Path: utzoo!attcan!uunet!samsung!brutus.cs.uiuc.edu!dali!milton!wcalvin
From: wcalvin@milton.acs.washington.edu (William Calvin)
Newsgroups: comp.ai
Subject: Re: Is there a clock in the brain?
Keywords: Experiment.  Time-resolution.  Counter/timer.
Message-ID: <2803@milton.acs.washington.edu>
Date: 11 Apr 90 16:13:38 GMT
References: <3376@uwm.edu>
Reply-To: wcalvin@milton.acs.washington.edu (William Calvin)
Organization: University of Washington, Seattle
Lines: 27



        Re human timing experiments:  10 millisec is pretty
good for jitter, about the limit of the motor neurons that run the
muscles.  But there are occasions when one can do much better, as in
throwing a ball, timing the release to less than a millisecond (you
can work backward from the physics to estimate a launch window).
        So how can the system perform better than the most precise of
its individual elements?  Averaging.  In sound localization, you can 
detect interaural arrival times down about 10 microseconds.  That's
because you can average over hundreds of repetitions of the waveform,
taking many milliseconds before arriving at the judgment of sound direction.
But for throwing, you can't average that way:  you must use an ensemble
average, averaging the recommendations of hundreds of timers that are all
trying to accomplish the launch window time.  To cut the jitter in half,
you just use four times as many timers.  Given the intrinsic jitter of
the best of individual neurons (when acting as pacemakers, rather than
merely reacting to an input) in the 10 millisecond range, you need a
ten-fold reduction to get the jitter under 1 msec -- and that means a hundred
such timers.
        If your library has the JOURNAL OF THEORETICAL BIOLOGY, see my
article on throwing in the September 1983 issue.  Or Mile 144-155 of my
book THE RIVER THAT FLOWS UPHILL (Sierra Club Books 1987).  Or Ch.10 in
my new book THE CEREBRAL SYMPHONY:  SEASHORE REFLECTIONS ON THE STRUCTURE
OF CONSCIOUSNESS (Bantam, 1989).

                  William H. Calvin             wcalvin@u.washington.edu