Psychology Colloquia

Abstract:: While several recent papers have argued that human perception and motor control can be understood in terms of rational statistical inference, the question of whether similar analyses apply to human cognition remains controversial. I will present analyses of two kinds of "everyday inductive leaps" - inductive inferences that people make every day, from small amounts of evidence. The first of these inferences is predicting the future, in which people predict the total duration or extent of a quantity based on its current value. The second is detecting coincidences, a cognitive capacity that is often used as an example of human irrationality. I will use a detailed investigation of these two aspects of everyday cognition to argue that human inductive inferences can be guided by accurate probabilistic knowledge of the world and a well-calibrated sense of statistical evidence.
About the Speaker:: People solve challenging computational problems every day, making predictions about future events, learning new causal relationships, or discovering how objects should be divided into categories. My research investigates how this is possible, first identifying the nature of the underlying computational problems, and then examining whether we can explain aspects of human behavior as the result of approximating optimal solutions to those problems. Since many of the problems people face in everyday life are problems of induction, requiring inferences from limited data to underconstrained hypotheses, these optimal solutions draw on methods developed in statistics, machine learning, and artifical intelligence research. Exploring how these methods relate to human cognition provides connections between these fields and cognitive science, as well as a way to turn insights obtained from studying people into new formal techniques.

For More Information About This Speaker:

http://psychology.berkeley.edu/faculty/profiles/tgriffiths.html
Dr. Griffiths's website.

University of California, San Diego
Homepage Calendar Directions Past Events Psych Homepage UCSD Homepage	The Department of Psychology is Honored to Present a Talk by Tom Griffiths University of California, Berkeley "Everyday Inductive Leaps: Making Predictions and Detecting Coincidences" Presented on May 10, 2007 Location: The Crick Conference Room Mandler Hall, room 3545 Abstract: While several recent papers have argued that human perception and motor control can be understood in terms of rational statistical inference, the question of whether similar analyses apply to human cognition remains controversial. I will present analyses of two kinds of "everyday inductive leaps" - inductive inferences that people make every day, from small amounts of evidence. The first of these inferences is predicting the future, in which people predict the total duration or extent of a quantity based on its current value. The second is detecting coincidences, a cognitive capacity that is often used as an example of human irrationality. I will use a detailed investigation of these two aspects of everyday cognition to argue that human inductive inferences can be guided by accurate probabilistic knowledge of the world and a well-calibrated sense of statistical evidence. About the Speaker: People solve challenging computational problems every day, making predictions about future events, learning new causal relationships, or discovering how objects should be divided into categories. My research investigates how this is possible, first identifying the nature of the underlying computational problems, and then examining whether we can explain aspects of human behavior as the result of approximating optimal solutions to those problems. Since many of the problems people face in everyday life are problems of induction, requiring inferences from limited data to underconstrained hypotheses, these optimal solutions draw on methods developed in statistics, machine learning, and artifical intelligence research. Exploring how these methods relate to human cognition provides connections between these fields and cognitive science, as well as a way to turn insights obtained from studying people into new formal techniques. For More Information About This Speaker: http://psychology.berkeley.edu/faculty/profiles/tgriffiths.html Dr. Griffiths's website.
Researchers and the general public are both welcome to attend the Psychology department's colloquia. Reservations are not required, and admission is free. If you have any questions regarding the department's colloquium series, then please write to colloquia@psy.ucsd.edu