A Roadmap for the Development of Applied Computational Psychiatry

Revisiting the role of computational neuroimaging in the era of integrative neuroscience

Computational models have become integral to human neuroimaging research, providing both mechanistic insights and predictive tools for human cognition and behavior. However, concerns persist regarding the ecological validity of lab-based neuroimaging studies and whether their spatiotemporal resolution is not sufficient for capturing neural dynamics. This review aims to re-examine the utility of computational neuroimaging, particularly in light of the growing prominence of alternative neuroscientific methods and the growing emphasis on more naturalistic behaviors and paradigms. Specifically, we will explore how computational modeling can both enhance the analysis of high-dimensional imaging datasets and, conversely, how neuroimaging, in conjunction with other data modalities, can inform computational models through the lens of neurobiological plausibility. Collectively, this evidence suggests that neuroimaging remains critical for human neuroscience research, and when enhanced by computational models, imaging can serve an important role in bridging levels of analysis and understanding. We conclude by proposing key directions for future research, emphasizing the development of standardized paradigms and the integrative use of computational modeling across neuroimaging techniques.

Test-retest reliability of reinforcement learning parameters

It has recently been suggested that parameter estimates of computational models can be used to understand individual differences at the process level. One area of research in which this approach, called computational phenotyping, has taken hold is computational psychiatry. One requirement for successful computational phenotyping is that behavior and parameters are stable over time. Surprisingly, the test-retest reliability of behavior and model parameters remains unknown for most experimental tasks and models. The present study seeks to close this gap by investigating the test-retest reliability of canonical reinforcement learning models in the context of two often-used learning paradigms: a two-armed bandit and a reversal learning task. We tested independent cohorts for the two tasks (N = 69 and N = 47) via an online testing platform with a between-test interval of five weeks. Whereas reliability was high for personality and cognitive measures (with ICCs ranging from .67 to .93), it was generally poor for the parameter estimates of the reinforcement learning models (with ICCs ranging from .02 to .52 for the bandit task and from .01 to .71 for the reversal learning task). Given that simulations indicated that our procedures could detect high test-retest reliability, this suggests that a significant proportion of the variability must be ascribed to the participants themselves. In support of that hypothesis, we show that mood (stress and happiness) can partly explain within-participant variability. Taken together, these results are critical for current practices in computational phenotyping and suggest that individual variability should be taken into account in the future development of the field.

Dynamic computational phenotyping of human cognition

Computational phenotyping has emerged as a powerful tool for characterizing individual variability across a variety of cognitive domains. An individual's computational phenotype is defined as a set of mechanistically interpretable parameters obtained from fitting computational models to behavioural data. However, the interpretation of these parameters hinges critically on their psychometric properties, which are rarely studied. To identify the sources governing the temporal variability of the computational phenotype, we carried out a 12-week longitudinal study using a battery of seven tasks that measure aspects of human learning, memory, perception and decision making. To examine the influence of state effects, each week, participants provided reports tracking their mood, habits and daily activities. We developed a dynamic computational phenotyping framework, which allowed us to tease apart the time-varying effects of practice and internal states such as affective valence and arousal. Our results show that many phenotype dimensions covary with practice and affective factors, indicating that what appears to be unreliability may reflect previously unmeasured structure. These results support a fundamentally dynamic understanding of cognitive variability within an individual.

Closing the loop in psychiatric deep brain stimulation: physiology, psychometrics, and plasticity

Deep brain stimulation (DBS) is an invasive approach to precise modulation of psychiatrically relevant circuits. Although it has impressive results in open-label psychiatric trials, DBS has also struggled to scale to and pass through multi-center randomized trials. This contrasts with Parkinson disease, where DBS is an established therapy treating thousands of patients annually. The core difference between these clinical applications is the difficulty of proving target engagement, and of leveraging the wide range of possible settings (parameters) that can be programmed in a given patient's DBS. In Parkinson's, patients' symptoms change rapidly and visibly when the stimulator is tuned to the correct parameters. In psychiatry, those same changes take days to weeks, limiting a clinician's ability to explore parameter space and identify patient-specific optimal settings. I review new approaches to psychiatric target engagement, with an emphasis on major depressive disorder (MDD). Specifically, I argue that better engagement may come by focusing on the root causes of psychiatric illness: dysfunction in specific, measurable cognitive functions and in the connectivity and synchrony of distributed brain circuits. I overview recent progress in both those domains, and how it may relate to other technologies discussed in companion articles in this issue.

Current status, challenges and future prospects in computational psychiatry: a narrative review

BACKGROUND

Computational psychiatry is an area of scientific knowledge which lies at the intersection of neuroscience, psychiatry, and computer science. It employs mathematical models and computational simulations to shed light on the complexities inherent to mental disorders.

AIM

The aim of this narrative review is to offer insight into the current landscape of computational psychiatry, to discuss its significant challenges, as well as the potential opportunities for the fields growth.

METHODS

The authors have carried out a narrative review of the scientific literature published on the topic of computational psychiatry. The literature search was performed in the PubMed, eLibrary, PsycINFO, and Google Scholar databases. A descriptive analysis was used to summarize the published information on the theoretical and practical aspects of computational psychiatry.

RESULTS

The article relates the development of the scientific approach in computational psychiatry since the mid-1980s. The data on the practical application of computational psychiatry in modeling psychiatric disorders and explaining the mechanisms of how psychopathological symptomatology develops (in schizophrenia, attention-deficit/hyperactivity disorder, autism spectrum disorder, anxiety disorders, obsessive-compulsive disorder, substance use disorders) are summarized. Challenges, limitations, and the prospects of computational psychiatry are discussed.

CONCLUSION

The capacity of current computational technologies in psychiatry has reached a stage where its integration into psychiatric practice is not just feasible but urgently needed. The hurdles that now need to be addressed are no longer rooted in technological advancement, but in ethics, education, and understanding.

Adaptive Design Optimization as a Promising Tool for Reliable and Efficient Computational Fingerprinting

A key challenge in understanding mental (dys)functions is their etiological and functional heterogeneity, and several multidimensional assessments have been proposed for their comprehensive characterization. However, such assessments require lengthy testing, which may hinder reliable and efficient characterization of individual differences due to increased fatigue and distraction, especially in clinical populations. Computational modeling may address this challenge as it often provides more reliable measures of latent neurocognitive processes underlying observed behaviors and captures individual differences better than traditional assessments. However, even with a state-of-the-art hierarchical modeling approach, reliable estimation of model parameters still requires a large number of trials. Recent work suggests that Bayesian adaptive design optimization (ADO) is a promising way to address these challenges. With ADO, experimental design is optimized adaptively from trial to trial to extract the maximum amount of information about an individual's characteristics. In this review, we first describe the ADO methodology and then summarize recent work demonstrating that ADO increases the reliability and efficiency of latent neurocognitive measures. We conclude by discussing the challenges and future directions of ADO and proposing development of ADO-based computational fingerprints to reliably and efficiently characterize the heterogeneous profiles of psychiatric disorders.

Amygdala response predicts clinical symptom reduction in patients with borderline personality disorder: A pilot fMRI study

Borderline personality disorder (BPD) is a prevalent, devastating, and heterogeneous psychiatric disorder. Treatment success is highly variable within this patient group. A cognitive neuroscientific approach to BPD might contribute to precision psychiatry by identifying neurocognitive factors that predict who will benefit from a specific treatment. Here, we build on observations that BPD is accompanied by the enhanced impact of the aversive effect on behavior and abnormal neural signaling in the amygdala. We assessed whether BPD is accompanied by abnormal aversive regulation of instrumental behavior and associated neural signaling, in a manner that is predictive of symptom reduction after therapy. We tested a clinical sample of 15 female patients with BPD, awaiting dialectical behavior therapy (DBT), and 16 matched healthy controls using fMRI and an aversive Pavlovian-to-instrumental transfer (PIT) task that assesses how instrumental behaviors are influenced by aversive Pavlovian stimuli. Patients were assessed 1 year after the start of DBT to quantify changes in BPD symptom severity. At baseline, behavioral aversive PIT and associated neural signaling did not differ between groups. However, the BOLD signal in the amygdala measured during aversive PIT was associated with symptom reduction at 1-year follow-up: higher PIT-related aversive amygdala signaling before treatment was associated with reduced clinical improvement at follow-up. Thus, within the evaluated group of BPD patients, the BOLD signal in the amygdala before treatment was related to clinical symptom reduction 1 year after the start of treatment. The results suggest that less PIT-related responsiveness of the amygdala increases the chances of treatment success. We note that the relatively small sample size is a limitation of this study and that replication is warranted.

Rumination Derails Reinforcement Learning with Possible Implications for Ineffective Behavior

How does rumination affect reinforcement learning-the ubiquitous process by which we adjust behavior after error in order to behave more effectively in the future? In a within-subject design (n=49), we tested whether experimentally manipulated rumination disrupts reinforcement learning in a multidimensional learning task previously shown to rely on selective attention. Rumination impaired performance, yet unexpectedly this impairment could not be attributed to decreased attentional breadth (quantified using a "decay" parameter in a computational model). Instead, trait rumination (between subjects) was associated with higher decay rates (implying narrower attention), yet not with impaired performance. Our task-performance results accord with the possibility that state rumination promotes stress-generating behavior in part by disrupting reinforcement learning. The trait-rumination finding accords with the predictions of a prominent model of trait rumination (the attentional-scope model). More work is needed to understand the specific mechanisms by which state rumination disrupts reinforcement learning.

Reinforcement learning modeling reveals a reward-history-dependent strategy underlying reversal learning in squirrel monkeys

Insight into psychiatric disease and development of therapeutics relies on behavioral tasks that study similar cognitive constructs in multiple species. The reversal learning task is one popular paradigm that probes flexible behavior, aberrations of which are thought to be important in a number of disease states. Despite widespread use, there is a need for a high-throughput primate model that can bridge the genetic, anatomic, and behavioral gap between rodents and humans. Here, we trained squirrel monkeys, a promising preclinical model, on an image-guided deterministic reversal learning task. We found that squirrel monkeys exhibited two key hallmarks of behavior found in other species: integration of reward history over many trials and a side-specific bias. We adapted a reinforcement learning model and demonstrated that it could simulate squirrel monkey-like behavior, capture training-related trajectories, and provide insight into the strategies animals employed. These results validate squirrel monkeys as a model in which to study behavioral flexibility. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Reinforcement Learning Disruptions in Individuals With Depression and Sensitivity to Symptom Change Following Cognitive Behavioral Therapy

IMPORTANCE

Major depressive disorder is prevalent and impairing. Parsing neurocomputational substrates of reinforcement learning in individuals with depression may facilitate a mechanistic understanding of the disorder and suggest new cognitive therapeutic targets.

OBJECTIVE

To determine associations among computational model-derived reinforcement learning parameters, depression symptoms, and symptom changes after treatment.

DESIGN, SETTING, AND PARTICIPANTS

In this mixed cross-sectional-cohort study, individuals performed reward and loss variants of a probabilistic learning task during functional magnetic resonance imaging at baseline and follow-up. A volunteer sample with and without a depression diagnosis was recruited from the community. Participants were assessed from July 2011 to February 2017, and data were analyzed from May 2017 to May 2021.

MAIN OUTCOMES AND MEASURES

Computational model-based analyses of participants' choices assessed a priori hypotheses about associations between components of reward-based and loss-based learning with depression symptoms. Changes in both learning parameters and symptoms were then assessed in a subset of participants who received cognitive behavioral therapy (CBT).

RESULTS

Of 101 included adults, 69 (68.3%) were female, and the mean (SD) age was 34.4 (11.2) years. A total of 69 participants with a depression diagnosis and 32 participants without a depression diagnosis were included at baseline; 48 participants (28 with depression who received CBT and 20 without depression) were included at follow-up (mean [SD] of 115.1 [15.6] days). Computational model-based analyses of behavioral choices and neural data identified associations of learning with symptoms during reward learning and loss learning, respectively. During reward learning only, anhedonia (and not negative affect or arousal) was associated with model-derived learning parameters (learning rate: posterior mean regression β = -0.14; 95% credible interval [CrI], -0.12 to -0.03; outcome sensitivity: posterior mean regression β = 0.18; 95% CrI, 0.02 to 0.37) and neural learning signals (moderation of association between striatal prediction error and expected value signals: t97 = -2.10; P = .04). During loss learning only, negative affect (and not anhedonia or arousal) was associated with learning parameters (outcome shift: posterior mean regression β = -0.11; 95% CrI, -0.20 to -0.01) and disrupted neural encoding of learning signals (association with subgenual anterior cingulate prediction error signals: r = -0.28; P = .005). Symptom improvement following CBT was associated with normalization of learning parameters that were disrupted at baseline (reward learning rate: posterior mean regression β = 0.15; 90% CrI, 0.001 to 0.41; loss outcome shift: posterior mean regression β = 0.42; 90% CrI, 0.09 to 0.77).

CONCLUSIONS AND RELEVANCE

In this study, the mapping of reinforcement learning components to symptoms of major depression revealed mechanistic features associated with these symptoms and points to possible learning-based therapeutic processes and targets.

Towards a pragmatist dealing with algorithmic bias in medical machine learning

Machine Learning (ML) is on the rise in medicine, promising improved diagnostic, therapeutic and prognostic clinical tools. While these technological innovations are bound to transform health care, they also bring new ethical concerns to the forefront. One particularly elusive challenge regards discriminatory algorithmic judgements based on biases inherent in the training data. A common line of reasoning distinguishes between justified differential treatments that mirror true disparities between socially salient groups, and unjustified biases which do not, leading to misdiagnosis and erroneous treatment. In the curation of training data this strategy runs into severe problems though, since distinguishing between the two can be next to impossible. We thus plead for a pragmatist dealing with algorithmic bias in healthcare environments. By recurring to a recent reformulation of William James's pragmatist understanding of truth, we recommend that, instead of aiming at a supposedly objective truth, outcome-based therapeutic usefulness should serve as the guiding principle for assessing ML applications in medicine.

TAPAS: An Open-Source Software Package for Translational Neuromodeling and Computational Psychiatry

Psychiatry faces fundamental challenges with regard to mechanistically guided differential diagnosis, as well as prediction of clinical trajectories and treatment response of individual patients. This has motivated the genesis of two closely intertwined fields: (i) Translational Neuromodeling (TN), which develops "computational assays" for inferring patient-specific disease processes from neuroimaging, electrophysiological, and behavioral data; and (ii) Computational Psychiatry (CP), with the goal of incorporating computational assays into clinical decision making in everyday practice. In order to serve as objective and reliable tools for clinical routine, computational assays require end-to-end pipelines from raw data (input) to clinically useful information (output). While these are yet to be established in clinical practice, individual components of this general end-to-end pipeline are being developed and made openly available for community use. In this paper, we present the Translational Algorithms for Psychiatry-Advancing Science (TAPAS) software package, an open-source collection of building blocks for computational assays in psychiatry. Collectively, the tools in TAPAS presently cover several important aspects of the desired end-to-end pipeline, including: (i) tailored experimental designs and optimization of measurement strategy prior to data acquisition, (ii) quality control during data acquisition, and (iii) artifact correction, statistical inference, and clinical application after data acquisition. Here, we review the different tools within TAPAS and illustrate how these may help provide a deeper understanding of neural and cognitive mechanisms of disease, with the ultimate goal of establishing automatized pipelines for predictions about individual patients. We hope that the openly available tools in TAPAS will contribute to the further development of TN/CP and facilitate the translation of advances in computational neuroscience into clinically relevant computational assays.

Computational approaches and machine learning for individual-level treatment predictions

RATIONALE

The impact of neuroscience-based approaches for psychiatry on pragmatic clinical decision-making has been limited. Although neuroscience has provided insights into basic mechanisms of neural function, these insights have not improved the ability to generate better assessments, prognoses, diagnoses, or treatment of psychiatric conditions.

OBJECTIVES

To integrate the emerging findings in machine learning and computational psychiatry to address the question: what measures that are not derived from the patient's self-assessment or the assessment by a trained professional can be used to make more precise predictions about the individual's current state, the individual's future disease trajectory, or the probability to respond to a particular intervention?

RESULTS

Currently, the ability to use individual differences to predict differential outcomes is very modest possibly related to the fact that the effect sizes of interventions are small. There is emerging evidence of genetic and neuroimaging-based heterogeneity of psychiatric disorders, which contributes to imprecise predictions. Although the use of machine learning tools to generate clinically actionable predictions is still in its infancy, these approaches may identify subgroups enabling more precise predictions. In addition, computational psychiatry might provide explanatory disease models based on faulty updating of internal values or beliefs.

CONCLUSIONS

There is a need for larger studies, clinical trials using machine learning, or computational psychiatry model parameters predictions as actionable outcomes, comparing alternative explanatory computational models, and using translational approaches that apply similar paradigms and models in humans and animals.

Can Machines Find the Bilingual Advantage? Machine Learning Algorithms Find No Evidence to Differentiate Between Lifelong Bilingual and Monolingual Cognitive Profiles

Bilingualism has been identified as a potential cognitive factor linked to delayed onset of dementia as well as boosting executive functions in healthy individuals. However, more recently, this claim has been called into question following several failed replications. It remains unclear whether these contradictory findings reflect how bilingualism is defined between studies, or methodological limitations when measuring the bilingual effect. One key issue is that despite the claims that bilingualism yields general protection to cognitive processes (i.e., the cognitive reserve hypothesis), studies reporting putative bilingual differences are often focused on domain specific experimental paradigms. This study chose a broader approach, by considering the consequences of bilingualism on a wide range of cognitive functions within individuals. We utilised 19 measures of different cognitive functions commonly associated with bilingual effects, to form a "cognitive profile" for 215 non-clinical participants. We recruited Welsh speakers, who as a group of bilinguals were highly homogeneous, as means of isolating the bilingualism criterion. We sought to determine if such analyses would independently classify bilingual/monolingual participant groups based on emergent patterns driven by collected cognitive profiles, such that population differences would emerge. Multiple predictive models were trained to independently recognise the cognitive profiles of bilinguals, older adults (60-90 years of age) and higher education attainment. Despite managing to successfully classify cognitive profiles based on age and education, the model failed to differentiate between bilingual and monolingual cognitive ability at a rate greater than that of chance. Repeated modelling using alternative definitions of bilingualism, and just the older adults, yielded similar results. In all cases then, using our "bottom-up" analytical approach, there was no evidence that bilingualism as a variable indicated differential cognitive performance - as a consequence, we conclude that bilinguals are not cognitively different from their monolingual counterparts, even in older demographics. We suggest that studies that have reported a bilingual advantage (typically recruiting immigrant populations) could well have confounded other key variables that may be driving reported advantages. We recommend that future research refine the machine learning methods used in this study to further investigate the complex relationship between bilingualism and cognition.

Impaired Evidence Accumulation as a Transdiagnostic Vulnerability Factor in Psychopathology

There is substantial interest in identifying biobehavioral dimensions of individual variation that cut across heterogenous disorder categories, and computational models can play a major role in advancing this goal. In this report, we focused on efficiency of evidence accumulation (EEA), a computationally characterized variable derived from sequential sampling models of choice tasks. We created an EEA factor from three behavioral tasks in the UCLA Phenomics dataset (n = 272), which includes healthy participants (n = 130) as well-participants with schizophrenia (n = 50), bipolar disorder (n = 49), and attention-deficit/hyperactivity disorder (n = 43). We found that the EEA factor was significantly reduced in all three disorders, and that it correlated with an overall severity score for psychopathology as well as self-report measures of impulsivity. Although EEA was significantly correlated with general intelligence, it remained associated with psychopathology and symptom scales even after controlling for intelligence scores. Taken together, these findings suggest EEA is a promising computationally-characterized dimension of neurocognitive variation, with diminished EEA conferring transdiagnostic vulnerability to psychopathology.

Computational Models of Interoception and Body Regulation

To survive, organisms must effectively respond to the challenge of maintaining their physiological integrity in the face of an ever-changing environment. Preserving this homeostasis critically relies on adaptive behavior. In this review, we consider recent frameworks that extend classical homeostatic control via reflex arcs to include more flexible forms of adaptive behavior that take interoceptive context, experiences, and expectations into account. Specifically, we define a landscape for computational models of interoception, body regulation, and forecasting, address these models' unique challenges in relation to translational research efforts, and discuss what they can teach us about cognition as well as physical and mental health.

Driven by Pain, Not Gain: Computational Approaches to Aversion-Related Decision Making in Psychiatry

Although it is well known that "losses loom larger than gains," computational approaches to aversion-related decision making (ARDM) for psychiatric disorders is an underdeveloped area. Computational models of ARDM have been implemented primarily as state-dependent reinforcement learning models with bias parameters to quantify Pavlovian associations, and differential learning rates to quantify instrumental updating have been shown to depend on context, involve complex cost calculations, and include the consideration of counterfactual outcomes. Little is known about how individual differences influence these models relevant to anxiety-related conditions or addiction-related dysfunction. It is argued that model parameters reflecting 1) Pavlovian biases in the context of reinforcement learning or 2) hyperprecise prior beliefs in the context of active inference play an important role in the emergence of dysfunctional avoidance behaviors. The neural implementation of ARDM includes brain areas that are important for valuation (ventromedial prefrontal cortex) and positive reinforcement-related prediction errors (ventral striatum), but also aversive processing (insular cortex and cerebellum). Computational models of ARDM will help to establish a quantitative explanatory account of the development of anxiety disorders and addiction, but such models also face several challenges, including limited evidence for stability of individual differences, relatively low reliability of tasks, and disorder heterogeneity. Thus, it will be necessary to develop robust, reliable, and model-based experimental probes; recruit larger sample sizes; and use single case experimental designs for better pragmatic and explanatory biological models of psychiatric disorders.

Improving the Reliability of Computational Analyses: Model-Based Planning and Its Relationship With Compulsivity

BACKGROUND

Computational models show great promise in mapping latent decision-making processes onto dissociable neural substrates and clinical phenotypes. One prominent example in reinforcement learning is model-based planning, which specifically relates to transdiagnostic compulsivity. However, the reliability of computational model-derived measures such as model-based planning is unclear. Establishing reliability is necessary to ensure that such models measure stable, traitlike processes, as assumed in computational psychiatry. Although analysis approaches affect validity of reinforcement learning models and reliability of other task-based measures, their effect on reliability of reinforcement learning models of empirical data has not been systematically studied.

METHODS

We first assessed within- and across-session reliability and effects of analysis approaches (model estimation, parameterization, and data cleaning) of measures of model-based planning in patients with compulsive disorders (n = 38). The analysis approaches affecting test-retest reliability were tested in 3 large generalization samples (healthy participants: n = 541 and 111; people with a range of compulsivity: n = 1413).

RESULTS

Analysis approaches greatly influenced reliability: reliability of model-based planning measures ranged from 0 (no concordance) to above 0.9 (acceptable for clinical applications). The largest influence on reliability was whether model-estimation approaches were robust and accounted for the hierarchical structure of estimated parameters. Improvements in reliability generalized to other datasets and greatly reduced the sample size needed to find a relationship between model-based planning and compulsivity in an independent dataset.

CONCLUSIONS

These results indicate that computational psychiatry measures such as model-based planning can reliably measure latent decision-making processes, but when doing so must assess the ability of methods to estimate complex models from limited data.

Mapping infant neurodevelopmental precursors of mental disorders: How synthetic cohorts & computational approaches can be used to enhance prediction of early childhood psychopathology

Bridging advances in neurodevelopmental assessment and the established onset of common psychopathologies in early childhood with epidemiological data science and computational methods holds much promise for identifying risk for mental disorders as early as infancy. In particular, we propose the development of a mental health risk algorithm for the early detection of mental disorders with the potential for high public health impact that applies and adapts methods innovated in and successfully applied to early detection of cardiovascular risk. Specifically, we propose methods to advance risk prediction of early developmental psychopathology by creating synthetic cohorts that contain complete behavioral and neural data in the first years of life, as the basis for a robust and generalizable risk algorithm. The application of computational approaches within synthetic cohorts, an approach increasingly applied in psychiatry, may be particularly well suited to advancing risk prediction in early childhood mental health. We propose new research directions using these methods to generate an early childhood mental health risk calculator that could significantly advance early mental health risk detection to direct preventive intervention and/or need for more intensive assessment within a pragmatic framework for maximal clinical utility. The availability of such a tool in early childhood, a period of high neuroplasticity, holds promise to reduce the burden of mental disorder by identifying risk early in the clinical sequence and delivering prevention that targets the neurodevelopmental vulnerability phase.

Neuroimaging reward, craving, learning, and cognitive control in substance use disorders: review and implications for treatment

Substance use disorder is a leading causes of preventable disease and mortality. Drugs of abuse cause molecular and cellular changes in specific brain regions and these neuroplastic changes are thought to play a role in the transition to uncontrolled drug use. Neuroimaging has identified neural substrates associated with problematic substance use and may offer clues to reduce its burden on the patient and society. Here, we provide a narrative review of neuroimaging studies that have examined the structures and circuits associated with reward, cues and craving, learning, and cognitive control in substance use disorders. Most studies use advanced MRI or positron emission tomography (PET). Many studies have focused on the dopamine neurons of the ventral tegmental area, and the regions where these neurons terminate, such as the striatum and prefrontal cortex. Decreases in dopamine receptors and transmission have been found in chronic users of drugs, alcohol, and nicotine. Recent studies also show evidence of differences in structure and function in substance users relative to controls in brain regions involved in salience evaluation, such as the insula and anterior cingulate cortex. Balancing between reward-related bottom-up and cognitive-control-related top-down processes is discussed in the context of neuromodulation as a potential treatment. Finally, some of the challenges for understanding substance use disorder using neuroimaging methods are discussed.

Identifying subtypes of Hypersomnolence Disorder: a clustering analysis

BACKGROUND

Patient heterogeneity is problematic for the accurate assessment and effective treatment of Hypersomnolence Disorder. Clustering analysis is a preferred approach for establishing homogenous subclassifications. Thus, this investigation aimed to identify more homogeneous subclassifications of Hypersomnolence Disorder through clustering analysis.

METHODS

Patients undergoing polysomnography (PSG) and multiple sleep latency test (MSLT) assessment for hypersomnolence were recruited as part of a larger investigation. A sample of patients with Hypersomnolence Disorder was determined based on a post hoc chart review protocol. After removing persons with missing data, 62 participants were included in the analyses. Self-report total sleep time, Epworth Sleepiness Scale (ESS) score, and Sleep Inertia Questionnaire (SIQ) score were chosen as clustering variables to mirror Hypersomnolence Disorder diagnostic traits. A statistically-driven clustering process produced two clusters using Ward's D hierarchical approach. Clusters were compared across characteristics, self-report measures, PSG/MSLT results, and additional objective measures.

RESULTS

The resulting clusters differed across a variety of hypersomnolence-related subjective metrics and objective measurements. A more severe hypersomnolence phenotype was identified in a cluster that also had elevated depressive symptoms. This cluster endorsed significantly greater daytime sleepiness, sleep inertia, and functional impairment, while displaying longer sleep duration and worse vigilance.

CONCLUSIONS

These results provide growing support for a nosological reformulation of hypersomnolence associated with psychiatric disorders. Future research is necessary to solidify the conceptualization and characterization of unexplained hypersomnolence presenting with-and-without psychiatric illness.

Machine learning and big data in psychiatry: toward clinical applications

Psychiatry is a medical field concerned with the treatment of mental illness. Psychiatric disorders broadly relate to higher functions of the brain, and as such are richly intertwined with social, cultural, and experiential factors. This makes them exquisitely complex phenomena that depend on and interact with a large number of variables. Computational psychiatry provides two ways of approaching this complexity. Theory-driven computational approaches employ mechanistic models to make explicit hypotheses at multiple levels of analysis. Data-driven machine-learning approaches can make predictions from high-dimensional data and are generally agnostic as to the underlying mechanisms. Here, we review recent advances in the use of big data and machine-learning approaches toward the aim of alleviating the suffering that arises from psychiatric disorders.

Bayesian computational markers of relapse in methamphetamine dependence

Methamphetamine use disorder is associated with a high likelihood of relapse. Identifying robust predictors of relapse that have explanatory power is critical to develop secondary prevention based on a mechanistic understanding of relapse. Computational approaches have the potential to identify such predictive markers of psychiatric illness, with the advantage of providing a finer mechanistic explanation of the cognitive processes underlying psychiatric vulnerability.

In this study, sixty-two recently sober methamphetamine-dependent individuals were recruited from a 28-day inpatient treatment program, and completed a Stop Signal Task (SST) while undergoing functional magnetic resonance imaging (fMRI). These individuals were prospectively followed for 1 year and assessed for relapse to methamphetamine use. Thirty-three percent of followed participants reported relapse.

We found that neural activity associated with two types of Bayesian prediction error, i.e. the difference between actual and expected need to stop on a given trial, significantly differentiated those individuals who remained abstinent and those who relapsed. Specifically, relapsed individuals exhibited smaller neural activations to such Bayesian prediction errors relative to those individuals who remained abstinent in the left temporoparietal junction (Cohen's d = 0.91), the left inferior frontal gyrus (Cohen's d = 0.57), and left anterior insula (Cohen's d = 0.63). In contrast, abstinent and relapsed participants did not differ in neural activation to non-model based task contrasts or on various self-report clinical measures.

In conclusion, Bayesian cognitive models may help identify predictive biomarkers of relapse, while providing a computational explanation of belief processing and updating deficits in individuals with methamphetamine use disorder.

•

Methamphetamine-dependent individuals (MDI) face a high rate of relapse after treatment.

•

Can a Bayesian learning modeling and fMRI be used to identify markers of relapse?

•

MDI who relapsed within 1 year have smaller activation to Bayesian model-based prediction errors.

•

Such neural pattern was observed in left temporo-parietal junction, IFG, and anterior insula.

•

MDI more likely to relapse show weaker tracking of uncertainty and updating of their belief model.

COMPASS: An Open-Source, General-Purpose Software Toolkit for Computational Psychiatry

Mathematical modeling of behavior during a psychophysical task, referred to as "computational psychiatry," could greatly improve our understanding of mental disorders. One barrier to the broader adoption of computational methods, is that they often require advanced statistical modeling and mathematical skills. Biological and behavioral signals often show skewed or non-Gaussian distributions, and very few toolboxes and analytical platforms are capable of processing such signal categories. We developed the Computational Psychiatry Adaptive State-Space (COMPASS) toolbox, an open-source MATLAB-based software package. This toolbox is easy to use and capable of integrating signals with a variety of distributions. COMPASS has the tools to process signals with continuous-valued and binary measurements, or signals with incomplete-missing or censored-measurements, which makes it well-suited for processing those signals captured during a psychophysical task. After specifying a few parameters in a small set of user-friendly functions, COMPASS allows users to efficiently apply a wide range of computational behavioral models. The model output can be analyzed as an experimental outcome or used as a regressor for neural data and can also be tested using the goodness-of-fit measurement. Here, we demonstrate that COMPASS can replicate two computational behavioral analyses from different groups. COMPASS replicates and can slightly improve on the original modeling results. We also demonstrate the use of COMPASS application in a censored-data problem and compare its performance result with naïve estimation methods. This flexible, general-purpose toolkit should accelerate the use of computational modeling in psychiatric neuroscience.

Prediction Models of Functional Outcomes for Individuals in the Clinical High-Risk State for Psychosis or With Recent-Onset Depression: A Multimodal, Multisite Machine Learning Analysis

IMPORTANCE

Social and occupational impairments contribute to the burden of psychosis and depression. There is a need for risk stratification tools to inform personalized functional-disability preventive strategies for individuals in at-risk and early phases of these illnesses.

OBJECTIVE

To determine whether predictors associated with social and role functioning can be identified in patients in clinical high-risk (CHR) states for psychosis or with recent-onset depression (ROD) using clinical, imaging-based, and combined machine learning; assess the geographic, transdiagnostic, and prognostic generalizability of machine learning and compare it with human prognostication; and explore sequential prognosis encompassing clinical and combined machine learning.

DESIGN, SETTING, AND PARTICIPANTS

This multisite naturalistic study followed up patients in CHR states, with ROD, and with recent-onset psychosis, and healthy control participants for 18 months in 7 academic early-recognition services in 5 European countries. Participants were recruited between February 2014 and May 2016, and data were analyzed from April 2017 to January 2018.

AIN OUTCOMES AND MEASURES

Performance and generalizability of prognostic models.

RESULTS

A total of 116 individuals in CHR states (mean [SD] age, 24.0 [5.1] years; 58 [50.0%] female) and 120 patients with ROD (mean [SD] age, 26.1 [6.1] years; 65 [54.2%] female) were followed up for a mean (SD) of 329 (142) days. Machine learning predicted the 1-year social-functioning outcomes with a balanced accuracy of 76.9% of patients in CHR states and 66.2% of patients with ROD using clinical baseline data. Balanced accuracy in models using structural neuroimaging was 76.2% in patients in CHR states and 65.0% in patients with ROD, and in combined models, it was 82.7% for CHR states and 70.3% for ROD. Lower functioning before study entry was a transdiagnostic predictor. Medial prefrontal and temporo-parieto-occipital gray matter volume (GMV) reductions and cerebellar and dorsolateral prefrontal GMV increments had predictive value in the CHR group; reduced mediotemporal and increased prefrontal-perisylvian GMV had predictive value in patients with ROD. Poor prognoses were associated with increased risk of psychotic, depressive, and anxiety disorders at follow-up in patients in the CHR state but not ones with ROD. Machine learning outperformed expert prognostication. Adding neuroimaging machine learning to clinical machine learning provided a 1.9-fold increase of prognostic certainty in uncertain cases of patients in CHR states, and a 10.5-fold increase of prognostic certainty for patients with ROD.

CONCLUSIONS AND RELEVANCE

Precision medicine tools could augment effective therapeutic strategies aiming at the prevention of social functioning impairments in patients with CHR states or with ROD.

Computational Phenotyping: Using Models to Understand Individual Differences in Personality, Development, and Mental Illness

This paper reviews progress in the application of computational models to personality, developmental, and clinical neuroscience. We first describe the concept of a computational phenotype, a collection of parameters derived from computational models fit to behavioral and neural data. This approach represents individuals as points in a continuous parameter space, complementing traditional trait and symptom measures. One key advantage of this representation is that it is mechanistic: The parameters have interpretations in terms of cognitive processes, which can be translated into quantitative predictions about future behavior and brain activity. We illustrate with several examples how this approach has led to new scientific insights into individual differences, developmental trajectories, and psychopathology. We then survey some of the challenges that lay ahead.

A Low-Level Perceptual Correlate of Behavioral and Clinical Deficits in ADHD

In many studies of attention-deficit hyperactivity disorder (ADHD), stimulus encoding and processing (perceptual function) and response selection (executive function) have been intertwined. To dissociate deficits in these functions, we introduced a task that parametrically varied low-level stimulus features (orientation and color) for fine-grained analysis of perceptual function. It also required participants to switch their attention between feature dimensions on a trial-by-trial basis, thus taxing executive processes. Furthermore, we used a response paradigm that captured task-irrelevant motor output (TIMO), reflecting failures to use the correct stimulus-response rule. ADHD participants had substantially higher perceptual variability than controls, especially for orientation, as well as higher TIMO. In both ADHD and controls, TIMO was strongly affected by the switch manipulation. Across participants, the perceptual variability parameter was correlated with TIMO, suggesting that perceptual deficits are associated with executive function deficits. Based on perceptual variability alone, we were able to classify participants into ADHD and controls with a mean accuracy of about 77%. Participants' self-reported General Executive Composite score correlated not only with TIMO but also with the perceptual variability parameter. Our results highlight the role of perceptual deficits in ADHD and the usefulness of computational modeling of behavior in dissociating perceptual from executive processes.

Psychiatric Symptom Dimensions Are Associated With Dissociable Shifts in Metacognition but Not Task Performance

BACKGROUND

Distortions in metacognition-the ability to reflect on and control other cognitive processes-are thought to be characteristic of poor mental health. However, it remains unknown whether such shifts in self-evaluation are due to specific alterations in metacognition and/or a downstream consequence of changes in decision-making processes.

METHODS

Using perceptual decision making as a model system, we employed a computational psychiatry approach to relate parameters governing both decision formation and metacognitive evaluation to self-reported transdiagnostic symptom dimensions in a large general population sample (N = 995).

RESULTS

Variability in psychopathology was unrelated to either speed or accuracy of decision formation. In contrast, leveraging a dimensional approach, we revealed independent relationships between psychopathology and metacognition: a symptom dimension related to anxiety and depression was associated with lower confidence and heightened metacognitive efficiency, whereas a dimension characterizing compulsive behavior and intrusive thoughts was associated with higher confidence and lower metacognitive efficiency. Furthermore, we obtained a robust double dissociation-whereas psychiatric symptoms predicted changes in metacognition but not decision performance, age predicted changes in decision performance but not metacognition.

CONCLUSIONS

Our findings indicate a specific and pervasive link between metacognition and mental health. Our study bridges a gap between an emerging neuroscience of decision making and an understanding of metacognitive alterations in psychopathology.

Human metacognition across domains: insights from individual differences and neuroimaging

Metacognition is the capacity to evaluate and control one's own cognitive processes. Metacognition operates over a range of cognitive domains, such as perception and memory, but the neurocognitive architecture supporting this ability remains controversial. Is metacognition enabled by a common, domain-general resource that is recruited to evaluate performance on a variety of tasks? Or is metacognition reliant on domain-specific modules? This article reviews recent literature on the domain-generality of human metacognition, drawing on evidence from individual differences and neuroimaging. A meta-analysis of behavioral studies found that perceptual metacognitive ability was correlated across different sensory modalities, but found no correlation between metacognition of perception and memory. However, evidence for domain-generality from behavioral data may suffer from a lack of power to identify correlations across model parameters indexing metacognitive efficiency. Neuroimaging data provide a complementary perspective on the domain-generality of metacognition, revealing co-existence of neural signatures that are common and distinct across tasks. We suggest that such an architecture may be appropriate for "tagging" generic feelings of confidence with domain-specific information, in turn forming the basis for priors about self-ability and modulation of higher-order behavioral control.

Adolescent neurocognitive development and impacts of substance use: Overview of the adolescent brain cognitive development (ABCD) baseline neurocognition battery

Adolescence is characterized by numerous social, hormonal and physical changes, as well as a marked increase in risk-taking behaviors. Dual systems models attribute adolescent risk-taking to tensions between developing capacities for cognitive control and motivational strivings, which may peak at this time. A comprehensive understanding of neurocognitive development during the adolescent period is necessary to permit the distinction between premorbid vulnerabilities and consequences of behaviors such as substance use. Thus, the prospective assessment of cognitive development is fundamental to the aims of the newly launched Adolescent Brain and Cognitive Development (ABCD) Consortium. This paper details the rationale for ABC'lected measures of neurocognition, presents preliminary descriptive data on an initial sample of 2299 participants, and provides a context for how this large-scale project can inform our understanding of adolescent neurodevelopment.

Multivariate Pattern Analysis of Genotype-Phenotype Relationships in Schizophrenia

Genetic risk variants for schizophrenia have been linked to many related clinical and biological phenotypes with the hopes of delineating how individual variation across thousands of variants corresponds to the clinical and etiologic heterogeneity within schizophrenia. This has primarily been done using risk score profiling, which aggregates effects across all variants into a single predictor. While effective, this method lacks flexibility in certain domains: risk scores cannot capture nonlinear effects and do not employ any variable selection. We used random forest, an algorithm with this flexibility designed to maximize predictive power, to predict 6 cognitive endophenotypes in a combined sample of psychiatric patients and controls (N = 739) using 77 genetic variants strongly associated with schizophrenia. Tenfold cross-validation was applied to the discovery sample and models were externally validated in an independent sample of similar ancestry (N = 336). Linear approaches, including linear regression and task-specific polygenic risk scores, were employed for comparison. Random forest models for processing speed (P = .019) and visual memory (P = .036) and risk scores developed for verbal (P = .042) and working memory (P = .037) successfully generalized to an independent sample with similar predictive strength and error. As such, we suggest that both methods may be useful for mapping a limited set of predetermined, disease-associated SNPs to related phenotypes. Incorporating random forest and other more flexible algorithms into genotype-phenotype mapping inquiries could contribute to parsing heterogeneity within schizophrenia; such algorithms can perform as well as standard methods and can capture a more comprehensive set of potential relationships.

Generative models for clinical applications in computational psychiatry

Despite the success of modern neuroimaging techniques in furthering our understanding of cognitive and pathophysiological processes, translation of these advances into clinically relevant tools has been virtually absent until now. Neuromodeling represents a powerful framework for overcoming this translational deadlock, and the development of computational models to solve clinical problems has become a major scientific goal over the last decade, as reflected by the emergence of clinically oriented neuromodeling fields like Computational Psychiatry, Computational Neurology, and Computational Psychosomatics. Generative models of brain physiology and connectivity in the human brain play a key role in this endeavor, striving for computational assays that can be applied to neuroimaging data from individual patients for differential diagnosis and treatment prediction. In this review, we focus on dynamic causal modeling (DCM) and its use for Computational Psychiatry. DCM is a widely used generative modeling framework for functional magnetic resonance imaging (fMRI) and magneto-/electroencephalography (M/EEG) data. This article reviews the basic concepts of DCM, revisits examples where it has proven valuable for addressing clinically relevant questions, and critically discusses methodological challenges and recent methodological advances. We conclude this review with a more general discussion of the promises and pitfalls of generative models in Computational Psychiatry and highlight the path that lies ahead of us. This article is categorized under: Neuroscience > Computation Neuroscience > Clinical Neuroscience.

Neuromarkers for Mental Disorders: Harnessing Population Neuroscience

Despite abundant research into the neurobiology of mental disorders, to date neurobiological insights have had very little impact on psychiatric diagnosis or treatment. In this review, we contend that the search for neuroimaging biomarkers-neuromarkers-of mental disorders is a highly promising avenue toward improved psychiatric healthcare. However, many of the traditional tools used for psychiatric neuroimaging are inadequate for the identification of neuromarkers. Specifically, we highlight the need for larger samples and for multivariate analysis. Approaches such as machine learning are likely to be beneficial for interrogating high-dimensional neuroimaging data. We suggest that broad, population-based study designs will be important for developing neuromarkers of mental disorders, and will facilitate a move away from a phenomenological definition of mental disorder categories and toward psychiatric nosology based on biological evidence. We provide an outline of how the development of neuromarkers should occur, emphasizing the need for tests of external and construct validity, and for collaborative research efforts. Finally, we highlight some concerns regarding the development, and use of, neuromarkers in psychiatric healthcare.

The Insular Cortex Dynamically Maps Changes in Cardiorespiratory Interoception

Palpitations and dyspnea are fundamental to the human experience of panic anxiety, but it remains unclear how the brain dynamically represents changes in these interoceptive sensations. We used isoproterenol, a rapidly acting peripheral beta-adrenergic agonist similar to adrenaline, to induce sensations of palpitation and dyspnea in healthy individuals (n=23) during arterial spin labeling functional magnetic resonance imaging (fMRI). We hypothesized that the right mid-insular cortex, a central recipient of viscerosensory input, would preferentially respond during the peak period of cardiorespiratory stimulation. Bolus infusions of saline and isoproterenol (1 or 2 μg) were administered in a blinded manner while participants continuously rated the intensity of their cardiorespiratory sensation using a dial. Isoproterenol elicited dose-dependent increases in cardiorespiratory sensation, with all participants reporting palpitations and dyspnea at the 2 μg dose. Consistent with our hypothesis, the right mid-insula was maximally responsive during the peak period of sympathetic arousal, heart rate increase, and cardiorespiratory sensation. Furthermore, a shift in insula activity occurred during the recovery period, after the heart rate had largely returned to baseline levels, with an expansion of activation into anterior and posterior sectors of the right insula, as well as bilateral regions of the mid-insula. These results confirm the right mid-insula is a key node in the interoceptive network, and inform computational models proposing specific processing roles for insula subregions during homeostatic inference. The combination of isoproterenol and fMRI offers a powerful approach for evaluating insula function, and could be a useful probe for examining interoceptive dysfunction in psychiatric disorders.

Premonitory urges and tics in Tourette syndrome: computational mechanisms and neural correlates

Tourette syndrome is characterized by open motor behaviors - tics - but another crucial aspect of the disorder is the presence of premonitory urges: uncomfortable sensations that typically precede tics and are temporarily alleviated by tics. We review the evidence implicating the somatosensory cortices and the insula in premonitory urges and the motor cortico-basal ganglia-thalamo-cortical loop in tics. We consider how these regions interact during tic execution, suggesting that the insula plays an important role as a nexus linking the sensory and emotional character of premonitory urges with their translation into tics. We also consider how these regions interact during tic learning, integrating the neural evidence with a computational perspective on how premonitory-urge alleviation reinforces tics.

Computational Psychosomatics and Computational Psychiatry: Toward a Joint Framework for Differential Diagnosis

This article outlines how a core concept from theories of homeostasis and cybernetics, the inference-control loop, may be used to guide differential diagnosis in computational psychiatry and computational psychosomatics. In particular, we discuss 1) how conceptualizing perception and action as inference-control loops yields a joint computational perspective on brain-world and brain-body interactions and 2) how the concrete formulation of this loop as a hierarchical Bayesian model points to key computational quantities that inform a taxonomy of potential disease mechanisms. We consider the utility of this perspective for differential diagnosis in concrete clinical applications.

The Stochastic Early Reaction, Inhibition, and late Action (SERIA) model for antisaccades

The antisaccade task is a classic paradigm used to study the voluntary control of eye movements. It requires participants to suppress a reactive eye movement to a visual target and to concurrently initiate a saccade in the opposite direction. Although several models have been proposed to explain error rates and reaction times in this task, no formal model comparison has yet been performed. Here, we describe a Bayesian modeling approach to the antisaccade task that allows us to formally compare different models on the basis of their evidence. First, we provide a formal likelihood function of actions (pro- and antisaccades) and reaction times based on previously published models. Second, we introduce the Stochastic Early Reaction, Inhibition, and late Action model (SERIA), a novel model postulating two different mechanisms that interact in the antisaccade task: an early GO/NO-GO race decision process and a late GO/GO decision process. Third, we apply these models to a data set from an experiment with three mixed blocks of pro- and antisaccade trials. Bayesian model comparison demonstrates that the SERIA model explains the data better than competing models that do not incorporate a late decision process. Moreover, we show that the early decision process postulated by the SERIA model is, to a large extent, insensitive to the cue presented in a single trial. Finally, we use parameter estimates to demonstrate that changes in reaction time and error rate due to the probability of a trial type (pro- or antisaccade) are best explained by faster or slower inhibition and the probability of generating late voluntary prosaccades.