Defining behavior

[From Bill Powers (950514.2000 MDT)]

Like Rick Marken, I got a call from Mark Abrams asking me to define
behavior.

OK. I agree with Rick's general discussion, so I'll try to go at it from
a few slightly different angles. Off the top of my head ...

I: Point of view.

For each of us there are two basic kinds of behavior: the behavior of
other people, and our own behavior. They are completely different.

When we see other people behaving, we see their movements, the way their
hands and arms and legs and bodies move about and affect things in the
environment. We hear the sounds they make and the expressions on their
faces. We see how they are standing. We see how their hair is parted. We
see, in other words, the impressions that their motor activities make on
our own perceptions.

When we see our own behavior, we see something quite different. The
movements of our arms, legs, bodies and so on, the expressions on our
faces, are seldom of any concern to us. What matters to us is the effect
that our own activities have on aspects of the world we perceive, that
we are paying attention to, that we care about. Of this world, other
people know practically nothing. They can see our movements, but they
can't see why we are making them. They can even see that our movements
are having effects on the environment, but which of those effects
matters to us they can only guess. They often guess wrong.

When, through artificial aids, we are shown how we look and what we do
as others see it, we are usually surprised, and often dismayed. In a
video or a right-angled mirror we see someone's face, but it is far less
familiar than the faces of other people we know -- the part in our hair
is on the wrong side and the assymetries of our faces are wrong, too.
When we raise a hand, the wrong hand moves. When we laugh, the face
exhibits wrinkles and distortions that have no connection with how it
feels to laugh. When we see ourselves in profile, a total stranger
appears. Hearing a tape recording of our own voices for the first time
is also unsettling; it is full of unfamiliar overtones, mumbling and
ungrammatical, interrupted by embarrassing mannerisms and hestitations.

But perhaps most surprising (especially in a video) is seeing ourselves
doing all kinds of things that we didn't know we were doing. We pick
things up and hold them, and put them down again. We scratch. We make
faces. We clear our throats. We sigh, and twitch. We knock things to the
floor, oblivious. When we hear ourselves speaking, we want to stop the
recording because what we hear ourselves saying is not what we meant, or
it's so incoherent that we can't imagine anyone else having understood
us. We interrupt other people, we seem unaware of what they are saying;
we seem completely unaware of the effects we're having.

When we see ourselves as others see us, it becomes obvious that what we
think of ourselves as doing is very likely to be different from what
others see us doing. And it becomes obvious that others can't see the
essential (to us) aspect of what we're doing: what it is that we intend
to happen in our own perceptions, looking with our eyes and sensing with
our other senses.

From such experiences, we can learn some things about behavior. We have

to separate the actions, the movements we make, from the effects that
they are supposed to have. A given action is just a movement, a change
of posture or bodily configuration. When we act, our attention is not on
the action itself, but on its effects, consequences, outcomes. We seldom
notice the effects that were not those we intended, as long as the ones
we intended are occurring. We are usually unaware of what our body parts
are actually doing, what efforts are really involved. When we see
ourselves from outside, we can't see any of the things that matter from
the inside.

II. Actions versus effects of actions.

Distinguishing actions from their effects is not made any easier by
language. We say "Close the door," not "exert a force on the door that
will result in its closing." We say "turn on the light" or "steer left"
or "pick up your socks" or "wait for me under the clock on the corner,"
as if these were actions rather than consequences of _unnamed_ actions.
This lends a rather magical air to the way we talk about behavior. We
name what is supposed to happen as a result of physical activities, and
talk as if somehow we could make these results occur without any actual
physical activities.

In ordinary discourse we can usually figure out what is meant when an
action is named in terms of its effects. We just imagine one of the
actions that would produce such an effect. But in scientific discourse,
this problem becomes a major impediment to understanding. What does it
mean to say that a rat "pressed a bar?" It means that the rat exerted
efforts of some kind which resulted in the bar moving downward. But it
does not actually tell us what the rat _did_. If behavior is supposedly
caused by stimuli, those stimuli must result in the creation of muscle
forces; there is no way they can act directly to cause the bar to move.
To describe a behavior as a "problem-solving response" is to suggest
that there is some physical action a person can take which has as its
main property the solving of a problem.

III: An organized way to think about behavior.

When we want to be exact about describing behavior, we have to consider
three things. The central consideration is what it is that measures the
behavior. If we are describing a behavior called "closing a door," the
central consideration is the door that is open, and then swings closed.
We measure it by measuring the angle of the door. We are clearly talking
about something happening in the environment, not about what an organism
is doing.

Next, we have to ask what brings this effect about. The closing of the
door is brought about by someone or something exerting a force on it.
When a person closes a door, therefore, the action produced by the
person is not the closing of the door, but the application of a force to
the door using arms and hands (or legs and feet, or heads or hips, or
whatever the means happens to be). The cause of the door's closing is
the action taken by the person. We describe that action not as "closing"
but as "pushing" or "pulling." We can describe the action (applying a
horizontal force of a certain amount) independently of what its effects
are (closing a door, knocking a vase off a table, pulling a puppy along
on a leash). The action is the immediate physical effect of what the
nervous system is doing via muscles. The effect is what happens in the
environment as a result of the action.

And finally, we have to ask which of the many effects of any given
action is the one being perceived and compared with a reference
condition. When the door is closed, its angle changes, the opening of
the door is visually blocked, and the transmission of sound through the
opening is much reduced. Also, if there is a mirror on the door, it
becomes oriented so you can see youself in it, and if there is a cat
following you, the cat is kept out. If the house is cold, closing the
door will keep heat in while you take a shower. And closing the door can
be a message: do not disturb.

All of these effects of the action, both direct effects on the angle of
the door and indirect effects on other variables, occur and can be seen
by someone else. So the question then becomes, which of these effects is
being perceived by the person producing the action? To ask which is
perceived is also to ask which is intended.

Of course the fact that one effect of action is intended while a number
of others is not does not prevent the other effects from occurring or
being perceived by others. This is why our own behavior looks so
different to us than it does to others. We know which effect is
intended, and we adjust our actions until the intended effect, as we
perceive it, matches what we wanted to perceive. But in doing that, we
unintentionally change the states of many other variables. An onlooker
does not necessarily know which of the variables that changed state was
the one that was intended to change, and which other changes are only
side-effects. So the onlooker may see us as keeping the cat out while
what we were really doing was checking our hair in a mirror.

If we get in the habit of parsing behavior into the physical action, the
effects of that action on the environment, and the perceived effect that
is intended, we will understand the behavior of other people much
better. More important, we will realize that there is more than one way
to understand another person's behavior, and we will thereby avoid
jumping to wrong conclusions. We will realize that when we see a person
"doing" something, that person may actually be doing something entirely
different. We will look more carefully at what is going on, instead of,
for example, overlooking the camera and lights when we see a mugger
attacking an old lady on the street.

ยทยทยท

-------------------------------------------------------------------
Best,

Bill P.