[From Rick Marken (950511.2100)]
Mark George (950510) --
During our discussion, we generally raise questions that the rest of the
class can answer, but this time we ran into a wall. Our question was,
on chapter 10, page 143, section 10.3.1, third paragraph, first sentence
The sentence in question is: "The control-theory model predicts that all
sequences, auditory, visual or proprioceptive (taps) are perceived at the
same level, and hence at the same rate."
I think that you are being very polite to call this sentence a "wall". It is
actually more like something you might find scawled on a wall. If I
knew people were going to read the book so carefully I would have
been more careful about what I said:-)
Let me try to parse it for you. The first part just says that all sequences
are perceived at the same level of the hierarchy of control; a sequence is
a sequence, whether it is a sequence of sounds, lights or feelings. The
idea is that there are perceptual functions at some level of the
hierarchy that compute the degree to which particular sequences are
occuring.
The second part says that sequences "are perceived at the same rate".
This is simply wrong. A sequence of finger taps can occur at many
different rates -- fast to slow. But there is a _maximum_ rate at which
the sequence can be perceived as a sequence. It turns out that this
maximum rate (about 4 components/sec) is the same regardless of the components
of sequence -- finger taps, sounds, lights, etc. What I should have said in
the sentence that caused the problem is something like "all types of sequences
can be perceived up to the same maximum rate". Once the elements of the
sequence exceed this rate, the sequence can no longer be perceived (and, as I
have shown in subsequent research, it can also no longer be controlled; you
can't control what you can't perceive).
The fact that all sequences can be perceived up to the same maximum
rate suggests that the perception of sequence (regardless of type) occurs
at the same level of the control hierarchy. This is because, according to
HPCT, control systems at the same level should operate at about the
same speed (have the same bandwidth). Perceptual integration time (the
minimum time within which a perception can be computed -- which determines
the maximum rate at which a sequence can be perceived) is an important
determinant of the bandwidth of a control system.
I have a very simple HyperCard demo of the perceptual effect if you are
interested. The demo shows how different aspects of the same event
become perceptible (and imperceptible) with changes in the speed of
the event.
Hope this helps; it's a complex topic.
Best
Rick