Abstract
Previous neurophysiological studies have suggested that visual response fields in the gaze control system primarily utilize an eye-centered frame of reference at both the single unit and population level (Sajad et al. 2015; Sadeh et al. 2016). We recently confirmed that visual responses to saccade targets in the monkey frontal (FEF) and supplementary eye fields (SEF) showed a preference for eye-centered coding at the population level, even in the presence of a large visual landmark (Bharmauria et al. 2020, 2021). Specifically, visual activity showed less variance when plotted relative to target-in-eye coordinates (Te) as opposed to Th (head) or Ts (space) coordinates. Here, we re-examined this database at the single unit level, and used a more sensitive spatial continuum analysis in the spatially tuned visual responses of 102 FEF neurons and 43 SEF neurons. We found that, consistent with previous reports, approximately 95 % of tuned FEF neurons preferentially coded for Te. However, in the case of SEF, neurons almost equally preferred Te, Th and Ts. Our continuum analysis (of fits between these three models, i.e., Te-Ts, Te-Th and Ts-Th) showed distribution peaks near Te, Th and Ts with a sparse scattering of best fits in between these continua. These data suggest that whereas FEF retains a fairly simple egocentric visual code, SEF visual responses can show more complex levels of spatial preprocessing (at least in the presence of a visual landmark). This result seems to complement the finding where it was shown that SEF stimulation is capable of producing gaze shifts toward eye, head, or space-fixed goals (Martinez-Trujillo et al. 2004).