| Sumario: | Among listeners with normal pure-tone sensitivity there is considerable variation in spectral and temporal discrimination abilities, as measured with nonspeech sounds. However, contrary to theories that associate deficits in auditory processing with degraded speech perception, individual differences in a battery of measures of spectral-temporal acuity for nonspeech sounds have little or no relation to individual differences in speech recognition under difficult listening conditions. Based on data collected with groups of over 500 college students and 465 first-graders, individual differences in speech processing are dependent on neither basic auditory discrimination abilities, as measured with a wide variety of nonspeech test stimuli, nor on general cognitive abilities, as reflected in IQ and SAT test scores, or college grades. In a related finding, speech-recognition skills among young children are not predictive of their academic achievement, including reading, in the first two years of elementary school. Other cognitive and intellectual abilities do, however, predict the academic accomplishments of the same children. It is proposed that speech recognition largely depends on two cross-modality mechanisms: an ability to recognize linguistic messages on the basis of fragmented information, presumably by the use of contextual and linguistic constraints, and a general ability to recognize familiar patterns.
|
|---|