JMIR Publications

ABSTRACT

Background:

Subclinical (i.e., threshold) social anxiety can greatly affect young people’s lives, but existing solutions appear inadequate considering its rising prevalence. Wearable sensors may provide a novel way to detect social anxiety and result in new opportunities for monitoring and treatment that would be greatly beneficial for sufferers, society and healthcare services. Nevertheless, indicators such as skin temperature from wrist-worn sensors have not been used in prior work on physiological social anxiety detection.

Objective:

This study aimed to investigate whether subclinical social anxiety in young adults can be detected using physiological data obtained from wearable sensors, including Heart Rate (HR), Skin Temperature (ST) and Electrodermal Activity (EDA).

Methods:

Young adults (N = 12) with self-reported subclinical social anxiety (measured by the widely used self-reported version of the Liebowitz Social Anxiety Scale, LSAS-SR) participated in an impromptu speech task. Physiological data was collected using an E4 Empatica wearable device. Using the pre-processed data and following a supervised machine learning approach, various classification algorithms such as Support Vector Machine (SVM), Decision Tree, Random Forest and K-Nearest Neighbours (KNN) were used to develop models for three different contexts. Models were trained to (1) classify between baseline and socially anxious states, (2) differentiate between baseline, anticipation anxiety and reactive anxiety states, and (3) classify between social anxiety experienced by individuals with differing social anxiety severity. The predictive capability of the singular modalities was also explored in each of the three supervised learning experiments. The generalisability of the developed models was evaluated using 10-fold cross validation as a performance index.

Results:

With modalities combined, the developed models yielded accuracies between 97.54% and 99.48% when detecting between baseline and socially anxious states. Models trained to differentiate between baseline, anticipation anxiety and reactive anxiety states yielded accuracies between 95.18% and 98.10%. Alongside this, the models developed to detect between social anxiety experienced by individuals with differing anxiety severity scores successfully classified with accuracies between 98.86% and 99.52%. Surprisingly, EDA was identified as the most effective singular modality when differentiating between baseline and social anxiety states, whereas ST was the most effective modality when differentiating between anxiety experienced by individuals with differing social anxiety severity.

Conclusions:

The results indicate that it is possible to accurately detect social anxiety as well as distinguish between levels of severity in young adults by leveraging physiological data collected from wearable sensors.