Perceptual Learning of Time-Compressed Speech: More than Rapid Adaptation
Figure 1
Pre- and post-test average verification thresholds.
From left to right thresholds are shown for the trained condition and four conditions designed to test generalization to untrained sentences (new tokens), untrained speakers (different male and different female) and to non-sentence materials (single words). Controls are marked with diamonds; trained listeners are marked with circles. Thresholds are expressed in fraction of original sentence duration, thus a value of 0.5 is equivalent to a sentence presented at twice its original rate. Pre-test thresholds are marked with empty symbols, post-tests with filled symbols. Error bars are ±1 standard error of the mean.