Summary: A new study from the Picower Institute at MIT mapped the rules neurons use to organize thousands of synaptic inputs in the primary visual cortex. By imaging both neuronal cell bodies (somas) and individual synapses on dendritic spines in awake mice, the researchers identified consistent organizing principles that determine how inputs are integrated to produce a neuron’s electrical output.
The results show that synaptic inputs are not randomly distributed across the dendritic tree. Instead, their functional impact depends on spatial factors, local clustering, and the input’s tuning for visual features such as orientation.
Key findings
- Proximity to the soma matters: Synapses closer to the cell body have stronger correlations with the neuron’s overall firing pattern. Proximal spines are more likely to reflect and be influenced by somatic activity.
- Local 5-micron neighborhoods: Dendritic spines form small, highly correlated clusters within roughly 5 microns. Spines inside these micro-enclaves tend to act together, while spines just beyond this boundary are less likely to participate in the same responses.
- Orientation selectivity is decisive: The degree to which a spine is tuned to a particular visual orientation is the single best predictor of how closely its activity matches the soma’s response.
- Dendritic branch differences: Visually responsive neurons carry more active spines on their long apical dendrites than unresponsive cells, though both apical and basal branches follow the same rules relating distance and local clustering to influence.
Source: Picower Institute at MIT
Even within the primary visual cortex—a brain region specialized for processing basic visual features—individual neurons do not always respond to visual stimuli. Each neuron receives thousands of inputs from diverse circuits, and only a subset of those inputs ultimately drive the neuron to fire an action potential. Understanding which synaptic inputs compel a neuron to join visual computations has been a central question in systems neuroscience.
In a study published in iScience and led by postdoctoral researcher Kyle Jenks with senior author Mriganka Sur, the team used two-photon in vivo calcium imaging to monitor activity at two scales: the soma and individual dendritic spines on layer 2/3 excitatory neurons in mouse visual cortex. They recorded responses while mice viewed drifting black-and-white grating patterns at different orientations, enabling direct comparison between the tuning of single synapses and the tuning of the neuron they contact.
The authors imaged both neurons that were visibly responsive to those stimuli and neurons that appeared unresponsive despite having visually responsive spines. Imaging both categories allowed the team to test which synaptic properties predicted alignment with somatic responses.
Key organizational rules
Distance from the soma: For visually responsive neurons, proximal spines—those located nearer the soma—showed higher correlation with somatic tuning than distal spines. Back-propagating signals from the soma, which can influence spine responsiveness, were also more detectable at shorter distances.
Local clustering: On responsive neurons, spines tended to form locally correlated groups. Spines within about 5 microns showed coordinated tuning, forming small functional neighborhoods. Surprisingly, spines just beyond that range were less likely than chance to share the same tuning, suggesting these micro-clusters sharpen the neuron’s feature selectivity.
Apical versus basal dendrites: Apical dendrites extend long distances and sample a wider range of cortical inputs, while basal dendrites generally receive more direct visual drive. Although basal branches overall showed more raw visual input, apical dendrites on visually responsive neurons contained significantly more stimulus-responsive spines than apical branches on unresponsive neurons. Both branch types obeyed the proximity and local-clustering principles.
Orientation selectivity is the strongest predictor: Using statistical modeling, the researchers evaluated several candidate predictors—distance from the soma, branch type, response reliability, and stimulus selectivity. By far the most important factor in determining how tightly a spine’s activity matched the soma was the spine’s orientation selectivity: spines that were sharply tuned to a particular orientation most reliably aligned with the neuron’s output.
As the authors summarize, excitatory layer 2/3 neurons in mouse visual cortex organize synaptic inputs in a non-random way that correlates with somatic responsiveness and tuning, branch identity, distance from the soma, and local spine-to-spine correlations. Documenting these organizing rules provides a concrete baseline for understanding how cortical circuits integrate information and how genetic or developmental perturbations might alter that integration.
The paper’s findings can guide future modeling of synaptic integration and give researchers a reference point for studying disorders that affect circuit connectivity and visual processing.
Authors: Kyle R. Jenks, Gregg R. Heller, Katya Tsimring, Kendyll B. Martin, Asrah Rizvi, Jacque Pak Kan Ip, and Mriganka Sur.
Funding: This research was supported by the National Institutes of Health, the Simons Foundation Autism Research Initiative, and the Freedom Together Foundation.
Key Questions Answered:
A: Neurons in visual cortex receive thousands of inputs from different circuits. Some of those inputs carry non-visual information, or the neuron’s internal synaptic organization may prioritize non-visual signals even when a subset of its synapses can respond to light.
A: The team used genetic tools to make dendritic spines and somas fluoresce when calcium surged, a proxy for neural activity. Two-photon imaging captured these signals in real time as mice viewed visual stimuli.
A: Many neurodevelopmental conditions alter how neurons connect. By establishing normal organizational rules for synaptic arrangement, researchers now have a baseline to identify precisely how circuit assembly and function diverge in disease models.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The journal paper was reviewed in full.
- Additional context was provided by staff.
About this visual neuroscience research news
Author: David Orenstein
Source: Picower Institute at MIT
Contact: David Orenstein – Picower Institute at MIT
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Functional organization of dendritic spines in mouse visual cortex layer 2/3 neurons” by Kyle R. Jenks et al., iScience.
DOI: 10.1016/j.isci.2026.115861
Abstract
Functional organization of dendritic spines in mouse visual cortex layer 2/3 neurons
Cortical neurons receive heterogeneous excitatory synaptic inputs, yet the organizational principles governing their distribution across the dendritic arbor are not well understood. Using two-photon in vivo calcium imaging, this study examined how synaptic visual inputs to dendritic spines of layer 2/3 excitatory neurons in mouse visual cortex are organized both globally and locally.
In visually responsive neurons, sharply tuned spines located proximally to the soma showed higher somatic tuning correlation than distal spines. Responsive neurons had more visually responsive spines on apical dendrites and higher pairwise tuning correlations between spines than unresponsive neurons. While pairwise tuning correlations between spines did not vary significantly with distance from the soma across the population, spines on responsive neurons were locally clustered based on correlated tuning.
These findings indicate that visual inputs are distributed non-randomly across dendritic arbors and are organized in relation to somatic responsiveness, distance from the soma, branch type, and local spine-to-spine correlations.