News,
A prevailing hypothesis in neuroscience is that individual neurons are selective for a function (Hirokawa et al., 2019), specialised for a particular task. However, more and more papers are published invoking differential functions and behavioural effects for the same neuron types. The categorical selectivity may not be the rule, but an exception, according to a recent study.
In July 2026, Posani and colleagues combed through the activity of more than 14,000 neurons scattered across 43 cortical areas in mice, drawn from the International Brain Laboratory’s openly released Brainwide Map dataset. They asked: is each neuron a specialist (devoted to one variable, like a hammer) or a generalist (responding to many, like a Swiss Army knife)?
They found that selectivity for function was primarily related to anatomical location, not to individual neurons. Within the categorical structure, neurons were generalists and could respond to many different behavioural events. However, the neuronal spiking responses remained distinguishable: a neuron’s response pattern, the authors showed, predicted its anatomical location, or brain module.
This could mean that functional information is processed at the population and circuit level, rather than by individually specialised neurons. The result lands in the middle of a long-running argument: some studies have described tidy, categorical groupings of neurons in frontal cortex, while others found category-free, thoroughly mixed responses in parietal and reward-related areas (Raposo et al., 2014). What Posani and colleagues add is scale: evidence from the entire mouse cortex pointing toward the mixed camp. This paper throws the field into an intellectual spin, especially for those searching for specific functions for specific neuron types. However, this does not mean that there aren’t any neuronal categories or cell types with differential, innate behaviours, morphological signatures and chemical orchestras. It just indicates that we should be careful about attributing behavioural functions to individual neurons or neuron categories, and perhaps that the neat categories we find sometimes owe more to the entanglement of the brain’s networks and our statistical methods.
Notably, the article drew a lot of attention within weeks of publication, a signal of how much the field needs this question settled.
The study has a few limitations, however. An interesting question is whether the generalist behaviour holds for both principal cells and interneurons. The authors themselves take this up in their discussion: the brain contains many distinct neuronal classes, inhibitory ones especially, which would seem like prime suspects for such functional groupings. Yet one explanation they offer for the absence of clusters is that the boundaries between cell types simply aren’t mirrored in how the cortex parcels out functions. Furthermore, the mice faced a fairly simple task, one they were drilled on extensively. If task complexity increases, how would that affect the results?
Whatever the mechanism turns out to be, the study’s broader lesson seems clear: the brain’s functional categories live in its large-scale organisation, not in the interiors of its individual regions.
Declaration of AI use: This article was written with the assistance of Lumo AI for grammar, spell checking, and proofreading.
References
Hirokawa, J., Vaughan, A., Masset, P., Ott, T., Kepecs, A., 2019. Frontal cortex neuron types categorically encode single decision variables. Nature 576, 446–451. https://doi.org/10.1038/s41586-019-1816-9
Posani, L., Wang, S., Muscinelli, S.P., Paninski, L., Fusi, S., 2026. Rarely categorical, highly separable representations along the cortical hierarchy. Nature. https://doi.org/10.1038/s41586-026-10668-4
Raposo, D., Kaufman, M.T., Churchland, A.K., 2014. A category-free neural population supports evolving demands during decision-making. Nat Neurosci 17, 1784–1792. https://doi.org/10.1038/nn.3865
Leave a Reply