Page 123 - Media Effects Advances in Theory and Research

112 D. ROSKOS-EWOLDSEN, B. ROSKOS-EWOLDSEN, F. DILLMAN CARPENTIER

However, the phenomena of interest to media scholars studying prim-
ing (e.g., violent media influencing aggressive behavior, political coverage
influencing what information is used to make judgments of the president)
cannot easily be explained by network-based theories of media priming.
At a basic level, the priming effect that network models of memory
address dissipates too quickly to explain many of the media priming
effects. Of course, the time course issue can be addressed by assuming, as
Price and Tewksbury (1997) do, that media portrayals increase the chronic
accessibility of constructs, and it is the chronic accessibility of the con-
structs that results in the media effects that are being studied (see also
Roskos-Ewoldsen et al., in press; Shrum, 1999; Shrum & O’Guinn, 1993).
Although we believe that chronic accessibility is important (e.g., Roskos-
Ewoldsen, 1997; Roskos-Ewoldsen, Arpan-Ralstin, & St. Pierre, in press;
Roskos-Ewoldsen & Fazio, 1992a, 1992b, 1997), we propose that the phe-
nomena of priming and of chronic accessibility should be incorporated
into a larger theoretical frame that involves mental models of memory.
As one example of how the mental models approach works, we can
look at the process model of the attitude–behavior relationship (Fazio,
1986, 1990; Fazio & Roskos-Ewoldsen, 1994; Roskos-Ewoldsen, 1997). In
this model, accessible attitudes influence behavior by influencing how the
current situation is defined. In other words, accessible attitudes influence
behavior by influencing the mental model that is constructed of the situa-
tion. We argue that many priming effects can be similarly reinterpreted to
operate in this manner. In brief, the prime influences how later informa-
tion is interpreted by influencing the type of mental model that is con-
structed to understand the situation.
Mental models can relate to media priming in two ways. First, when
dealing with a novel situation, there is the option of forming a new men-
tal model or accessing an existing mental model from memory. Obviously,
people have myriad mental models stored in long-term memory. If an
existing mental model is accessed from memory, the issue becomes which
mental models are accessed and used to understand/interpret a particu-
lar situation. Although the match between the mental models stored in
memory and the existing situation will influence which model(s) are
accessed, we argue further that, as with other constructs in memory, men-
tal models will vary in their accessibility from memory (Radvansky &
Zacks, 1997; Wyer & Radvansky, 1999). Consequently, mental models can
be primed by the media, increasing the likelihood that they will be
accessed. For example, when the media focused on the pending war over
Kuwait during the Bush presidency, this coverage could result either in
the creation of a new mental model of President Bush and the Kuwait cri-
sis or the modification of an existing mental model of President Bush. In
either case, the frequent coverage of the issue would increase the accessi-