Sign In

Communications of the ACM

ACM TechNews

The Brain's Crowdsourcing Software

View as: Print Mobile App Share:
A large crowd.

Neuroscientists have found that groups of cells in the brain cooperate to produce specific results, much like the way in which crowdsourcing software works.

Credit: James Cridland/Flickr

Neuroscientists are discovering the complex interactions of the human brain and how groups of cells cooperate to produce certain results, in a manner similar to crowdsourcing software.

Columbia University's Stefano Fusi, the Massachusetts Institute of Technology's Earl K. Miller, and their colleagues this year released a study in which they taught monkeys to remember and respond to a particular shape rather than another while brain activity was recorded. The team recorded the activity of multiple prefrontal neurons simultaneously, instead of focusing on each neuron individually. Perplexing "mixed selectivity" patterns emerged, in which one neuron might respond when the monkey remembered just one shape or only when it recognized the shape but not when it recalled it, while a nearby cell displayed a different pattern.

Borrowing concepts from computer scientists, the researchers discovered the monkey's brain relied on the same general technique that Google uses for its search algorithm. Instead of ranking search results by choosing a few features of each Web page, Google integrates the numerous, idiosyncratic choices of individual users to produce better results. If neurons focus on only limited features, the brain is restricted to capturing those features and combinations of features. The brain understands more complex patterns by integrating information from many different neurons with varying response patterns, in a method similar to crowdsourcing.

From The Wall Street Journal
View Full Article - May Require Registration


Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account