Speech-language pathologists in the Department of Speech, Language, and Hearing Sciences at the University of Connecticut report results of a pilot study using the Language Environment Analysis technology (LENA) to capture and analyze real-world samples of language use from a person with aphasia (PWA) before and after intensive language therapy. The authors goals were:
 to determine the feasibility of collecting and analyzing natural language samples occurring spontaneously within the home on 2 days separated by a period of therapy;  to scrutinize the resulting analyses of language use for occurrences of reportable, clinically meaningful quantitative differences.
The authors primary focus is on use of the LENA technology for aphasia research, with a secondary focus on outcome changes per se following use of Schuell’s stimulation therapy. The PWA here wore a 3.5”x2.5” LENA unit containing a digital recorder that captures vocalizations and the surrounding language environment within a 4-6 foot radius. Individual recording sessions last up to 16 continuous hours. These recordings are algorithmically post-processed by specialized software to identify speech events and speakers, types of conversations, social context, and mood. The PWA in this study was provided with two such units, instructed in their use, asked to activate and wear one of those units for a day preceding therapy, and do the same with the other unit for a day following intervention. The researchers also conducted manual analysis of the recordings to tally adult word counts and conversational turns, for comparison with the numbers produced algorithmically by the LENA software. To document outcome changes, the authors administered the Western Aphasia Battery-Revised (WAB-R), the Auditory Comprehension Test for Sentences (ACTS), and the Communication Effectiveness Index (CETI) at start and end.
Data analyses document that post-treatment improvements in real-world language use were indeed present in the subject’s behaviors, captured on the recordings, and revealed by the language analysis software algorithms: after treatment, the subject engaged in more conversational interactions, with more conversation partners, with higher adult word counts than before. Outcome analyses showed a corresponding significant improvement in WAB-R AQ scores, and on 5 of the 16 CETI items, though not on the ACTS. There was a low degree of agreement, however, between manually produced tallies and those produced algorithmically. Overall, the article reports a promising start to an approach that should help elucidate spontaneous language uses by PWA.
For further reading: L. B. Suting and Jennifer Mozeiko, 2021,
Analysis of real-world language use in a person with Wernicke’s aphasia.
ASHA SIG-2 Perspectives, 6(3): 553-565