The Effect of Flash Banners On Multiattribute Decision Making: Distractor or Source of Arousal?
The Effect of Flash Banners On Multiattribute Decision Making: Distractor or Source of Arousal?
The Effect of Flash Banners On Multiattribute Decision Making: Distractor or Source of Arousal?
on Multiattribute Decision
Making: Distractor or
Source of Arousal?
Rong-Fuh Day
Southern Taiwan University of Technology
Jyun-Cheng Wang
National Tsing Hua University
ABSTRACT
line while participants were making their decisions. With this approach,
direct observation of whether attention was allocated to the primary
task of decision-making or to the peripheral flash ads was possible.
Attention
Research on flash ad stems from two lines of investigation, namely, mar-
keting research of advertisement and visual search in the field of atten-
tion. In advertising research, researchers are mainly concerned with
whether or not flash would increase the recall of flash content or the
click rate (Cho, 2003; Li & Bukovac, 1999; Yoon, 2003; Zhang, 2000). In
attentional research on visual search, researchers have investigated
whether or not flash would increase or decrease the efficiency of infor-
mation search on the Web in terms of expended time and accuracy (Hong,
Thong, & Tam, 2004; Zhang, 2000). Although these two lines of research
clearly reflect different academic and practical interests, they both main-
tain the same theoretical assumption that attention plays an important
mediating role in information processing (Bettman et al., 1998; Kahne-
man, 1973; MacInnis & Jaworski, 1989; MacInnis et al., 1991).
Most researchers in the area of attention agree that attention can be
regarded as processing resources with limited supply when an individ-
ual is faced with multiple demands of concurrent mental activities (Kah-
neman, 1973; Luck & Vecera, 2002). Hence, issues regarding the allo-
cation and control of attentional resources are important. In general,
allocation and control of attention can be accomplished in two distinct
modes: voluntary and involuntary control (Kahneman, 1973). Voluntary
METHOD
Participants
A total of 30 college students were recruited as participants from the
Southern Taiwan University of Technology, Taiwan. Each participant
was paid a cash reward of NT$100 for their participation. In order to
encourage them to make decisions accurately, an additional incentive of
NT$80 was awarded based on decision accuracy.
Figure 1. Layout of the two information environments used in the experiment. The
top banner shown in the right panel is the peripheral flash advertisement. The central
portion of the both layouts is the decision information matrix about four skin-protec-
tion lotions. It describes four attributes, including SPF, Polished, Moisture, and Fresh-
ness. The values of each attribute are relative scores, ranging from 1 to 5. It is worth
noting that the original experimental layout was in Chinese. In this case, the EBA
decision tasks on the two decision information matrices are equivalent in terms of ele-
mentary information processes. The spatial distribution of eye fixations for the EBA
tasks in the two information environments is shown. As can be seen from the figure,
only a handful of eye fixations were laid on the peripheral banners.
Before any formal analysis, the eye movement data of each participant
were first inspected with a custom-made program to see whether or not
the data were valid (see Figure 1). Except for one participant whose data
was invalid, a total of 18,726 fixations from 29 participants were used for
further analysis, of which 12,036 fixations were recorded from the
no–flash-ad information environment, and 8,490 fixations were recorded
from the flash-ad information environment.
This study aims to clarify the role that flash peripheral ads may act as
a distractor or as a source of arousal in the on-line process of making
decisions. It was found that, with the same level of decision accuracy,
participants actually reached decisions with less time and exhibited a
higher level of arousal for equivalent decision tasks in the information
environment with peripheral flash ads than without. Moreover, it was also
found that participants fixated directly on the peripheral flash banners
only occasionally, with an average of less than 4% of the time. These find-
ings provide support to the view that flash peripheral ads tend to be a
source of arousal rather than a distractor. According to Kahneman’s
(1973) capacity model, the presence of flash peripheral ad first aroused
the participant, which in turn may have motivated the participant to
increase the supply of processing resources. Concomitantly, the decision-
making tasks benefited from sharing the additional supply of mental
resources and resulted in improved performance.
One of the most interesting aspects in these findings is that although
participants fixated their eyes on the peripheral ads only sparingly, they
were unable to isolate themselves completely from the influence of the
peripheral ad. Such influence is consistent with the nature of visual sys-
tem in that peripheral vision is dominated by the retinal magnocellular
system, which is sensitive to motion or signals of transient change (e.g.,
flash) arising from objects located in the peripheral area of the visual
field (Palmer, 1999). Therefore, in this study, the flash can exert its influ-
ence on the participants through peripheral vision, perhaps even with-
out their conscious awareness.
These findings also extend the popular Yerkes-Dodson (Yerkes & Dod-
son, 1908) law of psychological motivation to the on-line decision sce-
nario. The main idea suggested by the Yerkes-Dodson law is that arousal
is an important mediator or intervening variable in many types of behav-
ior. According to the Yerkes-Dodson law, when arousal level is posited on
a continuum where one end entails a state of calm and the other entails
an extreme heightened state, the relationship between the arousal level
and the cognitive performance can be characterized as an inverted-U
shape. In the present study, the information environment without flash
banners may represent a situation in which the participant’s inner state
was at the calm end. Upon the sudden presence of flash banners, how-
ever, the participant’s inner state may be shifted to the more heightened
end. As a consequence of heightened arousal, the participant accelerated
his or her speed in reaching a decision. The findings not only contribute