Page 401 - The Social Animal

Liking, Loving, and Interpersonal Sensitivity 383

A few years ago, I developed a theory of interpersonal attraction,
called the gain-loss theory, that makes a rather different prediction. 53
My theory is simple. It suggests that increases in positive, rewarding
behavior from another person have more impact on an individual
than constantly rewarding behavior from that person. Thus, if we
take being liked as a reward, a person whose liking for us increases
over time will be liked better than one who has always liked us. This
will be true even if the number of rewards was greater from the lat-
ter person. Similarly, losses in positive behavior have more impact
than constant negative behavior from another person.Thus, a person
whose esteem for us decreases over time will be disliked more than
someone who has always disliked us even if the number of negative
actions were greater from the latter person. To return to the party, I
would predict you will like the individual most in the gain situation
(where she begins by disliking you and gradually increases her lik-
ing), and you will like her least in the loss condition (where she be-
gins by liking you and gradually decreases her liking).
To test my theory, I needed an experimental analogue of the party
situation, but for reasons of control, I felt it would be essential to col-
lapse the several events into a single long session. In such an experi-
ment, it is important that the subject be absolutely certain that the
evaluator is totally unaware that she (the evaluator) is being overheard.
This eliminates the possibility of the subject’s suspecting the evaluator
of intentional flattery. This situation presents a difficult challenge for
the experimentalist.The central problem in devising a way to perform
the experiment was one of credibility: How can I provide a believable
situation in which, in a relatively brief period, the subject (1) interacts
with a preprogrammed confederate, (2) eavesdrops while the confed-
erate evaluates him or her to a third party, (3) engages in another con-
versation with the confederate, (4) eavesdrops again, (5) converses
again, (6) eavesdrops again, and so on, through several pairs of trials.
To provide any kind of a cover story would indeed be difficult; to pro-
vide a sensible cover story that would prevent subjects from becoming
suspicious would seem impossible. But, in collaboration with Darwyn
Linder, I did devise such a situation. The devices we used to solve
54
these problems are intricate, and they provide a rare opportunity to
look behind the scenes of an unusually fascinating experimental pro-
cedure. Accordingly, I would like to describe this experiment in some
detail, in the hope that it will give you an understanding of some of