The extent to which brain structure is influenced by sensory input during development is a critical but controversial question. A paradigmatic system for studying this is the mammalian visual cortex. Maps of orientation preference (OP) and ocular dominance (OD) in the primary visual cortex of ferrets, cats and monkeys can be individually changed by altered visual input. However, the spatial relationship between OP and OD maps has appeared immutable. Using a computational model we predicted that biasing the visual input to orthogonal orientation in the two eyes should cause a shift of OP pinwheels towards the border of OD columns. We then confirmed this prediction by rearing cats wearing orthogonally oriented cylindrical lenses over each eye. Thus, the spatial relationship between OP and OD maps can be modified by visual experience, revealing a previously unknown degree of brain plasticity in response to sensory input.
‘Democracy’ has been frequently praised by the Chinese political leadership, while liberal democratic institutions have remained conspicuously absent under CPC one-party rule. This study explores t ...
We performed optical intrinsic signal imaging of cat primary visual cortex (Area 17 and 18) while delivering bipolar electrical stimulation to the retina by way of a supra-choroidal electrode array. Using a general linear model (GLM) analysis we identified statistically significant (p < 0.01) activation in a localized region of cortex following supra-threshold electrical stimulation at a single retinal locus.(1) demonstrate that intrinsic signal imaging combined with linear model analysis provides a powerful tool for assessing cortical responses to prosthetic stimulation, and (2) confirm that supra-choroidal electrical stimulation can achieve localized activation of the cortex consistent with focal activation of the retina.
Previous studies investigating the response properties of neurons in the primary visual cortex of cats and primates have shown that prolonged exposure to optimally oriented, high-contrast gratings leads to a reduction in responsiveness to subsequently presented test stimuli. We recorded from 119 neurons in cat V1 and V2 and found that in a high proportion of cells contrast adaptation also occurs for gratings oriented orthogonal to a neuron's preferred orientation, even though this stimulus did not elicit significant increases in spiking activity. Approximately 20% of neurons adapted equally to all orientations tested and a further 46% showed at least some adaptation to orthogonally oriented gratings, whereas 20% of neurons did not adapt to orthogonal gratings. The magnitude of contrast adaptation was positively correlated with adapting contrast, but was not related to the spiking activity of the cells. Highly direction selective neurons produced stronger adaptation to orthogonally oriented gratings than other neurons. Orientation-related adaptation was correlated with the rate of change of orientation tuning in consecutive cells along electrode penetrations that traveled parallel to the cortical layers. Nonoriented adaptation was most common in areas where orientation preference changed rapidly, whereas orientation-selective adaptation was most common in areas where orientation preference changed slowly. A minority of neurons did not show contrast adaptation (14%). No major differences were found between units in different cortical layers, V1 and V2, or between complex and simple cells. The relevance of these findings to the current understanding of adaptation within the context of orientation column architecture is discussed.
Adaptation is a ubiquitous property of the visual system. Adaptation often improves the ability to discriminate between stimuli and increases the operating range of the system, but is also associated with a reduced ability to veridically code stimulus attributes. Adaptation to luminance levels, contrast, orientation, direction and spatial frequency has been studied extensively, but knowledge about adaptation to image speed is less well understood. Here we examined how the speed tuning of neurons in cat primary visual cortex was altered after adaptation to speeds that were slow, optimal, or fast relative to each neuron's speed response function. We found that the preferred speed (defined as the speed eliciting the peak firing rate) of the neurons following adaptation was dependent on the speed at which they were adapted. At the population level cells showed decreases in preferred speed following adaptation to speeds at or above the non-adapted speed, but the preferred speed did not change following adaptation to speeds lower than the non-adapted peak. Almost all cells showed response gain control (reductions in absolute firing capacity) following speed adaptation. We also investigated the speed dependence of contrast adaptation and found that most cells showed contrast gain control (rightward shifts of their contrast response functions) and response gain control following adaptation at any speed. We conclude that contrast adaptation may produce the response gain control associated with speed adaptation, but shifts in preferred speed require an additional level of processing beyond contrast adaptation. A simple model is presented that is able to capture most of the findings.
Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods References Decision letter Author response Article and author information Metrics Abstract The extent to which brain structure is influenced by sensory input during development is a critical but controversial question. A paradigmatic system for studying this is the mammalian visual cortex. Maps of orientation preference (OP) and ocular dominance (OD) in the primary visual cortex of ferrets, cats and monkeys can be individually changed by altered visual input. However, the spatial relationship between OP and OD maps has appeared immutable. Using a computational model we predicted that biasing the visual input to orthogonal orientation in the two eyes should cause a shift of OP pinwheels towards the border of OD columns. We then confirmed this prediction by rearing cats wearing orthogonally oriented cylindrical lenses over each eye. Thus, the spatial relationship between OP and OD maps can be modified by visual experience, revealing a previously unknown degree of brain plasticity in response to sensory input. https://doi.org/10.7554/eLife.13911.001 eLife digest The structure of the brain results from a combination of nature (genes) and nurture (environment). The brain’s ability to adapt to changes in the environment is known as plasticity, and the young brain is especially plastic. An animal’s sensory experiences in early life help to determine how its brain will process sensory input as an adult. One of the best sensory systems in which to study this process is the visual system. Within the visual system, some brain cells respond only to input from the left eye and others only to input from the right eye. Cells that respond to input from the same eye are arranged to form columns. Within each column, some cells respond only to lines with a particular orientation. Cells with different preferred orientations are grouped together in patterns that resemble pinwheels. The relative positions of the pinwheels and eye-specific columns within the brain tissue belonging to the visual system have so far been robust to changes in visual experience during development, suggesting that they are determined by an animal’s genes. However, Cloherty, Hughes et al. have now tested the unexpected predictions of a computer model. The model suggested that rearing animals so that they saw mostly vertical lines through one eye, and mostly horizontal lines through the other, would cause a form of plasticity that had never been observed before. Specifically, it would change the relative positions of the pinwheels and eye-specific columns within the visual parts of the brain. This prediction turned out to be correct. Young cats that wore special lenses – which slightly distorted what they saw but did not obviously affect their behavior – showed the predicted changes in brain structure. The results confirm that this aspect of brain structure is partly determined by nurture, as opposed to being entirely specified by nature. A key future challenge is to identify the chemical signaling that enables sensory input to have these effects on brain structure. It might then be possible to use drugs to restore normal brain activity in cases where abnormal sensory input has altered the brain, for example in the condition known as amblyopia (or “lazy eye”). https://doi.org/10.7554/eLife.13911.002 Introduction In cats and monkeys neurons in the primary visual cortices are selective for both the orientation of the visual input (orientation preference, OP) and its eye of origin (ocular dominance, OD) (Hubel and Wiesel, 1977). These feature preferences are arranged spatially in the form of OP and OD maps, with stereotypical structure within each map, and strong spatial relationships between them (Blasdel and Salama, 1986; Bonhoeffer and Grinvald, 1991; Bartfeld and Grinvald, 1992; Obermayer and Blasdel, 1993; Hübener et al., 1997; Nauhaus et al., 2012). While some aspects of the structure of OD and OP maps individually are plastic in response to altered visual input, such as monocular deprivation (Hubel et al., 1977; Shatz and Stryker, 1978; Farley et al., 2007) or stripe rearing (Sengpiel et al., 1999; Tanaka et al., 2006), none of these manipulations has succeeded in modifying the overall spatial relationships between OD and OP maps, which have appeared immune from environmental influence. In particular OP map pinwheels, where domains representing all orientations meet at a point, always tend to lie close to the center of OD regions. This is true even after manipulations of the visual input such as rearing animals with artificially induced strabismus (Hubel and Wiesel, 1965; Löwel, 1994; Löwel et al., 1998) or monocular deprivation (Crair et al., 1997). However, whether this relationship is a fundamental aspect of map structure that is determined by innate mechanisms (Godecke and Bonhoeffer, 1996; Crair et al., 1998; Kaschube et al., 2002; Katz and Crowley, 2002; Tomita et al., 2013), and thus beyond the limits of brain plasticity, is unclear. Computational models of map formation based on Hebbian plasticity principles have played an important role in understanding the mechanisms governing visual development. In particular ‘dimension reduction’ models predict negative correlations between the local gradient magnitudes of different maps (Durbin and Mitchison, 1990; Swindale, 1996), thus explaining why pinwheels, which have a high orientation gradient, normally tend to lie near the centre of OD regions, where the ocularity gradient is small. However, such models also suggest that the spatial relationship between pinwheels and OD regions might be sensitive to visual experience (Giacomantonio et al., 2010). Here we used a computational model to predict how rearing animals with visual input biased to vertical orientations in one eye and horizontal orientations in the other eye (cross-rearing) would change these relationships. We then confirmed this prediction by raising cats with weak (-10 dioptre) cylindrical lenses placed in front of their eyes throughout the critical period. This demonstrates a form of plasticity in the relationships between visual feature maps that has not previously been observed. Results Computational prediction The elastic net algorithm (Durbin and Mitchison, 1990) uses Hebbian learning to optimize a trade-off between coverage and continuity constraints, and can explain many aspects of visual map formation (Swindale, 1996; Goodhill, 2007). When simulating normal rearing (Erwin et al., 1995; Carreira-Perpinan and Goodhill, 2004; Carreira-Perpinan et al., 2005) this reproduces the experimental observations cited above that pinwheels tend to be located near the center of OD columns. However, previous simulations of map development (Giacomantonio et al., 2010) suggested that this relationship would be disrupted when horizontal orientations were over-represented in one eye and vertical orientations were over-represented in the other (Hirsch and Spinelli, 1970; 1971; Blakemore, 1976). Here we simulated this scenario using the elastic net algorithm to quantify further the degree to which the relationship between pinwheels and OD borders would change as a function of the strength of over-representation (see Materials and methods). To measure the relationship of pinwheels to OD regions we separately divided the left and right eye regions of simulated OD maps into five equally sized bins, which represented areas of the OD map from the centres of the OD columns to the border regions, similarly to Hübener et al. (1997). As the strength of over-representation (α) increased, the histograms became increasingly biased away from the center of OD regions, indicating a shift in the relative location of pinwheels towards the OD borders (Figure 1). To determine whether this is due to the movement of pinwheels, OD borders, or both, we then fixed the random seed in the algorithm and explored how the spatial relationships changed within reproducible maps as a function of α. An example is shown in Figure 1g, from which it is clear that both pinwheel positions and OD borders move in the cross-reared case. The average distance that pinwheels move from their original (α = 1) positions as a function of α is shown in Figure 1h. Thus, in the model, cross-rearing alters the spatial relationship between pinwheels and OD borders by causing movement of both. Figure 1 Download asset Open asset The elastic net model predicts changes in spatial map relationships under cross-rearing. (a) Simulated orientation preference map (colours), orientation pinwheels (black dots), and ocular dominance borders (black lines) under normal rearing (relative strength of over-representation of horizontal and vertical contours in the input α = 1). (b) Histogram of pinwheel locations relative to the OD borders under normal rearing, showing a preference for pinwheels located near the centre of OD regions as previously observed experimentally. Error bars show ± 1 SEM from 10 independent simulations. The dashed line shows the expected distribution if pinwheels were arranged randomly. (c) Simulated orientation preference map for the cross-reared condition (α = 3). (d) Histogram of pinwheel locations relative to OD borders for this case. (e,f) Simulated orientation preference maps and corresponding histogram of pinwheel locations relative to OD borders for a higher level of cross-rearing (α = 5). In the simulations of cross-rearing, pinwheels are shifted away from the centre and towards the border of OD regions. (g) A cropped region of a simulated OP and OD map produced with the same random seed but increasing strengths of over-representation. Circles show the location of a pinwheel and lines show the location of the adjacent OD border. Both the pinwheel and the OD border move under cross-rearing relative to their positions under normal rearing (α = 1), but the distance between them decreases. (h) The average distance that pinwheels move from their original positions (measured in units of average OP map wavelength) as a function of the strength of cross-rearing. Errors bars show ± 1 SEM across all pinwheels in the map. Scale bars in panels a, c, e and g indicate 15 pixels in the simulated feature maps. Source data for this figure are available in Figure 1—source data 1. https://doi.org/10.7554/eLife.13911.003 Figure 1—source data 1 This HDF5 file contains the numerical values shown in Figure 1. https://doi.org/10.7554/eLife.13911.004 Download elife-13911-fig1-data1-v1.zip However, plasticity of this type has not yet been observed experimentally, leaving open the possibility that the relationship between pinwheels and OD columns could instead be determined by intrinsic mechanisms and is not susceptible to environmental modification. Testing the prediction To directly test these predictions we reared cats from 3 weeks of age with -10 dioptre cylindrical lenses mounted comfortably in front of their eyes using soft neoprene masks. The lens covering the left eye had its axis aligned vertically, so that the left eye was exposed to high contrast contours with primarily horizontal orientation. Conversely, the lens covering the right eye had its axis aligned horizontally, so that the right eye was exposed to high contrast contours with primarily vertical orientation (Figure 2). The lens strength was chosen based on preliminary observations that animals wearing -10 dioptre lenses exhibited normal behavioural activity, while animals with higher power lenses (e.g., as used in Tanaka et al. [2006]) were noticeably less active. Animals wore the masks for 6 hr per day while the room was illuminated and were otherwise kept in darkness. During light periods the animals were monitored at least every 30 min to ensure the masks remained in place and to encourage active visual behaviours, such as chasing light patterns and balls. Paw striking behaviour towards objects in front of the animals appeared normal. From age 20 weeks we used intrinsic signal optical imaging followed by extracellular single unit recordings to map OP and OD preferences in cortical areas 17 and 18 of 5 cross-reared animals (10 hemispheres), and 6 normally reared control animals (11 hemispheres). Figure 2 Download asset Open asset Optical characteristics of the -10 dioptre cylindrical lenses. (a) Circular square wave test grating (1 cycle/°) viewed normally (no lens). (b) The same grating viewed through the -10 dioptre cylindrical lens, with the lens axis aligned horizontally, attenuating horizontal and preserving vertical contours. (c) The radially symmetric distribution of power over spatial frequency for the test grating viewed normally. (d) The radially asymmetric distribution of power over spatial frequency for the same grating when viewed through the -10 dioptre cylindrical lens with the lens axis aligned horizontally. Contours orthogonal to the axis are preserved while contours parallel to the axis are attenuated. Power spectra shown in (c) and (d) are normalized to the peak power (black). https://doi.org/10.7554/eLife.13911.005 Cross-rearing alters tuning properties of single units We recorded extracellular spiking responses to quantify the tuning properties of single units in both the normal and cross-reared animals. In the five cross-reared animals we recorded from 182 units from 20 electrode tracks, 76 with a preference for input from the right eye (which experienced predominantly vertical contours) and 106 with a preference for input from the left eye (which experienced predominantly horizontal contours). In the six control animals, we recorded from 86 units from 17 electrode tracks. Electrode tracks were positioned without reference to cortical map structure. Consistent with previous work (Coppola et al., 1998; Li et al., 2003) and theoretical predictions (Hunt et al., 2013), the distribution of preferred orientation for units from the control animals exhibited an over representation of the cardinal orientations (Figure 3a). This distribution was well described by a sine curve with period 90º (r2 = 0.69). In contrast, distributions of preferred orientation for units from the cross-reared animals showed clear biases depending on the eye providing the dominant input. Units with a preference for input from the left eye showed a bias for horizontal orientations (Figure 3b, black line), while those with a preference for input from the right eye showed a bias for near vertical orientations (Figure 3b, gray line). These distributions were consistent with the orientation of the lenses fitted over each eye and were well described by sine curves with period 180º (r2 = 0.75, left eye; r2 = 0.79, right eye). In contrast to the distribution from the control animals, sine curves with period 90º provided a relatively poor account of the data (r2 = 0.11, left eye; r2 = 0.03, right eye). Figure 3 with 3 supplements see all Download asset Open asset Tuning properties of single units. (a) The distribution of preferred orientation in control animals exhibited an over representation of cardinal orientations. The dashed line shows the expected distribution if all orientations were represented equally. The best-fitting sine curve with period 90° had peaks at 84° and 174° (thick line, r2 = 0.69). (b) Distributions of preferred orientation in the cross-reared animals, for units driven predominantly by input from the left (black) and right eye (grey). The best-fitting sine curves with period 180° peaked at 174° for the left eye (thick black line, r2 = 0.75) and 104° for the right eye (thick grey line, r2 = 0.79). Cross-rearing thus caused a systematic change in the distributions of preferred orientation of single units. Cross-rearing also caused an increase in monocularity (c) and an increase in preferred spatial frequency (d) of single units. Source data for this figure are available in Figure 3—source data 1. https://doi.org/10.7554/eLife.13911.006 Figure 3—source data 1 This HDF5 file contains the numerical values shown in Figure 3. https://doi.org/10.7554/eLife.13911.007 Download elife-13911-fig3-data1-v1.zip We also calculated the inter-ocular difference in preferred orientation, ∆OP, for each unit as the preferred orientation for the right eye minus the preferred orientation for the left eye. Consistent with previous reports (Nelson et al., 1977; Cooper and Pettigrew, 1979) we found a torsional disparity in the preferred orientation of the two eyes in both our control (mean ΔOP = 11.7°, p<0.001, two-tailed t-test) and cross-reared (mean ΔOP = 9.8°, p<0.001, two-tailed t-test) animals. We found no significant difference in the distribution of ΔOP for control and cross-reared animals (p=0.44, two-tailed, two-sample t-test; Figure 3—figure supplement 1). Units from cross-reared animals showed a higher level of monocularity than those from control animals (Figure 3c). The median monocularity index (MI, see Material and methods) of units from control and cross-reared animals was 0.24 and 0.38, respectively. This difference was significant (p=0.002, Kruskal-Wallis test). Cross-rearing therefore appears to induce subtle changes in the combination of input from the two eyes at the level of single neurons. In addition to each unit’s ocular dominance and orientation preference, we also quantitatively measured their tuning for spatial and temporal frequency (See Materials and methods). We found no significant differences in the monocular spatial or temporal frequency tuning for dominant vs non-dominant eyes or for left vs right eyes (regardless of dominance) in either control or cross-reared animals (p>0.05, Kruskal-Wallis tests). We therefore combined the populations of left- and right-eye dominant units within the two experimental groups (i.e., control and cross-reared) and compared their tuning parameters for the dominant eye. Units from cross-reared animals exhibited marginally higher preferred spatial frequencies compared to those from control animals (median preferred spatial frequency 0.24 and 0.16 cycles/° for cross-reared and control animals respectively; Figure 3d). While this difference was significant (p<0.001; Kruskal-Wallis test), the tuning curves were very broad and overlapped substantially. We found no difference in either the bandwidth (p=0.19; Kruskal-Wallis test) or skew (p=0.40; Kruskal-Wallis test) of the spatial frequency tuning curves for normal and cross-reared animals. Similarly, we found no significant difference in either the preferred temporal frequency (p=0.2, Kruskal-Wallis test; Figure 3—figure supplement 2), the temporal frequency bandwidth (p=0.28, Kruskal-Wallis test) or the skew of the temporal frequency tuning curves (p=0.63, Kruskal-Wallis test) of units from control and cross-reared animals. Using the optimal grating parameters (orientation, size, spatial and temporal frequency) for each unit we also measured their response as a function of stimulus contrast. We found no difference in either the maximum spike rate (p=0.56, Kruskal-Wallis test) or the semi-saturation contrast (p=0.81, Kruskal-Wallis test; Figure 3—figure supplement 3) of units from control and cross-reared animals. Selectivity for stimulus parameters, responsiveness and sensitivity to stimulus contrast therefore appeared normal in the cross-reared animals. Extended spatial decorrelation for intrinsic signal optical imaging While previous studies have usually used red light (wavelengths >600 nm) for intrinsic signal imaging, here we used green light (520 nm) due to the much higher signal to noise ratio obtained (Figure 4a). Conventional techniques for map generation from intrinsic signal imaging data are based on the calculation of difference images by either subtraction or division by a reference image that is independent of the stimulus of interest. We found that these techniques left strong blood vessel artifacts in the maps derived from green light data, particularly for ocular dominance maps. However we were able to overcome this problem by using extended spatial decorrelation (ESD), a more sophisticated analysis technique (Stetter et al., 2000) (see Materials and methods). This technique robustly separated the noise and mapping signals even in the green light data (Figure 4b–d). Consistent with a brief earlier report (Frostig et al., 1990), by imaging one of our control animals with both green (520 nm) and red (609 nm) wavelengths we found that the maps produced were very similar (Figure 4e–h). We also compared measures of orientation preference from the OP maps with corresponding measures obtained from single unit recordings from the superficial layers of the cortex at each electrode track location. Where we obtained robust estimates from both the imaging and unit recordings we found a high level of correlation between the two measures (r2 = 0.64, p=0.003) with a median absolute difference of 15.2°. Figure 4 Download asset Open asset Extended spatial decorrelation recovers OP and OD maps from green light imaging. (a) Time course of the relative change in reflectance (∆R/R) during a trial, averaged over all pixels and all trials, measured with red (609 nm) and green (520 nm) light. The shaded region shows the stimulus period. Consistent with earlier reports (Sirotin and Das, 2009; Sirotin et al., 2009), green light produced a much stronger signal. (b–d) Representative source coefficient time series from the extended spatial decorrelation algorithm (see Materials and methods). Sources 46–50 (as per legend) for (b) the real component of the OP map, (c) the imaginary component of the OP map, and (d) the OD map. The shaded region shows the stimulus period. The mean over the pre-stimulus period was subtracted from each source. It is clear in each case that one source (in this case source 50) most strongly represents the signal of interest. (e,f) OP maps generated from the green (e) and red (f) light responses. (g,h) OD maps generated from the green (g) and red (h) light responses. Over the region shown here (the one used for analysis), the correlation between red and green OP maps was r2 = 0.78 and between the red and green OD maps was r2 = 0.77. Colour encodes the preferred orientation in the OP maps (as per legend) and brightness encodes eye preference in the OD maps, with black and white representing the left and right eyes, respectively. Data from a control animal. Scale bars: 1 mm. Source data for this figure are available in Figure 4—source data 1. https://doi.org/10.7554/eLife.13911.011 Figure 4—source data 1 This HDF5 file contains the numerical values shown in Figure 4. https://doi.org/10.7554/eLife.13911.012 Download elife-13911-fig4-data1-v1.zip Cross-rearing alters the proportion of cortical area representing different orientations We measured OP and OD maps in cortical areas 17 and 18 of both normal and cross-reared animals using intrinsic signal optical imaging (typical maps for each case are shown in Figure 5a–f). Cross-rearing caused a very slight reduction in orientation selectivity, as derived from the OP maps (Figure 5—figure supplement 1). Although the selectivity distributions appear very similar, the difference was statistically significant (p<0.001, two-sample Kruskal-Wallis test). However cross-rearing induced profound changes in map structure. In control animals there was an over-representation of cardinal (horizontal and vertical) orientations in the OP map. This is consistent with the distribution of orientation preferences of single units described above, and with previous reports for ferrets (Coppola et al., 1998) and cats (Li et al., 2003). The proportion of the map devoted to different orientations was well fit by a sine curve with period 90º (r2 = 0.6; Figure 5g; Figure 5—figure supplement 2). However, in the cross-reared animals the OP maps calculated for each eye showed proportions that were now better fit by sine curves with period 180º (r2 = 0.64, left eye; r2 = 0.71, right eye). In each case these curves peaked close to the orientation of the lens covering that eye (Figure 5h; Figure 5—figure supplement 2). As a comparison, the best-fitting sine curves with period 90º had r2 values of 0.33 (left eye) and 0.13 (right eye). Thus, cross-rearing caused substantial shifts in the distribution of orientations across the cortex. Figure 5 with 2 supplements see all Download asset Open asset Cross-rearing changes the distribution of orientation preferences. (a) OP map, (c) OD map and (e) overlay of OD and OP contours for a control cat. (b) OP map, (d) OD map and (f) overlay of OD and OP contours for a cross-reared cat. While qualitatively the control and cross-reared maps look similar, quantitative analysis revealed differences. (g) Proportion of cortical area representing different orientations from binocular stimulation for all control hemispheres (thin line: mean ± 1 SEM, thick line: least-squares sine curve fit). The best-fitting sine curve with period 90° had peaks at 7° and 97°, and an r2 value of 0.6. For comparison the dashed line at a frequency of 1/8 represents equal proportions. (h) Data from left (thin black line) and right (thin grey line) monocular stimulation for all cross-reared hemispheres. The best-fitting sine curve with period 180° peaked at 0° for the left eye (horizontal orientations, thick black line) and 148° for the right eye (vertical orientations, thick grey line). The r2 values for the fits were 0.64 (left eye) and 0.71 (right eye). In contrast the best-fitting sine curves with period 90° (not shown) had r2 values of 0.33 (left eye) and 0.13 (right eye). Thus cross-rearing caused a systematic shift in the proportions of the maps occupied by each orientation, towards the orientation that each eye predominantly experienced. As in Figure 4, colour encodes the preferred orientation in the OP maps and brightness encodes eye preference in the OD maps. Scale bars: 1 mm. Source data for this figure are available in Figure 5—source data 1. https://doi.org/10.7554/eLife.13911.013 Figure 5—source data 1 This HDF5 file contains the numerical values shown in Figure 5. https://doi.org/10.7554/eLife.13911.014 Download elife-13911-fig5-data1-v1.zip Cross-rearing alters the spatial relationship between OP and OD maps The characteristic relationships between OP and OD maps are intersection angles and the distance of pinwheels to OD borders. There was no statistically significant change in pinwheel density (Kaschube et al., 2010) between rearing conditions (Figure 6a). The distribution of intersection angles between OD and OP maps was also similar in both rearing conditions (Figure 6—figure supplement 1). There were subtle changes in the spatial distribution of orientation selectivity: in control animals orientation selectivity was slightly greater near OD borders, while in cross-reared animals selectivity was slightly greater near the centre of OD regions (Figure 6—figure supplement 2). However there was a clear effect on the distance of pinwheels to OD borders. In our control animals, the distribution of pinwheels relative to OD borders was very similar to that of Hübener et al. (1997) (Figure 6b) but in cross-reared animals there was a significant shift of pinwheels away from the centre of OD regions (Figure 6c). Specifically, there was a significant under representation of pinwheels in the bin corresponding to the centre of OD columns (p=0.01, two-tailed t-test, power = 0.77, 95% confidence interval for difference in means = [0.06, 0.36]), as predicted in our model simulations (Figure 6d). Thus, the altered visual input during cross-rearing changed a fundamental aspect of the spatial relationship between OP and OD maps. The similarity between the model prediction and our data provides further evidence that models based on dimension reduction, such as the elastic net, capture essential elements of the mechanisms by which cortical maps develop. Figure 6 with 2 supplements see all Download asset Open asset Spatial relationship between pinwheels and ocular dominance is modified by rearing condition. (a) Pinwheel density relative to squared map wavelength was not significantly different between control and cross-reared animals, both being consistent with the theoretically predicted value of π (dashed line) (Kaschube et al., 2010). (b) Pinwheel locations relative to the centres/borders of OD regions were quantised into 5 bins similarly to Hübener et al. (1997). For control animals, pinwheels were disproportionately overrepresented at the centre of OD regions (n = 71 pinwheels total in control hemispheres), consistent with previous data (Hübener et al., 1997). (c) In strong contrast, for the cross-reared animals pinwheels were disproportionately underrepresented at the centre of OD regions (n = 55 pinwheels total in cross-reared hemispheres). (d) Computational simulations using the elastic net reproduced the shift of pinwheels away from the centres of OD regions in the cross-reared compared to control condition (data replotted from Figure 1b,f). For all graphs error bars show ± 1 SEM. p-values in (a) and (c) are from two-tailed, two-sample t-tests. Source data for this figure are available in Figure 6—source data 1. https://doi.o
Neurons in the visual cortex code relative changes in illumination (contrast) and adapt their sensitivities to the visual scene by centering the steepest regions of their sigmoidal contrast response functions (CRFs: spike rate as a function of contrast) on the prevailing contrast. The influence of this contrast gain control has not been reported at nonoptimal drift rates. We calculated the Fisher information contained in the CRFs of halothane-anesthetized cats. Fisher information gives a measure of the accuracy of contrast representations based on the ratio of the square of the steepness of the CRF and the spike-rate dependency of the spiking variance. Variance increases with spike rate, so Fisher information is maximal where the CRF is steep and spike rates are low. Here, we show that the contrast at which the maximal Fisher information (C MFI ) occurs for each adapting drift rate is at a fixed level above the adapting contrast. For adapting contrasts of 0 to 0.32 the relationship between C MFI and adapting contrast is well described by a straight line with a slope close to 1. The intercept of this line on the C MFI -axis is drift-rate dependent, although the slope is not. At high drift rates relative to each cell's peak the C MFI offset is higher than that for low drift rates. The results show that the contrast coding strategy in visual cortex maximizes accuracy for contrasts above the prevailing contrast in the environment for all drift rates. We argue that tuning the system for accuracy at contrasts above the prevailing value is optimal for viewing natural scenes.
'/%3e%3c/defs%3e%3cpath fill='%23012169' d='M0 0h10080v5040H0z'/%3e%3cpath stroke='%23fff' stroke-width='.6' d='m0 0 6 3m0-3L0 3' clip-path='url(%23b)' transform='scale(840)'/%3e%3cpath stroke='%23e4002b' stroke-width='.4' d='m0 0 6 3m0-3L0 3' clip-path='url(%23c)' transform='scale(840)'/%3e%3cpath stroke='%23fff' stroke-width='840' d='M2520 0v2520M0 1260h5040'/%3e%3cpath stroke='%23e4002b' stroke-width='504' d='M2520 0v2520M0 1260h5040'/%3e%3cg fill='%23fff'%3e%3cuse xlink:href='%23d' x='2520' y='3780'/%3e%3cuse xlink:href='%23a' x='7560' y='4200'/%3e%3cuse xlink:href='%23a' x='6300' y='2205'/%3e%3cuse xlink:href='%23a' x='7560' y='840'/%3e%3cuse xlink:href='%23a' x='8680' y='1869'/%3e%3cuse xlink:href='%23e' x='8064' y='2730'/%3e%3c/g%3e%3c/svg%3e)
