Re: comparison of vision and auditory system

["Charles S. Watson" <watson@xxxxxxxxxxx>]

Jont,
Years ago Hirsh and Sherrick (1961-JEP, 62, 423-432) concluded that within
and between-modality temporal acuity was pretty similar, based on temporal
ordering tasks.  I believe they did this using a classical psychophysical
method (probably the constant method, my memory is fuzzy on this and the
paper is not at hand). I recall more clearly that Don Sinex did
several similar inter-modality (and within) temporal acuity
experiments as his dissertation (Washington Univ., 1978), and found that
whenever vision was involved, the threshold were higher (worse).  That
is, the most accurate judgements of simultaneity or of inter-event times
were for auditory-auditory, tactile-tactile, and auditory-tactile
judgements, while the least accurate were for vision plus any modality.
That made sense to us, given the durations involved in photochemical
transduction, compared to those of mechanical or hydromechanical systems.
In general the subjects could detect arrival-time differences a good deal
smaller than 20 msec, the number associated with a lot of temporal-order
thresholds.

As to temporal acuity being "surprisingly bad," I think that depends on the
task, duration of training, etc..  For the most part listeners may not have
much reason to attend to very small temporal differences.  But when they are
trained to do so they can surprise you by performing almost up to the limits
imposed by the spectral spread of brief events.  We remarked on that in a
paper some years back:  Kelly and Watson (1986) "Stimulus-based limitations
on the discrimination between different temporal orders of tones."  JASA.,
79, 1934-1938.

Certainly, some binaural phenomena (e.g. being able to notice a 10 microsec
delay at one ear relative to the other) show that precise temporal (microsec)
information is carried as far as the olivary nuc., so at least at that level
the auditory system has far more precise temporal information available than
does the visual system at corresponding points...at least that has always
appeared to be a valid generalization.

Regards,

Chuck


On Thu, 29 May 1997, Jont Allen wrote:

> >Roger Watt <r.j.watt@STIR.AC.UK> wrote:
> >In vision, there is exquisite spatial precision as well as resolution
> >but much less precision in the time domain, despite reasonable temporal
> >resolution.
>
> I find this point interesting because this is also true in the auditory
> system. The frequency JND is very small, but the acuity of temporal events
> is suprisingly bad. I have always assumed that in the auditory system this
> was by design, because of the degrading effects of echos and reverberation,
> for example.
>
> If the temporal acuity is quantitatively similar (is it?) in the two system,
> then this might be saying something about how the CNS functions.
> For example, for the auditory case, it seems likely that the reason for the
> poor temporal acuity across frequency is that the information is being
> merged after many "layers" or stages of processing. This follows from
> the old observation that temporal acuity necessarily decreases for deeper
> layers.
>
> Does this make sense to anybody, or is this just the sound of soft wind?
>
> Jont Allen
>