Re: AUDITORY Digest - 9 Oct 2005 to 10 Oct 2005 (#2005-208)

[Antje Ihlefeld <ihlefeld@xxxxxx>]

In addition to what has been proposed by list members, a useful 
reference might be a review chapter by Durlach and Colburn 1978. It 
includes psychophysical data on JND of ITDs of pure-tones and pulsed 
noise. From looking at my somewhat blurry copy it appears that the JND 
for noise bursts ranges from 24usec to 4-6usec, depending on the 
duration of the stimulus. The JND decreases by roughly 2usec for every 
doubling of duration from 16msec signal duration until about 500msec, at 
which point it levels off. For pure tones, a plot is included that shows 
how JNDs depend on frequency. The highest sensitivity of about 12usec is 
reached around 1kHz, for f>1.5kHz the JNDs become unbounded. (The same 
book contains another excellent review chapter by Colburn and Durlach 
about psychophysical models of binaural hearing).

I am not sure whether it is really necessary to integrate across 
frequency in this type of task (integration as in e.g. the straightness 
and centrality model for localization). Instead, the subject may merely 
need to detect *some* decorrelation in the stimulus in order to do a 
discrimination task. A limited set of best frequencies should contain 
enough information to do this.

Antje
p.s. The precise reference is: Durlach, NI, and Colburn, HS (1978). 
“Binaural Phenomena,” Chapter 10 in Handbook of Perception, Vol. 4, 
Carterette and Friedman (Eds.), Academic Press, NY.

AUDITORY automatic digest system wrote:

> There are 3 messages totalling 184 lines in this issue.
>
> Topics of the day:
>
>  1. Integration of ITD information (3)
>
> ----------------------------------------------------------------------
>
> Date:    Mon, 10 Oct 2005 19:22:59 +0200
> From:    Bernhard Laback <Bernhard.Laback@xxxxxxxxxx>
> Subject: Integration of ITD information
>
> Dear list,
>
> Is there any evidence for integration of interaural time difference 
> (ITD) information across frequency channels in ITD detection / 
> discrimination tasks? More specifically, are ITD JNDs lower for a 
> broadband stimulus compared to any narrowband stimulus within the 
> spectral range covered by the broadband stimulus?
>
> Of course, I assume that the overall intensity is held constant.
>
> I would appreciate any hint,
>
> With best regards,
>
> Bernhard Laback
>
>