Abstract

Detection of signals from resonant systems is of interest both for the field of speech perception and for testing the applicability of Signal Detectability Theory to signals of complex wave form. Stimuli were the output of a recurrently-excited, simple resonant circuit. The impulse frequency was fixed at 100 cps, and the resonant frequency was 500 cps. The signal was, then, a train of decaying 500 cycle sinusoids. This signal was gated for various durations and presented binaurally in a background of white noise to three observers. A two-alternative forced-choice technique was used. Five different rates of decay, giving signal band widths of 10, 30, 100, 200, and 375 cps were used. Each signal was presented at five different durations; 0.01, 0.02, 0.05, 0.1, and 0.20 sec, keeping the energy constant by varying the signal power inversely with the duration. The efficiency (η) of each of three observers was estimated for all 25 combinations of band width and duration. Results show that for narrow-band-width signals efficiency remains fairly constant over the durations studied. On the other hand, wider band-width signals produce lower performance at longer durations. The results are considered in relation to various current descriptive models, and the usefulness of “auditory band width” concepts is weighed in the light of the data. (Supported by the Office of Naval Research and Air Force Cambridge Research Center.)

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call