Wednesday, September 7, 2011

Reaction: Attention and Visual Memory in Visualization and Computer Graphics

The main objective of the paper is to shed light on visual perception .i.e., how people look at an image and what they learn from it. The paper performs deep analysis of what is known as "preattentive processing" - data learned from a small portion before an eye movement occurs. The paper states that what we see depends on what is on our minds. But human mood is known to change, does changing mood affect the way we see things? The paper presents several models on preattentive processing which are all hypothesis. It would have been helpful if experiments were stated on these models to give them credibility.

In one of the models by Treisman, the properties that were preattentive were determined. But the paper does not state the method in which they were determined. I was left wondering if this was done scientifically or through a survey. One feature described in the paper, ensemble coding, was particularly interesting. It showed that it was easier to compute averages than remembering specific details. The amount of information present in the paper is staggering. A lot of concepts had correlation to each other which was not evident in the first reading.