Processing of second-order, contrast-modulated stimuli in mouse visual cortex

Abstract

Visual processing along the primate ventral stream takes place across a hierarchy of areas, characterized by an increase in both complexity of neuronal preferences and invariance to changes of low-level stimulus attributes. A basic type of invariance is form-cue invariance, where neurons have similar preferences in response to first-order stimuli, defined by changes in luminance, and global features of second-order stimuli, defined by changes in texture or contrast. Whether in mice, a now popular model system for early visual processing, visual perception can be guided by second-order stimuli is currently unknown. In this project, we asked whether mice can use second-order stimuli to guide visual perception in a cue-invariant way and assessed potential cue-invariant representations of stimulus orientation in two areas of mouse visual cortex. For both behavioral and electrophysiological experiments, we used a common set of luminance-modulated gratings (LGs) and contrast-modulated gratings (CGs), obtained by multiplying a contrast envelope with a noise carrier. We created two types of CG stimuli that differed in the Fourier energy distribution of the carrier: high-frequency noise and low-frequency noise. To examine a potential effect of different contrast of LGs and CGs, we matched the root-mean-square (RMS) contrast between first- and second- order gratings, in which case c was 0.335 for the LGs. We tested whether mice can generalize orientation discrimination learned with first- order, LGs to various untrained second-order, CGs. We first trained head-fixed mice in a classical conditioning paradigm to perform a coarse orientation discrimination on LGs. Once the animal reached stable and reliable orientation discrimination performance, we replaced LGs with CGs with low-frequency noise carriers to test the generalization of orientation discrimination to second-order gratings. We found that mice, after learning a coarse orientation discrimination involving only LGs, could readily generalize orientation discrimination to CGs, albeit with a substantial drop in performance. Then, we wondered whether the overall lower performance for CGs was related to their lower RMS contrast compared with LGs. To test this hypothesis, we probed mice with LGs whose RMS contrast was lowered to match that of CGs. We found that mice could perform well during the orientation discrimination task for LGs matched in RMS contrast. Indeed, across all animals tested, performance was similar for both levels of contrast. Finally, we tested mice with CGs with high-frequency noise carriers. Again, mice could see this type of CG and importantly, they could also discriminate between the two grating orientations, albeit performance was again considerably lower compared with that for LGs. Together, these results demonstrate that mice can use second-order stimuli to guide visual perception. We performed extracellular recordings, both during anesthesia and wakefulness, in mouse areas V1 and LM, where we compared orientation tuning curves to LGs and CGs. We found that neurons in area V1 and LM were less responsive and less selective to CGs than to LGs, both during anesthesia and wakefulness. Interestingly, this reduction was particularly prominent during anesthesia. We wondered whether our finding of weaker responses and broader orientation tuning for CG than LG responses could be explained by the lower RMS contrast of CGs. Thus, we performed control experiments, in which we measured responses to LGs that were matched in RMS contrast to the CGs. We found that neurons in both visual areas still responded more weakly to CGs than LGs but this reduction in responsiveness was less pronounced compared with conditions with full contrast LGs. Similar to our results with full-contrast LGs, orientation selectivity also decreased considerably between LGs matched in RMS contrast and CGs. Indeed, orientation selectivity did not differ significantly between responses to full-contrast and reduced-contrast LGs. Together, the reduced RMS contrast of CGs might contribute to the reduction in peak responsiveness to CGs but cannot account for the poorer orientation selectivity for CGs. We also investigated the underlying mechanism of responses to second-order stimuli in awake recordings: we tested orientation tunings of both V1 and area LM in response to a high-frequency noise CGs, in which the noise carrier’s spatial frequency distribution was concentrated beyond the passband of many V1 and LM neurons (Marshel et al., 2011). We first observed that less than half of the recorded neurons with significant responses to LGs also had visually evoked activity to CGs with high-frequency noise. This fraction of responsive neurons was considerably lower compared to that obtained for CGs with low-frequency noise. Interestingly, the difference in responsiveness between the two types of CG stimuli was stronger for area V1 than LM. To examine whether the CG representation might contribute toward cue- invariant perception of stimulus orientation, we also compared the neurons’ preferred orientation, separately for each grating type. We found that preferred orientations for CGs with low-frequency noise and LGs were correlated for both areas V1 and LM. Interestingly, for CGs with high-frequency noise, the distribution of differences in preferred orientation was non-uniform only for area LM and preferred orientations were only correlated for area LM. Together, the broad similarity of preferred orientations between grating types provides some evidence for a coarse cue- invariance, which might in turn be part of the neural basis for perceptual generalization of orientation discrimination.