Effects of simulated cataracts on speechreading
Authors
Advisors
Issue Date
Type
Keywords
Citation
Abstract
Watching a speaker‘s face can improve a listener‘s speech understanding, especially at poorer signal-to-noise ratios (SNRs). Little is known, however, about the effects of visual impairments on speechreading. In a series of studies, young adults‘ visual enhancement to speech intelligibility under normal vision and simulated cataract vision was tested. In Study 1, speech intelligibility was tested while Central Institute for the Deaf Everyday Speech Sentences were presented via live-voice at a fixed -13 dB SNR under normal vision and mild cataract conditions. In Study 2, speech intelligibility was tested while Speech in Noise (SIN) Sentences were presented via high luminance, recorded talker at SNRs ranging from 0 to -21 dB under normal vision and moderate-to-severe cataract vision. In Study 3, speech intelligibility was tested while SIN Sentences were presented via natural luminance, recorded talker at SNRs ranging from 0 to -21 dB under normal vision and simulated mild cataract vision. In Study 4, speech intelligibility was tested while SIN Sentences were presented via natural luminance, recorded talker at SNRs ranging from 0 to -21 dB under normal vision and simulated severe cataract vision. In Study 5, speech intelligibility was tested while SIN Sentences were presented via recorded talker at eight luminance levels using neutral density (ND) filters ranging from 0 to 4.2 ND at .6 steps under normal vision and simulated mild cataract vision. In Study 6, speech intelligibility was tested while SIN Sentences were presented via recorded talker at eight luminance levels using neutral density filters ranging from 0 to 4.2 ND at .6 steps under normal vision and simulated severe cataract vision. Participants‘ ability to use visual information to support speech understanding was significantly reduced under simulated mild cataracts and was nearly eliminated under simulated severe cataracts. This effect was observed under natural levels of luminance of the talker‘s face and was mitigated by high levels of luminance.