Flow stimuli reveal ecologically-appropriate responses in mouse visual cortex

Luciano Dyballa, Mahmood S. Hoseini, Maria C. Dadarlat, Steven W. Zucker, Michael P. Stryker

TL;DR

This paper challenges the conventional view of mouse vision as poor, based on traditional grating stimuli. By introducing "visual flow patterns" that mimic naturalistic optic flow, the researchers found that mouse primary visual cortex (V1) neurons respond robustly to much finer details than previously thought, suggesting a more sophisticated visual capacity relevant to natural behaviors.

What Problem Does This Paper Address?

The mouse visual system is a popular model for studying vision, development, and disease. However, a significant discrepancy exists: while mice exhibit complex, visually-guided behaviors like prey capture, standard assays using sine-wave gratings suggest they have very poor vision. Specifically, studies using these gratings indicate that V1 neurons have large receptive fields and are primarily tuned to very low spatial frequencies (below 0.08 cycles/degree). This raises a critical question: how can mice perform so adeptly in natural tasks if their visual system is so limited? The authors propose that traditional stimuli may not adequately capture the visual information relevant to a mouse's natural environment.

Methods

The researchers developed a novel stimulus class called "visual flow patterns." These patterns are inspired by the optic flow experienced by a mouse moving through its environment, resembling a "waterfall" of illuminated material. Unlike sine-wave gratings, these flow stimuli are comprised of drifting dots or dotted line-segments and are parametrically controllable in terms of orientation, spatial frequency, and direction of motion. They are more ecologically relevant than gratings but more analytically tractable than natural images.

The study involved recording the simultaneous responses of single neurons in the primary visual cortex (V1) of alert mice using 128-site silicon microelectrodes. They presented a diverse ensemble of stimuli, including drifting gratings and various flow patterns (single dots, oriented dotted segments) with different contrast polarities and spatial frequencies, while keeping the temporal frequency fixed at 4 Hz. Responses were analyzed using peristimulus time histograms (PSTHs) and tuning curves.

Datasets

Not discussed in the paper. The study utilized custom-generated visual stimuli.

Key Insights and Ideas

  • Ecological Relevance Matters: The study strongly suggests that the choice of visual stimulus is crucial for understanding neural responses. Stimuli that mimic naturalistic experiences (like optic flow) reveal capabilities that are masked by artificial stimuli (like gratings).
  • Mouse Vision is Better Than We Thought: Contrary to previous findings based on gratings, V1 neurons in mice show robust responses to high spatial frequencies (0.15–1.6 cycles/degree) when presented with flow stimuli.
  • Beyond Linear Models: The complex and varied responses of V1 neurons to the flow ensemble challenge conventional linear approaches to visual processing, indicating that a simple receptive field model is insufficient.
  • Multidimensional Selectivity: Many V1 neurons exhibit "mixed selectivity," responding to multiple stimulus dimensions (e.g., orientation, direction, contrast, spatial frequency) in a way that cannot be predicted by traditional linear models.
  • Layer-Specific Processing: There are hints of layer-specific differences in how V1 neurons process flow stimuli, with superficial layers potentially showing higher selectivity for differentiating flows from gratings, and deeper layers showing more invariance to geometry and contrast.

Key Results

  • Flow stimuli evoked strong visually-mediated responses in V1 neurons, extending well beyond what was predicted by spatial frequency analysis.
  • Flow responses were predominantly observed in higher spatial-frequency ranges (0.15–1.6 cycles/degree), significantly higher than the previously reported optimal range for mouse V1.
  • Many neurons showed orientation- or direction-selectivity to flow stimuli.
  • A significant proportion of neurons exhibited responses that depended strongly on the sign of contrast.
  • Many cells displayed distributed responses across the entire stimulus ensemble, indicating multidimensional selectivity.
  • When comparing stimulus classes, the study found that including low-frequency gratings led to a higher proportion of "MULTI" cells (responding well to both gratings and flows). However, when low-frequency gratings were excluded, flow stimuli and "MULTI" responses predominated over gratings alone, suggesting a preference for flow-related information at higher spatial frequencies.
  • For cells well-tuned to orientation and/or direction, there was a preference for oriented flow patterns over non-oriented ones.

Limitations and Future Work

  • Stimulus Ensemble Complexity: While the flow stimuli are more tractable than natural images, they still represent a complex ensemble. Further work is needed to fully disentangle the contributions of different stimulus features.
  • Operational Definitions: The authors acknowledge that their classification of cell types (GRATING, FLOW, MULTI) is a crude simplification for discussion, and the underlying neural computations are more complex.
  • Network Computations: The paper suggests that understanding network computations, rather than just individual features, should be the focus. Future research could explore how these flow responses are integrated across the visual network.
  • Ecological Validity: While flow stimuli are more ecologically relevant, they are still abstractions. Further investigation with more naturalistic stimuli could provide deeper insights.
  • Layer-Specific Mechanisms: The observed layer-specific differences in selectivity warrant further investigation to understand the underlying circuit mechanisms.
  • Contextual Effects: The study used wide-field stimuli, and while contextual interactions are mentioned as a potential factor, their specific role in mouse V1 processing of flow stimuli requires more detailed exploration.

Summary updated

Abstract

Abstract Assessments of the mouse visual system based on spatial frequency analysis imply that its visual capacity is low, with few neurons responding to spatial frequencies greater than 0.5 cycles/degree. However, visually-mediated behaviors, such as prey capture, suggest that the mouse visual system is more precise. We introduce a new stimulus class—visual flow patterns—that is more like what the mouse would encounter in the natural world than are sine-wave gratings but is more tractable for analysis than are natural images. We used 128-site silicon microelectrodes to measure the simultaneous responses of single neurons in the primary visual cortex (V1) of alert mice. While holding temporal-frequency content fixed, we explored a class of drifting patterns of black or white dots that have energy only at higher spatial frequencies. These flow stimuli evoke strong visually-mediated responses well beyond those predicted by spatial frequency analysis. Flow responses predominate in higher spatial-frequency ranges (0.15–1.6 cycles/degree); many are orientation- or direction-selective; and flow responses of many neurons depend strongly on sign of contrast. Many cells exhibit distributed responses across our stimulus ensemble. Together, these results challenge conventional linear approaches to visual processing and expand our understanding of the mouse’s visual capacity to behaviorally-relevant ranges. Significance Statement The visual system of the mouse is now widely studied as a model for development and disease in humans. Studies of its primary visual cortex (V1) using conventional grating stimuli to construct linear-nonlinear receptive fields suggest that the mouse must have very poor vision. Using novel stimuli resembling the flow of images across the retina as the mouse moves through the grass, we find that most V1 neurons respond reliably to very much finer details of the visual scene than previously believed. Our findings suggest that the conventional notion of a unique receptive field does not capture the operation of the neural network in mouse V1.
BibTeX
@inproceedings{dyballa2018flow,
  author = {Luciano Dyballa and Mahmood S. Hoseini and Maria C. Dadarlat and Steven W. Zucker and Michael P. Stryker},
  title = {{Flow stimuli reveal ecologically-appropriate responses in mouse visual cortex}},
  booktitle = {Proceedings of the National Academy of Sciences},
  year = {2018},
  doi = {10.1073/pnas.1811265115},
}
Figure