Page 38 - Media Effects Advances in Theory and Research

2. EXEMPLIFICATION THEORY OF MEDIA INFLUENCE 27

It is assumed that exemplar groupings, whether directly perceived or
retrieved from memory, are screened to discern the magnitude of the
groupings. As a rule, such screening is nonconsciously performed, but on
occasion, it may become conscious and deliberate. Irrespective of the level
of awareness involved, however, the screening is assumed to yield inci-
dent assessments in at least ordinal terms (e.g., few, many, a lot of cases).
Moreover, comparative assessments may be conducted, yielding a sense
of relative incidence rates (e.g., one grouping is larger than another) and
changes in incidence rates over time (e.g., a grouping is larger than
before). The degree of difference or change again can be at least ordinally
structured. Essentially, then, it is assumed that an archaic quantification
heuristic exists that continually monitors the prevalence of exemplars as
well as their relative distributions.
Exemplification theory further relies on two additional cognitive mech-
anisms: the representativeness and availability heuristics (Kahneman &
Tversky, 1973; Tversky & Kahneman, 1973). Both mechanisms can be con-
sidered well established by compelling research demonstrations (cf. Fiske
& Taylor, 1984; Sherman, Judd, & Park, 1989; Zillmann & Brosius, 2000).
The representativeness heuristic essentially projects that judgments about
event populations are extrapolations based on the scrutiny of exemplar
groupings, and that in this extrapolation the provision of abstract quanti-
tative information about exemplar distributions is immaterial. The
implicit devaluation of base-rate information for the assessment of event
populations is also known as the base-rate fallacy (Bar-Hillel, 1980). This
fallacy, it should be recognized, is entailed in the earlier-stated informa-
tion-processing assumption that expresses the coding dominance of con-
crete over abstract displays.
A secondary projection of the representativeness heuristic is that the
generalization from samples of events to populations of events is indepen-
dent of the size of the samples. Although generalizations from larger sam-
ples are, of course, known to be more reliable than generalizations from
smaller samples, a comparatively small exemplar group is thus expected
to have the same inferential power as all larger exemplar groups.
The availability heuristic projects that judgments about event popula-
tions are greatly dependent on exemplars that, at the time judgments are
rendered, are available in the sense of being cognitively manifest. This
availability, in turn, is considered to be a function of the ease with which
exemplars are accessed in memory and retrieved from it. In this context,
retrieval is thought to be mostly involuntary and spontaneous. Ready
retrieval thus may be characterized as a nondeliberate process by which
exemplars impose themselves from memory, thereby exerting dispropor-
tional influence on the contemplation and ultimately on the evaluation of
the exemplified event population.