Some comments on why you might not want to fixate a target for a
prolonged period of time before shooting at it:
Consider the geometry of pointing a gun (or a finger) at
a target. When you look at something its image falls on the retina.
In order to relate the image's position on the retina to the object's
position relative to anything else, you need to know the position of
the retina relative to the world. For example, let us look
at the following case. Imagine that the image of the target is 10
degs to left of the fovea (the retinal "straight ahead") and the eye
is deviated 10 degs to the left in the orbit and the head is deviated
10 degs left on the body and your shooting arm is deviated 10 degs
left with respect to the body. Then the target is 30 degs left from
the body straight ahead and the gun is already 10 degs deviated and
has to deviate a further 20 degs. Any errors in the representation of
any of these angles and a error will result. (see note 1)
The angles are measured and coded by the body's sensory systems. Most
sensory systems respond best to CHANGE and then ADAPT. So although
deviations of each of these things can be accurately represented,
MAINTAINED deviations lead to adaption and consequent
mis-representations, errors and inaccuracies.
For example, maintaining an eccentric eye position rapidly leads to
adaptation in the direction of interpretting the current (eccentric)
eye position as the straight ahead (ref 1 reviewed in ref 2).
I think, then, that this is the sense in which "prolonged fixed
staring ... makes the brain 'shift' the position". It is not
referring to shifting the physical location of the image (which would
require an eye movement) but to an (incorrect and unhelpful)
re-mapping of the image location on retina to its position relative
to the rest of the body which certainly happens.
Now it might be argued that this conversion of position from one
reference frame (the retina) to another (eg. the body or shooting
arm) is irrelevant since the system COULD solve the problem entirely
visually. The target and the sight (presumably with some kind of
cross hairs) are both in view and thus it could be argued that no
other sense than vision is required. You could just flail the gun
around until the cross-hairs and the target happened to line up and
then pull the trigger. It would not be necessary to know where the
eye, head, body, target or gun were at all!
If the gun was being controlled by servo motors and the operator were
sitting at a computer terminal moving the cross-hairs of the sights
on their screen, then I think that indeed, vision would be the only
sense needed for accurate target shooting (after the system was
calibrated!) But in the sport, tiny errors are highly significant and
they could easily arise by adaptation of any of the sensory systems
involved in the loop between the target being seen and the bullet
arriving.
REFERENCES
1 Morgan, CL (1978) Perception and Psychophysics 23: 61
2 Howard IP (1982) Human Visual Orientation. Wiley, NY
NOTES
In this example I have only considered the left/right dimension but
of course the up/down and clockwise/counter clockwise dimensions can
be considered in the same way.
I have also oversimplified by considering only the angles. Of course
the angles are not all around the same axes and this has large
geometric consequences.
I have also ignored the optical inversion of the image (which is a
red herring).
-------------------------------------------------
Laurence Harris phone: (416) 736-2100 x 66108
Department of Psychology fax: (416) 736-5814
York University
Toronto,Ontario M3J 1P3
CANADA
YORKVIS web page: http://www.yorku.ca/dept/psych/yorkvis/
-----------------------------------------------------