>OK we (philosopher, physicist and psychologist) are up here on the third
>floor wasting time by following the results from the international track and
>field event that is going on in Spain. As we looked at the times for the
>results of the men's 400 meter (a new world record) we started looking at
>the column that displays what we think is time to start - reaction time from
>the sound of the start pistol to "leaving" the starting block. Those times
>range from .133 to .284 seconds for the first five finishers. We checked a
>second event and found the fastest starting time to be .111 seconds. We also
>know that this measure must be used to indicate false starts. The questions:
>Do any of you know what the empirical limit for this reaction time is? Do
>you know what standard is used for a false start in events like this one? Do
>you know if that time is based on empirical work or theoretical
>considerations?
>Other than satisfying out curiosity, the answer to these questions could
>offer a good example for use in S&P, Physiological or Introductory.
>My colleagues and I offer our thanks in advance.
>Dennis
>
>Dennis M. Goff
>Dept. of Psychology
>Randolph-Macon Woman's College
>Lynchburg, VA 24503
Are you suggesting something _other than_ a rule such as, "Any reaction
time (or time to start) less than zero indicates a false start--leaving
before the starting gun has sounded. Any reaction time of zero or greater
seconds is accepted as a valid start."? For example, "Since it is
impossible to respond faster than .10 seconds, any start time faster than
that is considered a false start."?
I checked my old copy of Underwood's (1966) _Experimental Psychology_, in
which an experiment by Raab (1962) is described. Looking at simple RT, the
lower limits of RT appear to be about 116 msec, though the RT varies as a
function of stimulus duration (the shortest RT in this study occurred with
50 msec stimulus duration), and stimulus intensity (faster with 60 dB
stimulus than 40 dB stimulus) (p. 254). Underwood also goes on to discuss
the potential problem with "false reactions:" E may also occasionally
insert a false trial by giving the ready signal but not giving the RT
stimulus. Or, as another technique, E may discard all responses which are
apparently too fast on the grounds that these could not have been initiated
by the RT stimulus. For example, Drazin (1961) discarded all responses to
a visual stimulus if the response--the RT--was 100 msec. or less, or if
they occurred before the RT stimulus" (p. 255). Woodworth and Scholsberg
(1954) discuss "RT as being composed of [an] "irreducible minimum," plus a
remainder...a "reducible margin" (p. 21). In subsequent calculations they
select "105 msec as a likely limit in this instance" (p. 21), which they
say was "chosen as a fair estimate."
Do we DQ runners 'cause their start times are "too short" to have been
initiated by the starting gun (though still above zero)? Consider two
runners: A has a start time of .09 sec and an elapsed time of 10.09 (for a
_run_ time of 10.0 sec), and B has a start time of .11 sec and an elapsed
time of 10.15 (for a _run_ time of 10.04 sec). Both left the starting line
after the gun sounded, A got there sooner, and actually was running faster.
But do we DQ runner A because he/she left the blocks before some
"irreducible minimum" time of 105 msec (.105 sec), reasoning that A must
have jumped the gun?
Frankly, I don't want to be the official at _that_ track meet! Might make
some psychological sense, but it'd be tough to explain it to runner A.
Bob (whose time for getting to his office door following a knock might
approach 10.04 sec, but only with a tailwind)
=======================================================================
Robert T. Herdegen III
Department of Psychology
Hampden-Sydney College
Hampden-Sydney, VA 23943
=======================================================================
