AI, Communication, and Accessibility: A Scoping Review on Deaf/Hard-of-Hearing Learners with Learning Disabilities

• In Event: AI for Equity, Access, and Innovation in Postsecondary Writing and Learning Contexts

Abstract

Communication for individuals within the Deaf/Hard of Hearing population (DHH) has been investigated for centuries. ChatGPT, as an assistive technology for the Deaf and Hard of Hearing population, may improve accessibility opportunities; however, little empirical data are available. We provide a discussion of opportunities and constraints for the field.

Literature Review:
The communication use of individuals within the Deaf/Hard of Hearing population (DHH) has been researched for decades; however, little is known about individuals who are DHH and also have learning disabilities (LD) (Dostal et al., 2017; Mauk & Mauk, 1992; Wiley, 2012; Wolf Nelson & Crumpton, 2015). Notably, more literature is needed on the communication impact for individuals who are DHH/LD. The impact of Artificial Intelligence (AI) is increasing globally. Specifically, research on the use of AI in the DHH/LD population is limited (Papastratis et al., 2021). Thus, the importance and impact of AI on individuals who are DHH/LD warrant further investigation.

Methods:
Key terms will be identified for a scoping literature review on the topic of DHH/LD, AI, and the use of communication for this population. From selected databases (APA PsycInfo, Academic Search Premier, Communication & Mass Media Complete, Education Source, ERIC), a search using the key terms “deaf OR ‘hard of hearing’ AND ‘learning disabilities’ AND ‘artificial intelligence’” revealed only one article (Wang, 2017).

Another search including communication was conducted: (“deaf” OR “hard of hearing” AND “learning disabilities”) AND “communication” AND “artificial intelligence” yielded 47 preliminary results using the aforementioned databases. Articles will be screened for: a) population of DHH AND LD, b) use of communication (i.e., American Sign Language, verbal communication), and c) use of artificial intelligence.

Results:
A preliminary search revealed limited articles on DHH/LD and AI. Thorough results will be completed by the time of the presentation. In the event that a true intersection of research for individuals who are DHH/LD is not identified, the populations will be searched separately (i.e., one search on individuals who are DHH with AI; another search on individuals with LD and AI).

Discussion and Implications:
More research should be conducted on the intersection of DHH/LD and the use of AI for communication. The field acknowledges the need for more information on individuals who are DHH/LD in order to best teach, treat, and enhance communication. Nevertheless, research remains lacking in guidance on how to best address the needs of this population.

Author

• Juliana Hirn, University of Central Florida