An Internet-Based Intervention to Increase the Ability of Lesbian, Gay, and Bisexual People to Cope With Adverse Events: Single-Group Feasibility Study.

BACKGROUND: Lesbian, gay, bisexual, transgender, and queer (LGBTQ+) people are at higher risk of mental health problems due to widespread hetero- and cisnormativity, including negative public attitudes toward the LGBTQ+ community. In addition to combating social exclusion at the societal level, strengthening the coping abilities of young LGBTQ+ people is an important goal. OBJECTIVE: In this transdiagnostic feasibility study, we tested a 6-week internet intervention program designed to increase the ability of nonclinical LGBTQ+ participants to cope with adverse events in their daily lives. The program was based on acceptance and commitment therapy principles. METHODS: The program consists of 6 web-based modules and low-intensity assistance for homework provided by a single care provider asynchronously. The design was a single-group assignment of 15 self-identified LGB community members who agreed to participate in an open trial with a single group (pre- and postintervention design). RESULTS: Before starting the program, participants found the intervention credible and expressed high satisfaction at the end of the intervention. Treatment adherence, operationalized by the percentage of completed homework assignments (32/36, 88%) was also high. When we compared participants' pre- and postintervention scores, we found a significant decrease in clinical symptoms of depression (Cohen d=0.44, 90% CI 0.09-0.80), social phobia (d=0.39, 90% CI 0.07-0.72), and posttraumatic stress disorder (d=0.30, 90% CI 0.04-0.55). There was also a significant improvement in the level of self-acceptance and behavioral effectiveness (d=0.64, 90% CI 0.28-0.99) and a significant decrease in the tendency to avoid negative internal experiences (d=0.38, 90% CI 0.09-0.66). The level of general anxiety disorder (P=.11; d=0.29, 90% CI -0.10 to 0.68) and alcohol consumption (P=.35; d=-0.06, 90% CI -0.31 to 0.19) were the only 2 outcomes for which the results were not statistically significant. CONCLUSIONS: The proposed web-based acceptance and commitment therapy program, designed to help LGBTQ+ participants better manage emotional difficulties and become more resilient, represents a promising therapeutic tool. The program could be further tested with more participants to ensure its efficacy and effectiveness. TRIAL REGISTRATION: ClinicalTrials.gov NCT05514964; https://clinicaltrials.gov/study/NCT05514964.

Efficacy of the iJobs Web-Based Psychoeducational Intervention to Improve Job Search Behavior and Promote Mental Health Among Unemployed People: Protocol for a Waitlist Randomized Controlled Trial.

BACKGROUND: Unemployment affects millions of people worldwide and, beyond its economic impact, has severe implications for people's well-being and mental health. Different programs have been developed in response to this phenomenon, but to date, job-search interventions have proved to be most effective, especially the JOBS II program. The JOBS II program proved not only to be effective for re-employment but also has a positive impact on beneficiaries' mental health (ie, reduces anxiety or depression). However, by now, this evidence-based program has been delivered only on site in the various countries where it was implemented. In the digital era, web-based alternatives to such programs are highly needed because they have the advantages of scalability and cost-effectiveness. OBJECTIVE: In this context, we aim to investigate the efficacy of iJobs, the web-based adaptation of the JOBS II program, on job-search intensity and effort, the quality of job-search behaviors, and job-search self-efficacy. Further, 1 month after the intervention, we will also assess the employment status and the satisfaction with the job (if applicable). This study will also investigate the effect of iJobs on well-being and mental health (ie, anxiety and depression). METHODS: This study is a 2-arm randomized controlled trial. The 2 independent groups (intervention vs waiting list control group) will be crossed with 3 measurement times (ie, baseline, the postintervention time point, and 1-month follow-up). The design will be a 2 (intervention vs control) × 3 (baseline, the postintervention time point, and 1-month follow-up) factorial design. iJobs is a 2-week intervention consisting of 6 modules: an introductive module and 5 modules adapted from the original JOBS II program to the web-based setting and Romanian population. The web-based intervention also has a human component, as beneficiaries receive personalized written feedback after each module on the platform from a team of psychologists involved in the project. RESULTS: The enrollment of study participants started in June 2023 and is expected to end in May 2024. The data collection is expected to be completed by July 2024. The results are expected to be submitted for publication in the summer of 2024. CONCLUSIONS: This study is the first large-scale randomized controlled trial aiming to test the efficacy of a web-based adaptation of the JOBS II program. If our results support the efficacy of iJobs, they will offer the premise for it to become an evidence-based, accessible alternative for unemployed people in Romania and might be implemented in other countries. TRIAL REGISTRATION: ClinicalTrials.gov NCT05962554; https://clinicaltrials.gov/study/NCT05962554. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/55374.

Assessing the effectiveness of psychoeducational interventions on driving behavior: A systematic review and meta-analysis.

This review aimed to quantitatively summarize the evidence concerning the effectiveness of psychoeducational interventions on driving behavior. A final pool of 138 studies, totaling approximately 97,000 participants, was included in the analyses and covered all types of driving behavior targeted by the interventions. Using a random effects model, significant results were found for almost all driving outcomes, both post-intervention and long-term. The strongest effect was for reducing distracted driving at post-intervention (d = 1.87 [1.12, 2.60], Z = 4.94, p < 0.001). The only non-significant effects were for reducing errors in the long term (d = 0.50 [-0.87, 1.86], Z = 0.71, p = 0.48) and driving under the influence at post-intervention (d = 0.35 [0.00, 0.71], Z = 1.96, p = 0.05). Concerning which type of intervention was more effective, feedback, training and motivational ones appear to work best. Educational interventions show only weak effects, while awareness interventions seem mostly ineffective. Overall, our results show that most interventions can reduce different types of driving behaviors, but there are specific aspects to be considered based on the targeted behavior.

iJobs - An online implementation of the JOBS II program for fostering reemployment: A feasibility and acceptability study.

The current study aimed to test the feasibility and acceptability of iJobs, an online adaptation of the JOBS II program (Curran et al., 1999). iJobs is a two-week internet intervention for the unemployed, consisting of five modules. This study is an open-label trial with an uncontrolled, within-group, pre-posttest, and follow-up design. Out of the 56 participants allocated to the intervention, 36 completed (Mage = 25 years; 57.1 % females) the post-test (36 % dropout), and 34 the three months follow-up. The protocol-compliant participants followed the modules with great engagement (mean quality of assignments completion above 4 points out of 5 for each module). The online platform's usability was high (84.86 points out of 100). Participants reported high overall satisfaction with the program. Our results suggest that iJobs is a feasible intervention and was accepted by its beneficiaries. Relative to baseline, inoculation against setbacks (d = 0.64), job search self-efficacy (d = 0.50), and self-esteem (d = 0.28) increased significantly, while future career anxiety in the COVID-19 context decreased significantly (d = 0.34). No significant differences were found for depression, anxiety, and job-search behaviors. At three months follow-up, 55.9 % of the participants found employment, 5.9 % were in a job selection process, and 38.2 % were still unemployed. Job satisfaction was high among the employed.

"Give me a break!" A systematic review and meta-analysis on the efficacy of micro-breaks for increasing well-being and performance.

Recovery activities during short breaks taken between work tasks are solutions for preventing the impairing effects of accumulated strain. No wonder then that a growing body of scientific literature from various perspectives emerged on this topic. The present meta-analysis is aimed at estimating the efficacy of micro-breaks in enhancing well-being (vigor and fatigue) and performance, as well as in which conditions and for whom are the micro-breaks most effective. We searched the existent literature on this topic and aggregated the existing data from experimental and quasi-experimental studies. The systematic search revealed 19 records, which resulted in 22 independent study samples (N = 2335). Random-effects meta-analyses shown statistically significant but small effects of micro-breaks in boosting vigor (d = .36, p < .001; k = 9, n = 913), reducing fatigue (d = .35, p < .001; k = 9, n = 803), and a non-significant effect on increasing overall performance (d = .16, p = .116; k = 15, n = 1132). Sub-groups analyses on performance types revealed significant effects only for tasks with less cognitive demands. A meta-regression showed that the longer the break, the greater the boost was on performance. Overall, the data support the role of micro-breaks for well-being, while for performance, recovering from highly depleting tasks may need more than 10-minute breaks. Therefore, future studies should focus on this issue.

I Think I Should Get Vaccinated, I Feel I Should Not. Individual Differences in Information Processing and Vaccination Behavior (COVID-19).

Following the outbreak of the COVID-19 pandemic, the scientific community responded promptly by developing effective vaccines. Still, even though effective vaccines against COVID-19 became available, many people did not seem to be in a rush to become immunized. Community protection can be enhanced if more people decide to vaccinate, and thus it is necessary to identify relevant factors involved in vaccination behavior to find better ways of encouraging it. Vaccination behavior is the result of a decision process that might vary according to individual differences in information processing. We investigated the role of cognitive reflection ability and thinking styles in predicting self-reported vaccination behavior against COVID-19. A sample of 274 Romanian participants was surveyed for the present study, out of which 217 (Mage = 24.58, SD = 8.31; 53% female) declared they had the possibility to become vaccinated. Results showed that a higher level of cognitive reflection ability significantly increased the odds of becoming vaccinated. A rational thinking style was not linked to vaccination behavior. However, an experiential thinking style indirectly predicted vaccination behavior by means of attitudes towards vaccination. Since individual differences in information processing are, to a certain extent, linked to vaccination behavior, the design of vaccination campaigns could consider that people have specific information needs and address them as such.

Job Resources and Strengths Use in Relation to Employee Performance: A Contextualized View.

Job resources play a prominent role in employee performance literature, yet a fine-grained understanding of how resources are relevant for several performance types is still needed. Relying on the Job Demands-Resources and Conservation of Resources theories, the present study addresses this call in two ways. First, it examines the predictive effect of four job resources (i.e., role clarity, feedback, autonomy, and opportunities for development) on nine types of performance (i.e., proficiency, adaptivity, and proactivity as an individual, team, and organization member). Second, it tests the moderator role of strengths use in these relationships. Data was gathered from a sample of Romanian employees (N = 332) and analyzed via hierarchical multiple linear regression. The results indicate that the selected job resources are, indeed, predictors of different types of employee performance and not in a unitary manner. Role clarity and feedback appear to be the most relevant predictors for various performance types, while autonomy seems to be the least important. Also, strengths use moderates these relationships, but in a reinforcing manner only regarding opportunities for development. The interaction of strengths use with role clarity and feedback renders the latter two obsolete, indicating that individual strategies may act as substitutes for job resources. These findings add to the Job Demands-Resources theory's versatile nature and provide more clarity to practitioners who plan interventions to enhance specific performance types, taking individual strategies into account.

Can We Boost Treatment Adherence to an Online Transdiagnostic Intervention by Adding Self-Enhancement Strategies? Results From a Randomized Controlled Non-inferiority Trial.

Background: Internet-delivered psychotherapy represents an impactful large-scale solution for addressing psychological disorders. In spite of its flexibility and scalability, the fact that the ones in need have to initiate and sustain the curse of the treatment by themselves comes with considerable downsides in terms of treatment adherence. One solution could be to increase the ease of use and attractivity of the strategies and assignments from such programs. The present study aims to address this issue by incorporating a series of self-oriented strategies to the validated internet-delivered short version of the Unified Protocol (UP). By this mean we intend to complement the symptom-focused assignments, which may be more suitable in a therapist assisted context, with ones designed for self-enhancement, which may be easier approached as self-initiated. Based on a randomized controlled non-inferiority trial we compared the modified version of the UP with the standard short version. Method: The trial design was factorial, with two parallel arms and three measurement moments (baseline, post-intervention and 6-months follow-up). A total of 284 participants were randomly assigned to the intervention or the active control groups. The intervention group (baseline n = 142) received the self-enhanced nine modules of the UP (Self-enhanced 9UP) while the active control (baseline n = 142) received the standard nine modules (9UP). The newly added techniques were inspired by the acceptance and commitment therapy and were specific for self-concepts such as self-compassion or unconditional self-acceptance. Both programs lasted for 9 weeks. The non-inferiority of the Self-enhanced 9UP was tested against a margin of d = -0.35, on the following primary outcome measures: Patient Health Questionnaire 9 (PHQ9) - operationalization for depression; Generalized Anxiety Disorder 7 (GAD7) - operationalization for generalized anxiety or worry; Social Phobia Inventory (SPIN) - operationalization for social phobia; and Panic Disorder Severity Scale-Self Report (PDSS-SR) - that showed participants' level of panic. Treatment adherence was assessed through the drop-out analyses and the engagement in completing the homework assignments. Secondary outcome measures included several self-concept measures: Self-Compassion Scale (SCS); Rosenberg Self-Esteem Scale (RSES); Unconditional Self-Acceptance Questionnaire (USAQ); New General Self-Efficacy Scale (NGSE); and Self-Concept Clarity Scale (SCCS). On the secondary outcomes we explored the potential boost of effectiveness produced by the newly added self-enhancement components. Results: The dropout rates were similar in both groups (approximately 45%) and high overall. Adherence to treatment assignments was also modest and similar between groups (on average participants completed approximately half of the tasks), without a statistically significant bias toward the self-enhancement ones. Overall, both the intention-to-treat and completers analyses yielded no significant group by time interactions for any of the post-intervention and follow-up measurements, but a few non-inferiority analyses suggested that the Self-enhanced 9UP had a significantly weaker effectiveness than the standard 9UP. Within-group analyses showed significant alleviations on all the primary and secondary outcomes for both groups. The effect size estimates were mainly medium and high, and their magnitude tended to be kept also at 6-months follow-up. Discussion: We failed to increase treatment adherence, but we found support with some exceptions, for the non-inferiority hypothesis. Hence, the alterations performed to the 9UP protocol, although they did not boost the treatment attractiveness, they also did not decrease the treatment effectiveness as suggested by most non-inferiority analyses. Likewise, the gain on self-concepts was produced by both groups. Hence, the short version of the UP seems to have the potential of effectively alleviating a larger palette of psychological variables associated with mental health symptoms than previously known. Even though our main objective was only partially achieved, these secondary results are insightful and could open new avenues of research. Clinical Trial Registration: This trial has been registered at ClinicalTrials.Gov (NCT03917550; 17 April 2019; https://clinicaltrials.gov/ct2/show/NCT03917550).

Modelling digital and manual contact tracing for COVID-19. Are low uptakes and missed contacts deal-breakers?

Comprehensive testing schemes, followed by adequate contact tracing and isolation, represent the best public health interventions we can employ to reduce the impact of an ongoing epidemic when no or limited vaccine supplies are available and the implications of a full lockdown are to be avoided. However, the process of tracing can prove feckless for highly-contagious viruses such as SARS-CoV-2. The interview-based approaches often miss contacts and involve significant delays, while digital solutions can suffer from insufficient adoption rates or inadequate usage patterns. Here we present a novel way of modelling different contact tracing strategies, using a generalized multi-site mean-field model, which can naturally assess the impact of manual and digital approaches alike. Our methodology can readily be applied to any compartmental formulation, thus enabling the study of more complex pathogen dynamics. We use this technique to simulate a newly-defined epidemiological model, SEIR-T, and show that, given the right conditions, tracing in a COVID-19 epidemic can be effective even when digital uptakes are sub-optimal or interviewers miss a fair proportion of the contacts.

Personality predictors of speeding: Anger-Aggression and Impulsive-Sensation Seeking. A systematic review and meta-analysis.

INTRODUCTION: The purpose of this review was to identify the most influential personality predictors of speeding. METHOD: Thus, this study analyzed the associations of Anger-Aggression and Impulsive-Sensation Seeking with Speeding. Research studies included in online databases and papers identified in previous reviews were considered for inclusion. RESULTS: Using a random effects model, we found a small but significant effect for the relation between Anger-Aggression and Speeding (r = 0.12 [0.06-0.18],Z = 3.85,p < 0.001). We identified a significantly stronger effect for the relation between Impulsive-Sensation Seeking and Speeding, but still of low magnitude (r = 0.23 [0.16-0.29],Z = 6.54,p < 0.001). Moderator analysis revealed only one significant moderator: driver type. Namely, for professional drivers, the relation with Speeding was non-significant for both predictors (r = -0.004, p = 0.958 for Impulsive-Sensation Seeking and r = 0.02, p = 0.720 for Anger-Aggression, respectively) and significantly smaller than the associations for general population and young drivers. CONCLUSIONS: Overall, our results confirm Speeding's associations with both hypothesized most important predictors, but at a low magnitude.

Embracing Change: Continual Learning in Deep Neural Networks.

Artificial intelligence research has seen enormous progress over the past few decades, but it predominantly relies on fixed datasets and stationary environments. Continual learning is an increasingly relevant area of study that asks how artificial systems might learn sequentially, as biological systems do, from a continuous stream of correlated data. In the present review, we relate continual learning to the learning dynamics of neural networks, highlighting the potential it has to considerably improve data efficiency. We further consider the many new biologically inspired approaches that have emerged in recent years, focusing on those that utilize regularization, modularity, memory, and meta-learning, and highlight some of the most promising and impactful directions.

Beyond contingency awareness: the role of influence awareness in resisting conditioned attitudes.

Evaluative conditioning procedures change people's evaluations of stimuli that are paired with pleasant or unpleasant items. To test whether influence awareness allows people to resist such persuasive attempts, we conducted three experiments. In the first two experiments featuring low levels of influence awareness (N1 = 96, N2 = 93) we manipulated the degree of control people have in expressing their attitudes, by providing participants in one condition with the option to "pass" rather than respond, when they felt influenced in their evaluations of conditioned stimuli. In the third experiment (N3 = 240) we manipulated the level of influence awareness by using a warning instruction similar to the one found in prior controllability studies, while giving everyone the option to pass the evaluation when they felt influenced. All studies found that participants often failed to use the skip option to exert control over conditioned preferences. In some cases, this may be because participants failed to notice the pairings, but in most cases because participants lacked awareness that the pairings could influence them. Even when explicitly warned that the pairings could influence them, participants seemed to believe that they were not vulnerable to such effects.

Neural scene representation and rendering.

Scene representation-the process of converting visual sensory data into concise descriptions-is a requirement for intelligent behavior. Recent work has shown that neural networks excel at this task when provided with large, labeled datasets. However, removing the reliance on human labeling remains an important open problem. To this end, we introduce the Generative Query Network (GQN), a framework within which machines learn to represent scenes using only their own sensors. The GQN takes as input images of a scene taken from different viewpoints, constructs an internal representation, and uses this representation to predict the appearance of that scene from previously unobserved viewpoints. The GQN demonstrates representation learning without human labels or domain knowledge, paving the way toward machines that autonomously learn to understand the world around them.

A Web-Based Transdiagnostic Intervention for Affective and Mood Disorders: Randomized Controlled Trial.

BACKGROUND: Research increasingly supports a transdiagnostic conceptualization of emotional disorders (ie applying the same underlying treatment principles across mental disorders, without tailoring the protocol to specific diagnoses), and many international researchers are currently investigating this issue. OBJECTIVE: The aim of this study was to evaluate the efficacy and acceptability of a Web-based transdiagnostic program using a sample of Romanian adults diagnosed with anxiety and/or depression. METHODS: Volunteer participants registered for the study and completed a series of online self-report measures. Participants who fulfilled basic inclusion criteria on these measures were contacted for a telephone diagnostic interview using the Structural Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, 4th Edition Axis I Disorders (SCID-I). Enrolled participants were randomized to either the active treatment group (N=69) or the wait-list control group (N=36) using a 2:1 ratio. The transdiagnostic treatment was based on the Unified Protocol for Transdiagnostic Treatment of Emotional Disorders (UP; Barlow et al, 2011) that addresses common underlying mechanisms of anxiety and depression. Participants randomized to the active treatment condition received 10 weeks of Web-based treatment based on the UP. Throughout treatment, graduate students in clinical psychology provided guidance that consisted of asynchronous written communication on a secure Web platform. After the intervention, participants in both study conditions were invited to complete a set of self-report measures and a postintervention SCID-I interview conducted by a different team of graduate students blinded to participants' group and diagnostic status. Six months later, participants in the active treatment group were invited to complete an online follow-up assessment. RESULTS: During the intervention, active treatment participants completed on average 19 homework assignments (SD 12.10), and we collected data from 79.0% (83/105) at postintervention and 51% (35/69) at follow-up for self-report measures. Postintervention SCID-I interviews were collected from 77.1% (81/105) participants. Relative to the wait-list control group, the transdiagnostic intervention yielded overall medium to large effect sizes for the primary outcome measures (within-group Hedges g=0.52-1.34 and between-group g=0.39-0.86), and also for anxiety sensitivity (g=0.80), symptom interference (g=0.48), and quality of life (g=0.38). Significant within-groups effects only were reported for the active treatment group on Panic Disorder Severity Scale-Self Report (PDSS-SR, g=0.58-0.65) and Yale-Brown Obsessive Compulsive Scale (Y-BOCS, g=0.52-0.58). CONCLUSIONS: Insignificant between-group differences for the Y-BOCS and PDSS-SR could be explained by the small number of participants with the associated primary diagnostic (eg, only 3 participants with obsessive compulsive disorder) by the choice of outcome measure (PDSS-SR was not rated among the evidence-based measures) and by the fact that these disorders may be more difficult to treat. However, the overall results suggest that the transdiagnostic intervention tested in this study represents an effective treatment option that may prove easier to disseminate through the use of Web-based delivery systems. TRIAL REGISTRATION: ClinicalTrials.gov CT02739607; https://clinicaltrials.gov/ct2/show/study/NCT02739607 (Archived by WebCite at http://www.webcitation.org/6yY1VeYIZ).

Overcoming catastrophic forgetting in neural networks.

The ability to learn tasks in a sequential fashion is crucial to the development of artificial intelligence. Until now neural networks have not been capable of this and it has been widely thought that catastrophic forgetting is an inevitable feature of connectionist models. We show that it is possible to overcome this limitation and train networks that can maintain expertise on tasks that they have not experienced for a long time. Our approach remembers old tasks by selectively slowing down learning on the weights important for those tasks. We demonstrate our approach is scalable and effective by solving a set of classification tasks based on a hand-written digit dataset and by learning several Atari 2600 games sequentially.

Implicit attitudes towards risky driving: Development and validation of an affect misattribution procedure for speeding.

The present study aimed to develop and validate an instrument for measuring implicit attitudes towards speeding, based on the Affect Misattribution Procedure (AMP). The data collected from 140 drivers supported the reliability of the AMP (α=0.80). Results provided some support for the construct validity of the AMP, through its convergence with explicit attitudes towards speeding (r=0.22). Also, the AMP incrementally predicted driving violations (ΔR2=0.034) and traffic accidents (ΔR2=0.022), above and beyond demographic variables and explicit attitudes towards speeding, thus providing initial evidence for its criterion validity. The multiplicative interaction effect between implicit and explicit attitudes (in predicting the same criterion variables) was not supported. Overall, our results provide support for Speeding-AMP's validity and reliability, which can be successfully used in research concerning risky driving behavior.

A Meta-Analysis for the Efficacy of Hypnotherapy in Alleviating PTSD Symptoms.

A systematic review and meta-analysis of the efficacy of hypnotherapy in the treatment of PTSD used literature searches to obtain 47 articles. However, only 6 were experiments testing the efficacy of hypnosis-based treatments. A fixed-effects meta-analysis was applied to postintervention assessment results and 4-week follow-ups. A large effect in favor of hypnosis-based (especially manualized abreactive hypnosis) treatment was found for the studies that reported the posttest results (d = 1.17). The temporal stability of the effect remains strong, as reflected by the 4-week follow-up assessments (d = 1.58) and also by long-term evaluations (e.g., 12 months). Hypnosis appears to be effective in alleviating PTSD symptoms.

Human-level control through deep reinforcement learning.

The theory of reinforcement learning provides a normative account, deeply rooted in psychological and neuroscientific perspectives on animal behaviour, of how agents may optimize their control of an environment. To use reinforcement learning successfully in situations approaching real-world complexity, however, agents are confronted with a difficult task: they must derive efficient representations of the environment from high-dimensional sensory inputs, and use these to generalize past experience to new situations. Remarkably, humans and other animals seem to solve this problem through a harmonious combination of reinforcement learning and hierarchical sensory processing systems, the former evidenced by a wealth of neural data revealing notable parallels between the phasic signals emitted by dopaminergic neurons and temporal difference reinforcement learning algorithms. While reinforcement learning agents have achieved some successes in a variety of domains, their applicability has previously been limited to domains in which useful features can be handcrafted, or to domains with fully observed, low-dimensional state spaces. Here we use recent advances in training deep neural networks to develop a novel artificial agent, termed a deep Q-network, that can learn successful policies directly from high-dimensional sensory inputs using end-to-end reinforcement learning. We tested this agent on the challenging domain of classic Atari 2600 games. We demonstrate that the deep Q-network agent, receiving only the pixels and the game score as inputs, was able to surpass the performance of all previous algorithms and achieve a level comparable to that of a professional human games tester across a set of 49 games, using the same algorithm, network architecture and hyperparameters. This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.

Imagine all the people: how the brain creates and uses personality models to predict behavior.

The behaviors of other people are often central to envisioning the future. The ability to accurately predict the thoughts and actions of others is essential for successful social interactions, with far-reaching consequences. Despite its importance, little is known about how the brain represents people in order to predict behavior. In this functional magnetic resonance imaging study, participants learned the unique personality of 4 protagonists and imagined how each would behave in different scenarios. The protagonists' personalities were composed of 2 traits: Agreeableness and Extraversion. Which protagonist was being imagined was accurately inferred based solely on activity patterns in the medial prefrontal cortex using multivariate pattern classification, providing novel evidence that brain activity can reveal whom someone is thinking about. Lateral temporal and posterior cingulate cortex discriminated between different degrees of agreeableness and extraversion, respectively. Functional connectivity analysis confirmed that regions associated with trait-processing and individual identities were functionally coupled. Activity during the imagination task, and revealed by functional connectivity, was consistent with the default network. Our results suggest that distinct regions code for personality traits, and that the brain combines these traits to represent individuals. The brain then uses this "personality model" to predict the behavior of others in novel situations.