Even in the primary visual cortex, a brain region named for its specialized role in processing the basic features of what we see, not all neurons ultimately respond to the call to process properties of visual input. Perhaps that’s because each neuron receives a wide variety of inputs through thousands of circuit connections, or “synapses,” and must choose whether to respond to visual or other information.
In a new study in mice, neuroscientists at the Picower Institute for Learning and Memory at the Massachusetts Institute of Technology have uncovered how visual processing neurons bring order to this input and accomplish their tasks.
Neuroscientists are keenly interested in what inputs, out of a vast array of options, compel neurons to participate in brain computations and functions, said lead author Mriganka Sur, Newton Professor of Neuroscience in the Picower Institute and the Massachusetts Institute of Technology’s Department of Brain and Cognitive Sciences. Neurons ultimately participate in brain circuits by “firing” electrical action potentials.
The composition of inputs, the type of tissue, and the collection of neurons that modulate each other to generate action potentials are the essence of how brain circuits process information. These (visual cortex) cells are a microcosm of a very profound and larger picture of neuroscience. ”
Mriganka Sur, Senior Study Author, Newton Professor, Neuroscience, Picower Institute
In a study in iscience, The research team, led by postdoctoral fellow Kyle Jenks, made the discovery by closely imaging how not only the cell bodies of neurons, but also individual synapses formed on projections known as dendritic spines, respond as mice watch videos.
They performed this imaging not only on visually responsive neurons, but also on non-responsive neurons that have visually responsive spines. This allowed us to dissect a number of important properties that can influence where a particular synapse forms and how it affects the response of the cell body.
“This brings together many things that have been considered individually into one collective paper,” Jenks said. “We can compare how a neuron and the spines on that neuron respond to the same stimulus, and we can do this for both visually responsive and non-visually responsive neurons.”
reveal the rules
Jencks and his team genetically engineered neurons in layer 2/3 of the visual cortex to cause individual dendritic spines to light up when a spike in calcium indicated increased activity at the synapses on the spines. The scientists did the same thing with the cell body, or soma, to track how the cell responded and then signaled its overall response back to the synapses.
This way, while the mice watched the black and white grid move in different directions with their eyes at different angles, the scientists were able to track the overall response of each vertebrae and each cell to the patterned visual input.
In total, they tracked 11 neurons that responded to visual input and 11 others that appeared to ignore it. This allowed me to find some rules such as:
Distance from the soma matters: For cells responding to visual input, the closer the spines are to the soma, the more likely the response of individual spines will correlate with the activity of the soma. Similarly, somatic signals returning to the spine are thought to influence the preferred positional relationship between the spine and the soma, and were more likely to be detected closer to the soma than farther away.
Local clustering: In neurons responding to visual input, spines formed distinct small enclaves of correlated responses. Specifically, spines within 5 microns (five millionths of a meter) worked together. But just outside that 5-micron boundary, the spines were very unlikely to participate in that activity. Sarr speculates that these isolated pockets of activity may have sharpened the response from each enclave.
“apical” and “basal”: The neurons the researchers studied have two different types of dendrites. Apical dendrites are very long and protrude from the top, or “apex,” of the neuron and tend to receive a variety of inputs from throughout the cortex. Basal dendrites are shorter, extend from the bottom, and typically receive more raw visual input. Although basal dendrites received more visual input overall than apical dendrites, Jencks found that apical dendrites on visually responsive neurons had significantly more visually responsive spines than those on non-responsive neurons. And both types of dendrites followed the above rules regarding distance from the cell body equally.
Directional selectivity is most important: Jenks, Sur, and team used statistical modeling to determine which of a number of factors (stimulus selectivity, response reliability, distance of the spine from the soma, apical vs. basal, etc.) best explains the correlation between spine reactivity and soma reactivity. By far, the single factor that mattered most was how selective a spine was for its preferred lattice orientation.
“Our results reveal that synaptic inputs to excitatory layer 2/3 neurons (visual cortex) in mice are not randomly arranged, but are organized and distributed in a manner that correlates with multiple factors, including somatic reactivity, somatic tuning, branch type, distance from the soma, local correlation, and stimulus selectivity,” the researchers wrote.
Jenks and Sarr said their team’s findings could help advance the study of vision in the brain in a variety of ways. Certain genetic mutations that affect how neurons connect in circuits can affect neurons in the visual cortex and vision, Schurr said. Documenting these rules gives researchers a baseline to compare when examining the effects of such mutations. Jencks added that the findings could inform efforts to model how neurons integrate synaptic inputs in computations.
sauce:
Reference magazines:
Jenks, K.R. others (2026) Functional organization of dendritic spines in mouse visual cortex layer 2/3 neurons. iscience. DOI:10.1016/j.isci.2026.115861. https://www.cell.com/iscience/fulltext/S2589-0042(26)01236-8.
