Search Results (129)

Electroencephalography (EEG) is closely related to neural activity underlying emotional states. It has become an important input modality for emotion recognition in affective computing. However, many existing studies rely on complex experimental paradigms and full-channel EEG signals. They often construct high-dimensional features for discrete emotion classification and pay less attention to continuous emotional dynamics. To address these issues, this study uses six frontal EEG channels, including Fp1, Fp2, AF3, AF4, F7, and F8. These channels are relatively easy to acquire and are closely associated with emotional activity. A frontal hemispheric relation-aware temporal convolutional network (FHR-TCN) is proposed for continuous emotion regression and discrete emotion classification. Experiments on MAHNOB-HCI and DEAP evaluated FHR-TCN for continuous emotion regression and discrete emotion classification, respectively. Under the reported protocols, FHR-TCN achieved higher average scores than the evaluated baselines. It also showed lower parameter counts, MACs, and GPU inference latency than GRU. These findings support further deployment-oriented evaluation under limited-channel conditions. Full article

In recent years, physiological signal-based biometrics has gained increasing attention due to its resistance to spoofing attacks, suitability for continuous authentication, and compatibility with wearable devices. This study investigates a multimodal biometric framework based on three physiological signals: galvanic skin response (GSR), electrocardiogram (ECG), and photoplethysmography (PPG). Although GSR has demonstrated promising performance in emotion recognition and animal recognition studies, it remains relatively underexplored in biometric human recognition applications compared with ECG and PPG. As a non-invasive signal that can be easily acquired through simple skin contact, GSR offers several advantages, including low-cost sensing, ease of integration into wearable devices, user convenience, and suitability for continuous monitoring. To address this gap, the proposed framework combines features extracted from ECG, PPG, and GSR signals and employs machine learning algorithms for human recognition. The approach was evaluated using two publicly available datasets, CLAS and MAUS. Experimental results demonstrate that multimodal fusion significantly enhances recognition performance, achieving accuracies of 97% and 99% on the CLAS and MAUS datasets, respectively, with the K-Nearest Neighbors (KNN) classifier. These findings highlight the potential of integrating GSR with ECG and PPG signals to develop biometric systems for wearable and continuous authentication applications. Full article

Facial emotion recognition (FER) in real-world settings is limited by the semantic mismatch between discrete emotion categories and continuous Valence–Arousal–Dominance (V-A-D) dimensions and the lack of reliable uncertainty estimates under incomplete facial evidence. Existing uncertainty-aware FER studies mainly address annotation ambiguity or training-time reliability, leaving the behavior of predictive uncertainty under progressive input degradation insufficiently examined. This paper proposes BGDC (Bayesian Gaussian-mixture Distributional Consistency), a multi-task FER framework that integrates a GMM-based soft consistency module with a context-conditioned Bayesian regression head and explicitly models aleatoric and epistemic uncertainty. To evaluate predictive reliability, a controlled masking protocol is introduced to remove facial information under different spatial configurations. On FER2013-VAD, BGDC attains the highest classification accuracy of 0.6943 and the highest mean V-A-D CCC of 0.6079 among the compared configurations, and it yields a stronger epistemic uncertainty-error correspondence than MC Dropout in a single-model setting. Controlled masking further shows that the epistemic uncertainty of BGDC tracks task-relevant facial information loss rather than masking ratio alone: it rises with regression error when diagnostically important regions are removed, and it contracts when the masked region is largely task-irrelevant. Combining Bayesian uncertainty with the GMM-based distributional prior thus enables reliability-aware multi-task FER, in which controlled masking serves as a diagnostic intervention rather than as a benchmark of accuracy degradation alone. Full article

There is growing recognition of the health benefits that forests and green spaces provide to people. Forest bathing is a practice that promotes relaxation and human well-being through immersive, mindful experiences in forest environments. How forest bathing affects distinct dimensions of well-being is still not fully understood. In this study, we assessed changes in well-being before and after two and four forest bathing sessions and examined whether a brief introductory session on forest ecosystem services enhanced participants’ overall perception of well-being. Forty adults from a rural community in southern Chile completed the Warwick–Edinburgh Mental Well-being Scale to assess perceived well-being. Participants showed improvements in overall well-being after two sessions, with the most significant gains in relaxation, optimism, clarity of thought, and social connection. Scores remained stable between the second and fourth sessions, suggesting that initial exposure offers the most substantial benefits, while continued practice helps maintain them. Although the introductory session did not significantly affect overall well-being scores, it showed positive effects on optimism and social connection. These findings highlight forest bathing as an effective nature-based intervention to promote emotional and social well-being, with implications for policies advancing public health and sustainability goals. Full article

As core architectural environments for cultural heritage preservation and public education, museums are evolving from static exhibition spaces into immersive, multisensory interactive environments. The sensory attributes of the architectural environment—including multimodal information such as light, sound, and touch—exhibit a dynamic coupling with visitors’ emotional states. Responding to visitors’ growing emphasis on emotional enhancement, this study aims to improve the emotional experience of museum tours through multisensory compensation strategies. First, we conducted an experiment at the Shaanxi Archaeology Museum, capturing facial videos of participants during their tours and utilizing a facial expression analysis system for continuous emotion recognition. Subsequently, drawing on theories of multisensory interaction and emotion regulation, we constructed a multisensory emotion regulation model to guide the sensory compensation experiment. Visualization analysis of the results confirmed that multisensory compensation strategies within the architectural environment significantly increased positive emotions (from 48.23% to 60.78%). This study focuses on the mechanisms by which sensory compensation strategies in the architectural environment influence visitors’ emotional experiences, aiming to promote the transformation of cultural heritage spaces from “function-oriented” to “emotion-oriented” environments. Full article

We report a cross-station replication of endogenous circadian rhythms in plant bioelectric voltage, recorded continuously for 42 days at three independent sensor stations within a public science exhibition (Phänomena, Dietikon, Switzerland; March–April 2026). Three primrose (Primula vulgaris) stations were equipped with custom Biolingo bioelectric sensors (ESP32 + AD8232) and recorded autonomously through approximately 21,000 visitor interactions. We extracted DC-invariant spectral features from 5–10 s voltage windows (n = 78,431 quality-filtered files) and fitted two-stage cosinor models with bootstrap 95% confidence intervals. All three stations show a robust 24 h rhythm in the 1–5 Hz band power (bp_1–5), with peak-to-trough amplitudes between 0.35× and 1.19× of mesor (R²_med 0.72–0.87). Acrophase varies across stations from 05:00 to 11:00 local time. Critically, the rhythm survives an overnight-only restriction (18:00–09:00, no visitors) at all three stations, ruling out visitor presence as the rhythm driver. The most visitor-intensive station (faces of museum visitors triggering an emotion-recognition installation) additionally shows a sharp daytime amplitude collapse coincident with the exhibition opening at 09:00, during the hours of sustained visitor presence. This temporal coincidence is consistent with—though not by itself proof of—the cardiovascular-mechanosensory coupling characterized at single-subject resolution in a companion study. We argue that bp_1–5—the spectral band most directly related to plant action-potential activity—carries an endogenous circadian signal in Primula vulgaris and that this station-level signal co-varies with sustained nearby human presence in a manner consistent with frequency-selective mechanosensory coupling, although the observational design cannot establish this mechanism. From a biomimetic perspective, this suggests that the plant’s evolved bioelectric sensing apparatus might be leveraged as a live ambient biosensor for nearby human activity, complementing the more common biomimetic approach of replicating plant sensing in synthetic devices. Full article

Background/Objectives: Concussions occur across all sports globally; however, inconsistent symptom recognition continues to challenge diagnosis, management, and recovery. Although concussion effects extend beyond neurological dysfunction, gastrointestinal (GI) symptoms remain insufficiently captured within current assessment frameworks and may influence athletes’ fuelling choices and food tolerance during recovery. This study aimed to describe the prevalence and patterns of GI symptoms reported among athletes with a history of concussion and to explore whether symptom prevalence and burden differed by sex and country. Methods: A total of 401 adult athletes (225 males; mean age 32.4 ± 11.3 years) with a history of concussion were recruited from a multinational cohort (n = 123) and a US community rugby cohort (n = 278). Participants completed online surveys assessing concussion history, post-concussion symptoms (RPQ or SCAT-6), and GI-specific symptoms. Data were stratified by sex and country and analysed descriptively and comparatively. Results: Overall, 62.1% of athletes reported ≥1 GI symptom, with most reporting 1–5 symptoms (79.9%). The most frequent symptoms were nausea/vomiting (74.3%) and loss of/poor appetite (56.6%), followed by diarrhoea, abdominal pain/discomfort, and flatulence. Although GI symptom prevalence was higher in females (65.9% vs. 59.1%), no significant sex differences were observed (all p > 0.05). GI symptom burden varied by country (p < 0.001), with higher prevalence among Irish than US athletes (85.9% vs. 51.2%); however, these analyses were exploratory and unadjusted. RPQ and SCAT captured symptoms across somatic, cognitive, emotional, and sleep domains, however GI symptoms were underrepresented, with nausea/vomiting more frequently reported on RPQ/SCAT items (51.4%) than on GI-specific items (46.1%). Conclusions: GI symptoms were common following concussion, with variation by country and no statistically significant differences by sex. Findings indicate that concussion assessment tools (RPQ/SCAT) underrepresent the breadth of GI symptoms. Greater attention to GI assessment in concussion care is warranted. Incorporating simple GI screening alongside timely nutrition support may represent feasible additions to multidisciplinary, athlete-centred care pathways. Prospective studies are needed to clarify the clinical relevance of these findings and evaluate nutrition-related strategies. Full article

Facial expression recognition is a pivotal area in computer vision, traditionally focusing on categorical labels such as ‘Happy’, and ‘Sad’. Recent advancements have transitioned towards using the continuous indicators valence and arousal, reflecting the complicated nature of human emotions. This study introduces a method using semi-supervised learning to generate valence and arousal labels for existing categorical datasets, addressing challenges like racial bias. We propose a pseudo-label refinement framework, VAP-Refine (Valence-Arousal Pseudo-label Refinement), which enhances facial expression recognition by combining teacher model predictions and category-level statistics. These predicted labels are adjusted using ground truth category information and combined with actual category labels to train a more robust facial expression recognition model. Our approach improved accuracy from 69.3% to 72.26% and from 62.94% to 67.24% across two datasets. Fine-tuning with a teacher model predicting valence and arousal achieved 75.82% and 91.58% accuracy, with an F1-score of 0.9147, despite data imbalances. These results highlight the potential of semi-supervised learning to enhance facial expression recognition by incorporating continuous emotional indicators, improving model performance and contributing to more accurate affective computing applications. Furthermore, the proposed framework consistently improved performance across various backbone architectures, including ResNet50, SHViT, and DDAMFN++, highlighting its generalizability and versatility. Full article

Diabetes Mellitus is a chronic metabolic disorder characterized by elevated blood glucose due to defects in insulin secretion or action, or both, leading to serious short- and long-term complications if not effectively managed. However, there is limited qualitative evidence exploring how youths diagnosed with Type 1 Diabetes Mellitus (T1DM) experience disease onset, management, complications, emotional adaptation, and education within the South African public healthcare system. The study aims to investigate the lived experiences of youths living with T1DM in a selected public hospital in Limpopo province, South Africa. The objectives were to explore and describe the lived experiences of youths living with T1DM. A qualitative, explorative, descriptive, and contextual design was used to gain a thorough understanding of the experiences of youths living with T1DM. A non-probability sampling technique was used to select 12 participants using a pre-determined criterion. Data were collected through individual semi-structured interviews using an interview guide. The data were analyzed using Colaizzi’s method, where themes and sub-themes were developed with the inclusion of an independent coder. Measures to ensure trustworthiness and ethical considerations were adhered to throughout the study. The findings revealed that, despite the participants sharing the same diagnosis, they experience multiple interrelated barriers that significantly hindered effective self-care management, such as limited access to diabetic diet, glucometers and supplies, treatment and informational-related barriers, school-related challenges, transportation constraints and inadequate social support. Furthermore, the findings highlighted gaps in early recognition of symptoms, standardized diabetes education, psychosocial support, and continuity of care. The study recommends the need for holistic, patient-centred, and contextualized interventions that do not only address medical management but the socioeconomic, educational, and psychological needs of youths. Full article

Driver emotion recognition constitutes a fundamental pillar of intelligent cockpit systems, playing a pivotal role in enhancing driving safety and optimizing human–machine interaction. Despite the integration of vehicle sensor data in recent multimodal approaches, conventional fusion paradigms frequently encounter performance degradation due to the inherent noise and weak semantic correlation between vehicle telemetry and emotional states. To address these challenges, this study introduces a Dimensional Emotion-Guided Multi-task (DEGM) framework, a novel architecture designed to explicitly formalize the asymmetric roles of visual and vehicular modalities. Rather than employing simplistic feature concatenation, the proposed method maps multivariate vehicle data into a continuous Valence–Arousal–Dominance (VAD) space to characterize latent emotional tendencies within specific driving contexts. These predicted dimensions subsequently serve as semantic priors to conditionally modulate global facial representations through a Feature-wise Linear Modulation (FiLM) mechanism, facilitating robust and interpretable cross-modal interaction. Furthermore, the framework adopts a multi-task learning strategy that jointly optimizes discrete emotion classification and continuous dimension regression, leveraging the latter as a structural regularizer to refine the latent feature space. Comprehensive evaluations on the public PPB driving emotion dataset demonstrate that the proposed DEGM achieves a competitive accuracy of 87.50% and a weighted F1-score of 0.8727. The results validate that our framework provides a lightweight and robust paradigm for context-aware affect sensing, demonstrating strong potential for practical deployment in intelligent transportation systems. Full article

Nowadays, educational software across all learning levels is increasingly enhanced with Artificial Intelligence (AI), primarily through content generation or post-session learning analytics. However, most existing systems remain weakly connected to learners’ real-time affective states and rarely exploit emotional information as a direct control signal for instructional adaptation. In this work, we propose a proof-of-concept closed-loop affect-aware educational adaptation framework that integrates real-time facial emotion recognition into a dynamic learning control system. The proposed approach is built upon a dual-model ensemble architecture, combining a transformer-based model (CAGE) and a CNN-based model (DDAMFN++) trained on large-scale in-the-wild datasets. To bridge heterogeneous emotion representations, we introduce a probabilistic fusion strategy that aligns continuous valence–arousal predictions with discrete emotion classification via a Gaussian Mixture Model (GMM), enabling unified emotion inference in real time. Based on the fused emotional state, a temporal aggregation mechanism is applied to capture sustained affective trends rather than transient expressions. These aggregated signals are then mapped to instructional decisions through an emotion-driven adaptive control policy, which adjusts activity difficulty using an Average Emotion Score (AES). This establishes a fully automated closed-loop adaptation cycle, where detected learner affect directly influences the learning environment without requiring explicit user input or post-session questionnaires. The framework is integrated into an open-source educational platform (eduActiv8) to demonstrate feasibility and system-level behavior. Results from alpha-level validation show that the system can continuously monitor learner affect, generate interpretable emotional analytics, and dynamically adjust task difficulty in real time, while reducing user interaction overhead. This study contributes a modular architecture for affect-aware educational systems by combining real-time ensemble emotion recognition, probabilistic fusion of heterogeneous outputs, and closed-loop instructional adaptation. The proposed framework provides a foundation for future research in scalable, emotion-driven intelligent tutoring and adaptive learning environments. Full article

The increasing prevalence of work-related stress requires advanced, non-intrusive physiological monitoring solutions. As conventional methods are often impractical for continuous, real-world applications, this study investigates the deployment of artificial intelligence models on embedded systems for real-time emotion recognition from physiological signals. The study identified critical constraints for embedded implementation, including model size and memory capacity. An evaluation of various machine learning algorithms revealed that, while models like K-Nearest Neighbors (KNN) achieve high accuracy (88.8%), their excessive memory footprints make them unsuitable for resource-constrained hardware. Consequently, Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and recurrent neural network (RNN) architectures were deployed on an STM32F411 microcontroller, for which model compression proved essential. An experimental study validated the approach, achieving high recognition rates for pronounced emotions such as hatred (91%) and anger (85%), though with a lower accuracy for more subtle states. These results confirm the potential of embedded AI systems for physiological monitoring, highlighting the critical importance of feature selection and model compression for practical implementation. Full article

There are still challenges in speech emotion recognition, as the representation capability of single-modal information is limited, there are difficulties in capturing continuous emotional transitions in discrete emotion annotations, and the issues of modal structural differences and cross-sample alignment in multimodal fusion methods persist. To address these, this study undertakes work from both data and model perspectives. For data, a Chinese multimodal database STEM-E²VA was constructed, synchronously collecting four modalities of data: articulatory kinematics, acoustics, glottal signals, and videos. This covers seven discrete emotion categories and employs PAD continuous annotation. By integrating discrete and continuous dimensional annotations, it better represents the distinction between strong and weak emotions under the same discrete emotion label. Concurrently, to process the biases in PAD annotations, we employed the SCL-90 psychological questionnaire to analyze annotators’ cognitive and emotional perceptions, thereby ensuring data reliability. For model, this paper proposes a multimodal supervised contrastive fusion network incorporating PAD perception. It employs a PAD-enhanced hybrid contrastive loss function to optimize intra-model and inter-modal feature alignment. Utilizing a cross-attention mechanism combined with a GRU–Transformer network for temporal feature extraction, it achieves deep fusion of multimodal information, reducing inter-modal discrepancies and cross-class confusion. Experiments demonstrate that the proposed method achieves 85.47% accuracy in discrete sentiment recognition on STEM-E²VA, with a substantial reduction in RMSE for PAD dimension prediction. It also exhibits excellent generalization capability on IEMOCAP, providing a novel framework for integrating discrete and continuous sentiment representations. Full article

Existing research confirms associations between Big Five personality traits and emotional states, yet investigations into how personality traits modulate emotional dynamics and their gender-specific patterns remain limited. The present study developed a TFace-Bi-GRU-SE deep learning model that achieved a weighted accuracy of 63.50 ± 0.98% (peak single-run: 64.96%) and an F1 score of 65.21% in performance testing, with a single-inference time of 14.1 s, outperforming traditional methods. The model processed 10 min video recordings from 30 participants (19,262 observations), generating time-series data for valence (P) and arousal (A). Combined with Big Five personality assessments, continuous-time structural equation modeling (CTSEM) revealed distinct emotional dynamics: both P and A exhibited significant negative autoregression (−0.056 and −0.558, p < 0.001), with A reverting to baseline substantially faster (half-life: 1.2 s) than P (half-life: 12.3 s); cross-lagged effects were nonsignificant (P_A: 0.007; A_P: −0.026, p > 0.05). Arousal demonstrated greater instantaneous volatility (=0.339) than valence (=0.286, p < 0.001), with positive covariation between dimensions (0.218, p = 0.006). Exploratory analyses (N = 30) indicated that higher neuroticism and openness scores were associated with elevated arousal (Cohen’s d > 0.8), whereas higher agreeableness and conscientiousness scores were associated with elevated valence (d > 0.8). Gender moderated the neuroticism–arousal relationship, with more potent effects in females (r = 0.746, p = 0.008). Robustness analyses demonstrated high stability of core DRIFT parameters (P_P, A_A): bootstrap resampling (n = 50) yielded coefficients of variation < 0.35 with 100% directional consistency; subgroup validation confirmed cross-sample invariance. Sensitivity analyses revealed that an additional 8% measurement error induced less than 9% bias (8.3% for both P_P and A_A) in autoregressive parameters while preserving half-life ratios, confirming CTSEM’s capacity to extract reliable dynamics from moderately accurate AI outputs. Bootstrap and Bayesian analyses identified ten personality–DRIFT associations with directional consistency ≥ 70%; these constitute preliminary hypotheses for adequately powered future studies (N ≥ 61). This study provides methodological foundations for personalized affective intervention research. Data and code are publicly available (see Data Availability Statement). Full article

Emotion recognition from physiological signals has immense applications in healthcare and human–computer interaction. We developed an electrodermal activity (EDA)-graph signal processing pipeline that produces highly sensitive features for detecting the affective dimensions (arousal and valence) of emotions. Using the Continuously Annotated Signals of Emotion dataset, we compared our graph-based EDA features (EDA-graph) with traditional time- and frequency-domain EDA features and features derived from other signals (heart rate variability, pulse transit time, electromyography, skin temperature, and respiration) for detecting affective dimensions using machine learning regression models. The EDA-graph features showed superior performance in continuous affective dimension recognition compared to the most accurate state-of-the-art models, achieving RMSE values of 0.801 for arousal and 0.714 for valence. Furthermore, we used a variety of traditional and recently published datasets collected in laboratory and ambulatory settings to perform a comprehensive evaluation of the robust generalization capabilities of our approach across different emotional contexts. The models demonstrated exceptional performance in classifying emotional states across the datasets, achieving 98.2% accuracy in detecting positive, negative, and mixed emotions; 92.75% in discriminating between emotions (relaxed, amused, bored, scared, and neutral); and 86.54% in detecting stress vs. no stress. These results highlight the potential of a graph-based analysis of EDA in emotion recognition systems in different contexts, especially for real-world applications. Full article