Symposium Speakers

David Lewis (University of Pittsburgh)

A Neural Circuitry Basis for Impaired Cortical Network Oscillations and Cognitive Dysfunction in Schizophrenia

Deficits in cognitive control, the ability to adjust thoughts or behaviors in order to achieve goals, are now considered to be a core feature of schizophrenia and to be the best predictor of long-term functional outcome. Cognitive control depends on the coordinated activity of a number of brain regions, including the dorsolateral prefrontal cortex (DLPFC). Subjects with schizophrenia exhibit altered activation of the DLPFC, and reduced frontal lobe gamma band (~40 Hz) oscillations, when performing tasks that require cognitive control. Gamma oscillations require robust activity in the reciprocal connections between the parvalbumin-containing basket cell class of cortical GABA neurons and neighboring pyramidal neurons. Thus, alterations in either the excitatory or inhibitory synapses in this circuit could contribute to impaired gamma oscillations and cognition in schizophrenia. This presentation will review the evidence for these types of alterations in the DLPFC of subjects with schizophrenia, and the convergent findings indicating which alterations might be primary disturbances and which are compensatory responses. Together, the findings suggest a mechanistic model of "re-set" excitatory-inhibitory balance in DLPFC circuitry that could give rise to impaired gamma oscillations and could explain the developmental course of functional disturbances in individuals with schizophrenia.

Steven Silverstein (Rutgers Biomedical and Health Sciences)

Cognitive organization as a dimension of individual differences and psychopathology

Visual processing impairments are well established in schizophrenia. These include abnormalities in contrast sensitivity; various excitatory and inhibitory effects, including those involved in masking and surround suppression; and in form, motion, brightness, and color processing. Many of these effects have been demonstrated using sophisticated psychophysical paradigms that avoid generalized deficit confounds, and many have known neurobiological correlates as revealed by EEG/ERP or functional MRI. Recently, the clinical importance of these impairments has become clearer, with research indicating that they predict conversion to psychosis and are associated with poorer functional outcomes in patients. Moreover, it has been demonstrated that the disturbances in the computational mechanisms and neural circuitry involved in specific visual impairments in schizophrenia are clear examples of more widespread failures in: 1) predictive coding, which is thought to be the basis of positive symptoms; or 2) cognitive coordination, which is thought to be the basis of disorganized symptoms. As would be predicted by these models, positive and disorganized symptoms are significantly and differentially correlated with visual impairments that can be viewed as exemplars of their basic mechanisms. Several studies have also clarified the state/trait status of specific impairments, thereby increasing their utility for studies of treatment-related biomarkers or endophenotypes, respectively. Finally, recent evidence indicates that changes in the retina – the only neural tissue that can be viewed directly in fully awake subjects – are associated with illness progression and poorer cognitive functioning – suggesting their use as a proxy for brain structure and function, as has been demonstrated with other CNS diseases. Additionally, several known visual impairments in schizophrenia suggest either illness- or medication-related retinal effects. This presentation will review data on selected visual processing impairments to demonstrate their different neural bases (from retina to occipital lobe to frontal cortex), computational mechanisms, status as biomarkers, and treatment implications.

William Carpenter (University of Maryland)

Avolition in schizophrenia: A failure to translate reward information into motivated behavior

Avolition has been at the core of the schizophrenia construct since the earliest clinical conceptualizations of the disorder, and is a principal determinant of poor functional outcome that limits social and occupational success. Currently, there are no FDA approved treatments for avolition in schizophrenia, potentially because the cognitive and neural basis of this pathology is not well understood. The current presentation discusses recent developments in the etiology, assessment, and treatment of negative symptoms of schizophrenia, taking a translational neuroscience approach to explaining avolitional pathology. We begin by providing a historical overview of avolition in schizophrenia, highlighting early clinical conceptualizations proposed by Kraepelin, Bleuler, and Rado. These early descriptions are contrasted with modern views of negative symptoms originating in the 1970’s that were further refined in the 2000’s with the NIMH Consensus Conference on Negative Symptoms. Current issues in conceptualizing and measuring negative symptom pathology are discussed, including the primary vs. secondary distinction, whether the latent structure is continuous or categorical, evidence for the multi-dimensionality and 2 separable negative symptom domains, and recent developments in negative symptom assessment. Research on reward processing will also be reviewed, which has begun to provide important insights into the cognitive and neural mechanisms associated with motivational impairments in schizophrenia. Specifically, data will be presented on several aspects of reward processing that are impaired in schizophrenia, including: (1) dopamine-mediated basal ganglia systems that support reinforcement learning and the ability to predict cues that lead to rewarding outcomes; (2) orbitofrontal cortex-driven deficits in generating, updating, and maintaining value representations; (3) aberrant effort-value computations, which may be mediated by disrupted anterior cingulate cortex and midbrain dopamine functioning; and (4) altered activation of the prefrontal cortex, which is important for action selection and generating exploratory behaviors in environments where reward outcomes are uncertain. A new translational affective neuroscience model for understanding avolition is discussed, which proposes that aberrant cortico-striatal interactions may be a common factor underlying these various reward processing abnormalities that manifest clinically as avolition.

Raquel Gur (University of Pennsylvania)

Neurodevelopmental genomic strategies in the study of the psychosis spectrum

Consistent with the goals of precision medicine to re-define illness mechanistically through elucidating the pathophysiology from gene action to symptoms, large scale genomic studies have been linking genomic variation to continuous quantitative phenotypes. Such an approach can lead to early detection and pathological processes enabling early intervention. This paradigm shift is now applied in psychiatry with an increased focus in schizophrenia research on early identification of psychosis as the process emerges. Convergent approaches integrate phenotypic features with neurocognitive and neuroimaging measures in large -scale studies. Most studies have examined help seeking youths.

We have applied two complementary strategies to probe the underlying neurobiology of psychosis risk. The first is the study of a community-based sample of youths with no known neurogenetic syndrome; the second is the study of youths with a known genetic syndrome that confers significant increased risk for psychosis. Both samples underwent "deep phenotyping" and were compared to healthy participants.

The Philadelphia Neurodevelopment Cohort is a large (about 9, 500) prospective sample of genotyped youths where converging measures of brain and behavior have been obtained. Individuals with psychosis spectrum symptoms showed aberrant brain function across neurocognitive and neuroimaging measures.

The 22q11.2 Deletion Syndrome (22q11DS) is associated with ~25% risk of psychosis emerging in adolescence and early adulthood. In this sample we compare individuals with psychotic feature to those without to establish unique characteristic that modulate psychosis risk.

Both sample are followed longitudinally to establish intake predictors of clinical high-risk status and, ultimately transition to psychosis. As importantly factor indicative of resilience will be uncovered using these strategies.

Rubin Gur (University of Pennsylvania)

Multimodal brain and behavior indices of psychosis risk

With the increased emphasis on mechanistic accounts of psychosis that can lead to early detection and intervention, early indicators of psychosis risk can provide the tools for theoretical understanding that can inform treatment. Most studies have used clinical manifestations of subthreshold symptoms in help-seeking individuals to target samples for followup. The Philadelphia neurodevelopmental Cohort uniquely provides a community-based sample with clinical and neurocognitive data on about 9,500 youths and multimodal neuroimaging data on a subsample of about 1,600. Followup data at about 2 years were obtained, including neuroimaging, on a sample of 500 youths, of whom about half had reported psychotic symptoms at intake and the rest reported no psychiatric symptoms.
We present data from the intake evaluation, showing that individuals who report psychotic symptoms have neurocognitive deficits and delayed development of neurocognitive performance landmarks. These deficits were most pronounced for executive functions and social cognition but were also significant for memory and complex cognition. Predictors of maintaining high-risk status for psychosis were more specific. They had worse neurocognitive performance at intake for face memory and social cognition. For the subsample with neuroimaging data at intake, the best predictors of continued psychotic symptoms were brain volumes, which were lower for striatal, limbic and cortical regions also implicated in adults with psychosis.
It appears that the combination of clinical and brain-behavior data can provide a powerful set of predictors of persistence of psychotic symptoms. These measures can also provide mechanistic insights, especially when linked with genomic data, and suggest avenues for intervention by delineating vulnerable brain systems linked to circumscribed neurocognitive domains.

Bruce Cuthbert (National Institute of Mental Health)

The NIMH Research Domain Criteria Project: New approaches to classifying psychotic spectrum disorders.

Several factors have contributed to a renewed debate in recent years about the nature of schizophrenia. These include discussions about modifications to the criteria for the DMS-5 and ICD-11 revisions, data that schizophrenia and bipolar disorder do not "breed true," GWAS findings of common genetic risk among disorders, and endophenotype-based intermediate phenotypes that show considerable overlap across disorders. These factors accord with proposals that schizophrenia should be thought of not as a specific disease, but rather as a syndrome that represents one segment of a broad spectrum of serious mental illness. Testing such hypotheses requires a different approach to classification that transcends typical "disorder versus control" studies that preclude analysis of cross-cutting mechanisms. The NIMH Research Domain Criteria (RDoC) project was initiated to develop an experimental classification system based upon functional, behavioral/cognitive dimensions and neurobiological measures of the neural systems that implement these functions. We describe the ways in which the RDoC framework is designed for evaluating research grant applications, with particular reference to its role in facilitating explorations of heterogeneity and co-morbidity that can lead to more precise diagnosis and treatment for psychotic disorders.