In a study published online this week in Proceedings of the National Academy of Sciences, Michael Spivey, a psycholinguist and associate professor of psychology at Cornell University, tracked the mouse movements of 42 undergraduate students while working at a computer.
Students heard a word--such as "candy"--and were then shown two pictures. If the pictures were of different sounding objects--such as "candy" and "ziggurat"--the mouse moved in a straight line to the candy and clicked on it. If the words for the pictures sounded similar--"candy" and "candle"--they were slower to click on the correct answer, and the mouse trajectory was more curved. This indicates that, when faced with ambiguity, humans study what limited data they have before clicking.
Under the old metaphor, one would have expected subjects to rush to one solution and then correct the answer if they had chosen wrong.
Interestingly, the whole field of artificial intelligence has moved from a Boolean model, in which systems guide themselves through a series of embedded rules, to amodel, in which machines guide themselves by studying past experiences. Bayesian probability also underlies search engines.
"In thinking of cognition as working as a biological organism does, on the other hand, you do not have to be in one state or another like a computer, but can have values in between--you can be partially in one state and another, and then eventually gravitate to a unique interpretation, as in finally recognizing a spoken word," Spivey said.