This study investigated the relationship between cardiorespiratory health and cognitive functioning in young adults with a recent onset of schizophrenia. Participants (N=105) treated at the UCLA Aftercare Program were provided four weekly sessions of cognitive training conducted in a group over a six-month period. Half were assigned to a Cognitive Training plus Exercise group (CT+E) and received a 150 min/week structured aerobic exercise program, and the other half were provided Cognitive Training plus a Healthy Living Group (CT+HLG). Intervention frequency was reduced for months 7 - 10. All participants were assessed with the MATRICS Consensus Cognitive Battery at baseline, 6 months, and 12 months. Cardiorespiratory fitness was assessed with the Bruce protocol at the same time points. The overall MCCB composite score at 6 months within the CT+E group was significantly correlated with body fat percentage (r = −0.52, p = .006), absolute VO2 max uptake (r = .41, p = .048), and relative VO2 max (r = .55, p = .006). The pattern of correlations was similar at the 12-month point. There were fewer significant correlations within the CT+HLG participants, which strengthens the argument that structured aerobic exercise created meaningful physiological cognitive coupling. Improvements in aerobic fitness may enhance neurocognitive functioning through mechanisms related to greater cerebral blood flow, brain derived neurotrophic factor, and neurotransmitter and mitochondrial functioning.

Daughters of depressed mothers are at elevated risk for developing Major Depressive Disorder (MDD) by early adulthood, with mixed evidence suggesting that inhibitory control (IC), the ability to regulate automatic responses, is a heritable cognitive process. This study utilizes low- and high-risk daughters based on maternal depression history to examine whether intelligence and IC, assessed via Stroop, are associated with risk and depressive symptoms. Mothers and their biological daughters (dyads, N=45, ages 11-16) completed a modified Stroop, which facilitated fast and incorrect responses; IC performance is evaluated by reaction time (RT) to same- and switch-color conditions. Dyads also completed a brief assessment of intelligence using the WASI-II and relevant self-report depression measures (BDI-II for mothers and CESD for daughters). We hypothesize strong correlations between daughter and mother intelligence, and that reduced IC would be associated with greater depression (or risk) and lower intelligence. We test these hypotheses using linear regressions controlling for age and education, predicting 1) daughter’s intelligence from mother’s intelligence, and 2) Stroop RT from intelligence, group (high/low risk), depression severity, and Stroop condition. Understanding how maternal depression and cognitive functioning interact may inform early identification strategies and targeted prevention efforts for youth at elevated risk, ultimately contributing to reducing the onset and burden of depression across development.

Suicide is a leading cause of death among youth in the United States. Marginalized youth, defined by factors such as race, ethnicity, socioeconomic status, and gender identity, are disproportionately exposed to trauma and face increased risk for psychological distress, yet often encounter barriers to accessing mental health care. These disparities highlight the need for comprehensive assessment in adolescent inpatient settings, where standardized trauma screening at admission is often lacking. This study addresses this gap by developing a machine learning based approach to identify trauma exposure and predict suicide risk among youth admitted for suicidality. Using electronic health record data, we will apply methods including natural language processing and random forest models to develop and evaluate a trauma screening tool. We will also examine how socio ecological stressors such as neighborhood environment and discrimination contribute to suicide risk. Findings from this study aim to improve identification of high risk youth and support the development of targeted, evidence based interventions addressing both presenting symptoms and underlying risk factors.

Objective: Determine whether mechanisms of devalued positive affect (dampening) are identifiable in positive linguistic sentiment scores, and whether positive sentiment scores at baseline negatively predict depressive scores at follow-up. Method: 75 participants recounted memories at baseline, mid-treatment, post-treatment, and follow-up sessions as part of a virtual reality trial aimed at upregulating positive affect. Depression and anxiety were measured using DASS-21. Using a machine learning architecture, positive linguistic sentiment was extracted from recounted memories. Results: Positive sentiment at baseline and post-treatment predicted positive sentiment at follow-up (b = .556, p < .004; b = .777, p < .000). Positive sentiment was not correlated with dampening cross-sectionally/longitudinally. Positive sentiment was not correlated with depression cross-sectionally except at post-treatment, where an effect was observed controlling for baseline positive sentiment and pre/post-treatment dampening (b = 4.13, p < .017). Conclusions: Positive sentiment did not correlate with/predict dampening. Positive sentiment predicting same-session depression at post-treatment — controlling for initial positive sentiment and pre/post-treatment dampening – suggests that participants who are more depressed may recruit more positive sentiment, so as to offset higher dampening (dampening predicting depression at post-treatment supports this). Further analyses are necessary to parse this and other potential mechanisms.

This study examined how reward probability and response-reward contingency can affect the spatial variability of pecking in pigeons. Prior research has shown that pecking becomes more variable as rewards become less frequent–a pattern consistent with the Modified Law of Effect (Blaisdell et. al, 2016). In this experiment, pigeons experienced Instrumental (INS), Omission (OMS), and Pavlovian (PAV) contingencies across random ratio schedules (RR2, RR5, and RR20). On INS trials, food was delivered for meeting a peck-count criterion, on OMS trials meeting that criterion caused food to be withheld, and on PAV trials food was delivered regardless of behavior. The ratio value (RR2, RR5, RR20) determined how many pecks on average were required to reach criterion. As expected, all INS conditions and lower-ratio OMS conditions showed significant negative correlations (all ps < .05). However, RR20 OMS showed a significant positive correlation (p < .001), with higher reinforcement associated with higher variability. For this contingency, reinforced OMS trials showed more variability than non-reinforced trials, while the reverse was shown for INS trials–suggesting that omission contingencies inadvertently reinforced variability by rewarding pigeons for pecking below the criterion. These findings point to two distinct types of variability: unconditional variability driven by reward expectation, and learned variability sustained by contingency structure.