#	Pagina
attuale pagina	/open-h2020/projects/204826/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 2 - NEUROPOP (Large-scale analysis of neuronal population activity)

Teaser

Summary

What is the problem/issue being addressed?

This project is addressing the way large populations of neurons across the brain work together to process information.

Why is it important for society?

In the same way that a deep understanding of cell biology led to todayâ€™s rapid advances in treatment of cancer, a deep understanding how the brain functions will be essential to develop effective treatments for currently intractable disorders such as Alzheimerâ€™s disease. This project tackles the question of how neuronal populations compute information, one of the key questions that must be solved to achieve this objective. Additionally, discoveries made in how neurons process information will continue to inform design of artificially intelligent systems such as deep neural networks.

What are the overall objectives?

To characterize how large neuronal populations across the brain encode visual information, and the decisions that subjects make based on them.

Work performed

1. Overturning the textbook view that primary sensory cortex encodes primarily sensory information (Stringer et al Science 2019)
2. Demonstrating an quantitative regularity in neural population activity, revealing unexpected connections to pure mathematics (Stringer et al Nature 2019).
3. Characterizing the brain-wide correlates of Vision, choice, action, and engagement (Steinmetz et al bioRxiv 2018).

Final results

Leveraging new experimental technologies, these studies have recorded from larger neuronal populations that previously possible. This has led to revision of the textbook view of the function of visual cortex, and to unexpected connections with pure mathematics that may have important implications for the design of artificially intelligent systems.