Frequency and duration discrimination of short first-formant speechlike transitions.

Frequency and duration discrimination thresholds of short rising and falling one-formant speechlike transitions without a steady state were determined by means of same/different paired comparison tasks in two experiments. When frequency extent is varied (experiment 1), just noticeable differences decrease with increasing transition duration. Expressed in Hz, thresholds are, on average, 70, 63, and 58 Hz for 20, 30, and 50 ms, respectively. However, when transition duration is varied at a constant frequency extent (experiment 2), difference limens increase with increasing duration and are, on average, 2.7, 4.5, and 4.9 ms for standard transitions of 20, 30, and 50 ms, respectively. The thresholds determined in the two experiments indicate that different psychoacoustical cues are used depending on whether final frequency (experiment 1) or transition duration (experiment 2) are varied. Both experiments were performed at two different frequency regions (between 200 and 700 Hz and between 500 and 1000 Hz), but the results did not differ per region. In addition, no significant differences were found between rising and falling transitions. Particular attention was paid to a methodological issue, viz., the extent to which sensitivity changes as a result of different proportions of catch trials. It was found that the listeners maintained the same response strategies throughout the tests, as their performance is similar, irrespective of the number of catch trials included in the testing sessions.