Welcome to the Computational Vision and Learning Lab. This is the web site for Hongjing Lu's lab in the UCLA Psychology Department.

How do we guess right? We are not always lucky enough to get complete and sufficient information from the real world to make a judgment. For example, we have to infer a three-dimensional world from two-dimensional retinal images in visual perception; we have to make critical causal decisions based upon a few observations. The basic goal of our research is to investigate how humans make inferences from sparse and ambiguous data. More specifically, we aim to understand the representation and processing of visual and causal information that we acquire in everyday life. Although visual perception and causal reasoning are two distinct domains, the picture emerging from our work is that a key basis for human inference is heuristic priors — tacit general assumptions people make about the way the world works, which then guide their learning and inference from observed data.

Our approach includes the development of computational models to compare human behavior with model performance, and the use of empirical predictions derived from models to guide the design of experimental tests of perceptual and cognitive theories. Current areas of active study include: motion perception and biological motion, object recognition, causal learning, and diagnostic and analogical reasoning.

The Computational Vision and Learning (CVL) Lab

Franz 6550, Psychology Department, UCLA, Los Angeles, CA, 90095

Lab phone number: 310-206-4187