Intriguing little snippet in this weeks New Scientist about how Google’s search alogorithm may mimic the way that the human brain retrieves information. A study by Psychologist Tom Griffiths at The University of California, Berkeley, tested Pagerank’s methodology (of ranking sites by the number of sites that link to them) against other word-retrieval methodologies and found that it most closely matched the human model.
It seems that the way in which the brain stores our knowledge of words is very like a network, with different words represented as individual nodes, linked to other nodes that relate to it. Huge advances in the way in which humans can interact with computers have already been made – perhaps one day we might see Google launching a product that extends it’s search capabilities to our own brains? Scary thought.