward processing in non-affected siblings. These siblings do not have the illness, but share on average 50% of their genes with their ill relative^28,29 and have a 10-fold increased risk to develop schizophrenia.³⁰ Consequently, if siblings show impaired reward processing similar to that observed in patients, this would provide evidence in support for a genetic vulnerability underlying this phenotypic abnormality. Indeed, 1 recent study in first-degree relatives of schizophrenia patients reported on attenuated ventral striatal activity during reward anticipation.³¹ However, relatives of several generations were used, including siblings, parents, and adolescent offspring, ranging in age between 18 and 50. Moreover, they only tested activation in the ventral striatum during reward anticipation. Other regions as well as receipt of reward were not investigated. Impaired reward processing in siblings is anticipated given reports on fronto-striatal dysfunction during other cognitive functions such as inhibitory control,^2,9 emotion processing,³² and working memory.^33,34

Here, we measured brain activity using fMRI during reward processing in 27 unaffected siblings of schizophrenia patients and 29 matched healthy controls. All subjects performed a modified monetary incentive delay task.^35–37 Activation during reward anticipation and receipt of reward was investigated in regions involved in processing of reward: the bilateral ventral striatum, dorsal striatum, insula, SMA, and orbitofrontal cortex.^{22,36,38–40}

As outlined above, we hypothesized that siblings show fronto-striatal deficits during reward processing comparable to that of schizophrenia patients. Specifically, we hypothesized that, similar to patients, activation in the ventral striatum is decreased during reward anticipation. Furthermore, since schizophrenia patients show hyperactivation of the orbitofrontal cortex during receipt of reward,²⁰ we hypothesized that activity in this region is likely increased in siblings. Finally, because schizophrenia patients show associations between decreased ventral striatum activity and symptom severity, we hypothesize that reduced activation of the ventral striatum in siblings may be associated with sub-clinical symptoms as measured with the Community Assessment of Psychic Experiences (CAPE).⁴¹

Methods

Participants

Twenty-seven unaffected siblings of patients with schizophrenia and 29 healthy control subjects participated in this study (table 1). All individuals were participating in an ongoing longitudinal study at the Department of Psychiatry at the University Medical Center Utrecht.⁴² All subjects were right-handed. None of the participants received psychotropic medication, had any contraindications for MRI, suffered from alcohol or drug dependence, or had a neurological diagnosis. Sub-clinical symptoms in all participants were measured using the CAPE (see supplementary material for details).⁴¹ Seven siblings had a history of at least 1 depressive episode as verified by either the Mini International Neuropsychiatric Interview⁴³ or the Schedules for Clinical Assessment in Neuropsychiatry (SCAN 2.1).⁴⁴ None of the healthy controls had a history of psychiatric diagnosis. Healthy control subjects who had a first-degree relative suffering from a psychotic disorder were excluded. Participants received monetary compensation for participation. All gave written informed consent. The ethics committee of the University Medical Center of Utrecht approved this study.

Table 1.

Demographic, Clinical and Behavioral Characteristics of the Diagnostic Groups

	Healthy Controls (n = 29)	Siblings (n = 27)	P
Age (y)	30.3±1.7	31.7±1.2	.51
Gender (M/F)	12/17	14/13	.37
Edinburgh handedness	0.89±0.03	0.93±0.02	.22
IQ (WAIS)	117±3	119±2	.71
Participant’s education levela	7.0±0.22	7.0±0.23	.89
Father’s education level	5.4±0.40	5.2±0.45	.78
Mother’s education level	4.8±0.41	5.4±0.52	.18
Depression in medical history	0	7
Cigarette smokers	1	8	.01
Cigarettes per day	2±0	10±1	.07
Cannabis users for the last 12 mo	3	8	.07
CAPE positive dimension (range)	0.14±0.03 (0–0.4)	0.21±0.05 (0–1.1)	.43
CAPE negative dimension (range)	0.69±0.05 (0.4–1.0)	0.60±0.06 (0–1.5)	.44
CAPE depressive dimension (range)	0.81±0.06 (0.4–1.4)	0.86±0.08 (0–1.9)	.70
Amount won (€)b	14.41±0.21	14.76±0.19	.23
Reaction time potentially rewarding trials (ms)	303±7	311±7	.41
Reaction time nonrewarding trials (ms)	323±8	329±9	.60

Note: Values represent mean ± SEM. IQ, intelligence quotient; WAIS, Wechsler Adult Intelligence Scale; CAPE, Community Assessment of Psychic Experiences.

^a Level of education was measured on a 9-point scale ranging from no education (0) to university degree (8).

^b Note that the maximum amount participants could win was €15.

Monetary Incentive Delay Task

Participants performed a reward task^35–37 (figure 1) based on the monetary incentive delay task.²² This task allows the investigation of anticipation and receipt of reward, separately. At the beginning of each trial, a cue is presented for 750 ms signaling whether the subject can win money (potentially rewarding trial) or not (nonrewarding trial). For the potentially rewarding trials, this cue is a smiling face and for the nonrewarding trials a neutral face. Following this cue, subjects have to respond as fast as possible, by pressing a button, when a target stimulus (exclamation mark) appears on the screen. Subsequent feedback notifies participants of their performance, indicating if they have earned money on the trial, as well as their cumulative total at that moment. Subjects can win €1 during a potentially rewarding trial.

Fig. 1.

Schematic representation of the reward task, based on the monetary incentive delay task. There were 2 types of trials: a potentially rewarding (A) and a nonrewarding trial (B). The inter-trial-interval ranged from 1029 to 6979ms.

For both potentially rewarding and nonrewarding trials, subjects have to respond to the target stimulus within a certain time limit, ie, target duration. Responses are considered correct (correct feedback) if subjects responds within this time limit. Responses given after the time limit are considered incorrect (incorrect feedback).

The time limit is individually adjusted to ensure that each participant succeeds in 50% of the trials. This adjustment is based on 20 practice trials which are presented prior to the start of the experiment (when subjects were already in the scanner). From these practice data, the shortest reaction time to the target is used to determine the individual time limit for responses to the target. In 50% of the trials, 200ms is added to the duration of the individual time limit, enabling participants to be successful in these trials. In the remaining trials, 150ms is subtracted from the time limit, to make sure that participants cannot respond in time. This procedure resulted in about 50% correct feedback for both rewarding and nonrewarding trials, separately. The total amount of money won is presented at the end of the task. Participants are told that they receive the cumulative total amount of reward of the actual experiment in addition to the standard compensation for participation. The task consisted of 60 trials with a mean duration of 9571ms (range 4946–16 107ms, inter-trial-interval range 1029–6979ms), resulting in a total task duration of 9 minute 35 seconds.

Behavioral Data Analyses

Repeated-measures ANOVAs were performed to test for effects of condition (potentially rewarding trials, nonrewarding trials) and group (controls, siblings) on reaction time. The amount of money that participants won was compared between controls and siblings using a 2-sample t-test.

Functional Magnetic Resonance Imaging

Measurements.

All imaging was performed on a Philips 3.0-T Achieva whole-body MRI scanner (Philips Medical Systems). Functional images were obtained using a 2-dimensional echo planar imaging-sensitivity encoding (EPI-SENSE) sequence with the following parameters: voxel size 4mm isotropic; repetition time (TR) = 1600ms; echo time (TE) = 23ms; flip angle = 72.5°; 30-slice volume; SENSE-factor R = 2.4 (anterior-posterior). Three hundred twenty functional images were acquired during the task. Next, a whole-brain 3-dimensional fast field echo T1-weighted structural image was acquired for within-subject registration purpose, scan parameters: voxel size 1mm isotropic; TR = 25ms; TE = 2.4ms; flip angle = 30°; 150 slices.

Image Preprocessing.

Image preprocessing and analyses were carried out with SPM5 software (http://www.fil.ion.ucl.ac.uk/spm). After slice timing and realignment, the structural scan was coregistered to the mean functional scan. Next, using unified segmentation, the structural scan was segmented, and normalization parameters were estimated. Subsequently, all scans were registered to a Montreal Neurological Institute T1-standard brain using these normalization parameters and a 3-dimensional Gaussian smoothing kernel (8-mm full width at half maximum) was applied to all functional images.

Individual Analyses.

Individual datasets were analyzed using multiple-regression, to estimate brain activation time-locked to anticipation of reward, anticipation of nonreward, correct reward outcome, incorrect reward outcome, correct nonreward outcome, and incorrect nonreward outcome. To correct for head motion, the 6 realignment parameters were included in the design matrix as regressors of no interest. A high-pass filter was applied to the data with a cutoff frequency of 0.0058 Hz to correct for drifts in the signal. For each subject, we calculated brain activation related to reward anticipation (anticipation of reward vs anticipation of nonreward) and receipt of reward (correct reward outcome vs correct nonreward outcome). Head motion parameters were investigated to ensure there were no differences in motion between the groups and that for none of the subjects the maximum motion exceeded predefined thresholds (scan-to-scan: >3mm).

Region of Interest Analyses.

In the manner of previous fMRI studies investigating cognitive functioning,^18,32,36,45 region of interest (ROI) analyses were performed to test for group differences in 5 predefined regions that are known for their involvement in reward processing, including the bilateral ventral striatum, dorsal striatum, insula, SMA, and orbitofrontal cortices (also referred to as ventromedial prefrontal cortex) (figure 2).^22,36,38 Regions were created using the automated anatomical labeling-atlas.⁴⁶ For each ROI, the average level of brain activation (ie, percent signal change) was obtained for each subject. These subject-wise values were then submitted to independent samples t-test to investigate group differences (controls, siblings) in activation within ROIs of reward processing during anticipation and receiving. All results were corrected for multiple testing of 10 regions (5 regions left and right).

Fig. 2.

Region of interests (ROIs) in 5 predefined regions that are known on their involvement in reward processing.²² ROIs were the (1) supplementary motor area (SMA), (2) dorsal striatum, (3) ventral striatum, (4) orbitofrontal cortex, and (5) insula.

Whole-brain Analyses.

In addition to the ROI analyses, whole-brain group-wise analyses were performed to investigate brain activity outside the predefined ROIs during reward anticipation and receipt of reward.⁴⁷ To test group differences in activity between siblings and matched controls, whole-brain 2-sample t-tests were performed. All group activation maps were tested for significance at a family-wise-error corrected cluster level of P = .05 (cluster-defining threshold of P = .001, critical cluster size of 34 voxels).

Correlations.

Pearson correlations were used to test for associations between brain activation in all regions during reward processing, intelligence quotient (IQ) and sub-clinical symptoms as measured with the CAPE. In addition, we investigated the influence of past depressive episodes on brain activity in siblings. For this reason, we performed 2-sample t-tests to compare brain activity between siblings with and without a history of depression. Finally, we and others have shown that cigarette smoking and cannabis use impact activity during reward processing in the ventral striatum.^37,48 We therefore compared ventral striatum activation, using 2-sample t-tests, between cigarette smokers and non-smokers as well as between cannabis users and non-users. Importantly, none of the participants used other drugs for the last 12 months.

Results

Behavioral Results

All behavioral data of controls and siblings are presented in table 1. A main effect of reward on reaction time was found, with all subjects responding significantly faster to the target on potentially rewarding trials compared to nonrewarding trials (F[1,53] = 37, P < .01). This effect of reward did not differ between the groups (group by reward effect interaction: F[1,53] = 0.14, P = .71). There was no main effect of group (F[1,53] = 0.48, P = .49), indicating that there was no overall difference in response speed. Siblings and control subjects won the same amount of money (t[54] =1.21, P = .23). This was expected, because task difficulty was individually determined based on a training session prior to the experiment to ensure that all subjects won the same amount of money. In this way, potential performance confounds which may affect imaging results were excluded.

Imaging Results

Reward Anticipation.

During reward anticipation (anticipation of reward vs anticipation of nonreward), ROI analyses revealed that the ventral striatum, dorsal striatum, insula, and SMA were less active in siblings than in healthy controls (figure 3, supplementary table S1). After correcting for multiple comparisons for the number of ROIs, only the difference between groups in the left ventral striatum and the left SMA remained significant (figure 3). In the orbitofrontal cortex, no significant activation differences between siblings and controls were found.

Fig. 3.

ROI analyses revealed decreased activity in the left ventral striatum and the left SMA in siblings compared with controls during reward anticipation (anticipation of reward vs anticipation of nonreward) (A). During receiving reward (correct reward outcome vs correct nonreward outcome), siblings showed increased activity in the ventral striatum bilaterally and the left orbitofrontal cortex (B). Error bars indicate standard error of means. * = significant (P < .05) group differences. ** survived Bonferroni correction for multiple testing. HC = controls, SB = siblings.

To investigate whether there were activation differences between the groups outside these ROIs, we performed whole-brain analyses. As expected, healthy controls and siblings activated regions within the neural circuitry of reward anticipation including ventral and dorsal striatum, insula, parietal cortices, thalamus, and the SMA, corroborating the findings from the ROI analysis (figure 4). When compared directly by the 2-sample t-test, siblings showed significant less activation in the right anterior insula as compared to controls (figure 4).

Fig. 4.

Whole-brain activation during reward anticipation (anticipation of reward vs anticipation of nonreward) (A) and receipt of reward (correct reward outcome vs correct nonreward outcome) (B) in controls (HC) and siblings (SB). During reward anticipation, siblings showed decreased activity in the right anterior insula compared to the controls (center of mass in Talairach coordinates [30 32 0]). Significant activation clusters (P < .05, family-wise error-corrected) are displayed on the normalized and skull-stripped group-average brain (neurological orientation). Note that the colors for controls and siblings are scaled to the same colormap. Right = right.

Receipt of Reward.

During the receipt of reward, ROIs analyses revealed increased activity in siblings as compared to controls in the ventral striatum and orbitofrontal cortex (figure 3, supplementary table S1). It is important to note that we contrasted receipt of reward (ie, correct reward outcome) to correct nonreward outcome. For a complete overview of all outcome conditions, see supplementary figure S1. After correcting for multiple comparisons for the number of ROIs, the effect in bilateral ventral striatum and left orbitofrontal cortex remained significant (figure 3).

Whole-brain analyses showed that healthy controls and siblings activated circuits that are typical for receipt of reward including the orbitofrontal cortex (figure 4). However, a direct comparison of activation levels in the voxel-wise whole-brain analysis did not reveal significant group differences (figure 4).

Correlations.

Correlation analyses in siblings revealed that reduced levels of ventral striatum activation during reward anticipation were related to the increased levels of ventral striatum activation during the receipt of reward (striatum left: r = −.6, P = .001, striatum right: r = −.6, P = .0004) (supplementary figure S2). Next, left ventral striatal activity in siblings was negatively correlated with the effect of reward on reaction time (mean reaction time during nonrewarding trials minus mean reaction time during potentially rewarding trials) (r = −.6, P = .0005). This indicates that siblings who showed the most pronounced decrease in left ventral striatum activity did not respond faster on potentially rewarding trials compared to nonrewarding trials.

For siblings, we found a negative correlation between the CAPE negative dimension and activation of the left ventral striatum during reward anticipation (r = −.5, P = .01) (supplementary figure S3). We did not find such a correlation in healthy controls (r = −.01, P = .98). No correlations were found between brain activity during reward processing and IQ. There were no differences in brain activation for any of the regions during reward processing between siblings with and without a history of depression. Smoking status as well as cannabis use did not affect activation in the ventral striatum in siblings.

Discussion

This study investigated brain activity during reward processing in 27 unaffected siblings of schizophrenia patients and 29 matched controls using fMRI. During reward anticipation, siblings showed reduced activation in the left ventral striatum, left SMA, and the right anterior insula as compared to healthy controls. Moreover, decreased ventral striatal activity was correlated with the degree of sub-clinical negative symptoms only in siblings. In contrast, during the receipt of reward siblings showed increased activation in the bilateral ventral striatum and the left orbitofrontal cortex as compared to healthy controls. Since siblings share on average 50% of their genes with their ill relative, these results suggest that impaired reward processing may be related to the genetic vulnerability for schizophrenia.²⁹

Our finding of decreased ventral striatal activation during reward anticipation in siblings of schizophrenia patients as compared to healthy controls is consistent with several studies in medicated schizophrenia patients,^19,21,49 but not with others.^27,50,51 The latter inconsistency may be partly due to the varying effects of antipsychotics on activation in the fronto-striatal network.^16,52–55 Indeed, findings in medication-free^16,20 and medication-naïve patients^17,18 consistently show diminished ventral striatal activity during reward anticipation. In addition to what has been found in schizophrenia patients and consistent with our results in siblings, a recent study shows decreased ventral striatal activity during reward anticipation in first-degree relatives of schizophrenia, including parents, siblings, and adolescent offspring of schizophrenia patients.³¹ Here, we extend this finding by showing decreased striatal activity in unaffected siblings of schizophrenia patients as compared to controls. Consequently, relative to controls and schizophrenia patients, siblings show intermediate activation levels at best. The attenuated striatal activation during reward anticipation has been suggested to represent a ceiling effect due to an overall increased dopaminergic tone as shown in schizophrenia patients.^56–58 Whether decreased striatal activity in siblings reflects increased dopamine tone should be further explored by PET studies. For example, using PET, Brunelin et al.⁵⁹ reported increased ventral striatal dopamine release in siblings compared to healthy controls in response to metabolic stress.

Siblings with lower activation of the ventral striatum during reward anticipation showed higher scores of sub-clinical measures in the negative domain as measured with the CAPE.⁴¹ High scores in the negative dimension of the CAPE represent sub-clinical symptoms including apathy and lack of motivation. Our finding is consistent with results from Simon et al.²⁷ who reported similar correlations between decreased ventral striatal activity during reward anticipation and apathy in schizophrenia patients.²⁷ Moreover, apathy is associated with dysfunctional dopamine activation in the reward network.⁶⁰

In contrast to reduced activation in the ventral striatum during reward anticipation, we found increased activity in the ventral striatum during the receipt of reward in siblings compared to controls. This finding is consistent with 1 study showing increased ventral striatal activity in schizophrenia patients during successful reward feedback.²⁰ On the other hand, increased ventral striatal activation is not consistent with the study of Simon et al.,²⁷ who found no such an increase in 15 medicated schizophrenia patients during receiving reward. This discrepancy is likely caused by the fact that all patients received antipsychotic medication which has a direct effect on activation of the fronto-striatal network. Indeed, several studies reported normalization of striatal activation as a result of antipsychotic treatment.^16,52–55

We found that for siblings the increase in ventral striatal activation during the receipt of reward was related to the reduced activation during reward anticipation. This argues against a dopaminergic ceiling effect as explanation for reduced ventral striatum during anticipation in siblings of patients. Normally, ventral striatal activation occurs in response to an unpredicted reward.⁶¹ When a cue is coupled with a particular outcome (reward), activation shifts from the outcome period to cue presentation, ie, anticipation.^61,62 This activation shift is known to rely on adequate dopaminergic transmission⁶³ and involves the ventral striatum.^64,65 Reduced activity of the ventral striatum during reward anticipation in siblings as compared to controls suggest that this shift from receipt to anticipation of reward does not occur, possibly due to impairments in dopamine transmission.⁶¹ In other words, the hypothesized dopamine dysfunction in siblings may underlie the failure of cues signaling potential reward to trigger activation in the ventral striatum in siblings. As a consequence, striatal activity is relatively increased during receipt of reward. This hypothesis was confirmed by our behavior data indicating that those siblings who display the most pronounced decrease in ventral striatal activity during reward anticipation together with the most increased ventral striatal activity during receipt of reward, showed the least effect of reward on cues in response times. The failure in siblings to shift activation from outcome to cue is probably not specific for reward, but may reflect impaired cue processing in general. Indeed, we also observed similar striatal hypoactivation in siblings and patients during inhibition.^2,9 Although it is clear that dopamine is pivotal for reward processing,^64,66 abnormal ventral striatum functioning in siblings may, at least in part, also be attributed to abnormalities involving other neurotransmitters such as glutamate.^67,68

Our finding of decreased activity in the right anterior insula in siblings as compared to controls during reward anticipation is consistent with studies in schizophrenia patients showing reduced anterior insula activity during reward processing.^21,50 Given the fact that insular function is highly dependent on dopamine transmission,^69,70 aberrant insular functioning as shown here in siblings suggests impaired dopamine function in siblings similar to that in schizophrenia patients.^23,56

In addition to reduced activation in the anterior insula, we found decreased activity in the SMA during reward anticipation in siblings compared to controls. This is not surprising given the role of the SMA in motor preparation and the fact that this region is highly interconnected with the ventral striatum.^65,71,72 Indeed, SMA activity has been consistently found during reward anticipation.^73–76 Our finding therefore suggests decreased reward-based motor preparation in siblings.

Finally, we found increased activity in the orbitofrontal cortex during the receipt of reward in siblings compared to controls. This finding is consistent with 1 study showing exaggerated orbitofrontal cortex activation during reward feedback in schizophrenia patients.²⁰ However, they contrasted reward outcome with no-reward outcome (ie, incorrect reward trial), while we contrasted reward outcome with correct nonreward outcome. Post-hoc analyses of our data showed that this particular contrast yielded similar results (see supplementary figure S1).

Our findings of impaired fronto-striatal functioning in nonmedicated, disorder-free siblings may represent a genetic vulnerability for schizophrenia. However, common environmental factors cannot be ruled out. In order to fully investigate the genetic determinants of impaired reward processing, future studies should include monozygotic and dizygotic twins who are discordant for schizophrenia. In another cognitive domain, ie, spatial working memory, it has been shown that unaffected monozygotic co-twins show the greatest impairments compared to unaffected dizygotic co-twins and healthy control twin pairs, indicating a strong effect of genetic loading.^77–79

A potential limitation of this study is that we did not observe behavioral differences between siblings and controls: both groups responded faster on potentially rewarding trials compared to neutral trials. However, this lack of a direct association between brain deficits and performance has been reported on before.^2,80,81 In contrast, studies in schizophrenia patients using a similar task, have shown that patients respond significantly slower on potential rewarding trials compared to healthy controls.^18–20 However, reward experiments in siblings have never been performed, so it is unclear if behavioral deficits are to be expected.^2,33,82 Moreover, there have been various studies with other cognitive tasks showing brain activity abnormalities in siblings despite performance being equal to that of healthy participants. The task we used was relatively simple; more complex tasks may provoke behavioral deficits. Finally, using smiling and neutral faces may have had an effect apart from anticipatory processes we aimed to investigate. One could argue that a smiling face by itself is potentially rewarding. We opted to use these intuitive cues rather than the abstract cues developed by Knutson et al.,²² to minimize cognitive load, specifically for the siblings.

Here, we show fronto-striatal impairments during reward processing in unaffected siblings of schizophrenia patients. Specifically, ventral striatum activation was decreased during reward anticipation and increased during receipt of reward in siblings as compared to matched healthy controls. Moreover, decreased ventral striatal activation in siblings was correlated with sub-clinical negative symptoms. These data suggest that impaired cue processing, possibly as a result of dopamine dysfunction, may be related to the genetic vulnerability to develop schizophrenia.

Supplementary Material

Supplementary material is available at http://schizophreniabulletin.oxfordjournals.org.