Abstract
Recent studies of middle- and long-wavelength-sensitive cone responses have revealed neural activities from single-opponent neurons in the human visual cortex (Nunez et al., 2022; Qiao, Gegenfurtner & Chen, VSS 2022). However, it is still uncertain whether single-opponent activities driven by short-wavelength-sensitive cones (S-cones) can be observed in humans. Previous studies conducted on non-human primates have shown limited S-cone-driven single-opponent neurons (Lennie et al., 1990; Johnson et al., 2004; Solomon and Lennie, 2005). In human observers, Nunez et al. (2022) reported very weak or no S-cone-driven single-opponent signals using visual evoked potentials (VEPs), which may be due to a lack of single-opponent neurons in the human visual cortex or because VEP recordings are too noisy to detect weak signals. To address this issue, we conducted a study recording steady-state visual evoked potentials (SSVEPs) to S-cone stimuli flickering at 2 temporal frequencies (3Hz and 15Hz) and 6 spatial frequencies (0.2, 1, 2, 3, 4, and 8 c/deg) in 16 observers. Our results revealed that the response at 3 Hz has band-pass tuning (double-opponent) and the response at 15 Hz has low-pass tuning (single-opponent). We further investigated spatial tuning between 3Hz and 15Hz by measuring SSVEPs on 10 observers with S-cone stimuli at several temporal frequencies (3 Hz, 6 Hz, 10 Hz, and 15 Hz). The result indicates that spatial tuning is band-pass at low temporal frequencies (3, 6, 10 Hz) and low-pass at 15 Hz. Overall, our study suggests that there are S-cone-driven single-opponent neurons in the human visual cortex.