THE ROLE OF AFFECT IN INFORMATION SYSTEMS RESEARCH - Human-Computer Interaction and Management Information Systems: Foundations - page 284

et al., 1998) and reason-based assessments of the target are often regarded as having higher
normative/evidentiary status than feelings (Pham et al., 2001). However, affect sometimes pro-
vides judgmental responses that are potentially faster and more consistent across individuals, and
subsequently more predictive (Pham et al., 2001). Pham et al.'s four empirical studies showed that
affect performs better than cognition in predicting judgments. Similarly, Murry, Lastovicka, and
Singh (1992) also suggest that affective state has a direct effect upon attitudes toward advertising.
Researchers in marketing and consumer research have identified several conditions under which
affect has more influence. For example, Bagozzi et al. (1999) argued that when the processing of
information in a communication is low (e.g., due to low motivation, distraction, low need for cog-
nition, and weak arguments), emotional content in the communication gets processed directly and
transfers to, or influences, attitude toward the product or advertisement. In other words, the con-
sumer's affective mood has a direct effect on attitude when the likelihood of information process-
ing is low. When the likelihood of information processing is high, the affective mood influences
cognitive thoughts that are consistent with the mood.
Organizational behavior literature also demonstrates that affects such as mood (Weiss et al.,
1999) and beliefs tend to independently predict attitude such as job satisfaction, defined as the over-
all evaluation one makes about one's job and measured by “like/dislike” and other criteria (Weiss,
2002; Weiss et al., 1999). First, several researchers empirically proved that affects and beliefs are
distinct (Crites et al., 1994; Trafimow and Sheeran, 1998). Second, although they are independent,
affects and beliefs jointly predict users' reactions toward stimuli (Abelson et al., 1982; Breckler and
Wiggins, 1989; Crites et al., 1994). For example, Forgas (1995) argued that affects can influence
evaluative judgments directly through the principle of “affect-as-information.” Third, the relative
importance of affects and beliefs varies. Sometimes, affect has higher regression coefficients for atti-
tudes; at other times cognition has higher regression coefficients (Crites et al., 1994; Weiss, 2002).
The relationship between predictive capabilities of affect and cognition for behavioral inten-
tion is also reported in the literature. Trafimow and Sheeran (1998) conducted four analyses try-
ing to figure out which one accounts for more variance in behavioral intention for smoking. Their
results showed that affect accounts for significant unique variance in intention, but that cognition
has no such an impact on intention. However, in their studies, Trafimow and Sheeran (1998) fur-
ther tested the impacts of affect and cognition on the study habits of college students. As expected,
cognition, not affect, has significant impacts. Their findings suggest the strong moderating effects
of the task.

An Abstract Model of the Individual Interacting with Objects

Based on the above discussions, we present a general or abstract model of various elements
involved in the mental processes of an individual's interaction with an object (IIO) in his or her
environment. Figure 14.2 depicts this model; all concepts used in the model were defined earlier
in Table 14.1. The final dependent variables of the model are intention to interact with the object
or/and the actual behavior during the interaction. Studies in psychology have demonstrated the
strong relationship between these two dependent variables (Ajzen, 1985; Fishbein and Ajzen,
1975), which is beyond the scope of this paper.
The IIO model reflects the theoretical and empirical findings from existing literature discussed
above such as those about (1) trait

affective reaction (Brief, 1998; Fishbein and Ajzen, 1975;
Weiss, 2002; Weiss et al., 1999); (2) the interaction between affective and cognitive reactions
(Bagozzi et al., 1999; Berkowitz, 1993; Chaudhuri, 2002; Chen and Dubinsky, 2003; Epstein,
1993; Fishbein and Ajzen, 1975; Forgas, 1995; Forgas and George, 2001; Garbarino and Edull,

→