Aural Detection of a Gated Sinusoid as a Function of its Duration
J. C. R. Licklider, G. H. Flanagan · The Journal of the Acoustical Society of America · 1959
Using the two-interval forced-choice procedure, we conducted detection tests with gated sinusoids of 1000 cps in Gaussian-process noise. The signals varied in duration from 63 msec to 10 sec. We employed four different schedules or patterns of the time intervals within each trial. Duration was a highly significant variable. The temporal pattern was not. The results confirm and extend the findings of Hamilton and of Green, Birdsall, and Tanner [J. Acoust. Soc. Am. 29, 506–511 and 523–531 (1957)]. The ratio of signal energy to noise power density required for 75% correct detection falls from about 12.5 db at 5.0 msec to about 9.5 db at 20 msec (Green), is then constant at 9.5 db out to 160 reset, and then rises to about 13.5 db at 1.0 sec and 24 db at 10 sec. The question, whether or not these results indicate that the auditory parameters are variable and adjust themselves to the task at hand, will be discussed. [This research was supported in part under Air Force Contracts AF 18(600)-1219 and AF 19(604)-2051, monitored by the Operational Applications Laboratory, Air Force Cambridge Research Center, ARDC.]