Research Support for the Construct Validity of CEST 175The two inventories contributed independent, supplementary
variance to the prediction of many of the same variables and
unique variance to the prediction of other variables. More-
over, when the five NEO-FFI scales were entered into a re-
gression equation as predictors of the REI scales, they
accounted for only 37% of the variance of the Rationality
scale and 11% of the variance of the Experientiality scale.
This is of interest not only because it demonstrates that the
REI is mainly independent of the NEO-FFI, but also because
of the information it provides about the NEO-FFI. It suggests
that the NEO-FFI mainly measures attributes associated with
the rational system and is relatively deficient in measuring at-
titudes and behavior associated with preconscious, automatic
information processing.
Consistent with gender stereotypes, women report signifi-
cantly greater appreciation of and engagement in experiential
processing than men, and men report greater appreciation of
and engagement in rational processing. However, the mean
gender differences are small, and there is a great deal of over-
lap between the groups.
Given fundamentally different ways of processing infor-
mation, it might reasonably be expected that people with dif-
ferent thinking styles would be receptive to different kinds of
messages. To test this hypothesis, Rosenthal and Epstein
(2000) conducted a study with the REI in which they com-
pared the reactions of women with high scores on rationality
and low scores on experientiality with women with the oppo-
site pattern. The groups were subdivided according to
whether they received messages on the danger of breast can-
cer and the importance of self-examination in the form of
information designed to appeal to the rational or the experi-
ential mode of information processing. The rational mes-
sage emphasized actuarial and other objective information,
whereas the experiential message included personal appeals
and vivid individual cases. The dependent variable was the
intention to regularly conduct breast self-examinations. Both
groups reported a greater intention to conduct breast exami-
nations when the message they received matched their own
thinking style.
Individual Differences in Basic Beliefs About
the Self and the World
The Basic Beliefs Inventory (BBI; Catlin & Epstein, 1992) is
a self-report questionnaire that measures the four basic be-
liefs proposed in CEST. It includes a global scale of overall
favorability of basic beliefs and separate scales for measuring
each of the basic beliefs. The internal-consistency reliabili-
ties (coefficients alpha) of the scales are between .77 to .91.
The scales are moderately intercorrelated with a median cor-
relation of .42, thereby justifying combining them into an
overall scale of favorability of beliefs as well as considering
them individually.
You will recall that according to CEST, a person’s basic
beliefs are primarily derived from emotionally significant
personal experiences. To test this hypothesis, Catlin and
Epstein (1992) examined the relations of scores on the BBI
and self-reports of two kinds of highly significant emotional
experiences. The two kinds of experiences were extreme life
events, such as loss of a loved one, and the quality of rela-
tionships with parents during early childhood. In support of
hypothesis, both kinds of experiences were significantly and
coherently related to basic beliefs. Often, the two kinds of ex-
perience made independent, supplementary contributions to
the prediction of the same basic belief. Of additional interest,
the self-reported quality of childhood relationships with par-
ents moderated the influence of extreme life events on basic
beliefs.Summary and Conclusions Regarding Research
Support for CESTIn summary, the program of research on CEST has provided
impressive support for its construct validity. The following
basic assumptions of CEST have all been verified: There
are two independent information-processing systems that op-
erate in parallel by different rules. The systems are interac-
tive, with each influencing the other, and the interaction
occurs both sequentially and simultaneously. The influence
of experiential processing on rational processing is of partic-
ular importance, as it identifies a process by which people’s
automatic, preconscious, experiential processing routinely
biases their conscious rational thinking. The experiential sys-
tem is an associative, rapid, concretist, primarily nonverbal
system that is intrinsically highly compelling to the extent
that it can override the rational system, leading people to “be-
have against their better judgment.”
When people are aware of the maladaptive thoughts gen-
erated by their automatic experiential processing, they often
correct the thoughts through more deliberative reasoning in
their rational systems. There are reliable individual differ-
ences in the efficacy or intelligence of the experiential system.
The intelligence of the experiential system is independent of
the intelligence of the rational system and is more strongly as-
sociated with a variety of indexes of success in living than is
the intelligence of the rational system. Included are work suc-
cess, social facility, absence of drug and alcohol abuse, and
mental and physical well-being. There are reliable individual
differences in experiential and rational thinking styles. The
two thinking styles exhibit coherent patterns of relations
with a variety of criterion variables. There are also reliable