Page 122 - Media Effects Advances in Theory and Research

5. MEDIA PRIMING 111

states those entities are in), the interrelationships between the various
entities and the situation (including causality and intentionality), and
events that occur in that situation (Garnham, 1997; Johnson-Laird, 1983;
Radvansky & Zacks, 1997; Wyer & Radvansky, 1999; Zwaan & Radvan-
sky, 1998). Mental models are distinct from network models of memory,
but the entities and events within a mental model are hypothesized to be
linked to relevant representations in a network (Radvansky & Zacks,
1997; Wyer & Radvansky, 1999). In other words, mental models are
hypothesized to exist alongside and coupled with the semantic networks
that are hypothesized by network models of memory.
vanDijk (1998) argues that mental models involve the merger of
semantic memory (knowledge of the world) and episodic memory (mem-
ory for our past experiences). However, this argument can be misleading.
In particular, it might give the impression that mental models involve
only the representation of past situations we have personally experienced
(what vanDijk refers to as experience mental models). However, when defin-
ing mental models as cognitive representations of situations, the term sit-
uation is used very broadly. For example, we can have mental models of
ownership and the interrelationships of owners and the objects that are
owned (Radvansky & Zacks, 1997). Likewise, mental models can be used
in reasoning to represent possible worlds in which the premises of an
argument are true and to manipulate the possible worlds to discover what
occurs (Johnson-Laird, 1983).
Mental models have been used to understand a number of different
phenomena, including reasoning and problem solving (Greeno, 1984;
Johnson-Laird, 1983), language processing (Garnham, 1997), children’s
understanding of the world (Halford, 1993), text comprehension and
discourse (Graesser, Singer, & Trabasso, 1994; Morrow, Greenspan, &
Bower, 1987; vanDijk & Kintsch, 1983; Zwaan & Radvansky, 1998), chil-
dren’s implicit theories of physics (Gentner & Gentner, 1984), spatial
cognition (Radvansky, Spieler, & Zacks, 1993), message effects (Capella
& Street, 1989), political commercials (Biocca, 1991), and ideology
(vanDijk, 1998). We believe that they also can be used to understand
media priming.
We want to be clear that we are not arguing that priming, as conceptu-
alized by network models of memory, does not occur with the media.
Commercials clearly prime various concepts, and this priming can influ-
ence the interpretation of other commercials or the show that the com-
mercials were placed within (Yi, 1990a, 1990b). Likewise, watching a vio-
lent movie clip speeds the time it takes participants to pronounce
aggression-related words, compared to participants who watched a vio-
lence-free movie clip (Anderson, 1997). Both of these findings are consis-
tent with network models of priming.