Page 110 - Media Effects Advances in Theory and Research

5. MEDIA PRIMING 99

However, this priming effect appeared to lessen with time, because vio-
lent programming and cues did not influence aggression in the later peri-
ods of play as strongly as in the initial period of play.
In another study, Anderson (1997) investigated the influence of violent
media on the accessibility of aggression-related concepts. Undergraduate
students were randomly assigned to view either movie clips containing
violent scenes or movie clips featuring nonviolent content. After viewing
their assigned clips, the undergraduate participants completed a ques-
tionnaire assessing their state hostility level (Experiment 1) or their trait
and state hostility levels (Experiment 2).
Once the questionnaire was completed, participants were escorted to
another room to perform a task in which they read aloud 192 words that
appeared on a computer screen. These words were designed to elicit feel-
ings of aggression, anxiety, escape, or control. For example, attack would
be associated with aggression, whereas flight would be associated with
escape. Unbeknownst to the participants, the time it took them to pro-
nounce each word was recorded. Anderson (1997) hypothesized that par-
ticipants who had viewed the violent clips would be primed, such that
words associated with aggression (i.e., attack) would be more accessible,
and thus more quickly pronounced, than words not associated with
aggression (i.e., flight). In both experiments, participants who watched the
violent clip rated themselves as higher in state hostility than participants
who watched the nonviolent clips, but there were no differences in the
aggressive word reading times between the violent and nonviolent condi-
tions. This initial result suggested that, although participants’ aggressive
feelings were more accessible after the violent clip, their aggressive
thoughts did not appear to be more accessible. However, in the second
experiment, Anderson (1997) found that participants low in trait hostility
who had seen the violent clip reacted faster to aggressive words than did
low-hostility participants who had seen only the nonviolent clip. Partici-
pants high in trait hostility were not affected by the content of the film
clips, in terms of their reading times for aggressive words. In sum, Ander-
son showed across two studies that violent media can prime both aggres-
sive feelings (i.e., trait hostility) and aggressive thoughts (i.e., accessibility
of aggression-related words). The latter is true primarily for persons low
in trait hostility.
Consistent with Josephson’s (1987) and Anderson’s (1997) studies, the
meta-analysis by Roskos-Ewoldsen et al. (in press) found that depictions
of violence or violence-related concepts (e.g., weapons) prime violence
and aggression-related concepts. The study of boys’ aggression (Joseph-
son, 1987) also suggested that priming may dissipate over time.
Addressed next is the media priming literature regarding political news
coverage.