re: Self-Control

From: HARNAD Stevan (
Date: Mon May 19 1997 - 13:34:31 BST

> From: "Spencer Klair-Louise" <>
> I remember from last years py104, that an objection to behaviorism is
> its inability to account for differences in capacity, although in
> studying self-control, it appears that it is a behaviour like any other
> and therefore subject to contingencies of reinforcement etc etc. So
> surely any differences in this capacity would be due to a person's
> reinforcement history? I am trying to clarify my objections to the
> behavioural stance, but without much success and wondered if you would
> help.

Some of the things we do are already down there at the level of
reinforcement and shaping. Self-control may be one of these. A good
test of whether behaviourism has or has not explained something is
whether you could make a computer able perform as described using only
what you learned about reinforcement contingencies: Exercising
self-control may be such a behaviour, but naming objects is not.

In both cases, we get feedback from the consequencies of our behaviour,
but behaving, waiting and reward is what self-control is all about,
whereas the "behaviour" in naming objects is not explained by saying
that a rose gained behavioural control over the response "rose"
(it did, but that's the trivial part): It's explained by telling us what
the internal mechanism is that makes a system capable of learning to
call roses "roses" like that.

> What I do see is that behaviorism does not account for our mental
> processes or how we are able to be shaped by reinforcement in the first
> place, but then isn't that question answered by looking at our
> evolutionary history?

No; evolutionary history may give us a clue to what has shaped any
inborn mechanisms we may have, but it does not explain how they work.
So whether based on inborn mechanisms or learning, our behavioural
capacities need to be given a causal explanation. One way to think of it
is that we need to "reverse engineer" our behavioural capacities.
Experimental analysis of behaviour simply takes the engineering as

> My main objection centres around how it is that
> we are able to actually imagine or visualise the consequences to our
> actions, and then base a choice upon these forcasts, but am I not
> getting into deep water when talking about a non-corporeal force such
> as imagination?

Yes, conscious causation would be a problem; fortunately (or
unfortunately, as the case may be), we are not conscious of how we are
able to do things, so introspective evidence (e.g., visualising)
almost never explains how we are able to do things; it just tells us
what's going on in our minds when we do them. In other words,
we can't do the reverse engineering by introspection. Most of
what's going on in the head is not accessible to introspection anyay.

> The only other problem is the long standing debate over
> self and the possibility of dualism, although I do not see the point of
> dwelling on that as even by Skinner own admission "it is an ancient
> issue which has not yet been satisfactorily resolved" (1953). I'm sorry
> if this all sounds very confused, but I am, so if there are any clear
> objections to a behvioural explanation of self control for a cognitive
> stance such as your own, I would be very glad to hear them.

Self-control is just a reinforcement phenomenon, as far as I know:
It's about doing something now for a smaller reward, or later for a
bigger one. That sounds as if it is all at the behavioural level anyway:
Try to see how the go/no-go variables in self-control generalise to
whether or not to castle in a game of chess!

This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:47 GMT