High-Frequency SSVEP: Evidence for Task-Driven but Not for Stimulus-Driven Affective Attention

Choosing the appropriate signal parameter is paramount to producing informative and reproducible findings. In EEG-based affective neuroscience, it is useful to consider that affective processes can unfold over several seconds, which can limit the utility of event-related potentials (ERPs) that are most sensitive near the onset of a stimulus. One promising solution for probing affective attention over longer time-windows is to present stimuli in flickering mode that increases the number of stimulus 'onsets' in a unit of time, an approach known as Steady-State Visually Evoked Potentials (SSVEP). In this study (N = 44), we used a barely noticeable (and therefore less disturbing) periodic contrast modulation (42.5 Hz) to probe variation in attention towards the flickering stimuli modulated by stimulus-driven (negative vs. neutral) and task-driven (unregulated viewing vs. distraction via mental imagery) mechanisms. Time-frequency analysis based on rhythmic entrainment source separation revealed sensitivity to task-driven attentional manipulation, albeit in an unexpected direction. Surprisingly, affective valence did not modulate EEG power at the tagged frequency, diverging from previous reports based on low-frequency SSVEP. Meanwhile, the late positive potential (LPP) indicated sensitivity towards both task-driven and stimulus-driven attention, although the task-driven effect proved more local and did not generalize across time in a sliding-window robustness check. Together, these findings suggest that SSVEPs and LPPs index distinct aspects of affective attention. The potential origins of these findings are discussed, with emphasis on the involvement of eye movements and imagery-driven resource competition in the brain.