Effects of Emotional Expressiveness of a Female Digital Human on Loneliness, Stress, Perceived Support, and Closeness Across Genders: Randomized Controlled Trial

Effective interventions to reduce loneliness in big cities

PURPOSE OF REVIEW

Loneliness is a state of distress or discomfort between the desired and experienced level of connectedness to others. These feelings may be particular salient in urban environments that seemingly offer more opportunities for social contact, highlighting the discrepancy. The topic of loneliness has received increased attention because of its negative impact on mental and physical health combined with concerns of increased loneliness due to lockdowns and social distancing regulations during the coronavirus disease 2019 pandemic. We provide a bibliometric and random-effects meta-analysis of clinical trials published since 2020 and available via PubMed.

RECENT FINDINGS

Loneliness interventions have predominantly focused on elderly in the community. Adolescents and young adults as a second group at risk have received much less attention. On average across 44 effect sizes studied in 38 trials, interventions show moderate levels of effectiveness but are characterized by high heterogeneity and trials are often underpowered and use low quality designs. Multidimensional interventions show promise for alleviating loneliness, but the intervention context needs greater attention.

SUMMARY

Scalable and effective interventions for the general population and at-risk groups of loneliness are still scarce. Some promising interventions have been trialled and merit further attention.

A Neuro-Computational Model for Discrete-Continuous Dual-Task Process

Studies on dual-task (DT) procedures in human behavior are important, as they can offer great insight into the cognitive control system. Accordingly, a discrete-continuous auditory-tracking DT experiment was conducted in this study with different difficulty conditions, including a continuous mouse-tracking task concurrent with a discrete auditory task (AT). Behavioral results of 25 participants were investigated via different factors, such as response time (RT), errors, and hesitations (pauses in tracking tasks). In DT, synchronization of different target neuron units was observed in corresponding brain regions; consequently, a computational model of the stimulus process was proposed to investigate the DT interference procedure during the stimulus process. This generally relates to the bottom-up attention system that a neural resource allocates for various ongoing stimuli. We proposed a black-box model based on interactions and mesoscopic behaviors of neural units. Model structure was implemented based on neurological studies and oscillator units to represent neural activities. Each unit represents one stimulus feature of task concept. Comparing the model's output behavior with the experiment results (RT) validates the model. Evaluation of the proposed model and data on RT implies that the stimulus of the AT affects the DT procedure in the model output (84% correlation). However, the continuous task is not significantly changed (26% correlation). The continuous task simulation results were inconsistent with the experiment, suggesting that continuous interference occurs in higher cognitive processing regions and is controlled by the top-down attentional system. However, this is consistent with the psychological research finding of DT interference occurring in response preparation rather than the stimulus process stage. Furthermore, we developed the proposed model by adding qualitative interpretation and saving the model's generality to address various types of discrete continuous DT procedures. The model predicts a justification method for brain rhythm interactions by synchronization, and manipulating parameters would produce different behaviors. The decrement of coupling parameter and strength factor would predict a similar pattern as in Parkinson's disease and ADHD disorder, respectively. Also, by increasing the similarity factor among the features, the model's result shows automatic task performance in each task.