Page 90 - Media Effects Advances in Theory and Research

4. MEDIA CONSUMPTION AND PERCEPTIONS OF SOCIAL REALITY 79

sumption that television viewing is the causal factor. Although consider-
able evidence has accumulated that supports the existence of at least
a small-sized cultivation effect (Morgan & Shanahan, 1996), other
researchers have challenged the validity of this effect. Some research sug-
gests that the relationship between viewing and perceptions is not causal,
but rather a spurious one resulting from third-variable influences (e.g.,
direct experience, education, available time to view, personality) on both
television viewing and social perceptions (Doob & Macdonald, 1979;
Hirsch, 1980; Hughes, 1980; Wober & Gunter, 1988). Other research sug-
gests that the causal relation between viewing and social perceptions may
be reversed; that is, aspects of the individual (including preexisting social
perceptions) may influence the amount and content of viewing (Zillmann,
1980).
As noted earlier, one of the advantages of developing a cognitive
process model of media effects is that it has the potential to render
implausible certain alternative explanations for the effect (e.g., spurious-
ness, reverse causality). Two caveats should be noted, however. First, ren-
dering a particular alternative explanation implausible in a study merely
means that the explanation cannot completely account for a particular pat-
tern of results; it does not mean that the alternative explanation may not
be operating simultaneously but independent of other effects. Second, the
power of a process model is in the cumulative effect of a pattern of results,
not a focus on a single study. Thus, even though alternative explanations
may be possible for any one study, in the interest of parsimony, the alter-
native explanations should address the entire pattern of results to be an
effective challenge.

General Propositions of the Model

Two very simple and general propositions that are based on the princi-
ples of heuristic/sufficiency and accessibility form the basis of the
model. The first general proposition is that television viewing enhances
construct accessibility. As discussed earlier, aspects of television viewing
may plausibly be related to the accessibility of constructs encountered in
typical television fare. The second general proposition is that the social
perceptions that serve as indicators of a cultivation effect are constructed
through heuristic processing. Specifically, rather than constructing judg-
ments through an extensive search of memory for all available relevant
information (systematic processing), only a subset of relevant informa-
tion is retrieved, and specifically, the information retrieved is that which
is most accessible from memory. A corollary of this second general
proposition is that, at least for cases in which the judgments pertain to
perceptions of frequency of a class (set size) or likelihood of occurrence,