Cognitive bias modification for threat interpretations: using passive Mobile Sensing to detect intervention effects in daily life

ABSTRACT

Background

Social anxiety disorder is associated with distinct mobility patterns (e.g., increased time spent at home compared to non-anxious individuals), but we know little about if these patterns change following interventions. The ubiquity of GPS-enabled smartphones offers new opportunities to assess the benefits of mental health interventions beyond self-reported data.

Objectives

This pre-registered study (https://osf.io/em4vn/?view_only=b97da9ef22df41189f1302870fdc9dfe) assesses the impact of a brief, online cognitive training intervention for threat interpretations using passively-collected mobile sensing data.

Design

Ninety-eight participants scoring high on a measure of trait social anxiety completed five weeks of mobile phone monitoring, with 49 participants randomly assigned to receive the intervention halfway through the monitoring period.

Results

The brief intervention was not reliably associated with changes to participant mobility patterns.

Conclusions

Despite the lack of significant findings, this paper offers a framework within which to test future intervention effects using GPS data. We present a template for combining clinical theory and empirical GPS findings to derive testable hypotheses, outline data processing steps, and provide human-readable data processing scripts to guide future research. This manuscript illustrates how data processing steps common in engineering can be harnessed to extend our understanding of the impact of mental health interventions in daily life.

KEYWORDS:

• Social anxiety disorder
• passive sensing
• ecological momentary assessment
• cognitive bias modification for interpretation bias
• GPS

Acknowledgements

The authors would like to acknowledge Alexander R. Daros and Miranda L. Beltzer for their assistance in primary data collection.

Data availability statement

Data supporting the analyses are not openly available due to the identifying nature of GPS data. Feature extraction and data analysis scripts are available at https://osf.io/em4vn/ (Daniel et al., 2021). Non-identifiable data from the overall study collection are available at https://osf.io/eprwt/ (Daniel & Teachman, 2020).

Notes

1 Despite the pattern of null results for outcomes observed in Daniel et al. (2020), we decided to proceed with testing the present pre-registered hypotheses given that a more objective source of data (i.e., GPS vs. participant self-report) could uncover condition differences in mobility patterns that participants were not able to report on due to limitations in self-knowledge, demand characteristics, and/or memory bias.

2 All 49 CBM-I group participants who completed at least one CBM-I session and supplied GPS data throughout the five-week study period were included in analyses, regardless of how many additional CBM-I sessions they completed. This is consistent with the analytical approach used in the original paper from this data set (Daniel et al., 2020). Given that the current study requires GPS data to test for intervention effects on participant mobility patterns, and GPS data were not required for the Daniel et al. (2020) paper, there are slight discrepancies between the participants included in analyses across the two studies. Note that the 10 participants who were excluded from the current analyses due to a lack of GPS data did not opt out of providing GPS data; rather, the GPS sampling software was not functioning at the time of their enrollment in the study.

3 Contact the first author for a full list of measures included in the in-lab sessions and the EMA surveys.

4 Data collection was conducted throughout 2017 and 2018. As such, social distancing measures put in place to manage the COVID-19 pandemic were not in place at the time of data collection.

5 We set this threshold given that 50 m has been shown to provide the best performance for spatiotemporal clustering algorithms and semantic labeling of GPS traces (Boukhechba et al., 2018; Kang et al., 2005). Namely, smaller thresholds can trigger more false positives due to the noisy nature of GPS and given the low precision of GPS readings that can range from few meters to tens of meters depending on the user’s context (e.g., indoor vs outdoor, weather conditions, etc.).

Additional information

Funding

This work was supported by the National Institute of Mental Health [grant number R01MH113752]; University of Virginia Hobby Postdoctoral and Predoctoral Fellowship Grant; and Jefferson Scholars Foundation Fellowship Grant.