Oakyoon Cha¹ (); ¹Sungshin Women's University

Embedding oscillations in the decision criterion of the one-choice drift-diffusion model led to the emergence of oscillatory patterns in the response time distribution (Cha, 2024). Notably, even minimal noise in the oscillation amplitude greatly impacted these patterns, with the influence growing stronger at higher oscillation frequencies. An intuitive prediction from these results is that lower-frequency oscillations could provide better stability when used for sensory gating mechanisms. Consistent with this prediction, a study using binocular rivalry found a negative correlation between the peak alpha (7–13 Hz) frequency and stable percept durations (Katyal et al., 2019). This relationship between oscillation frequency and perceptual stability, although intuitive, has never been predicted by any model for multistable perception. To simulate binocular rivalry durations, the present study utilized the hierarchical Brownian model (HBM; Albert et al., 2017), which models perceptual evidence as continuously drifting between two boundaries representing competing percepts. I made three key modifications to this model: First, I introduced two intermediate boundaries to incorporate mixed perceptual dominance states, allowing the simulation of transitions where both percepts are partially dominant. Second, I embedded oscillations into the perceptual evidence and intermediate boundaries to reflect the rhythmic fluctuations in neural activity influencing perceptual dominance. Third, I reset the phases of these oscillations at the onset of each exclusive perceptual dominance period, mirroring the neural reinitialization that occurs with perceptual switches. This modified model confirmed the intuitive prediction that lower alpha peak frequencies lead to longer stable percept durations. Additionally, the model predicted that oscillations during mixed dominance periods would be stronger than during exclusive dominance periods, consistent with previous findings (Cha & Blake, 2019). These results suggest that lower oscillation frequencies afford advantages in achieving stable perception, enhancing our understanding of the neural mechanisms underlying perceptual stability.

Acknowledgements: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (Ministry of Science and ICT) (No. RS-2023-00211668).