MIT scientists have just released a paper detailing the potential for a new eye-tracking technology to gauge students’ reading proficiency. Study authors are hopeful this novel technology will help English as a Second Language (ESL) teachers better gauge how well their students understand reading assignments in the near future.
Researchers used data from 145 ESL students and over 35 native English speakers in this study. The ESL students spoke one of these four languages as their mother tongue: Japanese, Chinese, Portuguese, or Spanish.
After placing a camera on or near these students’ eyes, study authors gave each of them 156 sentences to read. Half of these sentences were exactly the same on everyone else’s tests. Scientists then examined the amount of time it took for each student to read the various sentences and compiled this data in what they called an “EyeScore.”
Once they had their EyeScores clearly listed, investigators compared the results from their eye-tracking study with more standardized tests like the Test of English as a Foreign Language. According to their results, the EyeScore showed similar results to the other tests, which proves this technology has the potential to be used as a fluency test.
Linguistic professors around the world are excited with the findings in this study. They hope the EyeScore will soon help uncover how reading affects the neurocircuitry of the brain and perhaps uncover how the brain evolved to be able to read and write in the first place.
Yevgeni Berzak, who is currently a postdoc in Computational Psycholinguistics at MIT, was the lead author on this study. A few other key researchers include Drs. Roger Levy and Boris Katz.
This research will be released in the next gathering of the Association for Computational Linguistics. Researchers entitled their work, “Assessing Language Proficiency from Eye Movements in Reading.”