A computational neuroimaging study of reinforcement learning and goal-directed exploration in schizophrenia spectrum disorders.

BACKGROUND: Prior evidence indicates that negative symptom severity and cognitive deficits, in people with schizophrenia (PSZ), relate to measures of reward-seeking and loss-avoidance behavior (implicating the ventral striatum/VS), as well as uncertainty-driven exploration (reliant on rostrolateral prefrontal cortex/rlPFC). While neural correlates of reward-seeking and loss-avoidance have been examined in PSZ, neural correlates of uncertainty-driven exploration have not. Understanding neural correlates of uncertainty-driven exploration is an important next step that could reveal insights to how this mechanism of cognitive and negative symptoms manifest at a neural level. METHODS: We acquired fMRI data from 29 PSZ and 36 controls performing the Temporal Utility Integration decision-making task. Computational analyses estimated parameters corresponding to learning rates for both positive and negative reward prediction errors (RPEs) and the degree to which participates relied on representations of relative uncertainty. Trial-wise estimates of expected value, certainty, and RPEs were generated to model fMRI data. RESULTS: Behaviorally, PSZ demonstrated reduced reward-seeking behavior compared to controls, and negative symptoms were positively correlated with loss-avoidance behavior. This finding of a bias toward loss avoidance learning in PSZ is consistent with previous work. Surprisingly, neither behavioral measures of exploration nor neural correlates of uncertainty in the rlPFC differed significantly between groups. However, we showed that trial-wise estimates of relative uncertainty in the rlPFC distinguished participants who engaged in exploratory behavior from those who did not. rlPFC activation was positively associated with intellectual function. CONCLUSIONS: These results further elucidate the nature of reinforcement learning and decision-making in PSZ and healthy volunteers.

Motivational deficits in schizophrenia relate to abnormalities in cortical learning rate signals.

Individuals from across the psychosis spectrum display impairments in reinforcement learning. In some individuals, these deficits may result from aberrations in reward prediction error (RPE) signaling, conveyed by dopaminergic projections to the ventral striatum (VS). However, there is mounting evidence that VS RPE signals are relatively intact in medicated people with schizophrenia (PSZ). We hypothesized that, in PSZ, reinforcement learning deficits often are not related to RPE signaling per se but rather their impact on learning and behavior (i.e., learning rate modulation), due to dysfunction in anterior cingulate and dorsomedial prefrontal cortex (dmPFC). Twenty-six PSZ and 23 healthy volunteers completed a probabilistic reinforcement learning paradigm with occasional, sudden, shifts in contingencies. Using computational modeling, we found evidence of an impairment in trial-wise learning rate modulation (α) in PSZ before and after a reinforcement contingency shift, expressed most in PSZ with more severe motivational deficits. In a subsample of 22 PSZ and 22 healthy volunteers, we found little evidence for between-group differences in VS RPE and dmPFC learning rate signals, as measured with fMRI. However, a follow-up psychophysiological interaction analysis revealed decreased dmPFC-VS connectivity concurrent with learning rate modulation, most prominently in individuals with the most severe motivational deficits. These findings point to an impairment in learning rate modulation in PSZ, leading to a reduced ability to adjust task behavior in response to unexpected outcomes. At the level of the brain, learning rate modulation deficits may be associated with decreased involvement of the dmPFC within a greater RL network.

Configurable NOR gate arrays from Belousov-Zhabotinsky micro-droplets.

We investigate the Belousov-Zhabotinsky (BZ) reaction in an attempt to establish a basis for computation using chemical oscillators coupled via inhibition. The system consists of BZ droplets suspended in oil. Interdrop coupling is governed by the non-polar communicator of inhibition, Br2. We consider a linear arrangement of three droplets to be a NOR gate, where the center droplet is the output and the other two are inputs. Oxidation spikes in the inputs, which we define to be TRUE, cause a delay in the next spike of the output, which we read to be FALSE. Conversely, when the inputs do not spike (FALSE) there is no delay in the output (TRUE), thus producing the behavior of a NOR gate. We are able to reliably produce NOR gates with this behavior in microfluidic experiment.

Latent structure of cognition in schizophrenia: a confirmatory factor analysis of the MATRICS Consensus Cognitive Battery (MCCB).

BACKGROUND: The number of separable cognitive dimensions in schizophrenia has been debated. Guided by the extant factor analytic literature, the NIMH Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) initiative selected seven cognitive domains relevant to treatment studies in schizophrenia: speed of processing, attention/vigilance, working memory, verbal learning, visual learning, reasoning and problem solving, and social cognition. These domains are assessed in the MATRICS Consensus Cognitive Battery (MCCB). The aim of this study was to conduct a confirmatory factor analysis (CFA) of the beta battery of the MCCB to compare the fit of the MATRICS consensus seven-domain model to other models in the current literature on cognition in schizophrenia. METHOD: Using data from 281 schizophrenia outpatients, we compared the seven correlated factors model with alternative models. Specifically, we compared the 7-factor model to (a) a single-factor model, (b) a three correlated factors model including speed of processing, working memory, and general cognition, and (c) a hierarchical model in which seven first-order factors loaded onto a second-order general cognitive factor. RESULTS: Multiple fit indices indicated the seven correlated factors model was the best fit for the data and provided significant improvement in model fit beyond the comparison models. CONCLUSIONS: These results support the assessment of these seven cognitive dimensions in clinical trials of interventions to improve cognition in schizophrenia. Because these cognitive factors are separable to some degree, it is plausible that specific interventions may have differential effects on the domains.

Cognition in schizophrenia and schizo-affective disorder: impairments that are more similar than different.

BACKGROUND: Cognition is increasingly being recognized as an important aspect of psychotic disorders and a key contributor to functional outcome. In the past, comparative studies have been performed in schizophrenia and schizo-affective disorder with regard to cognitive performance, but the results have been mixed and the cognitive measures used have not always assessed the cognitive deficits found to be specific to psychosis. A set of optimized cognitive paradigms designed by the Cognitive Neuroscience Test Reliability and Clinical Applications for Schizophrenia (CNTRACS) Consortium to assess deficits specific to schizophrenia was used to measure cognition in a large group of individuals with schizophrenia and schizo-affective disorder. METHOD: A total of 519 participants (188 with schizophrenia, 63 with schizo-affective disorder and 268 controls) were administered three cognitive paradigms assessing the domains of goal maintenance in working memory, relational encoding and retrieval in episodic memory and visual integration. RESULTS: Across the three domains, the results showed no major quantitative differences between patient groups, with both groups uniformly performing worse than healthy subjects. CONCLUSIONS: The findings of this study suggests that, with regard to deficits in cognition, considered a major aspect of psychotic disorder, schizophrenia and schizo-affective disorder do not demonstrate major significant distinctions. These results have important implications for our understanding of the nosological structure of major psychopathology, providing evidence consistent with the hypothesis that there is no natural distinction between cognitive functioning in schizophrenia and schizo-affective disorder.

The Premorbid Adjustment Scale as a measure of developmental compromise in patients with schizophrenia and their healthy siblings.

Schizophrenia is associated with subtle developmental compromise, but the degree to which this is also associated with heritability and genetic risk is uncertain. The goal of the current study was to investigate the childhood, adolescent, and early adulthood social and academic function of patients with schizophrenia, their healthy siblings, and normal controls, using the Premorbid Adjustment Scale (PAS). Generalized Estimating Equations were conducted to account for nesting of subjects within families. Patients (N=286) scored significantly worse than their healthy siblings (N=315) at every age period; siblings scored significantly worse than controls (N=261) at every age period. In probands, PAS scores got worse after early adolescence while control and proband scores improved after late adolescence. Furthermore, patient PAS scores significantly predicted the scores of their own discordant siblings in childhood and late adolescence. This effect approached significance in early adolescence and in the general scale. Thus, the most premorbidly impaired patients tended to have non-ill siblings with worse premorbid adjustment scores than the siblings of less impaired probands. The results suggest that both patients and many of their siblings share poor adjustment in childhood and adolescence, possibly due to shared genetic or environmental risk factors.

The effect of the physical characteristics of cues and targets on facilitation and inhibition.

The present experiment was conducted in order to examine the role of cue-target discriminability on early occurring attentional cuing effects and late occurring inhibition of return (IOR). The experiment used a single target stimulus in conjunction with three different cue stimuli. The cues were the same as the target, different in color, shape, and luminance to the target, or did not spatially overlap with the target. At shorter stimulus onset asynchronies (SOAs; 100 and 200 msec), attentional cuing effects were only found with the nonoverlapping cues. However, at longer SOAs (400 and 800 msec), approximately equal IOR effects were found with all three types of cues. The results indicated that the physical characteristics of the cues and targets affected the pattern of reaction times at the shorter SOAs but not at the longer SOAs. The conclusion is that the biphasic pattern of early facilitation and late inhibition following a peripheral cue should not be considered the definitive signature of the peripheral cuing paradigm.

Is position "special" in visual attention? Evidence that top-down processes guide visual selection.

The role of spatial position in selective visual processing has been the source of recent debate. Experiments by Tsal and Lavie (1988) and van der Heijden et al. (1996) have been designed to establish the status of position in visual selection. Tsal and Lavie found that observers tend to select letters from a briefly presented array according to position. Using the same paradigm, van der Heijden et al. found that observers tend to select according to colour, except under conditions of low contrast and unrestricted eye movements. The present study attempted to reconcile these findings by exploring the influence of top-down processes (task instructions) while explicitly controlling for eye movements. Experiment 1 demonstrated there was no inherent selection bias for stimuli similar to those used by van der Heijden et al., suggesting that the tendency to select according to colour found by van der Heijden et al. was due to task demands. Experiment 2 further established the role of top-down factors by replicating the results of van der Heijden et al. with our stimuli. Experiment 3 demonstrated that selection can be switched from colour to position by changing the demands of the task. These results suggest that selection may be accounted for by task demands (e.g., instructions) with no priority access for position information.

Generalization of training effects in schizophrenia.

The primary goal of this study was to investigate transfer of training (generalization) in patients with schizophrenia. We randomly assigned 33 schizophrenia subjects to one of three conditions: training on the Wisconsin Card Sort Test (WCST-T), training on the Halstead Category Test (CAT-T), or no training (No-T). The WCST and CAT were administered to all subjects at baseline. Subjects in the WCST-T and CAT-T groups then received training on the respective test, while the No-T group received additional untrained trials. All participants were subsequently retested on the WCST and CAT, and completed a brief neuropsychological battery. As hypothesized, the WCST-T and CAT-T groups exhibited large improvements on the trained test and moderate improvement on the untrained test, while the No-T group failed to show improvement on either test. These results suggest that the training paradigm did produce generalization, and that the changes were not due to practice effects. The extent of generalization across both training groups was strongly associated with neuropsychological test performance (Spearman's rho=0.56, P<0.05). The implications of these findings for rehabilitation programs were discussed, and recommendations were made for future research.

Correlates of insight in serious mental illness.

This study extends research into insight by examining its relationship to a variety of demographic, clinical, neurocognitive, and psychosocial variables among a broad diagnostic sample of 211 adults with serious mental illness. Participants completed a full battery of instruments measuring these variables. Results support a relationship between ratings of poor insight and a psychotic (vs. mood) diagnosis, increased psychiatric symptoms, poorer social skills, and negative medication attitudes. Minorities and those with a substance abuse diagnosis were also more likely to be rated as having poor insight. No relationship was found between level of insight and age, gender, education level, neurocognitive deficits, hospitalization history, size of one's social network, or quality of life measures. Results are discussed in the context of improving the measurement and assessment of insight, conceptualizing interventions aimed at addressing level of insight, and improving outcomes for patients with severe and persistent mental illness. Findings also support a need for continued investigation of how mental illness is understood, experienced, and expressed across diverse groups of people living with mental illness.

Correlates of long-term unemployment among inner-city adults with serious and persistent mental illness.

This study identified demographic, clinical, and vocational rehabilitation-related correlates of long-term unemployment among 219 adults with severe mental illness. Fifty-one percent of the sample had been unemployed five or more years before enrollment. Older age, a diagnosis of psychosis, severity of negative symptoms, and more previous hospitalizations were all significantly related to long-term unemployment. Gender, race, education, substance disorder diagnosis, severity of negative symptoms, and vocational training experience were not. The findings underscore the relevance of clinical and neurocognitive impairments to long-term unemployment and point to the need to critically reevaluate the effectiveness of traditional vocational rehabilitation services.

Perioperative beta-blocker withdrawal and mortality in vascular surgical patients.

OBJECTIVE: Our purpose was to determine the effect of postoperative beta-blocker withdrawal on mortality and cardiovascular events after vascular surgery. METHODS: Detailed data were collected on perioperative cardiovascular medication use and discontinuation and cardiovascular risk factors among consecutive major vascular surgical procedures at two university hospitals. RESULTS: A total of 140 patients received beta-blockers preoperatively. Mortality in the 8 patients who had beta-blockers discontinued postoperatively (50%) was significantly greater than in 132 patients who had beta-blockers continued (1.5%, odds ratio 65.0, P<.001). The effect of beta-blocker discontinuation was unaffected by adjustment by stratification for risk factors (all P< or =.01), for contraindications to restarting beta-blockers (P = .006), and by multivariable analyses adjusting for potential confounders (adjusted odds ratio 17.0, P =.01). beta-Blocker discontinuation also was associated with increased cardiovascular mortality (0% vs 29%, P =.005) and postoperative myocardial infarction (odds ratio 17.7, P =.003). CONCLUSION: Discontinuing beta-blockers immediately after vascular surgery may increase the risk of postoperative cardiovascular morbidity and mortality.

Short form of the WAIS-III for use with patients with schizophrenia.

The recent publication of the Wechsler Adult Intelligence Scale (WAIS-III), the most widely used standard test of intelligence, requires the development of a new short form for use with patients with schizophrenia for many clinical and research purposes. We used regression analyses of complete WAIS-III data on 41 outpatients with schizophrenia and 41 education-, and age-matched healthy subjects to determine the best combination of subtests to use as a short form. Excluding three subtests that are time-consuming to administer, and requiring that the solution includes one subtest from each of the four WAIS index scores, the combination that most fully accounted for the variance in full-scale IQ (FSIQ) for both participants with schizophrenia (R(2)=0.90) and healthy controls (R(2)=0.86) included the information, block design, arithmetic, and digit symbol subtests. When the restrictions regarding which subtests could enter were relaxed, the best four-subtest solution included information, block design, comprehension, and similarities. Although the latter explained 95% of the variance in FSIQ for schizophrenia participants and 90% of the variance for healthy controls, it consistently overestimated FSIQ for the schizophrenia group. We recommend the four-factor short form for use in future research and clinical practice in which a quick, accurate IQ estimate is desired.

Learning and forgetting in schizophrenia.

Recent studies of patients with schizophrenia have consistently demonstrated marked deficits on measures of initial learning. However, contradictory results have been reported concerning retention and forgetting. The present study examined the level of initial and delayed recall of stories and visual figures in a group of 76 patients with schizophrenia and 51 normal controls. Schizophrenia patients demonstrated marked impairments in initial and delayed recall as well as significantly worse percentage retention scores. However, schizophrenia patients and healthy controls individually matched on level of initial recall had nearly identical delayed recall performance. The results suggest a primary deficit in the initial acquisition of information rather than an accelerated rate of forgetting in schizophrenia.

Cognitive impairments in patients with schizophrenia displaying preserved and compromised intellect.

BACKGROUND: Although intellectual and neurocognitive deficits accompany schizophrenia, there are inconsistencies in the literature concerning issues of intellectual decline, premorbid deficits, a modal deficit pattern, and preserved abilities. METHODS: A battery of neuropsychological tests was administered once to 117 consecutively admitted patients with chronic schizophrenia and a group of 27 healthy control subjects to examine patterns of premorbid and current intellect (measured by means of reading scores and IQ, respectively) and the attendant cognitive profiles in schizophrenia using classification methods based on clinically derived (IQ levels) and atheoretical (cluster) techniques. RESULTS: Sixty patients (51%) with schizophrenia who displayed a general intellectual decline of 10 points or greater from estimated premorbid levels also exhibited deficits of executive function, memory, and attention. Twenty-eight patients (23%) with consistently low estimated premorbid intellect and current intellectual levels who displayed no evidence of IQ decline exhibited language and visual processing deficits in addition to deficits present in the intellectually declining group. The remaining 29 patients (25%) who displayed average estimated premorbid intellectual levels did not show IQ decline and exhibited a cognitive profile similar to normal, with the exception of executive function and attention impairment. Atheoretical analyses support the findings from clinically derived subgroups. CONCLUSIONS: These results suggest that IQ decline, although modal in schizophrenia, is not universally characteristic and that executive function and attention deficits may be core features of schizophrenia, independent of IQ variations.

Effects of contextual processing on visual conditional associative learning in schizophrenia.

BACKGROUND: We adapted visual conditional associative learning paradigms to assess the contextual processing deficit model of schizophrenic cognitive impairment proposed by J.D. Cohen and D. Servan-Schreiber in 1992. In this task subjects learn the associations between four sets of stimuli through the use of feedback. We administered two experimental conditional associative learning conditions: in one, the eight stimuli used to make four pairs were all different; in the other, the pairs were made from different combinations of four identical stimuli, requiring the use of contextual information to mediate correct performance. Two additional associative learning tasks were administered where subjects generated the stimulus pairings or observed the experimenter form the pairs, eliminating the need to learn from feedback. METHODS: We tested 37 patients with schizophrenia and 20 healthy control subjects in each conditional associative learning task condition. RESULTS: Patients demonstrated significant impairments on all four conditional associative learning tasks. The demand to process contextual information did not differentially impact patient performance. Patients were better able to learn associations if they generated or observed the pairings rather than utilized feedback to guide learning. CONCLUSIONS: Patients with schizophrenia demonstrate pronounced deficits in the ability to utilize feedback to guide learning. We found no evidence of an additional deficit in processing of contextual information.

A comparison of rCBF patterns during letter and semantic fluency.

To evaluate the functional neuroanatomies underlying letter and category fluency, 18 normal controls were studied with oxygen-15 water regional cerebral blood flow positron emission tomography. Three counterbalanced conditions each consisted of 6 trials (45 s each): letter fluency (generating words when cued with a particular letter), semantic fluency (generating words when cued with a particular category), and a control condition (generating days of the week and months of the year). Relative to the control, participants activated similar brain regions during both fluency tasks, including the anterior cingulate, left prefrontal regions, thalamus, and cerebellum; reductions were found in parietal and temporal regions. In a direct comparison of the 2 fluency tasks, inferior frontal cortex and temporoparietal cortex (hypothesized to participate in a phonologic loop for accessing word pronunciation) were activated more during letter than semantic fluency, whereas left temporal cortex (associated with access to semantic storage) was activated more during semantic than letter fluency. This study identifies subtle differences in the neural networks underlying letter and semantic fluency that may underlie the dissociation of these abilities in patients.