Objective assessment of depressive symptoms using eye-tracking features: An exploratory pilot study
Jiansen Lin1, Rongrong Zhong2, Lvchun Cui1
1Clinical Research Center & Division of Mood Disorders, Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China; College of Psychiatry and Clinical Psychology, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
Objective:
Major depressive disorder (MDD) exhibits significant symptom heterogeneity, challenging precise clinical assessment. Current evaluations rely on subjective patient and clinician reports, limiting objective monitoring. Eye-tracking technology offers a non-invasive approach to assessing and characterizing clinical features in depressive patients, enabling personalized assessment and management.
Methods:
This cross-sectional pilot study included 74 patients with MDD and 80 healthy controls. Participants completed a free-viewing task and a visual search task. Depressive, anxiety (including tension), anhedonia and sleep symptoms were assessed using the HAMD-17, HAMA, DARS, and PSQI scales. Correlations between symptom scores and eye-tracking variables were analyzed in an exploratory, hypothesis-generating framework, followed by regression models to identify associated eye-tracking variables.
Results:
Multiple eye-tracking features showed exploratory associations with specific symptoms, although none of these associations survived correction for multiple comparisons. Suicidal ideation was associated with longer average fixation duration for threatening pictures during the free-viewing task (AUC 0.753); anhedonia severity with first fixation latency/duration for angry faces in the visual search task; sleep disturbance with longer fixation durations and first fixation latency for positive faces; and tension with longer first fixation duration for negative faces (AUC 0.743).
Conclusion:
Specific eye-tracking features, particularly those related to processing threatening pictures and emotional faces, were associated with severity of depressive symptoms such as suicidal ideation, anhedonia, sleep disturbance, and tension in this exploratory pilot study. These preliminary findings require replication in independent samples before clinical application but support the potential utility of eye-tracking as an objective tool for assessing symptom domains in MDD.
