Very excited to share my PhD project, out today in Neuron! Check out the thread below for my best attempt at a high-level summary.

RT @lisa_giocomo: New paper officially out! Congratulations to Mari Sosa - amazing work - and just before she sets off to UC Boulder to st….

RT @LabPluta: ~30 years of research in non-behaving animals argue that bilateral touch integration likely occurs downstream of primary soma….

RT @DavidCRowland: If you eat something and get sick, you shouldn't eat that thing again. Problem is, there is a long delay between when yo….

RT @NeuroCellPress: Online now: Network influence determines the impact of cortical ensembles on stimulus detection

RT @NatureNeuro: Lateral inhibition from parvalbumin neurons in mouse primary visual cortex reduces sensitivity to contrast; lateral inhibi….

RT @ZuckermanBrain: Every moment we remember is encoded by changes in the synapses connecting brain cells. Now, in a mouse study in @nature….

RT @eunji_kkong: “How does local circuit architecture shape spatial representations in the hippocampal CA3?”. ✨Excited to share my first po….

Thanks @SchmidtFellows for hosting such a fun week with lots of thought-provoking sessions. Can’t wait for the next one!.

RT @wjeffjohnston: When does modular structure emerge in neural networks?.What are the consequences of this structure for learning and beha….

Congratulations @ianOldenburg !! So well-deserved for a lab doing very exciting work.

RT @ShahabBakht: Very cool work. Quite compatible with our recent work in humans: We also found that additional t….

This argues that downstream regions summate V1 activity without preferentially weighting more informative neurons to make sensory decisions in this task. Why use a suboptimal strategy? Perhaps this simple read-out is more generalizable and robust in unpredictable natural environs.

We found that impact of an ensemble on network activity, both through direct activation and the ensemble impacts on other local neurons, was strongly predictive of the behavioral impact. But the visual sensitivity of the ensemble was not related to behavioral effect (5/6)

If mice weight more informative, visually-sensitive neurons more strongly, then sensitive ensembles should have a larger effect on the visual detection performance. If an unweighted strategy is used, the change in overall population activity should determine effect (4/6).

Do mice use this more optimal, weighted read-out strategy? To test this, during task performance, we used 2-photon optogenetics to specifically stimulate ensembles of neurons that were either sensitive to the visual stimulus or random (mostly non-responsive) (3/6)

We trained mice to detect a grating of varying contrast and recorded V1 activity. V1 responds monotonically to contrast, so an unweighted mean decoder can detect stimuli. But a weighted decoder that weights more visually sensitive neurons outperforms the mean decoder (2/6)

How is V1 activity read-out to determine the presence of a visual stimulus? Our study using targeted 2-photon optogenetic stimulation in a detection task found that mice use a simple unweighted read-out strategy (1/6)

RT @HistedLab: Cool preprint; we just did a fun lab meeting on it. 🧠🧪. The core result is: Changing neural variability in V1 by injecting b….

RT @quintanad06: Timing is everything in life… and in the brain! Many theories suggest that the precise communication of sensory informatio….