Vignesh Subramanian ’24

Neural coding is the study of how neurons conduct information processing, with the aim of identifying relationships between stimuli and neuronal responses by examining electrical activity. One particular coding scheme, commonly known as population coding, involves generating spatiotemporal representations of activity in clusters of cells as opposed to individual cells. When such representations are mapped onto global topographic organization of an organism’s brain, scientists may “decode” the functional organization of various lobes and cortexes – what stimuli they respond to, how they respond, and the degree of precision in their responses. The organization of population code specific to visual stimuli has only recently been examined in mice. To close the gap between subjects used for visual perception studies, Stony Brook researcher Dr. Memming Park, in conjunction with the Allen Brain Observatory, explored how certain population responses in visual areas in mice project their stimulus preferences.

Researchers passively exposed Emx-IRES-Cre and Rbp4-Cre-KL100 transgenic mice – genetically modified with six tagged cell types to highlight forebrain function – to both synthetic and natural stimuli on a monitor. The stimuli intentionally varied in their spatiotemporal complexity and the associated signaling responses were screened in six key visual areas of the mouse brains by in vivo two-photon and epifluorescent calcium imaging. Statistical classifiers were then employed to discretize this data, comparing neural activity vectors and stimulus classification rates between populations of 128 neurons each. Researchers found that the decodability of stimuli from electrical activity in the secondary visual cortex (V2) of mice brains fared just as well as or worse than that of the primary visual cortex (V1). This suggests that the hierarchical structure of visual processing observed in primates – wherein visual area V1 elevates inputs to V2 – is present in mice as well.  Researchers also found that excitatory neurons in certain visual area populations displayed far greater decoding accuracy than individual cells, known as synergy, which suggests that representation of the causal electrical activity across neurons is area-specific. 

This study demonstrated a level of specialized organization of neural responses to visual stimuli in mice not previously documented, setting the animal up as a viable model for research on information processing. Ultimately, differences in how researchers were able to decode stimuli from the subjects’ neuronal electrical activity further detailed the biophysical separation of populations already known to be anatomically divergent, deepening understanding of this particular mammalian visual system.

Works Cited:

[1] I. Memming Park, et al., Organization of Neural Population Code in Mouse Visual System. eNeuro 5, 414-424 (2018). doi: 10.1523/ENEURO.0414-17.2018.

[2] Image retrieved from:  https://www.neuroscientificallychallenged.com/blog/know-your-brain-primary-visual-cortex