Search

Improvements in performance on many perceptual skills can occur with only a single training session. Of interest here is what aspects of the training experience are being learned during this brief exposure. Although there is considerable evidence that learning associated with specific feature values of the stimulus used in training (stimulus learning) contributes to these rapid improvements, there has been little direct investigation of the possibility that other types of learning do so as well. Here the authors show that not only stimulus learning but also learning of more general aspects of the training experience (procedure learning) contributed to rapid improvements in performance on interaural time difference discrimination. However, practice on the type of judgment to be made did not appear to aid performance (no task learning). These results are consistent with physiological reports that different neural mechanisms are engaged at different time points during even a brief training session, and imply that the circuits that are engaged and likely modified vary in the degree of their selectivity to the target condition. Such changes presumably enable further learning.

Training‐induced improvements on perceptual skills can be enhanced by increasing the amount of training and by resting between training and testing. However, how these two factors affect the generalization of learning from a briefly trained condition to an untrained one is unknown. Here, listeners were trained on an interaural‐level‐difference (ILD) discrimination condition (4‐kHz tones), then were tested for generalization to an interaural‐time‐difference (ITD) discrimination condition (0.5‐kHz tones). The amount of training and the time between training and testing differed across four groups. Listeners tested 10 hours after training had significantly lower ITD discrimination thresholds than naive listeners (n=94), regardless of whether training lasted for 20 min (n=14) or 2 hs (n=11). Thus, when there was a long time between training and testing, learning generalized from ILD to ITD discrimination regardless of the amount of training. In contrast, listeners tested immediately after training on ILD disc...

Talkers typically hyperarticulate when producing individual speech segments under conditions of sentence‐level, or narrow focus and of clear speech, which is speech produced in response to known speech perception difficulties on the part of the listener. This study investigated whether narrow focus and clear speech elicit similar changes in the speech signal. A comparison was conducted between the effects of these two phenomena on the durations of voiceless stops /k/ and /t/ in word‐initial and word‐medial positions of trochaic, disyllabic words. Both narrow focus and clear speech yielded longer closure and aspiration durations when /k/ and /t/ were in word‐initial position. For word‐medial stops, narrow focus yielded no change in closure duration and a lengthening of aspiration duration for both /k/ and /t/. In contrast, clear speech yielded no change (/t/) or an increase (/k/) in closure duration, and, more interestingly, significant decreases in aspiration duration for both /k/ and /t/. The different e...

Determining the training regimes that do and do not yield improvements in auditory discrimination provides information about the mechanisms underlying the trained task, and about learning itself. In a previous report, training with one standard stimulus yielded improvements in temporal‐interval discrimination that generalized to an untrained frequency at the trained interval [B. A. Wright et al., J. Neurosci. 17, 3956–3963 (1997)]. Here, the impact of training interval discrimination with two standard stimuli, as opposed to one, was investigated. Listeners were trained either with two standard intervals (100 and 350 ms) at one frequency (1 kHz; n=8), or with one standard interval (100 ms) at two frequencies (0.25 and 1 kHz; n=8). On average, listeners in both groups (1) learned less on each trained condition than listeners trained with only one stimulus (n=10), and (2) subsequently performed on the trained interval(s) at an untrained frequency (4 kHz) as though they had trained with only one of the stimuli—in one case, showing no improvement. These results indicate that, at some stage, temporal intervals are processed at frequency‐specific sites, and that training one task with two stimuli sharing all but one feature can interfere with both learning and generalization. [Work supported by NIDCD.]

Listeners often show rapid improvements when first performing auditory discrimination tasks. This learning was explored by investigating how training 300–1500 trials on one of three conditions (all 2IFC) influenced interaural‐time‐difference (ITD) discrimination on a target ITD condition (lateralization, 0‐μs ITD, 500 Hz). Listeners trained on ITD or interaural‐level‐difference (ILD) discrimination had equivalent ITD‐discrimination thresholds that were significantly lower than those of naive listeners. However, listeners trained on duration discrimination had ITD‐discrimination thresholds between those of naive and ITD‐trained listeners. These data suggest that rapid decreases in ITD‐discrimination threshold result partly from learning the general procedure (induced by the duration training) and partly from learning the lateralization task (induced by the ILD training). In contrast, compared to naive listeners, final within‐subject standard deviations of the ITD‐discrimination thresholds were significantly smaller for ITD‐trained, somewhat smaller for ILD‐trained, and no smaller for duration‐trained listeners. These data suggest that rapid reductions in threshold variabiilty result partly from learning the lateralization task (induced by the ILD training) and partly from learning something provided, here, only by ITD training. Thus, at least three different types of learning can contribute to initial improvements in performance on auditory discrimination tasks. [Work supported by NIDCD.]

It is generally accepted that training on one perceptual task does not yield learning on untrained tasks. If so, interleaved training on two tasks should not affect learning or stimulus generalization on either task. Contrary to this prediction, six listeners trained 2730 trials each on interval‐ and frequency‐discrimination tasks using the same stimulus (100 ms at 1 kHz) learned on the interval task 30% more with the trained stimulus, and 134% more with an untrained stimulus (100 ms at 4 kHz), than ten listeners trained 2934 trials on the interval task alone. Furthermore, listeners trained on both tasks with the same stimulus learned on the trained and untrained stimuli, respectively, (1) 105% and 136% more than 6 listeners trained 2160 trials each on both tasks but with a different stimulus (200 ms at 4 kHz) for the frequency task, (2) 19% and 42% more than 20 listeners trained 9675 trials on the interval task alone, and (3) 71% and 579% more than 6 listeners trained 9000 trials on the frequency task al...