Dynamic Tracking of State Anxiety via Multi-Modal Data and Machine Learning

Manipulating reinforcement probability changes the temporal overestimation of anticipated aversive stimuli

The interval timing can be distorted by emotions. Previous studies indicated that anticipated fear stimuli can lead to temporal overestimation, similar to the effects observed from direct exposure to fear stimuli. However, this time distortion may not always manifest when anticipated. This study aimed to systematically examine the effect of the reinforcement probability of anticipated fear stimuli on time perception in predictable emotional scenarios. The experiment established 100% fear conditioning by associating a conditioned stimulus (CS+) with an aversive unconditioned stimulus (US), electrical stimulation. Participants completed a temporal bisection task under different cues (threat CS+ and safe CS-) expectations. Participants were explicitly informed that an aversive electrical stimulus would always follow the threat cues (CS+) with 100% probability, though in reality, different blocks presented the threat cue with probability manipulation of 50 and 100%. Results showed that only in the 50% reinforcement probability, participants overestimated the duration when anticipating aversive electrical stimulation, while no significant differences were observed in the 100% reinforcement probability. Additionally, the effect of anxiety on temporal judgment failed to capture the overall trend as fixed effects but only contributed to the individual variations as random effects. The findings suggest that the anticipated aversive electrical stimulation may lead to temporal overestimation. Furthermore, the results indicate that a reliable approach for manipulating the effect of anticipated aversive electrical stimulation on temporal overestimation is to establish 100% fear conditioning and use a reinforcement probability like 50%.

Heart rate variability: Evaluating a potential biomarker of anxiety disorders

Establishing quantifiable biological markers associated with anxiety will increase the objectivity of phenotyping and enhance genetic research of anxiety disorders. Heart rate variability (HRV) is a physiological measure reflecting the dynamic relationship between the sympathetic and parasympathetic nervous systems, and is a promising target for further investigation. This review summarizes evidence evaluating HRV as a potential physiological biomarker of anxiety disorders by highlighting literature related to anxiety and HRV combined with investigations of endophenotypes, neuroimaging, treatment response, and genetics. Deficient HRV shows promise as an endophenotype of pathological anxiety and may serve as a noninvasive index of prefrontal cortical control over the amygdala, and potentially aid with treatment outcome prediction. We propose that the genetics of HRV can be used to enhance the understanding of the genetics of pathological anxiety for etiological investigations and treatment prediction. Given the anxiety-HRV link, strategies are offered to advance genetic analytical approaches, including the use of polygenic methods, wearable devices, and pharmacogenetic study designs. Overall, HRV shows promising support as a physiological biomarker of pathological anxiety, potentially in a transdiagnostic manner, with the heart-brain entwinement providing a novel approach to advance anxiety treatment development.

Machine Learning for Multimodal Mental Health Detection: A Systematic Review of Passive Sensing Approaches

As mental health (MH) disorders become increasingly prevalent, their multifaceted symptoms and comorbidities with other conditions introduce complexity to diagnosis, posing a risk of underdiagnosis. While machine learning (ML) has been explored to mitigate these challenges, we hypothesized that multiple data modalities support more comprehensive detection and that non-intrusive collection approaches better capture natural behaviors. To understand the current trends, we systematically reviewed 184 studies to assess feature extraction, feature fusion, and ML methodologies applied to detect MH disorders from passively sensed multimodal data, including audio and video recordings, social media, smartphones, and wearable devices. Our findings revealed varying correlations of modality-specific features in individualized contexts, potentially influenced by demographics and personalities. We also observed the growing adoption of neural network architectures for model-level fusion and as ML algorithms, which have demonstrated promising efficacy in handling high-dimensional features while modeling within and cross-modality relationships. This work provides future researchers with a clear taxonomy of methodological approaches to multimodal detection of MH disorders to inspire future methodological advancements. The comprehensive analysis also guides and supports future researchers in making informed decisions to select an optimal data source that aligns with specific use cases based on the MH disorder of interest.

Enhancing the conversational agent with an emotional support system for mental health digital therapeutics

As psychological diseases become more prevalent and are identified as the leading cause of acquired disability, it is essential to assist people in improving their mental health. Digital therapeutics (DTx) has been widely studied to treat psychological diseases with the advantage of cost savings. Among the techniques of DTx, a conversational agent can interact with patients through natural language dialog and has become the most promising one. However, conversational agents' ability to accurately show emotional support (ES) limits their role in DTx solutions, especially in mental health support. One of the main reasons is that the prediction of emotional support systems does not extract effective information from historical dialog data and only depends on the data derived from one single-turn interaction with users. To address this issue, we propose a novel emotional support conversation agent called the STEF agent that generates more supportive responses based on a thorough view of past emotions. The proposed STEF agent consists of the emotional fusion mechanism and strategy tendency encoder. The emotional fusion mechanism focuses on capturing the subtle emotional changes throughout a conversation. The strategy tendency encoder aims at foreseeing strategy evolution through multi-source interactions and extracting latent strategy semantic embedding. Experimental results on the benchmark dataset ESConv demonstrate the effectiveness of the STEF agent compared with competitive baselines.