From promise to practice: towards the realisation of AI-informed mental health care.

In this Series paper, we explore the promises and challenges of artificial intelligence (AI)-based precision medicine tools in mental health care from clinical, ethical, and regulatory perspectives. The real-world implementation of these tools is increasingly considered the prime solution for key issues in mental health, such as delayed, inaccurate, and inefficient care delivery. Similarly, machine-learning-based empirical strategies are becoming commonplace in psychiatric research because of their potential to adequately deconstruct the biopsychosocial complexity of mental health disorders, and hence to improve nosology of prognostic and preventive paradigms. However, the implementation steps needed to translate these promises into practice are currently hampered by multiple interacting challenges. These obstructions range from the current technology-distant state of clinical practice, over the lack of valid real-world databases required to feed data-intensive AI algorithms, to model development and validation considerations being disconnected from the core principles of clinical utility and ethical acceptability. In this Series paper, we provide recommendations on how these challenges could be addressed from an interdisciplinary perspective to pave the way towards a framework for mental health care, leveraging the combined strengths of human intelligence and AI.

The promise of a model-based psychiatry: building computational models of mental ill health.

Computational models have great potential to revolutionise psychiatry research and clinical practice. These models are now used across multiple subfields, including computational psychiatry and precision psychiatry. Their goals vary from understanding mechanisms underlying disorders to deriving reliable classification and personalised predictions. Rapid growth of new tools and data sources (eg, digital data, gamification, and social media) requires an understanding of the constraints and advantages of different modelling approaches in psychiatry. In this Series paper, we take a critical look at the range of computational models that are used in psychiatry and evaluate their advantages and disadvantages for different purposes and data sources. We describe mechanism-driven and mechanism-agnostic computational models and discuss how interpretability of models is crucial for clinical translation. Based on these evaluations, we provide recommendations on how to build computational models that are clinically useful.

Cerebellar anaplastic astrocytoma in adult patients: 15 consecutive cases from a single institution and literature review.

Adult cerebellar anaplastic astrocytomas (cAA) are rare entities and their clinical and genetic appearances are still ill defined. Previously, malignant gliomas of the cerebellum were combined and reviewed together (cAA and cerebellar glioblastomas (cGB), that could have possibly affected overall survival (OS) and progression-free survival (PFS). We present characteristics of 15 adult patients with cAA and compared them to a series of 45 patients with a supratentorial AA (sAA) in order to elicit the effect of tumor location on OS and PFS. The mean age at cAA diagnosis was 39.3 years (range 19-72). A history of neurofibromatosis type I was noted in 1 patient (6.7%). An IDH-1 mutation was identified in 6/15 cases and a methylated MGMT promoter in 5/15 cases. Patients in study and control groups were matched in age, sex and IDH-1 mutation status. Patients in a study group tended to present with longer overall survival (50 vs. 36.5 months), but the difference did not reach statistical significance. In both cAA and supratentorial AA groups presence of the IDH-1 mutation remains a positive predictor for the prolonged survival. The present study suggests that adult cAA constitute a group of gliomas with relatively higher rate of IDH-1 mutations and prognosis similar to supratentorial AA. The present study is the first to systematically compare cAA and supratentorial AA with respect to their genetic characteristics and suggests that both groups show a similar survival prognosis.

Obsessive-compulsive symptoms and information seeking during the Covid-19 pandemic.

Increased mental-health symptoms as a reaction to stressful life events, such as the Covid-19 pandemic, are common. Critically, successful adaptation helps to reduce such symptoms to baseline, preventing long-term psychiatric disorders. It is thus important to understand whether and which psychiatric symptoms show transient elevations, and which persist long-term and become chronically heightened. At particular risk for the latter trajectory are symptom dimensions directly affected by the pandemic, such as obsessive-compulsive (OC) symptoms. In this longitudinal large-scale study (N = 406), we assessed how OC, anxiety and depression symptoms changed throughout the first pandemic wave in a sample of the general UK public. We further examined how these symptoms affected pandemic-related information seeking and adherence to governmental guidelines. We show that scores in all psychiatric domains were initially elevated, but showed distinct longitudinal change patterns. Depression scores decreased, and anxiety plateaued during the first pandemic wave, while OC symptoms further increased, even after the ease of Covid-19 restrictions. These OC symptoms were directly linked to Covid-related information seeking, which gave rise to higher adherence to government guidelines. This increase of OC symptoms in this non-clinical sample shows that the domain is disproportionately affected by the pandemic. We discuss the long-term impact of the Covid-19 pandemic on public mental health, which calls for continued close observation of symptom development.

A Causal Role for the Pedunculopontine Nucleus in Human Instrumental Learning.

A critical mechanism for maximizing reward is instrumental learning. In standard instrumental learning models, action values are updated on the basis of reward prediction errors (RPEs), defined as the discrepancy between expectations and outcomes. A wealth of evidence across species and experimental techniques has established that RPEs are signaled by midbrain dopamine neurons. However, the way dopamine neurons receive information about reward outcomes remains poorly understood. Recent animal studies suggest that the pedunculopontine nucleus (PPN), a small brainstem structure considered as a locomotor center, is sensitive to reward and sends excitatory projection to dopaminergic nuclei. Here, we examined the hypothesis that the PPN could contribute to reward learning in humans. To this aim, we leveraged a clinical protocol that assessed the therapeutic impact of PPN deep-brain stimulation (DBS) in three patients with Parkinson disease. PPN local field potentials (LFPs), recorded while patients performed an instrumental learning task, showed a specific response to reward outcomes in a low-frequency (alpha-beta) band. Moreover, PPN DBS selectively improved learning from rewards but not from punishments, a pattern that is typically observed following dopaminergic treatment. Computational analyses indicated that the effect of PPN DBS on instrumental learning was best captured by an increase in subjective reward sensitivity. Taken together, these results support a causal role for PPN-mediated reward signals in human instrumental learning.

Computational noise in reward-guided learning drives behavioral variability in volatile environments.

When learning the value of actions in volatile environments, humans often make seemingly irrational decisions that fail to maximize expected value. We reasoned that these 'non-greedy' decisions, instead of reflecting information seeking during choice, may be caused by computational noise in the learning of action values. Here using reinforcement learning models of behavior and multimodal neurophysiological data, we show that the majority of non-greedy decisions stem from this learning noise. The trial-to-trial variability of sequential learning steps and their impact on behavior could be predicted both by blood oxygen level-dependent responses to obtained rewards in the dorsal anterior cingulate cortex and by phasic pupillary dilation, suggestive of neuromodulatory fluctuations driven by the locus coeruleus-norepinephrine system. Together, these findings indicate that most behavioral variability, rather than reflecting human exploration, is due to the limited computational precision of reward-guided learning.

How context alters value: The brain's valuation and affective regulation system link price cues to experienced taste pleasantness.

Informational cues such as the price of a wine can trigger expectations about its taste quality and thereby modulate the sensory experience on a reported and neural level. Yet it is unclear how the brain translates such expectations into sensory pleasantness. We used a whole-brain multilevel mediation approach with healthy participants who tasted identical wines cued with different prices while their brains were scanned using fMRI. We found that the brain's valuation system (BVS) in concert with the anterior prefrontal cortex played a key role in implementing the effect of price cues on taste pleasantness ratings. The sensitivity of the BVS to monetary rewards outside the taste domain moderated the strength of these effects. These findings provide novel evidence for the fundamental role that neural pathways linked to motivation and affective regulation play for the effect of informational cues on sensory experiences.

A Selective Role for Dopamine in Learning to Maximize Reward But Not to Minimize Effort: Evidence from Patients with Parkinson's Disease.

Instrumental learning is a fundamental process through which agents optimize their choices, taking into account various dimensions of available options such as the possible reward or punishment outcomes and the costs associated with potential actions. Although the implication of dopamine in learning from choice outcomes is well established, less is known about its role in learning the action costs such as effort. Here, we tested the ability of patients with Parkinson's disease (PD) to maximize monetary rewards and minimize physical efforts in a probabilistic instrumental learning task. The implication of dopamine was assessed by comparing performance ON and OFF prodopaminergic medication. In a first sample of PD patients (n = 15), we observed that reward learning, but not effort learning, was selectively impaired in the absence of treatment, with a significant interaction between learning condition (reward vs effort) and medication status (OFF vs ON). These results were replicated in a second, independent sample of PD patients (n = 20) using a simplified version of the task. According to Bayesian model selection, the best account for medication effects in both studies was a specific amplification of reward magnitude in a Q-learning algorithm. These results suggest that learning to avoid physical effort is independent from dopaminergic circuits and strengthen the general idea that dopaminergic signaling amplifies the effects of reward expectation or obtainment on instrumental behavior.SIGNIFICANCE STATEMENT Theoretically, maximizing reward and minimizing effort could involve the same computations and therefore rely on the same brain circuits. Here, we tested whether dopamine, a key component of reward-related circuitry, is also implicated in effort learning. We found that patients suffering from dopamine depletion due to Parkinson's disease were selectively impaired in reward learning, but not effort learning. Moreover, anti-parkinsonian medication restored the ability to maximize reward, but had no effect on effort minimization. This dissociation suggests that the brain has evolved separate, domain-specific systems for instrumental learning. These results help to disambiguate the motivational role of prodopaminergic medications: they amplify the impact of reward without affecting the integration of effort cost.

Learning to minimize efforts versus maximizing rewards: computational principles and neural correlates.

The mechanisms of reward maximization have been extensively studied at both the computational and neural levels. By contrast, little is known about how the brain learns to choose the options that minimize action cost. In principle, the brain could have evolved a general mechanism that applies the same learning rule to the different dimensions of choice options. To test this hypothesis, we scanned healthy human volunteers while they performed a probabilistic instrumental learning task that varied in both the physical effort and the monetary outcome associated with choice options. Behavioral data showed that the same computational rule, using prediction errors to update expectations, could account for both reward maximization and effort minimization. However, these learning-related variables were encoded in partially dissociable brain areas. In line with previous findings, the ventromedial prefrontal cortex was found to positively represent expected and actual rewards, regardless of effort. A separate network, encompassing the anterior insula, the dorsal anterior cingulate, and the posterior parietal cortex, correlated positively with expected and actual efforts. These findings suggest that the same computational rule is applied by distinct brain systems, depending on the choice dimension-cost or benefit-that has to be learned.