Background/Objectives: Imposed sleep restriction leads to increased feelings of appetite and hedonic eating behaviors (or food rewards). No study to date has assessed home-based measures of sleep with appetite and food rewards exclusively in Black emerging adults (ages 18–28 years), despite higher risks of short sleep and obesity in this population. We examined associations between 6-month changes in sleep with changes in appetite and food reward in Black emerging adults.

Methods: Fifteen Black emerging adults (12 females; age, 21 ± 2.5 years; body mass index, 25.7 ± 4.5 kg/m²; body fat, 25.8 ± 11.9%) completed two identical 7-day measurement bursts at baseline and 6 months. Sleep (duration, efficiency, and architecture) was captured via 7 days of actigraphy and 2 nights of in-home polysomnography. During a laboratory visit, participants completed appetite measures (desire to eat, hunger, fullness, and prospective food consumption) via visual analog scales before and for 3 h following standard breakfast intake. The food reward for the fat and sweet categories of food was measured before lunch with the Leeds Food Preference Questionnaire.

Results: Fasting fullness scores decreased from baseline to 6 months (−8.9 mm, p < 0.01) despite increases in body weight (2.6 kg, p < 0.01) and waist circumference (2.4 cm, p = 0.03). Increases in actigraph-measured sleep duration were associated with decreases in fasting desire to eat (r = −0.58, p = 0.04). Increases in actigraph-measured sleep efficiency were also associated with decreases in explicit liking for sweet foods (r = −0.60, p = 0.03).

Conclusions: Our findings suggest that improvements in sleep duration and sleep efficiency may lead to decreased feelings of appetite and food reward in Black emerging adults.

Background: Recent advancements in large language models (LLMs) have generated significant interest in their potential for assessing psychological constructs, particularly personality traits. While prior research has explored LLMs' capabilities in zero-shot or few-shot personality inference, few studies have systematically evaluated LLM embeddings within a psychometric validity framework or examined their correlations with linguistic and emotional markers. Additionally, the comparative efficacy of LLM embeddings against traditional feature engineering methods remains underexplored, leaving gaps in understanding their scalability and interpretability for computational personality assessment.

Objective: This study evaluates LLM embeddings for personality trait prediction through four key analyses: (1) performance comparison with zero-shot methods on PANDORA Reddit data, (2) psychometric validation and correlation with LIWC (Linguistic Inquiry and Word Count) and emotion features, (3) benchmarking against traditional feature engineering approaches, and (4) assessment of model size effects (OpenAI vs BERT vs RoBERTa). We aim to establish LLM embeddings as a psychometrically valid and efficient alternative for personality assessment.

Methods: We conducted a multistage analysis using 1 million Reddit posts from the PANDORA Big Five personality dataset. First, we generated text embeddings using 3 LLM architectures (RoBERTa, BERT, and OpenAI) and trained a custom bidirectional long short-term memory model for personality prediction. We compared this approach against zero-shot inference using prompt-based methods. Second, we extracted psycholinguistic features (LIWC categories and National Research Council emotions) and performed feature engineering to evaluate potential performance enhancements. Third, we assessed the psychometric validity of LLM embeddings: reliability validity using Cronbach alpha and convergent validity analysis by examining correlations between embeddings and established linguistic markers. Finally, we performed traditional feature engineering on static psycholinguistic features to assess performance under different settings.

Results: LLM embeddings trained using simple deep learning techniques significantly outperform zero-shot approaches on average by 45% across all personality traits. Although psychometric validation tests indicate moderate reliability, with an average Cronbach alpha of 0.63, correlation analyses spark a strong association with key linguistic or emotional markers; openness correlates highly with social (r=0.53), conscientiousness with linguistic (r=0.46), extraversion with social (r=0.41), agreeableness with pronoun usage (r=0.40), and neuroticism with politics-related text (r=0.63). Despite adding advanced feature engineering on linguistic features, the performance did not improve, suggesting that LLM embeddings inherently capture key linguistic features. Furthermore, our analyses demonstrated efficacy on larger model size with a computational cost trade-off.

Conclusions: Our findings demonstrate that LLM embeddings offer a robust alternative to zero-shot methods in personality trait analysis, capturing key linguistic patterns without requiring extensive feature engineering. The correlation between established psycholinguistic markers and the performance trade-off with computational cost provides a hint for future computational linguistic work targeting LLM for personality assessment. Further research should explore fine-tuning strategies to enhance psychometric validity.

Objective: This study examined accuracy and bias in daily perceptions of a romantic partner's autonomy support and undermining of health behaviors and the associations between misperceptions and both partners' daily experiences.

Design: A sample of 106 cohabiting couples from the community completed daily reports (N = 2377) of their own and their partner's autonomy support and undermining, and their daily experiences.

Results: We found tracking accuracy and projection in perceptions of provider's autonomy support and undermining, and evidence of misperceptions: people over-perceived the provider's undermining and under-perceived the provider's autonomy support. Misperceptions were linked with more negative experiences, indicating support and undermining may have the most benefit and least harm when more accurately perceived.

Conclusion: Although there is some accuracy in daily perceptions of provider support and undermining of health behaviors, there are also biases at play in perceptions and misperceptions are linked with more negative experiences, indicating that support and undermining have the most benefit and least harm to both perceiving and providing partners when such behaviors are accurately perceived.

Background
Trauma rooms are complex environments that require rapid actions and well-designed solutions to enhance patient care. Through simulation-based mock-up evaluations, this study examined design strategies for improved workflow, reduced interruptions, and enhanced technology integration in trauma rooms.

Methods
First, a design thinking session was conducted to develop design guidelines for a trauma room. Next, a full-scale room mock-up was built to accommodate scenario-based simulations by clinicians in various roles. Simulations were video recorded. After simulations, clinicians were interviewed to gather further insights.

Results
From the video analysis, 63 events were systematically coded, revealing variations in disruption frequency across different locations in the room. For instance, storage areas were the most frequently accessed spots, causing most disruptions to occur in transitional zones where nurses move around to retrieve medical supplies. The interviews highlighted the effectiveness of various design elements; for example, projecting medical procedures on a screen in the anteroom for students and observers was positively received by most participants.

Conclusion
Mock-up simulations are invaluable in the early design of healthcare facilities, particularly trauma rooms. They help identify workflow patterns, interruption risks, and technology interactions while incorporating input from diverse clinicians, ensuring an efficient and practical layout that meets real-world demands.

Counterintuitively, many people avoid self-relevant information. Numerous cross-sectional, lab-based empirical studies have found that people avoid information to avoid negative future emotions or when lacking coping resources; however, no studies have explored the day-to-day temporal dynamics of information avoidance. The present two-week daily diary study examined temporal dynamics of U.S. college students' (N = 181; approximately 2,200 diaries) information avoidance in everyday life across different contexts. Participants responded to measures assessing affect, coping resources, and behavior, which were predicted to be bidirectionally associated with information avoidance, building on existing models of information avoidance. Within- and between-person associations were assessed via multilevel models controlling for sociodemographic factors. The average participant avoided information on 30.49% of days, with more within-person than between-person variability. Participants were more likely to avoid information on days on which they reported greater negative affect and believed they had less self-efficacy in the avoided domain (e.g., participants avoided health information when lacking health self-efficacy). Information avoidance was not significantly associated with next-day affect, coping resources, or behavior, perhaps due to the design of the present study. Although null findings are inconsistent with the notion that information avoidance is maladaptive (at least in the short term), future research may explicate the consequences of information avoidance. Further understanding the temporal dynamics of information avoidance can help refine theories and best practices of information dissemination.