In DepthNeuroscience

Brain scientists dive into deep neural networks

See allHide authors and affiliations

Science  21 Sep 2018:
Vol. 361, Issue 6408, pp. 1177
DOI: 10.1126/science.361.6408.1177

You are currently viewing the summary.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution


At the Conference on Cognitive Computational Neuroscience this month, researchers presented new tools for comparing data collected from living brains with readouts from computational models known as deep neural networks. Such comparisons might offer up new hypotheses about how humans interpret sights and sounds, understand language, or navigate the world. Until recently, artificial intelligence couldn't come anywhere close to human performance on tasks like recognizing sounds or classifying images. But deep neural networks, loosely inspired by the brain, have logged increasingly impressive performances, especially on visual tasks. That has led some neuroscientists to wonder whether these models could yield insight into how our own brains process information.