Neural sources of letter and Vernier acuity

Visual acuity can be measured in many different ways, including with letters and Vernier offsets. Prior psychophysical work has suggested that the two acuities are strongly linked given that they both depend strongly on retinal eccentricity and both are similarly affected in amblyopia. Here we used high-density EEG recordings to ask whether the underlying neural sources are common as suggested by the psychophysics or distinct. To measure visual acuity for letters, we recorded evoked potentials to 3 Hz alternations between intact and scrambled text comprised of letters of varying size. To measure visual acuity for Vernier offsets, we recorded evoked potentials to 3 Hz alternations between bar gratings with and without a set of Vernier offsets. Both alternation types elicited robust activity at the 3 Hz stimulus frequency that scaled in amplitude with both letter and offset size, starting near threshold. Letter and Vernier offset responses differed in both their scalp topography and temporal dynamics. The earliest evoked responses to letters occurred on lateral occipital visual areas, predominantly over the left hemisphere. Later responses were measured at electrodes over early visual cortex, suggesting that letter structure is first extracted in second-tier extra-striate areas and that responses over early visual areas are due to feedback. Responses to Vernier offsets, by contrast, occurred first at medial occipital electrodes, with responses at later time-points being more broadly distributed—consistent with feedforward pathway mediation. The previously observed commonalities between letter and Vernier acuity may be due to common bottlenecks in early visual cortex but not because the two tasks are subserved by a common network of visual areas.


Reliable Component Analysis of 2F responses
Responses to local contrast changes manifest at 2F and the 2F response patterns across stimulus conditions are strikingly different from those for 1F for both Letter and Vernier tasks as can be seen by comparing Fig. 4 of the main paper to Fig. S1 here. For the letter targets, the earliest response manifested in RC2 which is focally distributed over the occipital pole (electrode 75) while at 1F RC1 is the earlier component and it is maximal on lateral occipital electrodes. The amplitude of the 2F RC2 response increases as a function of letter size up to a maximum at logMAR 0.84 rather than 0.6 for the RC2 response at 1F. The later, more broadly distributed 2F RC1 component is located over lateral occipital electrodes, with a maximum over right hemisphere (electrode 90), This pattern is consistent with a feed-forward pathway for 2F instead of a feedback pathway for 1F. The amplitude of RC1 does not change as a function of letter size, but it is monotonically increasing for 1F, further indicating that 1F and 2F are generated by different sources.
The Vernier RC2 at 2F is also distributed on the occipital midline, more anteriorly compared to that for letter targets, with maximum over electrodes 71 and 76 (Fig.S1). For the Vernier responses, RC2 leads RC1, which has distinct bilateral maxima at lateral occipital leads, with maxima over right electrodes (electrodes 90 and 91), consistent with a feedforward pathway. The amplitude of both RC1 and RC2 increased as a function of offset size, however, RC1 has a steeper slope.  Dynamics of the RC components of 2F. For letter targets, the response of RC2 leads the RC1 response by ~ 30 msec at the smallest letter size and by ~45 msec at the largest letter size (Fig. S2). For the Vernier targets, RC2 leads the RC1 response by ~45 msec for the smallest offset and by ~30 msec for the largest offset. The results further corroborate the feedforward nature of the 2F response.

Source estimation of letter 1F responses using MN + FACE
To estimate the sources of the letter 1F responses, we applied the MN+FACE inverse solution to the FFT coefficients of the 1F frequency bin, and averaged the FFT coefficients coherently over the source vertices within the following visual ROIs: VWFA and IOG, and early visual ROIs V1, V2, V3.
Source analysis recovered increasing current density as letter size increased, with larger responses in the left than right hemisphere in IOG and VWFA ROIs as was seen in the tuning of RC1. By contrast, Letter size tuning in the early visual ROIs V1 and V2 was similar to the tuning of RC2 (compare figure Fig. S3 to Fig. 4 in the text).

Validation of source estimations of letter responses using simulation
To further investigate whether RC1 scalp activity for intact letters vs scrambled letters at 1F was consistent with activation of IOG and/or VWFA, we simulated the scalp topography expected from the independent activation of IOG, VWFA and control ROIs, V1, V2 and V3 using EEGSourceSim 2 (ESSim.Simulate.ResolutionMatrix function). We used the BEM forward models from the 10 participants that were used in the source estimation of 1F letter responses for this simulation. Scalp activity was simulated by placing unit current dipoles, oriented perpendicular to the local cortical surface at each vertex of the individually defined ROIs. Surface topographies for group data were generated by averaging the individual participant forward projections. The similarity of the forward projections of the VWFA and IOG ROIs suggest that uniquely localizing the activity of the RC1 to one specific ROI is difficult. To quantify this, we computed cross-talk matrix using the same function from EEGSourceSim for VWFA, IOG and control ROIs. Briefly, the cross-talk matrix is computed by placing unit current dipoles at the vertices of the source ROI and then measuring current density in the target ROIs. Perfect localization would result in activity being restricted to the diagonal of the cross-talk matrix, and presence of non-diagonal elements indicates that EEG inverse solution cannot perfectly unmix the activity of the ROIs. The matrix (Fig S3c) shows strong off-diagonal elements for IOG/VWFA ROIs in the left hemisphere (Cross-talk = 0.31). This is mainly due to the ill-posed nature of the EEG inverse solution and the fact that the VWFA and IOG are located in ventral surface of the brain, which are distant from EEG electrodes, therefore their localization error is higher. Figure S3. Source estimation of letter 1F responses. A. Letter