Improving Reading Skills and Facilitating Web Browsing for Students with Learning Disabilities using Text-to-Speech Software

– Results of Studies in the U.S. and Japan –
Implications for Future Joint Research

Center on Disability Studies, University of Hawaii-Manoa:
Kiriko Takahashi; Kelly Roberts; Hye-Jin Park

Accessibility Center, Tokyo Research Laboratory, IBM Japan:
Reiko Nagatsuma; Shinji Iizuka; Takuya Ohko; Masakazu Takizawa; Takuji Wada; Takashi Saito

Two studies were highlighted on the use of Text to Speech (TTS) software by individuals with learning disabilities (LD) and discuss implications for future joint study by the U.S. and Japan. The empirical research conducted in Hawai`i looked at the use of TTS software as a tool for improving unaided reading comprehension and vocabulary of high school students with LD. The study conducted in Japan analyzed the effectiveness of added features to the existing IBM Easy Web Browsing (EWB) software by individuals with LD.

Many poor readers, who have LD, exhibit slow and effortful word decoding (Lundberg, 1995; Torgesen, 1998). Such “lower-level” decoding problems lead to poor comprehension of written materials. This is a definitive area of concern for students with LD both in the U.S. and in Japan.

TTS software, a computer based assistive technology, has become a familiar tool to access print for students with visual impairments in many countries. In fact EWB is one of the many TTS software initially developed for people with low vision and senior citizens. Now, TTS has become an emergent tool for students with LD, AD/HD, and other cognitive disorders. Several studies in the U.S. and in Europe have attested to the immediate impact of TTS use on reading comprehension and skills, but how well the impact is sustained for students with LD has not been thoroughly investigated. In Japan, this type of study on TTS is still limited.

In Hawai`i TTS study, data (Woodcock-Johnson III scores and independent use of the software) was collected at three time points over one academic year in two high schools that have different demographic, economic, and academic standings. The data was analyzed by general linear model-repeated measures using SPSS, and orthogonal polynominal transformation was chosen in transforming the within-subject factors into linear and quadratic variables to examine a trend of the observed measures. Results of this study showed TTS software has significant influence on unaided vocabulary building, but not on reading comprehension.

In Japan study, qualitative study was conducted on the effectiveness of EWB functions. One-on-one interviews were conducted with individuals with LD based on trial use of the EWB software. The results of the interviews were analyzed and matched with specific processing deficits. The functions found most helpful by individuals with LD were reading out loud, line ruler, colored overlay, and improved line spacing for both English and Japanese text. Specific to Japanese text, the furigana function and function to add spaces between words were identified as most helpful in comprehension and readability of the text.

One possible joint research is to implement EWB software for students with LD in Hawai`i and compare whether specific functions identified as most effective by individuals with LD in Japan will also be found most effective by students in Hawai`i. Data collected can also be analyzed to see whether use of one specific function improves vocabulary and reading comprehension more than other functions.

References

Lundberg, I. (1995). The computer as a tool of remediation in the education of students with reading disabilities-A theory-based approach. Learning Disabilities Quarterly, 18, 89-99.

National Assessment of Educational Progress (2007).　 Nation’s report card. Retrieved June 22, 2008, from http://nces.ed.gov/nationsreportcard/

Torgesen, J. K. (1998). Catch them before they fall: Identification and assessment to prevent reading failure in young children. American Educator, 22, 32-39.

Uno, Akira et al.Structure of cognitive impairment for dyslexia. The Japan Journal of Logopedics and Phoniatrics, 48, (2) 105- 112.