Page 115 - Media Effects Advances in Theory and Research

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

level of the node returns to its resting state and is no longer considered to
be activated.
Social psychologists began using priming procedures in the late 1970s
to study person perception, stereotyping, and attitude activation. The
general priming procedure in social psychological experiments involves
exposing participants to some priming event and then measuring
whether the priming event biased their interpretation of later ambiguous
information. For example, Srull and Wyer (1979) gave participants four
words (e.g., he, Sally, hit, kicked), and their task was to use three of the
words to construct a sentence. Unbeknownst to the participants, there are
only two sentences that can be constructed from these four words: “He hit
Sally” and “He kicked Sally.” In either case, according to network models
of memory, negative attitudes are activated, and the activation spreads to
the negative attitudes of other concepts (Fazio, 1986). When participants
next are asked to make various judgments of an ambiguously described
person or event, the negative aspects of the person or event fire sooner
than the positive aspects and subsequently influence the judgment more.
Consistent with the predictions of network models, the research in this
area typically finds that the ambiguous information is biased toward the
primes so that, if the primes are negative, the ambiguously described per-
son will be judged more harshly than if the primes are positive (Higgins,
Rholes & Jones, 1977; Srull & Wyer, 1979, 1980).
Research by both cognitive and social psychologists has demonstrated
two important characteristics of priming. First, the extent of a prime’s
effect on a target behavior or thought is a dual function of the intensity
and the recency of the prime (see the synapse model of priming, Higgins,
Bargh, & Lombardi, 1985). The intensity of a prime refers to either the fre-
quency of the prime (e.g., a single exposure vs. five exposures in quick
succession) or the duration of the prime. Higher-intensity primes produce
larger priming effects, and these effects dissipate more slowly than lower-
intensity primes (see Higgins et al., 1985). Recency simply refers to the
time lag between the prime and the target. Recent primes produce larger
priming effects than temporally distant primes.
A second important characteristic of priming is that the effects of a
prime fade with time. In lexical decision tasks (i.e., deciding whether the
target is a word or a nonword) and other related judgment tasks that use
reaction time as the dependent variable, the effect of the prime usually
fades within 700 milliseconds (Fazio, Sanbonmatsu, Powell, & Kardes,
1986; Neely, 1977). In tasks that involve judgments or evaluations of a
social stimulus, the effect of the prime also fades with time, though the
effect appears to fade more slowly (Srull & Wyer, 1979, 1980). In these
experiments, the priming effect can last up to 15 or 20 minutes, and possi-
bly up to one hour (Srull & Wyer, 1979). Srull and Wyer (1979, 1980) found