Replay and compositional computation.

Replay in the brain has been viewed as rehearsal or, more recently, as sampling from a transition model. Here, we propose a new hypothesis: that replay is able to implement a form of compositional computation where entities are assembled into relationally bound structures to derive qualitatively new knowledge. This idea builds on recent advances in neuroscience, which indicate that the hippocampus flexibly binds objects to generalizable roles and that replay strings these role-bound objects into compound statements. We suggest experiments to test our hypothesis, and we end by noting the implications for AI systems which lack the human ability to radically generalize past experience to solve new problems.

Stratification of adolescents across mental phenomena emphasizes the importance of transdiagnostic distress: a replication in two general population cohorts.

Characterizing patterns of mental phenomena in epidemiological studies of adolescents can provide insight into the latent organization of psychiatric disorders. This avoids the biases of chronicity and selection inherent in clinical samples, guides models of shared aetiology within psychiatric disorders and informs the development and implementation of interventions. We applied Gaussian mixture modelling to measures of mental phenomena from two general population cohorts: the Avon Longitudinal Study of Parents and Children (ALSPAC, n = 3018) and the Neuroscience in Psychiatry Network (NSPN, n = 2023). We defined classes according to their patterns of both positive (e.g. wellbeing and self-esteem) and negative (e.g. depression, anxiety, and psychotic experiences) phenomena. Subsequently, we characterized classes by considering the distribution of diagnoses and sex split across classes. Four well-separated classes were identified within each cohort. Classes primarily differed by overall severity of transdiagnostic distress rather than particular patterns of phenomena akin to diagnoses. Further, as overall severity of distress increased, so did within-class variability, the proportion of individuals with operational psychiatric diagnoses. These results suggest that classes of mental phenomena in the general population of adolescents may not be the same as those found in clinical samples. Classes differentiated only by overall severity support the existence of a general, transdiagnostic mental distress factor and have important implications for intervention.

Measuring self-regulation in everyday life: Reliability and validity of smartphone-based experiments in alcohol use disorder.

Self-regulation, the ability to guide behavior according to one's goals, plays an integral role in understanding loss of control over unwanted behaviors, for example in alcohol use disorder (AUD). Yet, experimental tasks that measure processes underlying self-regulation are not easy to deploy in contexts where such behaviors usually occur, namely outside the laboratory, and in clinical populations such as people with AUD. Moreover, lab-based tasks have been criticized for poor test-retest reliability and lack of construct validity. Smartphones can be used to deploy tasks in the field, but often require shorter versions of tasks, which may further decrease reliability. Here, we show that combining smartphone-based tasks with joint hierarchical modeling of longitudinal data can overcome at least some of these shortcomings. We test four short smartphone-based tasks outside the laboratory in a large sample (N = 488) of participants with AUD. Although task measures indeed have low reliability when data are analyzed traditionally by modeling each session separately, joint modeling of longitudinal data increases reliability to good and oftentimes excellent levels. We next test the measures' construct validity and show that extracted latent factors are indeed in line with theoretical accounts of cognitive control and decision-making. Finally, we demonstrate that a resulting cognitive control factor relates to a real-life measure of drinking behavior and yields stronger correlations than single measures based on traditional analyses. Our findings demonstrate how short, smartphone-based task measures, when analyzed with joint hierarchical modeling and latent factor analysis, can overcome frequently reported shortcomings of experimental tasks.

Shared Responsibility Decreases the Sense of Agency in the Human Brain.

Sharing responsibility in social decision-making helps individuals use the flexibility of the collective context to benefit themselves by claiming credit for good outcomes or avoiding the blame for bad outcomes. Using magnetoencephalography, we examined the neuronal basis of the impact that social context has on this flexible sense of responsibility. Participants performed a gambling task in various social contexts and reported feeling less responsibility when playing as a member of a team. A reduced magnetoencephalography outcome processing effect was observed as a function of decreasing responsibility at 200 msec post outcome onset and was centered over parietal, central, and frontal brain regions. Before outcome revelation in socially made decisions, an attenuated motor preparation signature at 500 msec after stimulus onset was found. A boost in reported responsibility for positive outcomes in social contexts was associated with increased activity in regions related to social and reward processing. Together, these results show that sharing responsibility with others reduces agency, influencing pre-outcome motor preparation and post-outcome processing, and provides opportunities to flexibly claim credit for positive outcomes.

Sexually divergent development of depression-related brain networks during healthy human adolescence.

Sexual differences in human brain development could be relevant to sex differences in the incidence of depression during adolescence. We tested for sex differences in parameters of normative brain network development using fMRI data on N = 298 healthy adolescents, aged 14 to 26 years, each scanned one to three times. Sexually divergent development of functional connectivity was located in the default mode network, limbic cortex, and subcortical nuclei. Females had a more "disruptive" pattern of development, where weak functional connectivity at age 14 became stronger during adolescence. This fMRI-derived map of sexually divergent brain network development was robustly colocated with i prior loci of reward-related brain activation ii a map of functional dysconnectivity in major depressive disorder (MDD), and iii an adult brain gene transcriptional pattern enriched for genes on the X chromosome, neurodevelopmental genes, and risk genes for MDD. We found normative sexual divergence in adolescent development of a cortico-subcortical brain functional network that is relevant to depression.

Ergodicity-breaking reveals time optimal decision making in humans.

Ergodicity describes an equivalence between the expectation value and the time average of observables. Applied to human behaviour, ergodic theories of decision-making reveal how individuals should tolerate risk in different environments. To optimize wealth over time, agents should adapt their utility function according to the dynamical setting they face. Linear utility is optimal for additive dynamics, whereas logarithmic utility is optimal for multiplicative dynamics. Whether humans approximate time optimal behavior across different dynamics is unknown. Here we compare the effects of additive versus multiplicative gamble dynamics on risky choice. We show that utility functions are modulated by gamble dynamics in ways not explained by prevailing decision theories. Instead, as predicted by time optimality, risk aversion increases under multiplicative dynamics, distributing close to the values that maximize the time average growth of in-game wealth. We suggest that our findings motivate a need for explicitly grounding theories of decision-making on ergodic considerations.

A Reduced Self-Positive Belief Underpins Greater Sensitivity to Negative Evaluation in Socially Anxious Individuals.

Positive self-beliefs are important for well-being, and are influenced by how others evaluate us during social interactions. Mechanistic accounts of self-beliefs have mostly relied on associative learning models. These account for choice behaviour but not for the explicit beliefs that trouble socially anxious patients. Neither do they speak to self-schemas, which underpin vulnerability according to psychological research. Here, we compared belief-based and associative computational models of social-evaluation, in individuals that varied in fear of negative evaluation (FNE), a core symptom of social anxiety. We used a novel analytic approach, 'clinically informed model-fitting', to determine the influence of FNE symptom scores on model parameters. We found that high-FNE participants learn more easily from negative feedback about themselves, manifesting in greater self-negative learning rates. Crucially, we provide evidence that this bias is underpinned by an overall reduced belief about self-positive attributes. The study population could be characterized equally well by belief-based or associative models, however large individual differences in model likelihood indicated that some individuals relied more on an associative (model-free), while others more on a belief-guided strategy. Our findings have therapeutic importance, as positive belief activation may be used to specifically modulate learning. AUTHOR SUMMARY: Understanding how we form and maintain positive self-beliefs is crucial to understanding how things go awry in disorders such as social anxiety. The loss of positive self-belief in social anxiety, especially in inter-personal contexts, is thought to be related to how we integrate evaluative information that we receive from others. We frame this social information integration as a learning problem and ask how people learn whether someone approves of them or not. We thus elucidate why the decrease in positive evaluations manifests only for the self, but not for an unknown other, given the same information. We investigated the mechanics of this learning using a novel computational modelling approach, comparing models that treat the learning process as series of stimulusresponse associations with models that treat learning as updating of beliefs about the self (or another). We show that both models characterise the process well and that individuals higher in symptoms of social anxiety learn more from negative information specifically about the self. Crucially, we provide evidence that this originates from a reduction in the amount of positive attributes that are activated when the individual is placed in a social evaluative context.

Human complex exploration strategies are enriched by noradrenaline-modulated heuristics.

An exploration-exploitation trade-off, the arbitration between sampling a lesser-known against a known rich option, is thought to be solved using computationally demanding exploration algorithms. Given known limitations in human cognitive resources, we hypothesised the presence of additional cheaper strategies. We examined for such heuristics in choice behaviour where we show this involves a value-free random exploration, that ignores all prior knowledge, and a novelty exploration that targets novel options alone. In a double-blind, placebo-controlled drug study, assessing contributions of dopamine (400 mg amisulpride) and noradrenaline (40 mg propranolol), we show that value-free random exploration is attenuated under the influence of propranolol, but not under amisulpride. Our findings demonstrate that humans deploy distinct computationally cheap exploration strategies and that value-free random exploration is under noradrenergic control.

Replay bursts in humans coincide with activation of the default mode and parietal alpha networks.

Our brains at rest spontaneously replay recently acquired information, but how this process is orchestrated to avoid interference with ongoing cognition is an open question. Here we investigated whether replay coincided with spontaneous patterns of whole-brain activity. We found, in two separate datasets, that replay sequences were packaged into transient bursts occurring selectively during activation of the default mode network (DMN) and parietal alpha networks. These networks are believed to support inwardly oriented attention and inhibit bottom-up sensory processing and were characterized by widespread synchronized oscillations coupled to increases in high frequency power, mechanisms thought to coordinate information flow between disparate cortical areas. Our data reveal a tight correspondence between two widely studied phenomena in neural physiology and suggest that the DMN may coordinate replay bursts in a manner that minimizes interference with ongoing cognition.

Brain micro-architecture and disinhibition: a latent phenotyping study across 33 impulsive and compulsive behaviours.

Impulsive and compulsive symptoms are common, tend to co-occur, and collectively account for a substantive global disease burden. Latent phenotyping offers a promising approach to elucidate common neural mechanisms conferring vulnerability to such symptoms in the general population. We utilised the Neuroscience in Psychiatry Network (NSPN), a cohort of young people (aged 18-29 years) in the United Kingdom, who provided questionnaire data and Magnetic Resonance Imaging scans. Partial Least Squares was used to identify brain regions in which intra-cortical myelination (measured using Magnetisation Transfer, MT) was significantly associated with a disinhibition phenotype, derived from bi-factor modelling of 33 impulsive and compulsive problem behaviours. The neuroimaging sample comprised 126 participants, mean 22.8 (2.7 SD) years old, being 61.1% female. Disinhibition scores were significantly and positively associated with higher MT in the bilateral frontal and parietal lobes. 1279 genes associated with disinhibition-related brain regions were identified, which were significantly enriched for functional biological interactions reflecting receptor signalling pathways. This study indicates common microstructural brain abnormalities contributing to a multitude of related, prevalent, problem behaviours characterised by disinhibition. Such a latent phenotyping approach provides insights into common neurobiological pathways, which may help to improve disease models and treatment approaches. Now that this latent phenotyping model has been validated in a general population sample, it can be extended into patient settings.

Examining the relationship between altered brain functional connectome and disinhibition across 33 impulsive and compulsive behaviours.

Impulsive and compulsive problem behaviours are associated with a variety of mental disorders. Latent phenotyping indicates the expression of impulsive and compulsive problem behaviours is predominantly governed by a transdiagnostic 'disinhibition' phenotype. In a cohort of 117 individuals, recruited as part of the Neuroscience in Psychiatry Network (NSPN), we examined how brain functional connectome and network properties relate to disinhibition. Reduced functional connectivity within a subnetwork of frontal (especially right inferior frontal gyrus), occipital and parietal regions was linked to disinhibition. Findings provide insights into neurobiological pathways underlying the emergence of impulsive and compulsive disorders.

Learning in anticipation of reward and punishment: perspectives across the human lifespan.

Learning to act to receive reward and to withhold to avoid punishment has been found to be easier than learning the opposite contingencies in young adults. To what extent this type of behavioral adaptation might develop during childhood and adolescence and differ during aging remains unclear. We therefore tested 247 healthy individuals across the human life span (7-80 years) with an orthogonalized valenced go/no-go learning task. Computational modeling revealed that peak performance in young adults was attributable to greater sensitivity to both reward and punishment. However, in children and adolescents, we observed an increased bias toward action but not reward sensitivity. By contrast, reduced learning in midlife and older adults was accompanied by decreased reward sensitivity and especially punishment sensitivity along with an age-related increase in the Pavlovian bias. These findings reveal distinct motivation-dependent learning capabilities across the human life span, which cannot be probed using conventional go/reward no-go/punishment style paradigms that have important implications in lifelong education.

How do the prevalence and relative risk of non-suicidal self-injury and suicidal thoughts vary across the population distribution of common mental distress (the p factor)? Observational analyses replicated in two independent UK cohorts of young people.

OBJECTIVES: To inform suicide prevention policies and responses to youths at risk by investigating whether suicide risk is predicted by a summary measure of common mental distress (CMD (the p factor)) as well as by conventional psychopathological domains; to define the distribution of suicide risks over the population range of CMD; to test whether such distress mediates the medium-term persistence of suicide risks. DESIGN: Two independent population-based cohorts. SETTING: Population based in two UK centres. PARTICIPANTS: Volunteers aged 14-24 years recruited from primary healthcare registers, schools and colleges, with advertisements to complete quotas in age-sex-strata. Cohort 1 is the Neuroscience in Psychiatry Network (n=2403); cohort 2 is the ROOTS sample (n=1074). PRIMARY OUTCOME MEASURES: Suicidal thoughts (ST) and non-suicidal self-injury (NSSI). RESULTS: We calculated a CMD score using confirmatory bifactor analysis and then used logistic regressions to determine adjusted associations between risks and CMD; curve fitting was used to examine the relative prevalence of STs and NSSI over the population distribution of CMD. We found a dose-response relationship between levels of CMD and risk of suicide. The majority of all subjects experiencing ST and NSSI (78% and 76% in cohort 1, and 66% and 71% in cohort 2) had CMD scores no more than 2 SDs above the population mean; higher scores indicated the highest risk but were, by definition, infrequent. Pathway mediation models showed that CMD mediated the longitudinal course of both ST and NSSI. CONCLUSIONS: NSSI and ST in youths reflect CMD that also mediates their persistence. Universal prevention strategies reducing levels of CMD in the whole population without recourse to screening or measurement may prevent more suicides than approaches targeting youths with the most severe distress or with psychiatric disorders.

Fractionation of impulsive and compulsive trans-diagnostic phenotypes and their longitudinal associations.

OBJECTIVE: Young adulthood is a crucial neurodevelopmental period during which impulsive and compulsive problem behaviours commonly emerge. While traditionally considered diametrically opposed, impulsive and compulsive symptoms tend to co-occur. The objectives of this study were as follows: (a) to identify the optimal trans-diagnostic structural framework for measuring impulsive and compulsive problem behaviours, and (b) to use this optimal framework to identify common/distinct antecedents of these latent phenotypes. METHOD: In total, 654 young adults were recruited as part of the Neuroscience in Psychiatry Network, a population-based cohort in the United Kingdom. The optimal trans-diagnostic structural model capturing 33 types of impulsive and compulsive problem behaviours was identified. Baseline predictors of subsequent impulsive and compulsive trans-diagnostic phenotypes were characterised, along with cross-sectional associations, using partial least squares. RESULTS: Current problem behaviours were optimally explained by a bi-factor model, which yielded dissociable measures of impulsivity and compulsivity, as well as a general disinhibition factor. Impulsive problem behaviours were significantly explained by prior antisocial and impulsive personality traits, male gender, general distress, perceived dysfunctional parenting and teasing/arguments within friendships. Compulsive problem behaviours were significantly explained by prior compulsive traits and female gender. CONCLUSION: This study demonstrates that trans-diagnostic phenotypes of 33 impulsive and compulsive problem behaviours are identifiable in young adults, utilising a bi-factor model based on responses to a single questionnaire. Furthermore, these phenotypes have different antecedents. The findings yield a new framework for fractionating impulsivity and compulsivity, and suggest different early intervention targets to avert emergence of problem behaviours. This framework may be useful for future biological and clinical dissection of impulsivity and compulsivity.

Personality dimensions emerging during adolescence and young adulthood are underpinned by a single latent trait indexing impairment in social functioning.

BACKGROUND: Personality with stable behavioural traits emerges in the adolescent and young adult years. Models of putatively distinct, but correlated, personality traits have been developed to describe behavioural styles including schizotypal, narcissistic, callous-unemotional, negative emotionality, antisocial and impulsivity traits. These traits have influenced the classification of their related personality disorders. We tested if a bifactor model fits the data better than correlated-factor and orthogonal-factor models and subsequently validated the obtained factors with mental health measures and treatment history. METHOD: A set of self-report questionnaires measuring the above traits together with measures of mental health and service use were collected from a volunteer community sample of adolescents and young adults aged 14 to 25 years (N = 2443). RESULTS: The bifactor model with one general and four specific factors emerged in exploratory analysis, which fit data better than models with correlated or orthogonal factors. The general factor showed high reliability and validity. CONCLUSIONS: The findings suggest that a selected range of putatively distinct personality traits is underpinned by a general latent personality trait that may be interpreted as a severity factor, with higher scores indexing more impairment in social functioning. The results are in line with ICD-11, which suggest an explicit link between personality disorders and compromised interpersonal or social function. The obtained general factor was akin to the overarching dimension of personality functioning (describing one's relation to the self and others) proposed by DSM-5 Section III.

Decreased value-sensitivity in schizophrenia.

Pathophysiology in schizophrenia has been linked to aberrant incentive salience, namely the dysfunctional processing of value linked to abnormal dopaminergic activity. In line with this, recent studies showed impaired learning of value in schizophrenia. However, how value is used to guide behaviour independently from learning, as in risky choice, has rarely been examined in this disorder. We studied value-guided choice under risk in patients with schizophrenia and in controls using a task requiring a choice between a certain monetary reward, varying trial-by-trial, and a gamble offering an equal probability of getting double this certain amount or nothing. We observed that patients compared to controls exhibited reduced sensitivity to values, implying that their choices failed to flexibly adapt to the specific values on offer. Moreover, the degree of this value sensitivity inversely correlated with aberrant salience experience, suggesting that the inability to tune choice to value may be a key element of aberrant salience in the illness. Our results help clarify the cognitive mechanisms underlying improper attribution of value in schizophrenia and may thus inform cognitive interventions aimed at reinstating value sensitivity in patients.

Emotional arousal and recognition memory are differentially reflected in pupil diameter responses during emotional memory for negative events in younger and older adults.

A better memory for negative emotional events is often attributed to a conjoint impact of increased arousal and noradrenergic modulation (NA). A decline in NA during aging is well documented but its impact on memory function during aging is unclear. Using pupil diameter (PD) as a proxy for NA, we examined age differences in memory for negative events in younger (18-30 years) and older (62-83 years) adults based on a segregation of early arousal to negative events, and later retrieval-related PD responses. In keeping with the hypothesis of reduced age-related NA influences, older adults showed attenuated induced PD responses to negative emotional events. The findings highlight a likely contribution of NA to negative emotional memory, mediated via arousal that may be compromised with aging.

Vigour in active avoidance.

It would be maladaptive to learn about catastrophes by trial and error alone. Investment in planning and effort are necessary. Devoting too many resources to averting disaster, however, can impair quality of life, as in anxiety and paranoia. Here, we developed a novel task to explore how people adjust effort expenditure (vigor) so as to avoid negative consequences. Our novel paradigm is immersive, enabling us to measure vigor in the context of (simulated) disaster. We found that participants (N = 118) exerted effort to avoid disaster-associated states, adjusting their effort expenditure according to the baseline probability of catastrophe, in agreement with theoretical predictions. Furthermore, negative subjective emotional states were associated both with threat level and with increasing vigor in the face of disaster. We describe for the first time effort expenditure in the context of irreversible losses, with important implications for disorders marked by excessive avoidance.

"Contemplating the Next Maneuver": Functional Neuroimaging Reveals Intraoperative Decision-making Strategies.

OBJECTIVE: To investigate differences in the quality, confidence, and consistency of intraoperative surgical decision making (DM) and using functional neuroimaging expose decision systems that operators use. SUMMARY BACKGROUND DATA: Novices are hypothesized to use conscious analysis (effortful DM) leading to activation across the dorsolateral prefrontal cortex, whereas experts are expected to use unconscious automation (habitual DM) in which decisions are recognition-primed and prefrontal cortex independent. METHODS: A total of 22 subjects (10 medical student novices, 7 residents, and 5 attendings) reviewed simulated laparoscopic cholecystectomy videos, determined the next safest operative maneuver upon video termination (10 s), and reported decision confidence. Video paradigms either declared ("primed") or withheld ("unprimed") the next operative maneuver. Simultaneously, changes in cortical oxygenated hemoglobin and deoxygenated hemoglobin inferring prefrontal activation were recorded using Optical Topography. Decision confidence, consistency (primed vs unprimed), and quality (script concordance) were assessed. RESULTS: Attendings and residents were significantly more certain (P < 0.001), and decision quality was superior (script concordance: attendings = 90%, residents = 78.3%, and novices = 53.3%). Decision consistency was significantly superior in experts (P < 0.001) and residents (P < 0.05) than novices (P = 0.183). During unprimed DM, novices showed significant activation of the dorsolateral prefrontal cortex, whereas this activation pattern was not observed among residents and attendings. During primed DM, significant activation was not observed in any group. CONCLUSIONS: Expert DM is characterized by improved quality, consistency, and confidence. The findings imply attendings use a habitual decision system, whereas novices use an effortful approach under uncertainty. In the presence of operative cues (primes), novices disengage the prefrontal cortex and seem to accept the observed operative decision as correct.

Parsing the Role of the Hippocampus in Approach-Avoidance Conflict.

The hippocampus plays a central role in the approach-avoidance conflict that is central to the genesis of anxiety. However, its exact functional contribution has yet to be identified. We designed a novel gambling task that generated approach-avoidance conflict while controlling for spatial processing. We fit subjects' behavior using a model that quantified the subjective values of choice options, and recorded neural signals using functional magnetic resonance imaging (fMRI). Distinct functional signals were observed in anterior hippocampus, with inferior hippocampus selectively recruited when subjects rejected a gamble, to a degree that covaried with individual differences in anxiety. The superior anterior hippocampus, in contrast, uniquely demonstrated value signals that were potentiated in the context of approach-avoidance conflict. These results implicate the anterior hippocampus in behavioral avoidance and choice monitoring, in a manner relevant to understanding its role in anxiety. Our findings highlight interactions between subregions of the hippocampus as an important focus for future study.