Eye Tracking in Human Language Processing
In another recent study to come out of the United Kingdom, researchers using eye tracking focused on human language and cognitively plausible models of language processing. Human language processing is considered “the ultimate gold standard for computational linguistics.” As humans can understand, interpret, and create language with speed and accuracy, as well as creativity, they’re able to deal with ambiguity and noise, adapting to new speakers, domains, and registers. They are able to achieve these levels of competency on the basis of limited training data, using learning algorithms that are largely unsupervised.
Thinking of humans as exceptional language processors is the idea behind the study, which worked with psycholinguistics, a discipline that studies human language processing as a source of information about design of efficient language processing systems. Computational linguists are often unaware of the literature that psycholinguists use, and so results from human language processing don’t often inform the design, implementation, or evaluation of artificial language processing systems.
By contrast, psycholinguistics is often oblivious of work in computational linguistics. In testing their theories, psycholinguists will construct computation models of human language processing. The models, however, fall short of engineering standards generally accepted in the computational linguistics community (standards like broad coverage, robustness, and efficiency).
The authors of the paper set out to propose a challenge, then, that would require the combination of research efforts in computational linguistics and psycholinguistics. They refer to this as the development of cognitively plausible models of human language processing. Once computational models of human processing language are available, they can be used to predict the difficulty that humans have when processing text or speech.
Cognitively Plausible Models of Human Language Processing
Related articles:
- Using Eye Tracking to Study Linguistic Annotation
- Eye Tracking: Challenging the Biological Monopoly
- Pupil Tracking in Airport Security: Can Body Language Indicate Terrorist Intent?
- Gaze Tracking Used to Improve Human-Robot Interactions
- Eye tracking study suggests movie subtitles as language learning tool