Previous studies posited the effectiveness of stimulated recall. However, few studies explored how SR is implemented in a relatively static context, for example, online self-directed learning, or took human factors, for example, cognitive style and gender, into consideration in such a context. To fill this gap, the current study, aims to introduce biofeedback as a stimulus for learners to engage in retrospection regarding their learning behavior. A quasi-experimental design study was carried out over a 12-week set of EFL self-regulated online reading activities. Pretest and posttest on reading performance and their cognitive taxonomy were assessed through a developed scale instrument, whereas physiological signals (e.g., gazing duration, verbal fixation, and brain wave) were captured via eye-tracking and electroencephalograph (EEG) technology. The results emphasized that (a) students' reading ability and cognitive hierarchy significantly improved through biofeedback stimulation. Moreover, (b) learners in single level-one cognitive hierarchic groups had significant improvements in both cognitive abilities and reading comprehension, whereas learners in multilevel hierarchic groups had no significant enhancements. Finally, (c) the optical data results and EEG reports showed that males favor procedural feedback and females have a preference for a conclusive assessment.