Jason Ralph/Michel Brudzinski and Jinrong Li; Grad Student Presentations

 

Jason Ralph/Michel Brudzinski and Jinrong Li; Grad Student Presentations

Sage 4101

May 7, 2008 12:00 PM - 1:30 PM

Michel Brudzinski:

Title: Visual Similarity is ObViS

Abstract:

Visual similarity can be an easy perceptual judgment for humans.

However, while similarity can sometimes be described in terms of a small number of dimensions like color and size, often, real world often are similar in ways that are more difficult to describe explicitly. When searching for a particular object in a scene cluttered with real-world objects, the visual similarity of objects within the scene may be the best predictor of eye-movements. Thus, to model visual attention during visual search, it is important to develope a measure of the visual similarity of parts of a visual scene. Our research project, ObViS, is attempting to evaluate the similarity of parts of a visual scene using low levels features such as intensity, color, and orientation. Data analysis from our first experiment has just begun, but some initial results will be presented.

 

Jason Ralph:

Title: Cognitive Workload in Multi-modal, Dual task environments.

Abstract:

Previous research into the understanding of cognitive workload and the way in which multi-modal interfaces affect performance have yet to yield predictive measures of task demands.

Our research has a goal of developing a model-based predictive measure of the amount of workload elicited by a task. We began by studying the performance of participants in the Simple Navigation Task, a simple multi-modal dual task environment. Preliminary results indicate that when directions are presented visually, participants learn to time their eye-movements and over time perform as well as participants in a multi-modal (auditory) condition. I will discuss the results of our research and future plans for incorporating EEG data analysis and modifications of the task environment.

 

Jinrong Li:

Title: Status Report on the Reason Programming Environment

Abstract: There are many, many computer programming languages (PL).  Some are designed to teach general programming, independent of specific application areas; a prominent example in this category is Logo, invented to make programming easy to learn, and to make the learners in questions better thinkers.  But the results have ben profoundly mixed; the evidence is unconvincing.  Clear thinking is in short supply, and Logo doesn't appear to be a remedy.

 We introduced a new programming language/environment called Reason, invented by Bringsjord, firmly based on Logic, and specifically aligned with the core skills constituting part of clear thinking.  The talk will give brief background and motivation with respect to Reason, including a brief history and categorization of programming languages, and (brief) introduction to Logo.  We will discuss some of the main Reason features and future design and implementation plans.

 

Add to calendar
Share|