Jp. Madden, THE ROLE OF FREQUENCY RESOLUTION AND TEMPORAL RESOLUTION IN THE DETECTION OF FREQUENCY-MODULATION, The Journal of the Acoustical Society of America, 95(1), 1994, pp. 454-462
The experiment investigated subjects' ability to detect short-duration
changes in frequency. In an adaptive, 2AFC task, three normal-hearing
subjects were asked to distinguish a sinusoidal signal that increased
in frequency in a series of discrete steps from a standard that was i
dentical except that its frequency increased essentially continuously.
The signals were 60 ms in duration with center frequencies of 0.25, 0
.5, 1, 2, 3, 4, and 6 kHz. The smallest frequency increase between ste
ps (FI) at which the stepped signal could be distinguished from the st
andard was determined as a function of the number of steps in the sign
al. As the number of steps increased and the step duration decreased,
the Fl at first decreased and then reached a roughly asymptotic level.
Eventually, however, at a certain number of steps, the FI increased r
apidly. The data were analyzed using a model of auditory temporal reso
lution that included a bank of bandpass filters, a nonlinearity, a tem
poral integrator, and a decision device. The analysis yielded ERDs tha
t ranged from 3.8 to 5.0 ms and did not change systematically with fre
quency. Detector efficiency varied considerably, being greatest at 0.5
and 1 kHz, and declining at higher and lower center frequencies.