Shao-Min (Sean) Hung¹, Yen-Ju Feng², Po-Jang (Brown) Hsieh²; ¹Waseda Institute for Advanced Study, Waseda University, ²Department of Psychology, National Taiwan University

Although psychophysical experiments have revealed a wide range of behavioral effects driven by invisible stimuli, the underlying neural machinery remains largely unknown. Previous attempts at localizing subliminal, especially high-level, signals suffered from low signal-to-noise ratios due to weak raw signals and/or active suppression. In this study, we implemented dynamic faces, a novel variant of interocular suppression (discontinuous flash suppression), as well as regions-of-interest (ROI) analyses to enhance the stimulus and statistical power. A total of forty-three participants was included based on a pilot study with n = 10. At the whole-brain level, our univariate analysis first replicated a typical widespread activation by conscious faces, which was absent by unconscious faces. However, with multivariate analysis, the classification was successful in the occipital-temporal regions between static/dynamic faces and their counterpart scenes. ROI analysis further showed that bilateral FFA and OFA successfully differentiated unconscious dynamic faces and scenes. Together, these results indicate in-depth processing of unconscious faces in the ventral visual stream. Our successful localization of unconscious high-level signals highlights the importance of the selection of stimulus, paradigm, and analysis.

Acknowledgements: The Ministry of Education (MOE), Yushan Young Scholar Program (NTU-110V0202), Ministry of Science and Technology (109-2410-H-002-004-MY3), National Taiwan University (111L9A00701), and Waseda Grant Program for Promotion of International Joint Research.