Simay Uner^1,2 (simayuner.nsc@gmail.com), Irem Akdogan^1,2, Hulusi Kafaligonul^1,2,3; ¹Department of Neuroscience, Aysel Sabuncu Brain Research Center, Bilkent University, Ankara, Turkiye, ²National Magnetic Resonance Research Center (UMRAM), Bilkent University, Ankara, Turkiye, ³Neuroscience and Neurotechnology Center of Excellence (NÖROM), Department of Anatomy, Faculty of Medicine, Gazi University, Ankara, Turkiye

Transcranial random noise stimulation enhances visual perception, a benefit most consistently observed with high-frequency (hf-tRNS) compared to low-frequency (lf-tRNS) protocols. Most existing work typically tests single stimulus parameters, limiting our understanding of how electrical noise influences broadband visual sensitivity. In addition, it remains unclear whether hf-tRNS and lf-tRNS modulate vision via similar mechanisms or if they induce distinct alterations in cortical tuning. To characterize frequency-dependent effects on overall visibility, we measured the foveal Contrast Sensitivity Function (CSF) in 57 participants across two experiments, using the quick CSF paradigm. tRNS was applied over the visual cortex at 1.5 mA. Participants received either lf-tRNS (n = 26) or hf-tRNS (n = 31) while performing an orientation discrimination task to estimate the full CSF. We analyzed key morphological parameters, including maximum contrast sensitivity (CSmax), peak spatial frequency (SFpeak), bandwidth, and area under the curve (AUCSF). Results indicated a dissociation between stimulation protocols, such that hf-tRNS did not uniformly boost contrast gain. It rather selectively enhanced sensitivity at mid-spatial frequencies while suppressing sensitivity at the low- and high-frequency tails, effectively narrowing the CSF bandwidth. In contrast, lf-tRNS failed to induce such morphological changes, signaling a fundamental divergence in efficacy. This sharpening effect in the hf-tRNS group was confirmed by a significant negative correlation between the changes in bandwidth and CSmax, a relationship absent in the lf-tRNS group. These findings demonstrate that hf-tRNS but not lf-tRNS alters the fundamental tuning of the visual system. Unlike the broad effects often attributed to general excitability changes, hf-tRNS appears to sharpen visual channels around mid-spatial frequencies, suggesting a stochastic resonance mechanism that selectively enhances signal-to-noise ratios in optimally tuned processing channels. We discuss potential mechanisms underlying this dissociation, suggesting that high- and low-frequency noise may interact differently with cortical processing dynamics to modulate visual sensitivity.

Acknowledgements: This work was supported by The Scientific and Technological Research Council of Türkiye (ARDEB 424K278, BIDEB 2211 Program) and the BAGEP Award of the Science Academy-Türkiye.