Paper Reading #25: Using Language Complexity to Measure Cognitive Load for Adaptive Interaction Design

Commentary

See what I have to say about ___'s and ___'s work.

References

Khawaka, M. A., Chen, F., and Marcus, N. (2010). Using language complexity to measure cognitive load for adaptive interaction design. Proceeding of the Acm conference on intelligent user interfaces. . Hong Kong: http://www.iuiconf.org/.

Article Summary

The researchers addressed the problem of adaptive interfaces in this paper, specifically adaptive interaction design. They proposed a method of controlling the level of adaptation of an interface via the monitoring of speech patterns. Different patterns were mapped to different levels cognitive load, or how much of the brain's processing power is being used to compute a task. Current speech recognition capabilities make real-time implementation infeasible; instead, the researchers performed quantitative analysis on transcriptions of several training exercises at an Australian bushfire management facility. They measured semantic difficulty, or the use of words, and syntactic complexity, or sentence length. Their hypothesis proved correct save for one prediction. Using statistical analysis, they were able to apply different language complexity measures to accurately map sentences used in different situations to the appropriate level of cognitive load for that situation.

For lack of a more salient picture, here is a table of their results.

As speech recognition capabilities become more advanced, they hope to test their implementation in a real-time situation.

Discussion

I really enjoyed this paper. Sometimes, I think I should have gone into English, specifically linguistics. Don't get me wrong: I like to chill out a little bit in this here discussion section, but I feel that crafting a literary piece or a technical document is akin to creating a work of art. Languages are so easy to use in a practical sense, but they can be wonderfully complex if they are truly understood. These researchers understand language (or at least how to analyze it). What's more is that they understand the importance of language and communication. Once speech recognition is up to par with the vision that these researchers have laid out, our interactions with our machines will be as seamless as our interactions with each other. Technology is a tool, and nothing more, but capable, efficient tools only make sense; why would we not strive to reach the limits of our potential to create and design?