Temporal frequency masking interactions in V1 depend on stimulus size

Presentation of counter-phasing visual stimuli generates a steady-state-visually-evoked-potential (SSVEP), which has been widely used in electroencephalogram (EEG) studies of attention. When two stimuli are presented at different temporal frequencies (TFs), EEG shows two peaks that are modulated by attention. However, when the stimuli are spatially overlapping, they also interact in complex ways. Specifically, previous studies using invasive recordings from V1 have shown that the simultaneous presentation of full-screen gratings at different TFs leads to more attenuation when the TFs are nearby. Further, this suppression is asymmetric if gratings are parallel in orientation, with lower frequencies producing more suppression. However, these studies used full-screen gratings that did not elicit strong spiking activity, so it was unclear whether the same interactions hold for spiking activity or when small stimuli are used. To address this, we presented smaller (1.5°) gratings and plaids (two superimposed gratings) to generate robust spiking in V1 while a bonnet macaque passively fixated on the monitor. We found that (i) small gratings had SSVEP peaks at lower frequencies (~4Hz) compared to the full-screen grating (~8-16Hz), (ii) although nearby TFs produced most suppression, the asymmetry was lost, suggesting that the asymmetry could be due to long-range connections and (iii) such interactions were not readily observed in spiking activity which instead responded to the beat frequency (difference between the two TFs). Our study reveals size-dependent interactions between multiple TFs that are useful for interpreting SSVEP studies and also sheds light on how the brain represents multiple stimuli.