Page 183 - Media Effects Advances in Theory and Research

172 PETTY, PRIESTER, BRIÑOL

sufficient to increase personal involvement and processing of the message
arguments (see right panel of Fig. 7.3). That is, when the messages con-
tained the self-relevant pronouns, strong arguments were more persua-
sive and weak arguments were less persuasive than when third-person
pronouns were used. Yet another way to increase self-relevance is to frame
a message to comport either with people’s values or self-conceptions. For
example, if a person is attuned to the image value of a product, framing
the message as dealing with image can increase message processing (Petty
& Wegener, 1998b; see Petty, Wheeler, & Bizer, 2000, for a review).
Although increasing the perceived personal relevance of a message is
an important way to increase thinking (see Petty, Cacioppo, & Haugtvedt,
1992, for a review), it is hardly the only one. For example, the degree to
which a source is perceived to be of questionable or low trustworthiness
has also been found to increase the extent of elaboration (Priester & Petty,
1995). In this research, the extent to which a source could be trusted to
convey accurate information was manipulated while keeping source
expertise high. In one study, source trustworthiness was manipulated by
either providing message recipients with background information that
suggested that the speaker was honest and could be trusted or was dis-
honest and could not always be trusted to provide accurate information.
In another study, trustworthiness was manipulated by having the source
either advocate a self-serving position (relatively untrustworthy) or a
position that violated the source’s own self-interests (relatively trustwor-
thy). Regardless of how source trustworthiness was manipulated, sources
of questionable trustworthiness engendered greater elaboration than
sources perceived to be trustworthy.
This increase in elaboration occurred primarily for individuals who are
not intrinsically motivated to elaborate (i.e., low in need for cognition;
Cacioppo & Petty, 1982), prompting them to elaborate when they would
normally forgo such effortful processing. That is, an untrustworthy source
increased elaboration under conditions when individuals would likely
not normally have elaborated. In contrast, individuals who intrinsically
enjoy elaboration (i.e., high need for cognition individuals) elaborated the
messages equally regardless of source trustworthiness. Kaufman, Stasson,
and Hart (1999) uncovered a similar pattern of results. Low need for cog-
nition participants were more likely to elaborate the information pre-
sented by an untrustworthy (i.e., National Enquirer) than trustworthy (i.e.,
Washington Post) source.
Why does source trustworthiness influence elaboration? The ELM pos-
tulates both that (a) individuals are motivated to hold correct attitudes and
that (b) although individuals are motivated to hold correct attitudes, the
amount and nature of the elaboration on which these attitudes are based
varies. In combination, these two postulates offer the explanation that
source trustworthiness influences assurance of accuracy, and it is this assur-