Mental Representation: Structure and Process 261The central process that is assumed to operate on this type of
representational structure is thespreading of activation.Each
of the nodes in a network can vary in its degree of activation.
When activation levels are minimal, the information contained
in a node is essentially dormant in long-term memory, exercis-
ing no influence over the ongoing course of social cognition.
However, when the level of activation rises above a critical
threshold, the information contained in the node is assumed to
enter working memory and to begin to influence ongoing cog-
nition. For example, if our hypothetical friend George were
suddenly encountered on the street, theGeorgenode in long-
term memory would be activated and thereby brought into
working memory. It is important that the activation that is in-
fused into the centralGeorgenode is assumed to spread along
the available links to connected concepts, with more activation
flowing along the stronger links. Whenever this activation is
sufficiently high, the connected links will also enter working
memory. Although there are variants on these assumptions, this
brief summary provides a reasonable description of the core
ideas of the associative network models.
The assumptions of the associative network models have
been used to illuminate a wide variety of social-cognitive
phenomena. To provide a representative sample, in this chap-
ter we focus on three domains in which such models have been
influential: attitudes, stereotypes, and memory for expectancy-
relevant material. Fazio (1986) proposed a model of attitude
structure that follows from the principles of the associative net-
work models. In his view, an attitude consists of a simple asso-
ciative structure: a node representing the attitude object, an
evaluative node, and a link connecting the two nodes. Of criti-
cal importance is the strength of the connecting link. For strong
attitudes, the link between the two nodes will be very strong,
and any time the node representing the attitude object gets ac-
tivated, the activation will be likely to spread to the evaluative
node, thereby activating the associated attitude. Weak atti-
tudes, however, will tend not to be automatically activated in
this way, because the link connecting the attitude object to the
evaluation is not likely to conduct enough activation to the
evaluation node when the attitude object node gets activated.
On the basis of this set of assumptions, Fazio was able to con-
struct a compelling model of the determinants of attitude-
behavior consistency. When attitudes are highly accessible
(i.e., when the link between the attitude object and the evalua-
tive node is strong), encountering the attitude object is likely to
be sufficient to activate the attitude. After it is brought into
working memory, the activated attitude can influence the on-
going stream of information processing by biasing the process
of interpreting the subjective meaning and perceived behav-
ioral affordances of the immediate situation. But none of this
will happen if the attitude is not sufficiently accessible.
Some prominent models of stereotyping also assume the
operation of an associative network structure (e.g., Devine,
1989; Dovidio, Evans, & Tyler, 1986). From this perspec-
tive, stereotypes consist of a central node representing a
particular social group (e.g., elderly people) that is linked to
various concepts that are assumed to characterize group
members (e.g.,slow, forgetful). When a member of the rele-
vant category is encountered, activation can spread along the
links from the central identity node to the associated stereo-
typical concepts. After these concepts enter working memory,
they can influence subsequent impressions and reactions. One
especially influential example is a study by Devine (1989).
In her experiment, concepts that are part of the African
American stereotype were activated (via a subliminal prim-
ing procedure). It is important that none of these concepts
dealt with the concept of hostility. However, because hostility
is assumed to be part of the cultural stereotype of African
Americans, Devine assumed that activating other parts of the
stereotype would also result in the activation of the concept of
hostility, through the spread of activation. In line with this
assumption, it was found that priming the African American
stereotype resulted in elevated perceptions of hostility on the
part of a subsequently encountered, ambiguous target. This
finding fits with the assumption that after a sufficient level of
activation reached the hostility node (by traversing the links
connecting it to the rest of the associative network), this
concept entered working memory and influenced subsequent
impressions. Devine argued that these associations are part
of culturally ingrained belief systems, and even when people
do not consciously endorse the relevant belief (e.g., even
when they donotbelieve that African Americans are hostile),
they are still prone to being influenced by the culturally
learned association.
Perhaps the most extensive development of associative
network models by social cognition researchers has occurred
as part of efforts to understand the impact of expectancies on
social memory (e.g., Hastie, 1980; Srull, 1981; Wyer & Srull,
1989). Researchers working in this tradition have attempted
to specify the factors that determine both the strength and the
types of links that form among activated pieces of infor-
mation, and they have also developed models addressing how
these associative structures are used in the process of memory
retrieval. In the typical experiment, participants learn some
initial facts about a particular target that establish a general
expectancy about him or her (e.g.,Tina is smart, intellectual,
well-educated,etc.). After an expectancy has been induced,
participants then read more detailed descriptions of the tar-
get’s behavior. These descriptions contain three classes of be-
haviors: those that are consistent, inconsistent, and irrelevant
to the general expectancy about the target. When a consistent