RESULTS
Comprehension scores were analyzed using mixed-effects regression and pairwise comparisons on the estimated marginal means. In study 1, the comprehension difference between text or hard-copy braille and the voice actor formats was assessed to address the first aim. Sighted participants had better comprehension with text (mean, 74.8%; 95% confidence interval [CI], 70.5 to 79.1%) than with a voice actor (mean, 69.7%; 95% CI, 65.4 to 74.0%; P = .02), and blind participants had superior comprehension with hard-copy braille (mean, 70.4%; 95% CI, 63.3 to 77.5%) than with a voice actor (mean, 61.9%; 95% CI, 54.7 to 69.0%; P = .03). In study 2, the comprehension differences among blind participants between the four formats were investigated to address the second aim. Comprehension was better with hard-copy braille (mean, 70.6%; 95% CI, 63.4 to 77.7%) than with a screen reader (mean, 60.7%; 95% CI, 53.5 to 67.9%; P = .02) and better with a braille display (mean, 69.7%; 95% CI, 62.5 to 76.9%) than with a screen reader (P = .04).
CONCLUSIONS
Study 1 supports the hypothesis that more physically engaging tasks enhance comprehension, and study 2 suggests that listening to scientific materials using a synthesized voice may reduce comprehension ability compared with hard-copy braille and braille displays.