What is it about?

In today's highly digitized and networked world, people interact with computer systems of different kinds (e.g., personal computers, laptops, mobile devices, IoT devices, video games, online social media) all the time. To help researchers and user interface / experience (UI/UX) designers to develop more usable computer user interfaces (UIs), cognitive modeling software tools have been widely used, CogTool being one of them. Existing cognitive modeling tools have limited features supporting large-scale modeling and simulation tasks, especially for UIs involving time-varying and programmatically controlled components. One example of such complicated UIs are user login systems we are using everyday: they often have dynamically created components and need to programmatically check correctness of the user's input to determine the next UI. This paper proposes a general software framework for enabling more powerful cognitive modeling tools, and describes a research prototype we developed -- CogTool+. A pedagogical example and two real-world user-authentication systems are used to illustrate how CogTool+ can be used to benefit UI/UX researchers and designers. The source code of CogTool+ is made publicly available at https://github.com/hyyuan/cogtool_plus.

Featured Image

Why is it important?

The proposed software frameworks and research prototype add important features to support large-scale cognitive modeling and simulation of human performance tasks, including (1) a higher level of parameterization and automation; (2) algorithmic components; (3) interfaces for using external data; and (4) a clear separation of tasks, allowing programmers, psychologists and UI/UX researchers and designers to work more effectively together. Our experiments showed that CogTool+ can help reduce human efforts on large-scale modeling tasks significantly.

Perspectives

The most important perspective from me as a co-author is the generalizability of the proposed framework and the prototype. The way how CogTool+ works is much like today's web programming: HTML, CSS, JavaScript and different supporting tools allow complicated web development tasks done by UI/UX designers (who often don't know much programming), programmers, and content creators to work together. I hope to see the cognitive modeling ecosystem can grow in the same way like the web, including possible webification of cognitive modeling tools.

Professor Shujun Li
University of Kent

Read the Original

This page is a summary of: CogTool+, ACM Transactions on Computer-Human Interaction, April 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3447534.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page