Central luminance flicker can activate peripheral retinotopic representation (Study III)

Originally study III aimed at localization of cortical regions that are more sensitive to luminance flicker than to other visual stimuli. Previous results had been suggested that human V6 is particularly sensitive to luminance (Portin et al., 1998; Dechent and Frahm, 2003). Retinotopic mapping and second control experiment were included to confirm that possible luminance sensitivity arises from a separate area.

5.3.1. Methods

Two visual stimulus systems were designed for study III. First, we aimed to maximise the contrast between the stimulus and the surround and to exclude all light reflections from the stimulus surround. This is the desired setup in studying a response to luminance stimuli, where one wants to exclude aberrant indirect stimulation of the retina. The subjects wore a black matte mask which covered the peripheral parts of the visual field, and the interior of the head coil was covered with the same cloth to diminish light reflections. In the second setup, the aim was to minimise the contrast of light scattered inside the eye. The subjects wore a white mask which was illuminated with bright light. Light was directed inside the magnet bore with two bundles of optic fibers, one for each eye. The idea to illuminate stimulus surround was adopted from the studies of cortical blindness (Barbur et al., 1994;

Weiskrantz et al., 1995). The parameters for the visual stimulus viewed through a 30-deg aperture inside the mask was similar in both measurements and it consisted of luminance flicker (black = 0.4 cd/m², white = 25 cd/m², flickering at 4 Hz), checkerboard pattern

reversal (reversing at 8 Hz, no change in the mean luminance), and rest (a fixation point with a grey background 11 cd/m²) blocks in interleaved order. The data were analysed with SPM2 and BALC. Both pattern reversal and luminance flicker blocks were contrasted against rest and the results were compared both in group and individual level and in 3D and on cortical surface. Responses on cortical surface were localised in relation to retinotopic visual areas.

Figure 8. The results of one representative subject projected on his left medial occipital surface. The responses to pattern reversal stimulus (pattern-rest, pFWE < 0.05) and the responses to luminance flicker that extend beyond the pattern responses [(luminance-rest, pFWE < 0.05) excluding (pattern-rest, p<0.05)] with both dark and illuminated stimulus surround. The borders between functional areas have been defined according to the results of study II.

5.3.2. Results

Figure 8 visualises the results of one representative subject on his medial cortical surface.

Retinotopic areas V1-V3 responded strongly to the pattern reversal stimulus and luminance flicker activated cortex outside the pattern-related responses only when the stimulus

surround is dark. This additional luminance response was located in the peripheral visual field representations of the areas V1-V3. Figure 9 shows the signal changes related to all stimulus conditions at the different eccentricities of V1. Responses to pattern reversal were stronger in stimulus representations, but the responses to luminance flicker exceeded the pattern responses in the periphery. Analysis of all data confirmed the results of the

representative subject. Both individual and group level results showed that the responses to luminance flicker with dark stimulus surround reached further anteriorly than the responses to pattern reversal. The mapping of retinotopic areas up to 50 degrees of eccentricity confirmed that these additional luminance responses were located mainly in the peripheral visual field representation of V1 and some activation was also detected in the peripheral visual field representations of V2 and V3. When the illumination around the stimulus was increased in the second experiment, these peripheral responses to luminance flicker

disappeared whereas the responses to pattern reversal stimulus behaved similarly in both conditions.

Figure 9. The perceptual signal changes in left V1 of the same subject as presented in Figure 8. The lines represent signal increases related patter-rest and luminance-rest contrasts during both dark and illuminated stimulus periphery. I marked regions-of-interest on V1 surface and calculated the mean signals within the ROIs. I delineated V1 according to the results of study II and placed the ROIs at regular intervals.

5.3.3. Discussion

Against a previous proposal (Dechent and Frahm, 2003), our results showed that luminance responses in PO sulcus originate from the peripheral visual field representations in visual areas V1, V2, and V3 and not from a separate luminance sensitive functional area.

Moreover, we did not find any cortical region more sensitive to luminance flicker than to pattern reversal stimuli. Thus a central luminance flicker stimulus cannot be used for

localisation of human V6 with fMRI. We suggest that the luminance responses in peripheral representations result from intraocular scattering of light even though light reflection may have a minor contribution to our responses. When the mean luminance difference between the stimulus and the surround of the stimulus is large, scattered light has high contrast and it

can stimulate the retina outside the region which optically corresponds to the stimulus. The intensity of stray light decreases with inverse square relationship with increasing angular distance from the light source (Vos 2003). Because low contrast gives relatively high signal in human V1 (Boynton et al., 1996), possibly enhanced by weighting of magnocellular processing stream in the periphery, scattered luminance contrast could result in spread of the response with linear amplitude decay as a function of distance detected in our study.

5.4. Peripheral visual field representation activates during saccades in darkness (Study