Short-term plasticity of the human adult visual cortex measured with 7T BOLD

Visual cortex, particularly V1, is considered to be resilient to plastic changes in adults. In particular, ocular dominance is assumed to be hard-wired after the end of the critical period. We show that short-term (2h) monocular deprivation in adult humans boosts the BOLD response to the deprived eye, changing ocular dominance of V1 vertices, consistently with homeostatic plasticity. The boost is strongest in V1, present in V2, V3 & V4 but absent in V3a and MT. Assessment of spatial frequency tuning in V1 by a population Receptive-Field technique shows that deprivation primarily boosts high spatial frequencies, consistent with a primary involvement of the parvocellular pathway. Crucially, the V1 deprivation effect correlates across participants with the perceptual increase of the deprived eye dominance assessed with binocular rivalry, suggesting a common origin. Our results demonstrate that visual cortex, particularly the ventral pathway, retains a high potential for homeostatic plasticity in the human adult.


13
To interact efficiently with the world, our brain needs to fine-tune its structure and function, 14 adapting to a continuously changing external environment. This key property of the brain, called 15 neuroplasticity, is maximal early in life, within the so called critical period (Berardi,Pizzorusso,& 16 it is thought to decline dramatically during adulthood, especially at the level of the primary sensory 18 cortices. During development, the plastic cortical response to abnormal visual experience is so 19 strong that occluding one eye for a few days induces a dramatic and permanent shift in ocular 20 dominance (the amount of V1 neurons responding to each eye) in favor of the open eye (Berardi et 21 al., 2000;Gordon & Stryker, 1996;Hubel & Wiesel, 1970;Hubel et al., 1977;Wiesel & Hubel, 22 1963), while the deprived eye becomes functionally blind or very weak, a phenomenon known as 23 amblyopia (Gordon & Stryker, 1996 & Hubel, 1963). After the closure of the critical period, V1 is thought to 25 become fundamentally hard-wired (Mitchell & Sengpiel, 2009 . The effect of short-term monocular deprivation is long-lasting, particularly for 43 chromatic equiluminant stimuli optimized for stimulation of the parvocellular pathway, for which a 44 significant boost is still observed up to 3h after the end of the short-term deprivation (Lunghi et al., 45 2013). In addition, the effect of 2h of deprivation can be retained across 6h of sleep (Menicucci,46 Lunghi, Zaccaro, Morrone, & Gemignani, 2018) and can lead to permanent visual changes in 47 amblyopic patients (Lunghi et al., 2016). All this evidence strongly suggests the involvement of a 48 plastic reorganization of visual cortical processes. 49 The boost of the deprived eye after short-term monocular deprivation is consistent with homeostatic 50 plasticity, an initial compensatory reaction of the visual system to deprivation aimed at maintaining 51 the average cortical activity constant despite the impoverished incoming visual input (G. 52 Turrigiano, 2012). Homeostatic plasticity was first reported in animal models during the 53 developmental critical period, following many days of monocular deprivation ( human perception, the macaque V1 deprivation effect is strongest when deprivation mainly affects 59 the parvocellular activity (Begum & Tso, 2016). 60 In adult humans, the neural substrates of short-term monocular deprivation effects have been 61 indirectly studied with MR spectroscopy (showing a GABA concentration change in the occipital 62 cortex, Lunghi, Emir, et al., 2015) and Visual Evoked Potentials (showing a modulation of the early 63 visual response components, Lunghi, Berchicci, et al., 2015). Here we directly measure the changes 64 in early visual cortical areas using 7T fMRI in adult humans, before and after two hours of 65 monocular deprivation. Assessing the BOLD change and its selectivity to spatial frequency with a 66 newly developed approach (conceptually similar to the population Receptive Field method, 67 Dumoulin & Wandell, 2008), we demonstrate a change of ocular drive of BOLD signals in primary 68 visual cortex, selective for the higher spatial frequencies and strongest along the ventral pathway, 69 consistent with a stronger plasticity potential of the parvocellular pathway in adulthood. 70

71
Monocular deprivation boosts V1 responses to the deprived eye and shifts BOLD ocular dominance 72 To investigate the visual modulation of BOLD signal by short term deprivation, we performed 73 ultra-high field (UHF, 7T) fMRI during the presentation of high contrast dynamic visual stimuli, 74 delivered separately to the two eyes, before and after 2h of monocular contrast deprivation (see 75 schematic diagram in Fig. 1A). 76 The reliability and high signal-to-noise ratio of our system allow us to obtain significant activations 77 of all the main visual areas, with only two blocks of stimulation ( We measured the plasticity effect by comparing activity before/after deprivation in response to 82 stimulation in the two eyes with low-and high-spatial frequency bandpass stimuli that differentially 83 stimulate the magno-and parvocellular pathways. Consistent with prior evidence suggesting higher 84 susceptibility to plasticity of the parvocellular pathway (Lunghi, Berchicci, et al., 2015;Lunghi et 85 al., 2011;Lunghi, Emir, et al., 2015;Lunghi & Sale, 2015), we observe a strong effect of 86 Monocular Deprivation on BOLD responses to stimuli of high spatial frequency (peak 2.7 cycles 87 per degree, high-frequency cut-off at half-height 7.5 cpd). Fig. 1D shows that the V1 response to 88 the high spatial frequency stimuli presented in the left and right eye is nearly equal before 89 deprivation ("PRE"). However, after deprivation ("POST"), the response in the two eye changes in 90 opposite directions, with a boost of the BOLD response (measured as GLM Beta values, expressed 91 in units of % signal change) of the deprived eye and a suppression of the non-deprived eye (see also 92 supplementary Fig. S1). This was formally tested with a two-way repeated measure ANOVA, 93  Ugurbil, 2007)). Before deprivation, the Ocular Dominance index is symmetrically distributed 105 around zero, indicating a balanced representation of the two eyes before deprivation (yellow 106 distribution in Fig.1E). After deprivation (black distribution in Fig.1E), the Ocular Dominance 107 distribution shifts to the right of 0, indicating a preference for the deprived eye (non-parametric 108 Wilcoxon sign-rank test comparing the PRE and POST Ocular Dominance medians, z = 115.39, p < 109 0.001). 110 In principle, the boost of responses to the deprived eye seen in Fig. 1D could be produced by 111 enhancing the response of vertices that originally preferred the deprived eye (without shifting ocular 112 dominance) or by changing Ocular Dominance of vertices that originally preferred the non-deprived 113 eye, driving them to prefer the deprived eye. The shift of the Ocular Dominance histogram in Fig.  114 1E is more compatible with the latter case, implying a recruitment of cortical resources for the 115 representation of the deprived eye. To investigate this further, we monitored the final POST-116 deprivation Ocular Dominance of individual vertices that, PRE-deprivation, preferred the deprived 117 eye (yellow half distribution in Fig 2A). The majority of vertices continue to prefer the same eye 118 before and after deprivation. The median Ocular Dominance is significantly larger than 0 both PRE 119 and POST (Wilcoxon sign-rank test, z > 101.54, p < 0.0001 in both cases) and the correlation 120 between Ocular Dominance indices before and after deprivation is strong and positive (Pearson's 121 R(32236) = 0.22 [0.21-0.23], p < 0.0001). Note that a completely random reassignment of Ocular 122 Dominance after deprivation would have produced a histogram centered at 0 and no correlation 123 between Ocular Dominance indices PRE-and POST deprivation. This is not consistent with the 124 results of Fig. 2B, which thereby provide evidence that our estimates of Ocular Dominance before 125 and after deprivation are congruent, even though they were collected in different fMRI sessions 126 separated by 2h. In addition, the distribution of Ocular Dominance after deprivation is well 127 predicted by adding only a small amount of noise to the original half distribution (Gaussian noise 128 with 0.12 standard deviation, black line), suggesting that these vertices were largely unaffected by 129 monocular deprivation. This is also supported by the repeated measure ANOVA of individual 130 subject data ( Fig. 2A), revealing a strong main effect of eye (F(1,18) = 48.28901, p < 10 -5 ): the 131 response to the deprived eye is stronger than the non-deprived eye, both before deprivation (due the 132 selection, t(18) = -8.616, p < 10 -5 ), and after deprivation (t(18) = -4.281, p < 10 -5 ), with no effect of short term monocular deprivation. This is confirmed with the repeated measure ANOVA (Fig. 2C), 144 where the time × eye interaction is significant (F(1,18) = 44.82812, p < 10 -5 ), implying a different 145 modulation PRE and POST deprivation. In addition and crucially, POST-deprivation BOLD 146 responses to the deprived eye are significantly larger than POST-deprivation responses to the non-147 deprived eye (t(18) = -2.775 p = 0.012; whereas, by selection, the opposite is true before 148 deprivation: t(18) = 12.034, p < 10 -5 ). 149 In summary, Ocular Dominance before deprivation defines two similarly sized sub-regions of V1 152 vertices (44.58 ± 5.38% and 55.42 ± 5.38% of analyzed V1 vertices; 44.84 ± 5.12% and 55.16 ± 153 5.12% of all V1 vertices) with radically different behaviors that are not consistent with an artifact 154 induced by vertex selection. The sub-region that originally represents the deprived eye does not 155 change with deprivation; the sub-region that originally represents the other non-deprived is 156 rearranged with deprivation, as a large portion of vertices turn to prefer the deprived eye. 157 If plasticity were not eye-specific and/or we failed to match our V1 vertices before/after 158 deprivation, we would expect that splitting the distribution of V1 ocular dominance generates 159 opposite effects in the two subpopulations: vertices preferring the deprived eye before deprivation 160 should swap to prefer the other eye, mirroring the effect seen in the vertices preferring non-deprived 161 eye. This is not seen, implying that we did successfully match vertices across the 2h of deprivation 162 and that the selective Ocular Dominance shift, observed for about half of our vertices, is not an 163 artifact. 164 We also measured the perceptual effects of short-term monocular deprivation affects using i.e. the expected effect; a value less than 1 indicates the opposite effect and a value of 1 indicates no 176 change of mean phase duration across eyes. All but two subjects have values larger than 1, 177 indicating a strong effect of deprivation. However, the scatter is large with values ranging from 0.7 178 to 3, suggesting that susceptibility to visual plasticity varies largely in our pool of participants. 179 Capitalizing on this variability, we tested whether the size of the psychophysical effect correlates 180 with the BOLD effect across participants. Using the same Eq. 6 to compute the deprivation effect 181 on BOLD responses (DIBOLD), we observed a strong correlation between the effect of monocular 182 deprivation on psychophysics and BOLD (shown in Fig. 3B). Subjects who showed a strong 183 deprivation effect at psychophysics (DIpsycho > 2) also showed a strong deprivation effect in BOLD 184 responses (DIBOLD = 1.85 ± 0.42). Given that the psychophysics was measured only for central 185 vision and at 2 cpd stationary grating, whereas BOLD responses were pooled across a large portion 186 on V1 and were elicited using broadband dynamic stimuli, the correlation suggests that the 187 psychophysical effect may be used as a reliable proxy of a general change of cortical excitability, 188 which can be measured by fMRI. 189 Thus, monocular deprivation produces a change of the spatial frequency selectivity of the V1 204 BOLD response. Before deprivation, the BOLD response shows a broad band-pass selectivity for 205 our stimuli, with a preference for the stimulus peaking around 1 cpd, and a slight attenuation at 206 higher spatial frequencies, similar for the two eyes (Fig. 4A). After deprivation (Fig. 4B), the non-207 deprived eye shows similar selectivity and an overall decrease of responses. For the deprived eye, 208 the shape of the curve changes: from band-pass to high-pass, implying that the enhancement affects 209 primarily the higher spatial frequencies. 210 To model this effect, we assume that each vertex on the cortical surface subtends a multitude of 211 neuronal channels, each with narrow tuning for spatial frequency and collectively spanning a large 212 range of spatial frequencies -an approach conceptually similar to the population Receptive Field 213 model for retinotopic mapping (Dumoulin & Wandell, 2008). Independently of the exact bandwidth 214 and peak preference of the neuronal population contributing to the final BOLD selectivity, we find 215 that the shape of all these curves is captured with a simple one-parameter model: what we term the 216 population tuning for Spatial Frequency. This is given by a Difference-of-Gaussians (DoG) function 217 with one free parameter, the spatial constant (while the excitation/inhibition spatial constant ratio is 218 fixed; see eq. 4 in the Methods and curves in Supplementary Fig. S4). The free parameter sets the 219 high spatial frequency cut-off at half-height of the filter. The continuous lines in Fig. 4 show how 220 the model fits the grand-average of V1 responses, with best fit cut-off around 5 cpd similar for all 221 conditions except for the POST-deprivation deprived eye, where the cut-off is 6.  shows that this behavior is systematically observed across V1 vertices, but only for the deprived 247 eye. Here the average cut-off is plotted as function of eccentricity, and the roll-off is consistent with 248 the map in Fig. 5A. For the non-deprived eye, there is no effect of deprivation on spatial frequency 249 selectivity (Fig. 5C). In contrast, for the deprived eye (Fig. 5D), there is a shift towards preferring 250 higher spatial frequencies, at all eccentricities, which is captured by an increased value of the cut-251 off frequency parameter leading to an increased acuity of the BOLD response to the deprived eye. 252 Note that the change of spatial frequency selectivity for the deprived eye is most evident at 253 eccentricities of 4 deg and higher (see Fig. 5D), where vertices have peak sensitivity at mid-to-low 254 spatial frequencies before deprivation. In the fovea, where many vertices already prefer the highest 255 spatial frequency stimulus before deprivation, our fitting procedure is likely to underestimate the 256 change of spatial frequency selectivity. Importantly, the spatial frequency selectivity for the non-257 deprived eye is unchanged at all eccentricities, corroborating the eye and stimulus-specificity of the 258 short-term monocular deprivation effect. 259 Given that the time × eye interaction in the full V1 region is not significant, and to minimize noise 272 contamination, we evaluated the effect of deprivation on spatial frequency cut-off at the individual 273 level by a "Deprived Eye Change (DepCcutoff)" index (Eq.7 in the methods), i.e. taking the POST vs. 274 PRE-deprivation ratio of the spatial frequency cut-off for the deprived eye alone. As this ratio 275 varies widely across participants, over more than 3 octaves, we asked whether this variability 276 correlates with our psychophysical probe of plasticity: binocular rivalry. We used the same Eq. 7 to 277 index the psychophysical change of the deprived eye (DepCpsycho), the POST to PRE-ratio of mean 278 phase duration for the deprived eye, and found a strong positive correlation (Fig. 6B).  deprivation, the deprived eye shows an increase of mean phase duration (in binocular rivalry) and 280 an increase of the spatial frequency cut-off (best fit of the BOLD responses): participants showing a 281 stronger increase of phase duration, also showed a larger shift of selectivity towards higher spatial 282 frequency. The correlation is consistent with the result of Fig. 3 showing that the enhancement of 283 BOLD responses is correlated with the change of binocular rivalry and selective for the highest 284 spatial frequency stimulus. 7E). The boost is present also in V3 and V4. In V4 the boost appears to be present also for lower 290 spatial frequencies, but again only for the deprived eye ( Fig. 7A-B), possibly reflecting the larger 291 spatial frequency bandwidth of V4 neurons compared to V1. 292 This result suggests a preferential involvement of the parvocellular vs. magnocellular pathway, 306 leading to the differential plasticity effect in extra-striate visual areas of the ventral and dorsal 307 pathway. Interestingly, the plasticity effect is robust in areas where the majority of cells are 308 binocular (like V3 and V4), indicating that the effect does not require segregated representations of 309 the two eyes (e.g. ocular dominance columns). 310

311
We demonstrate that two hours of abnormal visual experience has a profound impact on the neural 312 sensitivity and selectivity of V1. BOLD activity across the V1 cortical region paradoxically 313 increases for the eye that was deprived of contrast vision, with the strongest boost for the higher 314 spatial frequency stimuli, and decreases for the eye exposed to normal visual experience. We also observe an antagonistic suppression of the non-deprived eye BOLD response; together, the 327 two effects lead to a shift of ocular preference of individual vertices in favor of the deprived eye. 328 However, this effect is only observed in those V1 vertices that, before deprivation, responded 329 preferentially to the non-deprived eye. No such change of ocular preference is seen in vertices that 330 already prefer the deprived eye before deprivation, which maintain their eye-preference after 331 deprivation. This pattern of results cannot be explained by any overall gain increase; rather, it is 332 consistent with the idea that the representation of the deprived eye recruits cortical resources (which 333 may or may not correspond to cortical territory) normally dedicated to the other eye. 334 A similar antagonist effect on the two eyes (boosting the deprived eye and suppressing the non-335 deprived eye) was also observed in the VEP responses after short-term monocular deprivation 336 (Lunghi, Berchicci, et al., 2015) and could be implemented through a modulation of the 337 excitatory/inhibitory circuitry. Regulation of the excitation/inhibition balance through GABAergic 338 signaling is considered to be a key factor for cortical plasticity, including homeostatic plasticity 339 (Maffei & Turrigiano, 2008). Interestingly, the involvement of GABAergic signaling in the effect 340 of short-term monocular deprivation is directly supported by MR Spectroscopy data in adult 341 humans, showing that resting GABA in a large region of the occipital cortex is specifically reduced 342 after short-term monocular deprivation (Lunghi, Emir, et al., 2015). While ventral area V4 still shows a strong plasticity effect, area V3a, located at a similar tier in the 393 dorsal stream, shows no modulation with short-term monocular deprivation. V4 is a primary target 394 of the parvocellular system, which is best stimulated by our highest spatial frequency stimulus; V3a 395 and MT are preferential targets of the magnocellular system, which respond more strongly to our 396 lower spatial frequency stimuli (see Fig. 7). The different plasticity response of the ventral and 397 dorsal stream, together with the selectivity for the high spatial frequencies of the V1 plasticity, 398 suggests that the parvocellular pathway is most strongly affected by short-term plasticity. This fits 399 well with several sources indicating that selective deprivation of the stimuli that optimally drive the 400 parvocellular system is sufficient to produce a reliable plasticity effect (Begum & Tso, 2016). Twenty healthy volunteers with normal or corrected-to-normal visual acuity were examined (8 423 females and 12 males, mean age = 27 years) after giving written informed consent. 424

Experimental design 426
Each participant underwent two scanning sessions separated by two hours, during which they were 427 subject to the short-term monocular deprivation procedure described below. Just before each 428 scanning section, their binocular rivalry was measured psychophysically. One (male) participant 429 was excluded because of strong eye dominance tested with binocular rivalry before the deprivation. 430 This left 19 participants (8 females and 11 males) whose complete datasets were entered all 431 analyses. Sample size was set to enable testing for correlations between neuroimaging and 432 psychophysical data. Previous work (Lunghi, Emir, et al., 2015) reveals a correlation between MR 433 spectroscopy data and binocular rivalry measures r = 0.62 (or higher), which implies a minimum of 434 17 participants to detect a significant correlation at 0.05 significance level, with test power of 435 80% (Lachin, 1981). light to reach the retina (attenuation 0.07 logUnits, at least 3 times smaller than the threshold for 443 discriminating a full-field luminance decrement (Knau, 2000) and more than ten times smaller than 444 the minimum photopic luminance decrement required for shifting the spatial (Van Nes & Bouman, 445 1967) or temporal contrast sensitivity function (Kelly, 1961)). The patch prevents pattern vision, as 446 assessed by the Fourier transform of a natural world image seen through the eye-patch. During the 2 447 hours of monocular deprivation, observers were free to read and use a computer. where P is the peak spatial frequency, q is the filter half-width at half maximum in octaves. We 475 generated five band-pass noise stimuli, by setting q = 1.25 octaves and P = 0.1 cpd, 0.2 cpd, 0.4 476 cpd, 1.1 cpd, 2.7 cpd. Each stimulus was presented for a block of 3TRs, during which the image 477 was refreshed at 8Hz (randomly resampling a 800 × 600 window from the original matrix). Stimuli 478 were scaled to exploit the luminance range of the display, and this yielded very similar RMS 479 contrast values (shown in supplementary Fig. S2). Stimulus blocks were separated by 4TRs blanks, 480 consisting of a mid-level gray screen. The five band-pass noise stimuli blocks were presented in 481 pseudo-random order, twice per run, for a total of 70 TRs. In each run, stimuli were only presented 482 to one eye, while the other was shown a mid-level gray screen. Each eye was tested once, before 483 and after deprivation. 484 Immediately upon application of the monocular patch, we performed two additional scans to 485 perform retinotopic mapping of visual areas. Meridian and ring stimuli were presented monocularly 486 (to the non-patched eye) and were defined as apertures of a mid-level gray mask that uncovered a 487 checkerboard pattern, 1 deg at 1 deg eccentricity to 2.5 deg at 9 deg eccentricity, rotating and 488 contracting at a rate of one check per second. Meridians were defined by two 45° wedges centered 489 around 0° or around 90°. The horizontal and vertical meridian were presented interchangeably for 5 490 TRs each (without blanks) and the sequence was repeated 6 times for a total of 60 TRs. Rings 491 partitioned screen space into six contiguous eccentricity bands (0-0.9 deg, 0.9-1.8 deg, 1.8-3.3 deg, 492 3.3-4.7 deg, 4.7-6.48 deg, 6.48-9 deg). Odd and even rings were presented in two separate runs. In 493 each run, the three selected rings and one blank were presented in random order for 5 TRs each, and 494 the sequence was repeated (with different order) 6 times for a total of 120 TRs. 495

MR system and sequences 496
Scanning was performed on a Discovery MR950 7 T whole body MRI system (GE Healthcare,  Anatomical images were corrected for intensity bias using SPM12 (Friston, 2007) and processed by 524 a standard procedure for segmentation implemented in Freesurfer (recon-all: Fischl et al., 2002). In 525 addition, each hemisphere was aligned to a left/right symmetric template hemisphere 526 (fsaverage_sym: Greve et al., 2013). 527 Functional images were corrected for subject movements (Goebel, Esposito, & Formisano, 2006) 528 and undistorted using EPI images with reversed phase encoding direction (Brain Voyager COPE 529 plug-in Jezzard & Balaban, 1995). We then exported the preprocessed images from BrainVoyager 530 to NiFTi format. These were aligned to each participant's anatomical image using a boundary based 531 registration algorithm (Freesurfer bbergister function) and projected to the cortical surface of each 532 hemisphere. All analyses were conducted on data in the individual subject space. In addition, for 533 visualization purposes, we also aligned the results of timecourse analyses (GLM and subsequent 534 pRF and spatial frequency tuning estimates) to the left/right symmetric template hemisphere. 535 Averaged results across the 18x2 hemispheres are shown in the maps of Fig. 1B, Fig. 5A  eq. 2 545 with parameters n=3, t=1.5 s, and d=2.25 s (Boynton, Engel, Glover, & Heeger, 1996). Beta 546 weights of the stimuli predictors were taken as estimates of the BOLD response amplitude and 547 normalized by the predictor amplitude to obtain a measure that directly corresponds to % signal 548 change; beta weights were also scaled by an error measure to obtain t-values, following the same 549 procedure as in (Friston et al., 1994). Computing BOLD responses for each individual vertex of the 550 cortical surface leads to up-sampling the functional data (each 1.5 x 1.5 x 1.5 mm functional voxel 551 projecting on an average of 3 vertices). We ensured that this does not affect our statistical analyses 552 by first averaging data from all vertices within a region of interest (e.g. V1), thereby entering all 553 ANOVAs with a single value per subject and region of interest. 554 555

Population Receptive Field mapping 556
The pRFs of the selected voxels were estimated with custom software in Matlab, implementing a 557 method related to that described by Dumoulin and Wandell (Dumoulin & Wandell, 2008). We 558 modeled the pRF with a 1D Gaussian function defined over eccentricity, with parameters D and E as 559 mean and standard deviation respectively, and representing the aggregate receptive field of all 560 neurons imaged within the vertex area. We defined the stimulus as a binary matrix S representing 561 the presence of visual stimulation over space (here, eccentricity between 0 and 10 deg with 40 steps 562 per deg) for each of 6 ring stimuli. We used the results of our GLM analysis to estimate the vertex 563 response to each of our 6 rings (as t-values; using beta values yields very similar results). We 564 assumed that each vertex response is the linear sum over space (eccentricity) of the overlap between 565 the pRF of the voxel and the input stimulus, which is mathematically equivalent to the matrix 566 multiplication between the stimulus and the pRF. 567 where K is the index to ring number and varies between 1 and 6. 569 We used this equation to predict the response to our six rings for a large set of initial pRF 570 parameters; for each vertex, we measured the correlation (our goodness-of-fit index) between the 571 predicted response and the observed t-values. If the highest correlation was < .7 the vertex was 572 discarded; otherwise, the parameters yielding the highest correlation were used to initialize a 573 nonlinear search procedure (MATLAB simplex algorithm), which manipulated D and E to 574 maximize goodness-of-fit, with the constraint that D could not exceed 20 deg or be smaller than 1 575 deg, and E could not be smaller than .1 deg. Successful fits were obtained for 72.00 ± 1.86% of V1 576 vertices, for which the initial coarse grid search gave a correlation > 0.7 and the nonlinear search 577 settled within the constraints. All analyses (on average and distribution of responses and tuning 578 parameters) considered the sub-region of V1 for which a successful fit was obtained. We used D to 579 estimate the preferred eccentricity of each vertex. 580 The main modifications of our procedure relative to that described by Dumoulin and Wandell 581 (Dumoulin & Wandell, 2008) are the following: (a) fMRI data were acquired in a block design with 582 only six stimulus types (six eccentricity bands) rather than varying stimulus position at each TR; 583 this allowed us to use a standard GLM approach to estimate each vertex response to the six stimuli 584 (assuming a standard hemodynamic response function) and then use the pRF model to predict these 585 six time-points -much faster than predicting the full fMRI series of 120x2 TRs; (b) our stimuli and 586 consequently our pRFs were defined in one dimension (eccentricity) -whereas the standard pRF is 587 defined in 2D, eccentricity and polar angle (or Cartesian x and y); (c) we maximized the correlation 588 between the predicted and observed fMRI response time-courses rather than minimizing the root 589 mean square error; this eliminates the need to estimate a scale factor to account for the unknown 590 units of the BOLD signal. 591 Population Tuning for Spatial Frequency 592 Using a similar logic, we also estimated the population tuning for Spatial Frequency, which 593 represents the aggregate Spatial Frequency tuning of the population of neurons imaged within each 594 vertex area. We modeled the population tuning using a family of functions that includes the 595 psychophysical Contrast Sensitivity Function (CSF) and can be specified by the following one-596 parameter equation (Difference-of-Gaussians): 597 Like we did for the pRF mapping, we defined a stimulus matrix S representing the Fourier spectra 599 of our five bandpass noise stimuli, i.e. the energy of visual stimulation in the frequency domain 600 (here, between 0.03 cpd and 12.5 cpd) for each stimulus. We used the results of our GLM analysis 601 to estimate the vertex response to each of our five bandpass noise stimuli (as t-values; using beta 602 values yields very similar results). We assumed that each vertex response is the linear sum over 603 frequency of the overlap between the pSFT of the voxel and the input stimulus, which is 604 mathematically equivalent to the matrix multiplication between the stimulus and the pSFT. 605 Like for pRFs, we estimated the best-fit E parameter of each vertex pSFT with a two-step 606 procedure: a coarse-grid search followed by the simplex search. We used the matrix multiplication 607 of the pSFT and the stimulus to predict the response to our five bandpass noise stimuli for a large 608 set of initial E values (between 1 and 1,000 in 100 logarithmic steps); for each vertex, we measured 609 the correlation (our goodness-of-fit index) between the predicted response and the observed t-610 values. If the highest correlation was < .5, the voxel was discarded, otherwise the parameter 611 yielding the highest correlation were used to initialize a nonlinear search procedure (MATLAB 612 simplex algorithm), which manipulated E to maximize goodness-of-fit, with the constraint that E 613 could not be smaller than .3 and larger than 10,000. Successful fits were obtained for 88.84 ± 1.28% 614 of V1 vertices for which we obtained a successful eccentricity fit (86.77 ± 1.25% of all V1 615 vertices). 616 We express the E parameter in terms of the high-spatial frequency cutoff of the filter (highest 617 spatial frequency at half maximum), JLUV for each vertex: 618 JL UV= 1.26[ \ ] -0.045 eq. 5 619 Indices defining the effect of deprivation 620 We computed the effects of short-term monocular deprivation on both the dynamics of binocular 621 rivalry and our fMRI results, estimating the degree to which the two measures are correlated. In all 622 cases, the same equation was applied to psychophysical and fMRI data. 623 The first index, called "Deprivation Index" or DIpsycho and DIBOLD is given by eq. 6 624 bc = =