[From Chris Malcolm]
There is an interesting control observation to be made concerning
motorcycle steering. A motorcycle can only stably turn a corner while
leaned over so as to counteract the centrifugal force. The geometry of
motorcycle steering is so set up that when the bike is travelling
forwards and leaning over it "wants" to turn its wheel appropriately for
the turn; though some of them are better than others at this. The
problem is how to start turning, since if you simply turn the bars in
the appropriate direction the bike will fall over to the other side. The
fastest method is simply to turn the bars in the opposite direction to
the required turn. This causes the bike to fall over rapidly in the
required direction. When the bike has fallen over enough, the fall is
"caught" by turning the bars in the required direction of travel, thus
producing the balancing centrifugal force. In practice the bars
usually do not require to be turned by force, merely permitted to line
up as they wish to, i.e., the turning of the bars into the turn is
accomplished by relaxing the force applied to produce the initial
counter-steer.
This initial turning of the bars the "wrong" way is known as
"counter-steering". What is interesting about it is that most
experienced motorcyclists have learnt to do this, but that, unless they
have been exposed to explicit training, not only do they not realise
that this is what they do, but they usually strongly assert that they do
the opposite. Frequently considerable practical experiment is required
to disabuse them of their wrong hypothesis, and permit them to correctly
observe what they really do. Some find it impossible to learn this
perception, and continue to assert strongly that they do not do what a
simple video can show them to be doing.
Now if a motorcyclist actually starts a turn by counter-steering, but
strongly asserts that they do exactly the opposite, needing quite a lot
of practical experiment to manage to observe correctly what they do,
this means not only that they learnt this skill without being conscious
of what they were doing, but were able to learn and practise one
behaviour _while_ believing that they were doing the opposite. In fact,
having had this experience myself, I can attest that oddly enough one
has to _learn_ to perceive how one is turning the bars, and that this is
difficult, despite the fact that when rapidly changing direction one
exerts considerable force to apply the counter-steer.
One of the difficulties in doing this learning is that it is very
difficult indeed to experimentally impose variation on the arm movements
demanded by these learnt skills. For example, try riding a bike and
giving a capricious twist to the bars. It requires a supreme effort of
will, beyond many people's capability. And what you cannot will it is
difficult to become conscious of.
Clearly these apparent paradoxes arise because the "folk" theories of
how control is learnt and executed are wrong. What is required is a
theory of control which can be applied to the phenomena of
counter-steering in which the paradoxes become predicted. Can CSG supply
this? Can the "conventional" theories of control supply this?
Chris Malcolm