Predicting Personality with Smartphone Cameras: A Pilot Study

Abstract

Heart rate variability (HRV) provides essential mental health information for clinical diagnosis, telemedicine, preventive medicine, and public health. However, the lack of a convenient detection method limits its potential. This study aimed to investigate the feasibility and credibility of using smartphone Photoplethysmogram (PPG)-based HRV analysis for personality prediction. Ninety-five records were collected from students and university employees in Shenzhen, China. An app recorded five-minute films of their fingertips and converted the frames into HRV measures. Participants who were more extraverted and stable had a higher root mean square of successive differences (rMSSD; p=0.03 and 0.005, respectively), and a higher percentage of successive normal-to-normal (NN) intervals that differed by more than 50 ms (pNN50; p=0.05 and 0.004, respectively), and standard deviation of NN intervals (SDNN; p=0.02 and 0.01, respectively). Stable people also had higher log high-frequency HRV (p=0.008). The results from correlation coefficients and the Bland–Altman analysis verified the accuracy of smartphone PPG in HRV assessment. The correlation coefficients of all HRV measures obtained using smartphone PPG and reference ECG were higher than 0.9. Moreover, the Bland–Altman ratios were less than 0.2 for all HRV measures except pNN50. Taken together, the results of this study provide the first empirical evidence that supports the usability of smartphone PPG as a predictor of personality.