line of research, there are two major approaches. The first examines how af-
fect influences the storage and retrieval of information from long-term
memory (e.g., Bower, 1981; Forgas, 2000a). The second area focuses more
specifically on how affect influences the way in which information is proc-
essed and the way in which one approaches a particular situation. For in-
stance Ellis and Ashbrook (1988), Schwarz (1990), and more recently Bless
(2000) and Fiedler (2000) proposed theories suggesting that people in a pos-
itive versus negative mood may process information either analytically or
heuristically based on their mood. In addition, Fredrickson (1998) exam-
ined the role of positive emotions and suggested that positive emotions help
to broaden and build one’s resources. Given the complexity and breadth of
this research on affect and cognitive processing, we focus on how affect in-
fluences the way in which information is stored and processed, as we feel
this has the most profound implications for how affect influences learning
and achievement in academic settings (see Boekaerts, 1993; Pekrun, 1992;
Pekrun et al., 2002, for a general review of the influence of affect on learning
in academic settings).
Affect and Storage–Retrieval
From Long-Term Memory
Research relating long-term memory and affect has been seminal in incorpo-
rating affect into cognitive processing (Forgas, 2000b). The original research
in this area conducted by Bower investigated how mood influenced both the
encoding and retrieval of information from long-term memory. More specifi-
cally, Bower (1981) proposed an associative network theory suggesting that
mood was associated with information stored in long-term memory. Accord-
ingly, one would expect that a person’s current mood helps to activate infor-
mation that is congruent with this mood, thus making that information more
accessible (mood congruency effect). For instance, if a person is in a positive
mood, he is more likely to retrieve positively valenced information. In addi-
tion to mood congruency effects, Bower also proposed that a match between
mood during encoding and retrieval (state-dependent hypothesis) would fa-
cilitate recall. Bower’s associative network theory was tested in a variety of
studies, many of which did not consistently find results in line with the asso-
ciative network theory (Bower & Mayer, 1991; Forgas, 2000a).
In an attempt to reconcile the disparate findings regarding the relation of
mood to the retrieval of information from long-term memory, Forgas (1995,
2000a) proposed the affect infusion model (AIM). According to this model,
mood is only infused into a person’s thinking under situations where elabora-
tion and construction of knowledge is required. That is, one only recalls
mood congruent information from long-term memory when open, construc-
tive processes are required; thus the recall of mood congruent information
60 LINNENBRINK AND PINTRICH