Feist−Feist: Theories of
Personality, Seventh
EditionV. Learning Theories 15. Skinner: Behavioral
Analysis(^474) © The McGraw−Hill
Companies, 2009
Some of these studies have been concerned with the relationship between long-term
behavior patterns (i.e., personality) and contingencies of reinforcement. These studies
are generally of two kinds: They have asked either how conditioning affects person-
ality or how personality affects conditioning.
How Conditioning Affects Personality
In Chapter 1, we said that the key elements of personality are stability of behavior
over time and across different situations. By these criteria, personality change occurs
when new behaviors become stable over time and/or across different situations. One
domain in which personality change may be evidenced is in psychotherapy. In fact,
a major goal of therapy is to change behavior, and if the changes are stable over time
and situations, then one could talk about changing personality. We say this to make
clear that whereas Skinner discussed changing long-term behavior, he never really
discussed changing personality.
One basic assumption of Skinnerian conditioning is that reinforcement shapes
behavior. Yet, what are the factors that change reinforcement; that is, can certain
stimuli become more or less reinforcing for an individual over time? This is an im-
portant question in treating people with drug problems, because successful treatment
requires that a reinforcer (drug) lose its reinforcing value. For smokers, for example,
nicotine gradually becomes a negative reinforcer, as mild states of tension are re-
moved by the effects of this drug.
Some evidence has shown that psychomotor stimulants (such as cocaine or
d-amphetamines) increase smoking levels in those who smoke. There are two possible
explanations for the effect: First, perhaps the stimulant specifically increases the re-
inforcing effect of nicotine; second, perhaps psychomotor stimulants simply increase
activity levels in general, and smoking is just one of them. In order to test these two
competing explanations, Jennifer Tidey, Suzanne O’Neill, and Stephen Higgins
(2000) conducted a study with 13 smokers and put them through an elaborate test-
ing procedure (12 separate 5-hour sessions), in which they received either a placebo
or the drug d-amphetamine. Ninety minutes later the smokers had to choose between
two different reinforcers, money ($0.25) or smoking (two puffs). If they chose money,
a running total of the accumulated amount was shown on a computer screen and par-
ticipants were paid that amount at the end of the testing session. If they chose the cig-
arette, they were allowed two puffs immediately after doing the desired behavior. If
the stimulant simply increases general activity levels, there should be no systematic
preference for one reinforcer over the other (compared to baseline preferences). Ad-
ditionally, after the experimental session ended, they were allowed a period in which
they could smoke as much or as little as they wished (free smoking session).
However, results showed that smoking levels in both the experimental choice
(compared to money) and in the free smoking sessions increased in proportion to
d-amphetamine. The higher the dose of d-amphetamine, the more the participants
smoked. Even more importantly, however, smoking was chosen over money in the
choice session in direct proportion to the amount of d-amphetamine administered.
Therefore, the stimulant must increase the reinforcing value of nicotine specifically
and not the other reinforcer (money). In short, the answer to the question of whether
468 Part V Learning Theories