When processing degraded speech, highly time-compressed speech, or when listening to speakers with foreign accents, people need some time to adapt or adjust to this speech before it becomes fully intelligible. In this sense, adaptation is a higher order phenomenon: people learn to make certain transformations or criterion shifts, which can be exemplified by the shifting of phoneme boundaries.
When a subject repeatedly hears a stimulus at one end of a speech continuum (e.g. /da/ from a /da/ to /ta/ continuum), and is then presented with the randomised stimuli from the /da/ to /ta/ continuum, the boundary between the two phonemes is shifted towards the /da/-end of the continuum. After the presentation of so many /da/-sounds, the subject identifies more stimuli as /ta/-like (Rosen, 1979). However, the term adaptation is also used to cover lower order (physiological) phenomena, such as auditory fatigue. Fatigue results from the application of a stimulus which is usually considerably in excess of that required to sustain the normal physiological response of the receptor, and it is measured after the stimulus is removed (Hood, 1972). One could measure a subject’s hearing threshold at a certain frequency, and then present the subject with a fatiguing tone of a certain frequency and intensity for a period of time. The threshold could then be measured again, and the shift in threshold would be the measure of fatigue. The recovery of this post-stimulatory auditory fatigue may be quite rapid. If one assumes that feature detectors play a role in speech discrimination and classification tasks, the phoneme boundary shift resulting from repeated presentation of one of the stimuli could be explained by the fatigue of certain feature detectors rather than as a criterion shift (Eimas & Corbit, 1973).
- Eimas, P.D. & J.D. Corbit. 1973. Selective adaptation of linguistic feature detectors. Cognitive Psychology 4, 99-109.
- Hood, J.D. 1972. Fundamentals of identification of sensorineural hearing loss. Sound 6, 21-26
- Rosen, S.M. 1979. Range and frequency effects in consonant categorization. Journal of Phonetics 7, 393-402.