Multimodal detection of concealed information in rapid serial visual presentation using EEG and pupillometry

Concealed Information Tests (CITs) aim to detect recognition of critical information that individuals attempt to hide, typically using physiological markers. Although previous CIT studies have combined several autonomic indicators such as heart rate and pupil change, it is still unclear whether adding neural measures to pupillometry can capture complementary aspects of recognition and attentional control. To investigate this, we examined whether electroencephalography (EEG) and pupillometry jointly provide reliable markers of concealed information in a rapid serial visual presentation (RSVP) paradigm, which constrains strategic processing and emphasizes rapid recognition. Participants viewed streams of faces containing a critical probe-either a personally familiar (parent's) face or an unfamiliar face-while performing a sex-based target detection task. EEG and pupil size were recorded concurrently, focusing on four measures: ERP amplitude at Pz, theta-band (4-8 Hz) power, absolute pupil size, and pupil size change. At the group level, all measures differentiated probes from controls, reflecting attentional orienting and recognition. At the individual level, detection varied across modalities. To integrate evidence, we applied Fisher's method to combine p-values across measures. Although overall detection did not substantially improve, combined inference compensated for limitations of single measures.