International Journal of Electronic Commerce / Fall 2012, Vol. 17, No. 1, pp. 119–137.
Copyright © 2012 M.E. Sharpe, Inc. All rights reserved. Permissions: www.copyright.com
ISSN 1086-4415 (print) / ISSN 1557-9301 (online)
Attention to Banner Ads and Their Effectiveness:
An Eye-Tracking Approach
JooWon Lee and Jae-Hyeon Ahn
ABSTRACT: As with all forms of advertising, exposure is a necessary prerequisite for In-
ternet banner ad effectiveness. However, exposure does not guarantee a user’s attention,
an issue especially relevant to the Internet, where ad avoidance occurs most frequently.
And if an ad is noticed, the message may or may not remain in the consumer’s memory
after cognitive processing. However, even if the advertising message is not consciously
remembered, the exposure can be unconsciously processed and subsequently change
the user’s af fective state. To investigate how attention levels inﬂuence users, this study
uses eye tracking to measure the level of attention that results from an advertisement
exposure and explores how different levels of attention inﬂuence users in conscious and
unconscious ways. Also, we examine the effect of animation—one of the most popular
attention-grabbing tools—on at tention and how it moderates cognitive processing.
By measuring and analyzing users’ actual eye-movement data, we found that animation
in banner ads not only attracts less attention than static ads but also reduces the positive
effect of at tention on memory. In addition, although more than half of the participants
could not recognize the advertised brand, the animated banner ad was unconsciously
processed and did inﬂuence attitudes toward the brand. The results suggest that animation
in banner ads does not necessarily increase user attention, but that even if a user does not
consciously notice a banner ad, the user’s attitude toward the brand is inﬂuenced.
KEY WORDS AND PHRASES: Animation, at tention, banner ads, eye tracking, Internet
Spending for banner ads was projected to grow by $12.3 billion in 2011 and
to reclaim the top online ad segment from search ads within a few years .
However, due to a disappointing click-through rate (CTR) of less than one
out of one thousand Internet users, the effectiveness of banner ads has been
questioned. The growth in banner ad spending, despite this very low CTR,
implies that advertisers are beginning to understand that banner ads work like
other types of advertising, which is to say that exposure itself even without a
direct, immediate response may ultimately inﬂuence users’ brand preferences
and purchase choices.
However, exposure from the advertisers’ standpoint must be distinguished
from that from the consumers’ standpoint . Especially on the Internet,
where ad avoidance occurs most frequently , a signiﬁcant gap exists be-
tween advertisers’ assessment of exposure and Internet users’ assessment of
exposure. In order for the advertised message to be perceived and memorized,
gaining and preferably holding viewers’ attention is required. Since attention
is limited and selective [2, 67], not all the information on a Web page can be
The authors thank C.W. Park and Gratiana Pol for their valuable comments on an
earlier version of the paper.
120 LEE AND AHN
exposed and understood. Therefore, measuring attention rather than exposure
more accurately estimates advertising effectiveness.
Attention alone, however, is not the whole story: people do not remember
everything they notice . Sometimes an advertising message is cognitively
processed and remains in the consumer’s memory, but at other times it does
not. Clearly, a comprehensive investigation is needed to measure the level of
attention banner ads garner and how the attention exerts inﬂuence on users’
One of the most popular attention-grabbing tools employed in Internet
banner ads is animation, which is known to make objects salient and stimulate
higher levels of user involvement . Paradoxically, animation may alert
Internet users to the location of a banner ad, triggering ad avoidance behavior.
In addition, animated ads are known to require more of the reader’s cognitive
resources than static images , resulting in weaker memory performance. A
number of studies have shown that animation in banner ads is not an effective
tool. According to these studies, animation either does not affect memory [3,
14] or worsens it [6, 25].
Considering these ﬁndings, the ﬁrst objective of our study is to measure
the effect of animation on actual attention to assess ad avoidance. The second
objective is to investigate the effect of attention on memory moderated by the
existence of animation from the cognitive resource perspective. Which leads
us to the question: What happens when a banner ad is attended to but not per-
ceived? Because Internet users are known to pay minimal attention to banner
ads and devote minimal cognitive resources to ad processing , banner ads
are more likely to be processed unconsciously [38, 69]. Thus, memory mea-
sures may underestimate the effect of banner ads. A number of studies have
proved that in low-attention situations, advertising exposure is unconsciously
processed and does affect viewers’ judgment [38, 41], attitude [19, 69], or ad
evaluation . Does the unconscious effect occur without any attention at
all? Does more attention, even when not acknowledged, have a positive or a
negative effect? With these questions in mind, the third objective of our study
is to investigate the unconscious effects of attention with actual attention data,
which has not been studied empirically before.
For an advertising message to be most effective, both consumer attention
and cognitive processing are equally important [21, 40]. However, many
studies investigating memory-based cognitive processing assume salient
stimulus and repetitive exposure naturally result in attention . Yet recent
studies ﬁnd that it is difﬁcult to gain and hold the attention of consumers in
competitive advertising environments . The importance of attention in
a world of advertising overload cannot be overemphasized. Therefore, our
study focuses on how animation in banner ads inﬂuences attention by actually
measuring users’ attention with an eye-tracking device and how this attended
advertising message is memorized from the cognitive-processing perspective.
Furthermore, we investigate how attention without conscious perception exerts
an unconscious inﬂuence on a user’s affective state.
The rest of this paper is organized as follows: We examine the theoreti-
cal background and present the research hypotheses. Then, we describe the
research method and experimental procedure. Next, we report, analyze, and
INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 121
discuss the results of the experiment. We conclude with suggestions for future
Theoretical Background and Hypotheses
Selective Attention and Executive Functions
Because attention is limited and selective, only a portion of the information on
a Web page attracts people’s attention [40, 66, 67]. The directing of attention
is inﬂuenced by both bottom-up and top-down factors. It has been proven
that larger, brighter, and faster-moving objects or a unique object among
homogeneous distracters can be found more easily . This behavior is
attributed to bottom-up factors, whereby attention is automatically shifted
toward salient visual features. However, attention is not entirely responsive;
people can voluntarily guide their attention in accordance with their objec-
tives. These internal goals and intentions are known as top-down factors and
are closely related to executive functions [10, 58, 66]. Executive functions
are the high-level cognitive processes that are invoked in situations when
automatic processing does not sufﬁce and are considered to be closely related
to cognitive controls such as selective attention, behavioral inhibition, and
goal-directed behavior [18, 39]. According to this theory, neural pathways
between an attention-grabbing stimulus in the external environment and the
corresponding response are established innately or gradually with experience,
and the behavior of paying attention to that stimulus is automatically elicited
(bottom-up control). However, when this behavior conﬂicts with people’s
internal goals, top-down control is needed to willfully bias their attention
toward a goal-relevant stimulus that is in competition with a stronger but
goal-irrelevant stimulus, and this is when the executive functions are invoked
in order to override this otherwise automatic bottom-up processing .
Inattention to Animated Banner Ads
Although advertisers cannot control top-down factors, they can manipulate
bottom-up factors to increase stimuli salience in the hopes of catching users’
attention. Banner ads compete with editorial content as well as other banner
ads, so advertisers have employed a variety of attention-grabbing tools, such
as large size, vivid colors, and animation. Among these tools, animation is
the most common, and it has received much attention from academics and
practitioners [12, 25, 57].
The attention-attracting aspect of animation is supported by motion effect
theory, which claims that human beings tend to quickly direct their attention
toward moving objects and process the relevant information because they
regard moving objects in their peripheral vision as either threats or opportu-
nities . Despite the inherent attention-attracting aspect of animation, dif-
ferent users pay different amounts of attention to the same animated banner
ad because attention is determined by top-down factors as well as bottom-up
122 LEE AND AHN
factors. Top-down factors are particularly relevant to Internet users because
Internet users have been found to be more goal oriented  and to judge on-
line advertising more negatively than users of other media . Furthermore,
online ads that are embedded in editorial content are not Internet users’ main
interest. Therefore, attention is challenged when an Internet user whose goal
is to read editorial content rather than watch an advertisement (negative top-
down factor) faces a banner ad equipped with attention-grabbing animation
(positive bottom-up factor). Because Internet users are frequently in this situ-
ation, investigating the actual attention paid to banner ads should generate
valuable insight into users’ behavior.
According to executive function theory, it is initially necessary to elicit
executive function in order to select goal-relevant objects against strong, goal-
irrelevant objects; however, repeated selection will strengthen the pathway
from stimulus to response and ultimately automate the response . Like-
wise, in the context of Internet advertising, users learn to assume an animated
picture embedded in editorial content is a banner ad by past experience, that
is, even though animation is the important attention-grabbing tool supported
by motion effect theory, Internet users tend to develop a negative response to
banner ad–like objects, according to the executive function theory. This may
enable users to automatically avoid paying attention to animated banner ads,
contrary to advertisers’ expectations. Fixation duration and ﬁxation frequency
are the most commonly used measures of attention , and we propose the
following hypotheses based on the preceding discussion:
Hypothesis 1: Attention to an animated banner ad is lower than attention to
a static one.
Hypothesis 1a: Fixation frequency to an animated banner ad is lower than
ﬁxation frequency to a static one.
Hypothesis 1b: Total ﬁxation duration to an animated banner ad is shorter
than total ﬁxation duration to a static one.
Attention and Memory
In order for an advertising message to achieve its goal of consumer persua-
sion, attention alone is not enough; consumers must process what they have
seen . However, without attention, no further processing can occur to
inﬂuence subsequent consumer decision making. Correspondingly, more at-
tention leads to more opportunity to encode and store messages, and a positive
relationship between attention and memory has been found by a number of
eye-tracking studies [19, 27, 37, 47].
In addition, it can be inferred from the methodology employed in a number
of ad repetition studies [29, 68] that more attention will yield higher memory
performance: users were required to ﬁx their gaze on the advertising in most
experiments. Therefore, in conjunction with H1, claiming that banner anima-
tion depresses attention levels, we hypothesize a positive correlation between
attention and memory, which will complete our analysis of the effect of
INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 123
animation on memory through attention. Therefore, we can logically propose
the following hypothesis:
Hypothesis 2: More attention to a banner ad leads to better memory
Hypothesis 2a: More banner ad ﬁxation frequency leads to better memory
Hypothesis 2b: More banner ad total ﬁxation duration leads to better
Moderating Role of Cognitive Load
Once users pay attention to the marketing message, their cognitive and affec-
tive processes are triggered. This results in changes in their behavior as well
as in their psychological state, including memory, attitude, and preference .
Wedel and Pieters  showed that all these processes occur simultaneously
with and are accurately reﬂected in users’ eye movements; hence, it is said that
attention contributes strongly to the effectiveness of advertising [66, 67]. The
information gleaned from attention, however, is not all translated into memory;
the amount of information far exceeds what users’ brains can process .
According to theories of limited capacity [2, 34], people have limited cogni-
tive resources, which are allocated to each task depending on the available
resources and the user’s intention to process a message. Every cognitive task
requires a certain level of resources, and the amount depends on both the task’s
complexity and the person’s experience with the task. For successful message
processing, the resources allocated must meet the demand.
Internet users tend to devote minimal resources to processing banner ads.
If the advertisement itself requires too many cognitive resources to interpret,
resources will be lacking for further message processing, such as encoding and
storing, which are the necessary steps for memory . Animated banner ads,
which present new information in each frame, require more cognitive resources
than static banner ads. Therefore, we hypothesize that even among users who
pay the same amount of attention, memory performance will vary depending
on animated ad versus static ad exposure. In light of the preceding discussion,
we propose the following hypothesis concerning the moderating role of anima-
tion’s heavy cognitive load (i.e., the complexity of the message).
Hypothesis 3: A banner ad requiring more cognitive resources reduces the
positive effect of attention on memory.
Attention and Attitude: Mere Exposure Effect
Internet banner ads share space with editorial content and occupy only a
small fraction of the screen, but Internet users perform mostly goal-oriented
tasks: reading news, looking for information, and socializing. Thus, Internet
124 LEE AND AHN
banner ads are considered to be a classic example of unconsciously processed
messages, and Internet users are likely to be faced with persuasion in a very-
low-involvement situation .
Traditional theories have not addressed persuasion in this very-low-
involvement situation. The elaboration likelihood model, one of the most
widely adopted persuasion models, proposes that consumers follow a
central route in a high-involvement situation and a peripheral route in a
low-involvement situation. A central route requires extensive elaboration of
the information, resulting in a large amount of cognitive-processing effort,
whereas a peripheral route involves a meager amount of elaboration, result-
ing in less processing effort . As such, persuasion was once viewed as the
function of a consumer’s cognition of the message content . However, in
a very-low-involvement situation, where consumers pay little attention to
advertising and cognitive capacity is severely constrained, persuasion occurs
without cognition . Mere exposure effect is frequently proposed to explain
attitudinal changes that occur when ad exposure is so brief that its presence
is hardly recognized. This theory suggests that brief and repeated exposure
to a stimulus can encourage people to have familiarity and a more favorable
attitude toward that stimulus at an unconscious level, that is, even when they
cannot recollect being exposed to it [33, 71]. As such, affect and cognition are
proved to be processed independently , and a number of studies have
shown this unconscious effect of exposure [8, 17, 19, 35, 54]. Because mere
exposure effect tends to emerge in a low-attention situation [22, 70] and the
inﬂuence of incidental mere exposure is stronger when subjects are not aware
of the exposure , the level of attention has been proven to be negatively
associated with attitude . More speciﬁcally, Bornstein and D’Agostino have
shown that attitude toward the merely exposed stimulus is higher when expo-
sure durations are shorter and exposure frequency is higher, whereas longer
exposure weakens attitude but increases recognition . Taken together, we
propose the following hypotheses:
Hypothesis 4: Users with more frequent attention have more favorable at-
titudes toward the advertised brand.
Hypothesis 5: Users with longer attention duration per attention have less
favorable attitudes toward the advertised brand.
All the hypotheses are summarized in Figure 1.
Both static and animated versions of target banner ads were created. Animated
ads were set at two different speeds to check whether animation speed has
any effect on attention. The slow version consisted of four scenes per four
seconds, and the fast version consisted of ten scenes per four seconds; in
both cases each scene contained new information. One of the scenes from the
INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 125
animated version was employed in the static version to ensure that the static
version required less cognitive load than the animated version. All versions
the contained both text and graphics: text for the brand name and graphics
for the advertised product and related items. Figure 2 shows some examples
of banner ads.
Three different ads were designed for each animation speed to eliminate
ad-speciﬁc effects, which led to a total of nine target ads. Three common
product categories—toothpaste, coffee, and shampoo—were selected for the
target ads [1, 41]. To eliminate the effects of brand familiarity, ﬁctitious brand
names were used. These names were chosen after a pretest ensured brand
name neutrality in both meaning and attitude.
The participants were exposed to one out of nine target ads while going
through 20 Web pages. These pages were developed to be a replica of one of
the most popular online news portal sites. Twenty news stories about health,
travel, and cultural information were selected, and each Web page contained
one of these news stories accompanied by one banner ad on the upper right
side. Eight pages out of the 20 contained one of the target ads, and the other 12
pages contained one of the ﬁller ads created for this experiment. To guarantee
the salience of the banner ad, the banner ad was the only graphic on an all-text
page. Order of banner ad exposure was randomized across participants, and
target ads never appeared ﬁrst or last, so as to reduce primacy and recency
effects, respectively. The same ads were not shown to viewers consecutively,
but separated by at least one different banner ad.
Twenty news stories were randomly paired with either target or ﬁller
ads; each combination made one set, and other sets of news story–banner ad
combinations were formed to avoid any news story–speciﬁc effects and to
Participants and Procedure
One hundred and eighteen men and women from a business school volun-
tarily participated in the experiment; each received a lunch coupon in return.
Figure 1. Research Model with Hypotheses
126 LEE AND AHN
Participants were randomly assigned to one of the three animation speeds (fast,
slow, or static) and instructed to read 20 news pages as they normally would
while reading online news. The task was given to ensure that their attention
was under control of the top-down factors and their cognitive resources were
occupied  by the task, meaning that there would be less attention and fewer
cognitive resources available to the participants.
When going through news pages, participants could progress at their own
pace by pressing a “Next Page” button at the bottom of each page. Unlike
participants under forced-exposure conditions, in which they are asked to
pay attention to the advertising for a certain amount of time, participants
under this type of free-viewing condition are expected to pay little attention
to banner ads . Therefore, to ensure that enough attention data were col-
lected, the same target ad appeared 8 times as participants went through the
20 news pages. Some of the ﬁller ads were repeated as well so that the target
ad was not the only one appearing repeatedly.
The experiment lasted approximately 20 minutes. Attention data were
collected while participants viewed online news pages, but the participants
were not informed of this until the experiment was ﬁnished. Immediately
following the experiment, participants were instructed to ﬁll out a paper-
and-pencil questionnaire for memory, brand attitude, and demographic data
Figure 2a. Example of a Static Banner Ad
Figure 2b. Snap Shots from an Animated Banner Ad
INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 127
The most common approach for measuring the amount of attention paid to
advertising is to use self-reported memory measures by asking questions
such as “To what extent do you pay attention to the ads?” . However,
memory measures are a poor metric for measuring consumers’ attention
to advertising . Citing the problem of memory measures, Molosavljevic
and Cerf  argued that there are at least two problems: either a stimulus
is attended to but the awareness stage is not reached, making it impossible
to keep the stimulus in the memory and report it, or even when a stimulus
is stored in the memory, people may forget it, along with most of the stimuli
they have processed. Thus, to avoid the problem of memory measures, physi-
ological responses such as eye movements, which are tightly linked to shifts in
attention [11, 67], are considered more reliable indicators than self-reporting
or memory measures [31, 63]. In addition, findings from cognitive psychology
hold a great deal of promise for advertising research . To show how
cognitive processes work in various information-processing conditions ,
many eye-tracking studies have been conducted in the ﬁelds of information
systems [6, 9, 14, 23, 42] and marketing [45, 46, 65].
For collecting data on eye movement, an eye-tracking device is needed.
A Tobii T120 (Danderyd, Sweden) was used in this study. This device uses
infrared light to illuminate viewers’ eyes, the resulting reﬂections are picked
up by infrared sensors on the monitor, and the software then uses these data
to estimate eye position [13, 59]. Eye movements are collected with 120 Hz
frequency (or every 8.3 milliseconds) and then processed to calculate eye
ﬁxation frequency and duration. The Tobii T120 can identify where users are
gazing within less than a centimeter’s accuracy . Furthermore, it does not
require participants to mount any device on their head or eyes and looks like
a normal computer monitor, which allows for an unobtrusive and natural
environment in which to measure eye movement.
An eye-tracking device can measure ﬁxation frequency (i.e., number of eye
ﬁxations on target stimuli), ﬁxation duration (i.e., total duration of eye ﬁxation
on target stimuli), scan path (ﬁxation sequence), location of the ﬁrst ﬁxation,
time of the ﬁrst ﬁxation, and so forth . Because our interest is in processing
intensity at a speciﬁc location, ﬁxation duration and ﬁxation frequency were
chosen for the measurement .
Memory has been measured with recall, cued recall, and recognition in
many studies. Among these, recognition has been proven to be sensitive and
128 LEE AND AHN
discriminating enough when assessing memory  and is considered superior
to recall and cued recall . Furthermore, recall is reputed to underestimate
the effect of exposure in measuring ad effectiveness in low-involvement
situations ; thus, recognition was chosen to measure memory.
Recognition data were collected via questionnaire; participants were pre-
sented with four choices and asked to choose one banner ad they recognized
from the experiment. The brand names of the distracters used in the choices
were pretested for equivalent meaning: name familiarity, beneﬁt associations
implied by the name, and overall quality perceptions .
Attitude Toward a Brand
Attitude toward the advertised brand was collected via questionnaire and
assessed by averaging the scores of ﬁve nine-point measures: likable/unlik-
able, unpleasant/pleasant, appealing/unappealing, attractive/unattractive,
and bad/good [17, 28].
Effect of Animation on Attention
As stated in H1, we expected that attention paid would be lower for animated
banner ads than for static ads. The effect of animation on attention was tested
with a one-way ANOVA (analysis of variance) employing total ﬁxation dura-
tion and ﬁxation frequency as dependent variables. A signiﬁcant main effect
of animation on ﬁxation frequency (F(2, 115) = 5.410, p < 0.01, η2 = 0.08) and
total ﬁxation duration (F(2, 115) = 5.948, p < 0.01, η2 = 0.09) emerged, and
therefore H1a and H1b were supported (see Table 1). Static ads drew users’
eyes more frequently and for longer periods of time than ads with either
slow or fast animation. A post hoc test (Bonferroni test) showed a statistically
signiﬁcant difference between static ads and animated ads for both ﬁxation
frequency and total ﬁxation duration, but no difference between slow anima-
tion ads and fast ads. The Duncan test also divided the animation speed into
two homogeneous subsets: one consisting of static ads and the other of slow
or fast animation ads.
Effect of Attention on Memory and Moderating Effect of
Cognitive Load from Animation
We hypothesized in H2 that more attention to a banner ad would result in
better memory performance and in H3 that animation would reduce the
positive effect of attention on memory. Because of the dichotomous nature of
the recognition scores, we ran a logit regression using total ﬁxation duration,
ﬁxation frequency, and their interaction terms with animation as independent
variables and recognition as a dependent variable. Animation took on the value
INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 129
of 0 for static ads and 1 for animated ads. As shown in Table 2, participants’
recognition was positively affected by total ﬁxation duration (p < 0.01) but not
by ﬁxation frequency (p > 0.2), so H2b was supported but H2a was not. As
indicated by the negative coefﬁcient of the interaction of animation and total
ﬁxation duration in Table 2, animation weakened the effect of total ﬁxation
duration on recognition (p < 0.05). Thus, H3 was supported. Since ﬁxation
frequency did not have a signiﬁcant effect on recognition, the moderating
effect of animation on this relationship is not considered.
To determine whether visual attention fully mediated the relationship
between animation in banner ads and viewers’ memory, we conducted an
additional binary logistic regression using a bootstrapping technique .
Recognition was entered as a dependent variable, animation as a predictor
variable, and total ﬁxation duration as a mediator. The result revealed that
the total effect of animation on recognition (total effect = –0.77, p < 0.05) be-
came not signiﬁcant when the total ﬁxation duration mediator was included
in the model (direct effect of animation = –0.36, p > 0.3). A bootstrap analysis
showed that the 95 percent bias-corrected conﬁdence interval for the size of
the indirect effect excluded zero (–1.20, –0.18), indicating that there exists a
signiﬁcant indirect effect . Thus, we can say that total ﬁxation duration
fully mediated the relationship between animation and recognition.
Effects of Attention on Brand Attitude
In H4 and H5 we hypothesized a positive effect on brand attitude from fre-
quent attention and a negative effect on brand attitude from longer attention
duration per attention. Since ﬁxation duration per each attention matters in
these hypotheses, rather than total ﬁxation duration, we used the average of
all the ﬁxation durations for the analysis.
To test H4 and H5, we conducted a regression analysis with average ﬁxa-
tion duration and ﬁxation frequency as independent variables and attitude
toward the advertised brand as a dependent variable. Because mere exposure
effect was shown to emerge when the exposure is not acknowledged , this
analysis was conducted with data from participants who did not recognize
Table 1. Means and Standard Deviations for Total Fixation Duration
and Fixation Frequency.
Note: Numbers in parentheses are standard deviations.
130 LEE AND AHN
the banner ads to which they were exposed (n = 68). The results are shown
in Table 3 (R2 = 0.228, F(2, 66) = 9.76, p < 0.01). The negative coefﬁcient of the
average ﬁxation duration (p < 0.01) and the positive coefﬁcient of ﬁxation
frequency (p < 0.01) support H4 and H5.
Summary of Results
Our study examined the following: (1) the effect of visual stimuli (animation)
on attention; (2) the effect of various attention measures (total ﬁxation duration,
average ﬁxation duration, and ﬁxation frequency) on both memory (recogni-
tion) and attitude change; and (3) the moderating role of banner ads’ cognitive
requirement (animation). The results are summarized in Figure 3.
In this experiment, the argument that people instinctively pay attention
to visually salient stimuli did not hold true. Animated banner ads attracted
less attention than did static ads. The results of our eye-tracking study of-
fer an explanation for why animation in banner ads elicited worse memory
performance in past studies [3, 6], that is, it can be inferred from our study
that most Internet users employ executive functions to avoid visual objects
that are distinct from editorial content, regarding them as irrelevant to their
goal. In addition, animation speed itself did not make any difference in users’
ad avoidance behavior. This suggests that users have a tendency to avoid
animated objects regardless of the speed while surﬁng on the Web.
Evidence for an effect of attention on memory was found—the longer users
pay attention to a banner ad, the better their recognition performance. How-
ever, only total ﬁxation duration, not ﬁxation frequency, inﬂuences recogni-
tion performance. This implies that there is little difference between one eye
ﬁxation with long gaze duration and a number of short ﬁxations in terms of
Table 2. Logit Regression Results: Attention on Recognition.
Coefﬁcients 0.99 –0.12 0.77 –0.89 0.11
p0.01 0.20 0.11 0.03 0.34
Table 3. Regression Results: Attention on Attitude (N = 68)
Variable BSE(B) ß tSig. (p)
–4.08 1.60 –0.28 –2.55 0.01
Fixation frequency –0.05 0.01 0.45 4.06 0.00
2 = 0.228.
INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 131
The moderating role of cognitive load required by animated banner ads
was also found. Because of users’ limited cognitive capacity, animation in
banner ads negatively affects the relationship between attention (total ﬁxation
duration) and memory. Therefore, users may not be able to recognize a visu-
ally complex banner ad, despite attention paid, because of the extra cognitive
resources required for processing the ad.
Although more than half the participants could not recognize the advertised
brand, the banner ads imprinted unconsciously caused attitudinal changes.
This study supports the mere exposure effect by showing that frequent eye
ﬁxation improved participants’ attitude toward the advertised brand, whereas
longer average ﬁxation duration had a negative effect on attitude.
Discussion and Conclusion
In this study we aimed to investigate Internet banner ad processing and ef-
fectiveness from start to ﬁnish. Initially, we measured attention paid during
exposure to a message; next we assessed subsequent cognitive processing;
and ﬁnally, we analyzed the conscious and unconscious effects of attention.
Speciﬁcally, by utilizing a modern eye-tracking device, we collected actual
attention data from participants viewing the Internet at their own pace in a
natural setting. Findings from self-controlled exposure settings are considered
more reliable than those from forced-exposure settings, especially in the case of
advertising in very-low-involvement situations such as banner ads [40, 49].
Despite increasing understanding of the importance of attention in studying
consumers’ information processing, surprisingly little research exists on this
subject . A number of researchers have pointed out the need for additional
research on the role of visual stimuli in attracting attention [24, 30], particularly
in the context of Internet advertising, where static and dynamic visual features
are combined [66, 67]. From this study, by directly measuring and analyzing
Figure 3. Summary of Results
132 LEE AND AHN
actual attention to banner ads, we can draw several interesting conclusions.
First, we found that animation, the Internet’s most popular attention-attracting
tool, drives user attention away. These ﬁndings conﬁrm that, with the help of
executive functions, people are able to cognitively control their attention, se-
lecting goal-relevant objects over attention-grabbing stimuli. Second, we found
that various attention measures—speciﬁcally, ﬁxation frequency and ﬁxation
duration—have different effects on cognitive processing and subsequent
attitudinal changes. This would not have been uncovered without measuring
the actual attention. Third, we found that when banner ads are attended but
not remembered, they still can inﬂuence users in an unconscious way. Eye-
tracking data show that in natural exposure conditions many Internet users
pay very little attention to the banner ads. This means that banner ads are
placed in a mere exposure situation contrary to advertisers’ desires. However,
when repeated, very brief eye ﬁxations did change users’ attitudes even when
exposure was hardly recognized. Consequently, we veriﬁed the mere exposure
effect in the context of banner ads. These ﬁndings are particularly meaningful,
since most Internet users spare little to no attention and cognitive resource
on banner ads.
From a practical perspective, several strategies for banner ads can be derived
from the results of this study. First, advertisers should pay careful attention
when using animation in banner ads because animation may repel users’ at-
tention and hinder the cognitive processing of the exposed message. Second,
since ﬁxation frequency and ﬁxation duration have different effects, advertisers
should adopt different strategies according to their objectives. If brand recog-
nition is the main purpose of the advertisement, then only ﬁxation duration
matters and advertisers should try to extend the total ﬁxation duration either
by holding users’ attention as long as possible or by catching their attention as
often as possible. However, if a positive change toward brand attitude is the
main purpose, they should try to attract users’ attention frequently without
holding attention too long. Therefore, for the same total ﬁxation duration, it
would be more effective to attract attention with many short frequencies rather
than a few longer frequencies for both better memory and favorable attitude.
Advertisers should also be aware that even when users do not recognize—or
do not notice the existence of—a banner ad, it can still have an effect on users’
subsequent behaviors via unconscious inﬂuence.
This study has a couple of limitations. First, even though the experimental
situation was closer to real-life Internet use than experiments conducted in a
forced-exposure situation, the experiment was conducted in a laboratory set-
ting, and consequently the ﬁndings have limited potential for generalization.
Second, when testing the unconscious effect of attention on attitude toward
the brand (H4 and H5), 68 samples were used, which is not a large enough
sample size for a regression analysis. Although this was an inevitable result of
ensuring that only the unconscious effect was taken into account, additional
research with a larger sample size could generate more robust results. Third,
some variables that might be related to subjects’ news-reading behavior were
not strictly manipulated. By asking about and controlling their involvement
in reading online news as well as their prior Internet usage experience, the
results could have been more reliable.
INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 133
There are several areas that warrant further research. It is known that
repetitive exposures can easily engender irritation and subsequently inﬂuence
users’ attitudes toward an advertisement . Related topics to investigate
include (1) the relationship between eye-movement pattern and the feeling
of irritation and (2) the moderating role of irritation on the unconscious effect
of attention on attitude. In addition, it has been shown that task type on the
Internet, that is, information search versus online shopping, for example, af-
fects behavioral responses to advertising . This study employed only one
task type (news reading); thus, further research on how different tasks affect
Internet users’ behavior is warranted.
1. Baker, W.B. When can affective conditioning and mere exposure directly
inﬂuence brand choice? Journal of Advertising, 28, 4 (1999), 31–46.
2. Basil, M.D. Multiple resource theory I: Application to TV. Communication
Research, 21, 2 (1994), 177–207.
3. Bayles, M.E. Designing online banner advertisements: Should we ani-
mate? In L. Terveen (ed.), Proceedings of CHI 2002 ACM Conference on Human
Factors in Computer Systems, New York: ACM Press, 2002, pp. 363–366.
4. Bornstein, R.F. Exposure and affect: Overview and meta-analysis of
research, 1968–1987. Psychological Bulletin, 106, 2 (1989), 265–289.
5. Bornstein, R.F., and D’Agostino, P.R. Stimulus recognition and the mere
exposure effect. Journal of Personality and Social Psychology, 63, 4 (1992),
6. Burke, M.; Hornof, A.; Nilsen, E.; and Gorman, N. High-cost banner
blindness: Ads increase perceived workload, hinder visual search, and are
forgotten. ACM Transactions on Computer–Human Interaction, 12, 4 (2005).
7. Cho, C.H., and Cheon, H.J. Why do people avoid advertising on the
Internet? Journal of Advertising, 33, 4 (2004), 89–97.
8. Coates, S.L.; Butler, L.T.; and Berry, D.C. Implicit memory and consumer
choice: The mediating role of brand familiarity. Applied Cognitive Psychology,
20, 8 (2006), 1101–1116.
9. Cyr, D.; Head, M.; Larios, H.; and Pan, B. Exploring human images
in Website design: A multi-method approach. MIS Quarterly, 33, 3 (2009),
10. Desimone, R., and Duncan, J. Neural mechanisms of selective visual at-
tention. Annual Review of Neuroscience, 18 (1995), 193–222.
11. Deubel, H., and Schneider, W.X. Saccade target selection and object
recognition: Evidence for a common attentional mechanism. Vision Research,
36, 12 (1996), 1827–1837.
12. Diao, F., and Sundar, S.S. Orienting response and memory for Web ad-
vertisements: Exploring effects of pop-up window and animation. Commu-
nication Research, 31, 5 (2004), 537–567.
13. Djamasbi, S.; Siegel, M.; and Tullis, T. Generation Y, Web design, and
eye tracking. International Journal of Human–Computer Studies, 68, 5 (2010),
134 LEE AND AHN
14. Dreze, X., and Hussher, F.X. Internet advertising: Is anybody watching?
Journal of Interactive Marketing, 17, 4 (2003), 8–23.
15. Duchowski, A.T. Eye Tracking Methodology: Theory and Practice. London:
16. Fredricksen, C. Online advertising market poised to grow 20% in 2011.
eMarketer, June 8, 2011 (available at www.emarketer.com/PressRelease
17. Gardiner, J.M., and Richardson-Klavehn, A. Remembering and knowing.
In E. Tulving and F.I.M. Craik (eds.), The Oxford Handbook of Memory. New
York: Oxford University Press, 2000, pp. 229–244.
18. Gilbert, S.J., and Burgess, P.W. Executive function. Current Biology, 18, 3
19. Goodrich, K. Anarchy of effects? Exploring attention to online advertis-
ing and multiple outcomes. Psychology & Marketing, 28, 4 (2011), 417–440.
20. Grifﬁth, D.A.; Krampf, R.F.; and Palmer, J.W. The role of interface in
electronic commerce: Consumer involvement with print versus on-line
catalogs. International Journal of Electronic Commerce, 5, 4 (summer 2001),
21. Harrington, N.G.; Lane, D.R.; Donohew, L.; and Zimmerman, R.S. An
extension of the activation model of information exposure: The addition of
a cognitive variable to a model of attention. Media Psychology, 8, 2 (2006),
22. Heath, R.G.; Brandt, D.; and Nairn, A. Brand relationships: Strengthened
by emotion, weakened by attention. Journal of Advertising Research, 46, 4
23. Hervet, G.; Guérard, K.; Tremblay, S.; and Chtourou, M.S. Is banner
blindness genuine? Eye tracking Internet text advertising. Applied Cognitive
Psychology, 25, 5 (2010), 708–716.
24. Hoffman, D.L., and Novak, T.P. Marketing in hypermedia computer-
mediated environments: Conceptual foundation. Journal of Marketing, 60, 3
25. Hong, W.; Thong, J.Y.L.; and Tam, K.Y. Does animation attract online
users’ attention? The effects of ﬂash on information search performance and
perceptions. Information Systems Research, 15, 1 (2004), 60–86.
26. Hsieh, Y.-C., and Chen, K.-H. How different information types affect
viewer’s attention on Internet advertising. Computers in Human Behavior, 27,
2 (2011), 935–945.
27. Intraub, H. The role of implicit naming in pictorial encoding. Journal of
Experimental Psychology: Human Learning and Memory, 5, 2 (1979), 78–87.
28. Janiszewski, C. Preattentive mere exposure effects. Journal of Consumer
Research, 20, 3 (1993), 376–392.
29. Kirmani, A. Advertising repetition as a signal of quality: If it’s adver-
tised so much, something must be wrong. Journal of Advertising, 26, 3 (1997),
30. Kozinets, R.V. E-tribalized marketing? The strategic implications of
virtual communities of consumption. European Management Journal, 17, 3
INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 135
31. Krugman, H.E. A comparison of physical and verbal responses to televi-
sion commercials. Public Opinion Quarterly, 29, 2 (1965), 323–325.
32. Krugman, H.E. Memory without recall, exposure without perception.
Journal of Advertising Research, 40, 6 (2000), 49–54.
33. Kunst-Wilson, W.R., and Zajonc, R.B. Affective discrimination of stimuli
that cannot be recognized. Science, 207, 4430 (February 1, 1980), 557–558.
34. Lang, A. The limited capacity model of mediated message process. Jour-
nal of Communication, 50, 1 (2000), 46–70.
35. Lee, A.Y. Effects of implicit memory on memory-based versus stimulus-
based brand choice. Journal of Marketing Research, 34, 4 (2002), 440–454.
36. Li, H.; Edwards, S.M.; and Lee, J. Measuring the intrusiveness of adver-
tisements: Scale development and validation. Journal of Advertising, 31, 2
37. Loftus, G.R., and Kallman, J.J. Encoding and use of detail information in
picture recognition. Journal of Experimental Psychology: Human Learning and
Memory, 5, 3 (1979), 197–211.
38. Meyers-Levy, J., and Malaviya, P. Consumers’ processing of persuasive
advertisements: An integrative framework of persuasion theories. Journal of
Marketing, 63, (1999), 45–60.
39. Miller, E.K., and Cohen, J.D. An integrative theory of prefrontal cortex
function. Annual Review of Neuroscience, 24, 1 (2001), 167–202.
40. Molosavljevic, M., and Cerf, M. First attention then intention: Insights
from computational neuroscience of vision. International Journal of Advertis-
ing, 27, 3 (2008), 381–398.
41. Nordhielm, C.L. The inﬂuence of level of processing on advertising rep-
etition effects. Journal of Consumer Research, 29, 3 (2002), 371–381.
42. Owens, J.W.; Chaparro, B.S.; and Palmer, E.M. Text advertising blind-
ness: The new banner blindness? Journal of Usability Studies, 6, 3 (2011),
43. Perfect, T.J., and Askew, C. Print adverts: Not remembered but memo-
rable. Applied Cognitive Psychology, 8, 7 (1994), 693–703.
44. Petty, R.E.; Cacioppo, J.T.; and Schumann, D. Central and peripheral
routes to advertising effectiveness: The moderating role of involvement.
Journal of Consumer Research, 10, 2 (1983), 135–146.
45. Pieters, R., and Warlop, L. Visual attention during brand choice: The
impact of time pressure and task motivation. International Journal of Research
in Marketing, 16, 1 (1999), 1–17.
46. Pieters, R., and Wedel, M. Attention capture and transfer in advertising:
Brand, pictorial and text-size effects. Journal of Marketing, 68, 2 (2004), 36–50.
47. Pieters, R.; Warlop, L.; and Wedel, M. Breaking through the clutter:
Beneﬁts of advertisement originality and familiarity for brand attention and
memory. Management Science, 48, 6 (2002), 765–781.
48. Preacher, K.J., and Hayes, A.F. Asymptotic and resampling strategies
for assessing and comparing indirect effects in multiple mediator models.
Behavior Research Methods, 40, 3 (2008), 879–891.
49. Ray, M.L. When does consumer information processing research ac-
tually have anything to do with consumer information processing? In
136 LEE AND AHN
W.D. Perreault Jr. (ed.), Advances in Consumer Research, vol. 4. Atlanta: As-
sociation for Consumer Research, 1977, pp. 372–375.
50. Rayner, K.; Miller, B.; and Rotello, C.M. Eye movements when looking
at print advertisements: The goal of the viewer matters. Applied Cognitive
Psychology, 22, 5 (2008), 697–707.
51. Reeves, B., and Nass, C. The Media Equation: How People Treat Computers,
Television, and New Media Like Real People and Places. New York: Cambridge
University Press, 1996.
52. Rosbergen, E.; Pieters, R.; and Wedel, M. Visual attention to advertising:
A segment-level analysis. Journal of Consumer Research, 24, 3 (1997), 305–314.
53. Ryu, G.; Lim, E.A.C.; Tan, L.T.L.; and Han, Y.J. Preattentive processing
of banner advertisements: The role of modality, location, and interference.
Electronic Commerce Research and Applications, 6, 1 (2007), 6–18.
54. Shapiro, S.; Macinnis, D.J.; and Heckler, S.E. The effects of incidental ad
exposure on the formation of consideration sets. Journal of Consumer Re-
search, 24, 1 (1997), 94–104.
55. Singh, S.N.; Rothschild, M.L.; and Churchill, G.A. Recognition versus
recall as measures of television commercial forgetting. Journal of Marketing
Research, 25, 1 (1988), 72–80.
56. Stern, B.B.; Zinkhan, G.M.; and Holbrook, M.B. The netvertising image:
Netvertising image communication model (NICM) and construct deﬁnition.
In M.R. Stafford and R.J. Faber (eds.), Advertising, Promotion, and New Media.
Armonk, NY: M.E. Sharpe, 2005, pp. 30–50.
57. Sundar, S.S., and Kalyanaraman, S. Arousal, memory, and impression-
formation effects of animation speed in Web advertising. Journal of Advertis-
ing, 33, 1 (2004), 7–17.
58. Theeuwes, J. Top-down and bottom-up control of visual selection. Acta
Psychologica, 135, 2 (2010), 77–99.
59. Tobii eye tracking: An introduction to eye tracking and Tobii Eye Track-
ers. White Paper, Tobii Technology, Danderyd, Sweden, January 27, 2010
(available at www.scribd.com/doc/25907389/Tobii-Eye-Tracking-An-
60. Treisman, A., and Gormican, S. Feature analysis in early vision: Evi-
dence from search asymmetries. Psychological Review, 95, 1 (1988), 15–48.
61. Tsang, M.M.; Ho, S.; and Liang, T. Consumer attitudes toward mobile
advertising: An empirical study. International Journal of Electronic Commerce,
8, 3 (spring 2004), 65–78.
62. Vakratsas, D., and Ambler, T. How advertising works: What do we re-
ally know? Journal of Marketing, 63, 1 (1999), 26–43.
63. Vertegaal, R., and Ding, Y. Explaining effects of eye ﬁxation on mediated
group conversations: Amount or synchronization? In E.F. Churchill (ed.),
Proceedings of the 2002 ACM Conference on Computer Supported Cooperative
Work. New York: ACM Press, 2002, pp. 41–48.
64. Wang, K.; Wang, E.T.G.; and Farn, C. Inﬂuence of Web advertising
strategies, consumer goal-directedness, and consumer involvement on Web
advertising effectiveness. International Journal of Electronic Commerce, 13, 4
(summer 2009), 67–96.
INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 137
65. Wedel, M., and Pieters, R. Eye ﬁxation on advertisements and memory
for brands: A model and ﬁndings. Marketing Science, 19, 4 (2000), 297–312.
66. Wedel, M., and Pieters, R. Eye tracking for visual marketing. Foundations
and Trends in Marketing, 1, 4 (2006), 231–320.
67. Wedel, M., and Pieters, R. A review of eye-tracking research in market-
ing. In N. Malhotra (ed.), Review of Marketing Research. Armonk, NY: M.E.
Sharpe, 2007, pp. 123–147.
68. Yaveroglu, I., and Donthu, N. Advertising repetition and placement is-
sues in on-line environments. Journal of Advertising, 37, 2 (2008), 31–43.
69. Yoo, C.Y. Unconscious processing of Web advertising: Effects on implicit
memory, attitude toward the brand, and consideration set. Journal of Interac-
tive Marketing, 22, 2 (2008), 2–18.
70. Yoo, C.Y., and Kim, K. Processing of animation in online banner adver-
tising: The roles of cognitive and emotional responses. Journal of Interactive
Marketing, 19, 4 (2005), 18–34.
71. Zajonc, R.B. Attitudinal effects of mere exposure. Journal of Personality
and Social Psychology Monograph Supplement, 9, 2 (1968), 1–27.
72. Zajonc, R.B. Mere exposure: A gateway to the subliminal. Current Direc-
tions in Psychological Science, 10, 6 (2001), 224–228.
JOOWON LEE (firstname.lastname@example.org) is a marketing manager at SK Marketing & Company.
She received her Ph.D. in management engineering from the Korea Advanced Institute
of Science and Technology (KAIST) Business School in Seoul, Korea. Her research
interests include visual marketing, consumer psychology, and marketing strategies
for the new media industry.
JAE-HYEON AHN (email@example.com) is a professor of IT management at
the KAIST Business School, in Seoul, Korea. He received his Ph.D. in management
science and engineering from Stanford University. After graduation, he worked as a
senior researcher at AT&T Bell Labs from 1993 to 1998. His current research interests
are focused on strategy analysis of eWOM, content strategy in the new media industry,
design effectiveness and the evaluation of Internet advertisement through eye-tracking
approaches. He has published papers in various journals, including MIS Quarterly,
Management Science, Decision Support Systems, Journal of Information Technology, and
European Journal of Operational Research.