[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: comodulation release of masking (CMR)


You may like to take a look at the recent ASA/ICA98 proceedings

Todd, N.P.McAngus. (1998) A model of auditory image flow I: Architecture.
 Proc. 16th International Congress of Acoustics. Vol. II. Seattle, June, 1998.

Todd, N.P.McAngus and Brown, D. (1998) A model of auditory image flow II:
AM and FM detection. Proc. 16th International Congress of Acoustics. Vol.
II. Seattle, June, 1998.

This has similarities to the Dau model in that detection is based on
monitoring the modulation spectrum accross all cochlear channels. The
essential difference is that the templates used were derived from the
spatio-temporal power spectrum of the sequence of modulation images, rather
than the time-sequence itself.

I suggested that this may have implications for cross-channel effects
since the spatio-temporal power spectrum is computed by 'cortical' receptive
fields (RFs) which span a number of critical bands. So any decision procedure
 based on the whole power spectrum will be subject to interaction effects from
channels which are remote.

I have not attempted to apply this to CMR, but I did have a shot at MDI. I
found that the model did indeed predict some MDI, but the strength of the
effect was much less than that reported in the experimental work. At the
time I took this to imply bad news for the model. However, I understand
(personal communication) that many of the experimental effects reported
fade away with learning.

You may also like to have a look at

Todd, N.P.McAngus (1996) An auditory cortical theory of auditory stream
segregation. Network : Computation in Neural Systems. 7, 349-356.

In this paper I showed that it was possible to provide an account of streaming
by looking at the cross-correlation between the modulation spectrum (including
phase information) of cochlear channels. This may have implications
for CMR.



Lecturer in Psychology
University of Manchester
M13 9PL
Tel. +44 (0)161 275 2557

>Date:          Thu, 1 Oct 1998 17:05:35 +0000
>Reply-to:      jba@RESEARCH.ATT.COM
>From:          Jont Allen <jba@RESEARCH.ATT.COM>
>Organization:  AT&T Labs - Research, Florham Park NJ
>Subject:       comodulation release of masking (CMR)
>Dear List,
>Has anybody done a computer simulation of CMR that can accurately
>simulate results from real experiments. I mean, starting from the
>time wave forms, compute the envelopes, and then estimate the
>detection threshold?
>If not, what is the present `state of the art' of CMR models?
>Jont B. Allen, Room E161
>AT&T Labs-Research
>Shannon Laboratory
>180 Park Ave.
>Florham Park NJ 07932-0971
>973/360-8545voice, x7111fax
>McGill is running a new version of LISTSERV (1.8d on Windows NT).
>Information is available on the WEB at http://www.mcgill.ca/cc/listserv

Email to AUDITORY should now be sent to AUDITORY@lists.mcgill.ca
LISTSERV commands should be sent to listserv@lists.mcgill.ca
Information is available on the WEB at http://www.mcgill.ca/cc/listserv