NSCS 344, Week 1

Modeling a simple perceptual decision

Perceptual decisions

For our first model, we are goint to consider a relatively "simple" type of perceptual decision. As simple as this is, and as simple as the simplest models are, decisions like these and the models we use to describe them are a mainstay of mathematical psychology and computational neuroscience and have been applied to everything from memory retrieval to economic decisions.

Which way are these dots moving?

The particular decision we will focus on involves moving dots, an example is below. This stimulus consists of a bunch of dots that flicker in to and out of existence. When each dot appears it moves in a "semi-random" direction. Most of the time it moves in a preferred direction (in this case to the left), but some of the time it moves in a non-preferred direction (in this case to the right).
randomDots_50_loop.gif
The job of the participant is to look at this stimulus and report the preferred direction that most of the dots are moving in. For this simple stimulus, the decision is fairly easy. But the beauty of these stimuli is that we can control the difficulty by changing the fraction of dots moving in the preferred direction f. For example, by setting we can get a really easy decision ...
loopingImage2.gif
And by setting we can get a decision that is more difficult.