Computers in Human Behavior 48 (2015) 463–472

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier.com/locate/comphumbeh

The cross-platform synergies of digital video advertising: Implications

for cross-media campaigns in television, Internet and mobile TV
Joon Soo Lim a,⇑, Sung Yoon Ri b, Beth Donnelly Egan c, Frank A. Biocca d,e
a
Department of Public Relations, S.I. Newhouse School of Public Communications, Syracuse University, 215 University Place, Syracuse, NY 13244, United States
b
YTN (Korea’s 24-hour News Channel), Department of News Production II, 76 Sangamsanno, Newsquare, Mapo-gu, Seoul 121-904, Republic of Korea
c
Department of Advertising, S.I. Newhouse School of Public Communications, Syracuse University, Syracuse, NY 13244, United States
d
Media Interface and Network Design (M.I.N.D.) Labs, S.I. Newhouse School of Public Communications, Syracuse University, United States
e
Interaction Science, Sungkyunkwan University, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: This study examines the synergy effect of digital video advertising through television, mobile TV, and the
Available online 24 February 2015 Internet on general outcomes of advertising effectiveness. In a 3 (paired media conditions for ad repeti-
tion)  2 (product involvement) mixed factorial design, we examined empirical outcomes of the cross-
Keywords: media synergy effect.
Media planning The results show that participants exposed to repetitive ads on paired media of television, Internet, and
Cross-media mobile TV have greater perceived message credibility, ad credibility, and brand credibility than counter-
Synergy
parts exposed to repetitive ads from a single medium. The multiple-media repetition also generated more
Mobile advertising
Internet advertising
positive cognitive responses, attitude toward the brand, and higher purchase intention than the single-
Mobile TV medium repetition. Finally, the cross-platform synergy effect remained robust for different levels of pro-
duct involvement.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction tablets, and smartphones. The consumption of television and com-

mercial messages these days increasingly takes place on non-linear
Marketers have long used multiple channels to deliver impact- media through mobile TV, a system that conveys the television
ful levels of reach to their campaigns. Each medium was carefully content to the mobile phone via wireless or cellular networks
planned to optimize reach and frequency within that medium and (Jung, Perez-Mira, & Wiley-Patton, 2009). According to Nielsen’s
then aggregated to deliver a total, synergistic impact to the con- Cross-Platform Report for Q2 2014 (Nielsen, 2014), 72% of Ameri-
sumer. Abernethy, Cannon, and Leckenby (2002) note that adver- cans own a smartphone and 39% own a tablet device. Total time
tisers in this era of integrated marketing communications are spent watching video is increasing but driven solely by the
making more effort to drive synergy across media campaigns. increase in digital video viewing. The adoption of mobile TV is
The benefits of harnessing synergy across multiple media to especially widespread in Asian countries where the smartphone
build brand equity of products and services have been extensively penetration rate is greater than that of other countries and the
examined in marketing research on media planning (Lin, mobile phone has become the dominant communication tool.
Venkataraman, & Jap, 2013; Naik & Raman, 2003). With the advent Developed in South Korea in 2005 as the next-generation digital
of the Internet and satellite technology, the ability for advertising mobile television system, Digital Multimedia Broadcasting (DMB)
firms to engage with consumers who serially consume multiple has broadened a concept of mobility in consuming television con-
media (Lin et al., 2013) is increasingly important. These consump- tent, allowing viewers to store the recipe while watching a cooking
tion patterns have generated a variety of new media viewing habits television program and to pay for the order made through home-
ranging from a serial viewing of small, incomplete chunks of mul- shopping channels (Shin, 2009). In a study by Kim and Jun
timedia called media multiplexing (Lin et al., 2013) to a non-linear (2008), it was reported that about 80% of the South Korean popula-
viewing of multimedia on portable devices such as mobile TV, tion is equipped with DMB devices. As the importance of mobile
advertising is growing, DMB is wielding considerable leverage as
a key player for effective advertising on mobile platforms (Kim &
⇑ Corresponding author. Tel.: +1 315 443 8046. Jun, 2008). Despite the importance of non-linear media in media
E-mail addresses: jlim01@syr.edu (J.S. Lim), risungyoon@naver.com (S.Y. Ri), and commercial message consumption, the relative impact of
beegan@syr.edu (B.D. Egan), frank.biocca@gmail.com (F.A. Biocca).

http://dx.doi.org/10.1016/j.chb.2015.02.001
0747-5632/Ó 2015 Elsevier Ltd. All rights reserved.
464 J.S. Lim et al. / Computers in Human Behavior 48 (2015) 463–472

potential synergy between DMB, Internet, and traditional media is more cognitive responses that eventually mediated the repetition
not well understood. effect on persuasion.
As with provision of new media, multiple uses of media have Chang and Thorson (2004) changed the course of ad repetition
become increasingly common. Nowadays, one watches TV at research by addressing the synergistic effects of ad repetition on
home while he or she watches DMB on the move and can surf general outcomes of advertising effectiveness. This line of research
the Internet anytime and anywhere. As consumers are getting on the synergy effect is increasingly important as the scope and
skillful at simultaneous media use, advertisers are growingly reach of cross-media advertising is ever increasing.
interested in the impact of synergy of effective cross-platform
campaigns. As this shift is occurring, it is becoming increasingly 2.2. The synergy effect of cross-media advertising
imperative to understand how messages across these media
interrelate. Synergy is a fundamental concept in media planning (Lin et al.,
The current study is interested in examining the effect of adver- 2013) and understanding the impact of synergy has become
tising synergies between linear (i.e., television) and non-linear (i.e., increasingly important in the age of media convergence
Internet and DMB) media platforms. Examining the effect of DMB (Voorveld et al., 2013). Sheehan and Doherty (2001) once asserted
in the context of synergy is important because previous research that the ‘‘strategic synthesis of strategy and tactics across multiple
implied that the synergy effect could be affected by the screen size channels is the hallmark of integrated marketing communication
(Varan et al., 2013). (IMC)’’ (p. 48). The synergy effect is important because IMC funda-
Although numerous studies (Chang & Thorson, 2004; Havlena, mentally pursues the maximization of ‘‘the benefits of harnessing
Cardarelli, & De Montigny, 2007; Jessen & Graakjær, 2013; synergy across multiple media to build brand equity of products
Sheehan & Doherty, 2001; Stolyarova & Rialp, 2014; Varan et al., and services’’ (Naik & Raman, 2003, p. 1).
2013; Voorveld, Smit, & Neijens, 2013) examined the synergy Media synergy arises when the combined effect or impact of a
effect of advertising on cross devices, little is known about whether number of media activities creates added value beyond the sum
the synergy effect can also be found in cross-media advertising of their individual effects on individual consumers (Schultz,
employing mobile devices (Bart, Stephen, & Sarvary, 2014). In a Block, & Raman, 2011). Among such media synergies, television
replication of Chang and Thorson’s (2004) study, we examined and Internet advertising synergies are regarded as the most popu-
the synergy effect of cross-advertising when an ad of both a high lar fit. In the case of the Internet, advertisers are starting to exploit
and a low-involvement product was repeated on a unichannel vs. the communication potential of this technology. This is because
on multiple channels of Internet, mobile TV (i.e., DMB) and the Internet is most effective in motivating consumers to par-
television. ticipate in information processing with a lot of cognitive effort
(Sheehan & Doherty, 2001). On the other hand, television is most
effective in influencing consumers without a lot of cognitive effort.
2. Theoretical background Such complementary traits of two different media can contribute
to create a synergy effect.
2.1. From mere exposure theory to two-factor theory In this regard, Chang and Thorson (2004) examined the adver-
tising effect when the ad was simply presented with repetition
The question that repeated exposure to the advertising message as opposed to when it was presented in synergistic conditions such
without involvement or learning process would result in sales via as an Internet ad followed by a television ad. For instance, par-
attitude change has been the long-sought research agenda (cited ticipants in their experiment were exposed to either an Internet
more than 1817 times) in advertising (Belch, 1982; Campbell & or television ad only twice while counterparts in the synergetic
Keller, 2003; Hawkins & Hoch, 1992; Krugman, 1965; Malaviya, conditions viewed the same ad on television followed by on the
2007) since Krugman’s arresting question in his address to the Internet or vice versa. They found that individuals who were
American Association for Public Opinion Research in 1965. The exposed to the ad on the synergetic conditions paid more attention
main thesis in Krugman’s thought-provoking question was how to the ad than those who viewed it in repetition. The synergy effect
advertising could change consumer attitude followed by purchase was also found across different measures such as message credibil-
behavior despite the fact that most messages are easily forgotten ity and the number of cognitive thoughts. However, they failed to
over time. This big question was followed by a groundbreaking find the direct effects of television–Internet ad synergy on brand
study in which Zajonc (1968) demonstrated what Krugman credibility, ad credibility, and attitude change. Rather, the effect
(1965) described as the effect of ‘‘much of advertising content . . . of television–Internet synergy on persuasion occurred as a result
learned as meaningless nonsense material’’ (p. 351) in his mere of differential processing routes. In other words, participants in
repeated exposure experiment. This seminal study by Zajonc the synergy condition exhibited attitude change via central pro-
(1968) in fact provided theoretical grounds on the persuasive effect cessing of the ad message in that the persuasive ad effect from
of mere exposure in inducing positive attitude change. the multiple media was mediated by increased attention and cog-
In proposing the two-factor theory, however, Berlyne (1970) nitive responses to the message itself. In contrast, attitude change
demonstrated that an exposure to a noble message initially among participants exposed to repetitive ad conditions (i.e., only to
increased attitude toward the product due to the positive learning Internet ads or television ads with repetition) was observed
factor. But repeated exposures to a homogenous stimulus resulted through the focus on a peripheral cue (i.e., advertiser credibility)
in diminished positive attitudes. The first positive affect results instead of message credibility. These results support Harkins and
from positive habituation, while the decreased affect is due to Petty’s (1987) theoretical assumption that the multiple sources
the boredom factor. Several studies that examined the relationship would enhance message-centric processing and ultimately lead
between repetition and attitudes found that the inverted U-rela- to persuasion via central processing of the message.
tionship between the ad repetition and attitudes could be moder- A substantial body of empirical research provided theoretical
ated by a message’s complexity (Anand & Sternthal, 1990; Cox & underpinnings of the potential effect of the cross-media synergy
Cox, 1988), an audience’s cognitive responses (Batra & Ray, (Naik & Peters, 2009; Naik & Raman, 2003; Voorveld, 2011;
1986), and brand familiarity (Machleit & Wilson, 1988). Other Voorveld & Valkenburg, 2014). They are (1) the multiple-source
researchers (Cacioppo & Petty, 1979; Rethans, Swasy, & Marks, effect, (2) differential attention hypothesis, (3) forward encoding
1986) showed that the repeated exposures to an ad could generate hypothesis, and (4) repetition-variation theory.
 J.S. Lim et al. / Computers in Human Behavior 48 (2015) 463–472 465

2.2.1. The multiple-source effect than an ad repeated in a different execution. The decreased atten-
Chang and Thorson (2004) first tested the synergy effect on the tion is the result of repetition wearout that occurs when indi-
theoretical background of the multiple-source effect. They revealed viduals do not perceive anything new in the coming advertising
that advertising through multiple sources can bring about more message (Naik, Mantrala, & Sawyer, 1998). Recent studies have
effects such as more positive thoughts and information processing demonstrated that cross-media advertising can slow down the
through a central processing route than repetitive ads in a single process of repetition wearout because varying the media can
source. reduce the repetition wearout phenomenon by motivating indi-
However, under the condition of repetitive ads in a single viduals to pay more attention (Voorveld et al., 2013), reducing
source, information was processed through a peripheral route: ad boredom and reviving consumer interest in the ad (Grass &
credibility affected by advertiser credibility elicited positive Wallace, 1969). As a result, varied repetition of a promotional mes-
thoughts (Chang & Thorson, 2004). Even though TV and Internet sage in two different media resulted in higher attention from the
synergies influence a consumer’s cognition, they exert less influ- audience than repetitive promotion from a single medium (Tang
ence on affective and conative states (Chang & Thorson, 2004). Par- et al., 2007).
ticularly, in cases of low-involvement ads of products such as soda Previous research revealed that the varied repetition of an ad, as
or a cleaning product, impact on cognition is insufficient to induce opposed to the same repetition of a single ad, made individuals
greater attitudinal and behavioral effects. undertake more effortful processing of the repeated message and
According to the Elaboration Likelihood Model (ELM hereafter), a resulted in more recall of the message (Unnava & Burnkrant, 1991).
person’s elaboration likelihood is high when the person has the
motivation to scrutinize the argument. Thus, when elaboration like- 2.2.3. The forward encoding hypothesis
lihood is high, a person will be attentive to message arguments and Previous studies have demonstrated that the mere exposure to
the person’s attitude will change via a central route to persuasion. ad messages was effective in inducing more cognitive responses
When elaboration likelihood is low, a person will be attentive to (Belch, 1982; McCullough & Ostrom, 1974). Synthesizing the previ-
peripheral cues such as advertiser credibility and the person’s attitude ous research on the information processing of cross-platform
will change via a peripheral route to persuasion (Petty, Cacioppo, & advertising, Voorveld and his colleagues (Voorveld, Neijens, &
Schumann, 1983). When information is processed through a central Smit, 2011) proposed the so-called forward encoding hypothesis.
route, people concentrate on the perceived merits of the ad’s message The forward encoding hypothesis is a cognitive account for
and related ad contents and consider them thoroughly (Petty & Zajonc’s (1968) mere exposure effect. In a nutshell, the forward
Cacioppo, 1996). This leads to an increase in ad credibility. encoding hypothesis assumes that the ad aired in the first device
Research findings also suggest that the use of cross-media inte- may leave a memory trace for the ad that will be presented in
gration enhances perceived media engagement of overall advertis- another device. Voorveld et al. (2011) compared this function to
ing messages (Tang, Newton, & Wang, 2007; Wang, 2011) such the role of a teaser in that the ad ahead of the second exposure
that multi-channel communication tends to enhance message can grab the audience attention and interest. The forward encoding
credibility of an ad. hypothesis is grounded on the memory trace hypothesis in previ-
Thus, we posit the following three hypotheses: ous research (Edell & Keller, 1989; Keller, 1987). Edell and Keller
(1989) said: ‘‘. . .when a person is exposed to an ad for a second
H1. Participants exposed to an ad in multiple media will have time, the ad may serve as a retrieval cue for the stored ad memory
higher perceived message credibility than those who viewed the trace or as a second encoding opportunity’’ (p. 150). A closely con-
same ad in a single medium. nected but slightly different theoretical account for the synergy
effect is the so-called the encoding-variability hypothesis. This
H2. Participants exposed to an ad in multiple media will have hypothesis postulates that the same ad message from multiple
higher perceived advertiser credibility than those who viewed media yields greater memory than from a single medium
the same ad in a single medium. (Stammerjohan, Wood, Chang, & Thorson, 2005). This enhanced
recall performance occurs due to the fact that the varied encoding
via different medium adds a new piece of information to episodic
H3. Participants exposed to an ad in multiple media will have
representation of the advertised message (Jin, Suh, & Donavan,
higher perceived ad credibility than those who viewed the same
2008).
ad in a single medium.
Although this forwarding encoding hypothesis accounts for the
Researchers have explained the mechanism that might explain effect of repeated exposure in general, Voorveld et al. (2011) sug-
why a cross-media campaign results in more persuasion than a gested that the forward encoding would give a greater advantage
repetitive campaign through a single medium. When individuals for the sequential exposure from multiple devices than from
are exposed to an ad message on multiple media, they tend to per- repeated exposure from a single device. In explaining the differen-
ceive each medium as an independent source of information. tial advantage, they reasoned that the repeated exposure to an ad
Because messages from multiple independent sources are per- from a single device would make individuals less motivated to pro-
ceived to be more credible, multiple sources can enhance the cred- cess the repeated message (Batra & Ray, 1986), especially when
ibility of the brand as well as the ad (Chang & Thorson, 2004; there is a low level of involvement in the promoted message
Harkins & Petty, 1987; Voorveld, 2011). Therefore, we proposed (Cacioppo & Petty, 1979).
the following hypotheses: Therefore, we posit H5:

H4. Participants exposed to an ad in multiple media will have H5. Participants exposed to the same ad in multiple media will
higher perceived brand credibility than those who viewed the have more positive cognitive responses than those who viewed the
same ad in a single medium. same ad in a single medium.

2.2.2. The differential attention hypothesis 2.2.4. The repetition-variation theory

Unnava and Burnkrant (1991) also raised a possibility that indi- The repetition-variation theory accounts for the effect of vary-
viduals’ attention paid to a single ad that is repeated will be lower ing the information modality on attitude toward brand
466 J.S. Lim et al. / Computers in Human Behavior 48 (2015) 463–472

(Schumann, Petty, & Clemons, 1990; Stammerjohan et al., 2005). involvement is controlled, the ad credibility is influenced by ad
This theory postulates that information modality can be varied message credibility and advertiser credibility (MacKenzie & Lutz,
either by cosmetic variation or substantive variation. The former 1989).
is about the variation in nonsubstantive features of the ad, whereas Cognitive responses are both positive and negative ad-related
the latter deals with the variation in message content thoughts that are generated while watching the advertisement.
(Stammerjohan et al., 2005). The repetition-variation theory is a Previous research has revealed that the message credibility affects
useful theoretical framework since it can explain that the positive the cognitive responses especially when the consumer is highly
attitude change for brand is not so much the number of exposures involved in the product (Chang & Thorson, 2004). The cognitive
to an ad as the modality of ad presentation. response is also affected by ad credibility particularly under the
On the basis of the repetition-variation theory, a growing body low-involvement condition (Petty, Cacioppo, & Schumann, 1983).
of research suggests that when people are exposed to a message in Finally, attitudinal measures are composed of the most pivotal
multiple media, instead of being exposed to a message in the same two constructs (Spears & Singh, 2004): attitudes toward the brand
medium repetitively, they have more positive affective reactions to and purchase intent. Several studies (Rethans et al., 1986) have
the brand (Voorveld, 2011). In an experimental study, Voorveld measured the cognitive and attitudinal measures for advertise-
et al. (2011) found that participants in a cross-media campaign ments as outcomes of synergetic advertisement repetition. Accord-
condition had a greater interest in processing the message than ing to MacKenzie and Lutz (1989), the ad perception (i.e., ad
those who were in the TV–TV condition, which in turn led them credibility) is one of the important antecedents to observing the
to have more positive attitudes toward the brand. attitude toward brand and purchase intention. The ELM explains
Janiszewski (1993) earlier demonstrated how the incidental that the ad credibility is influenced by the message credibility
exposure to a brand name can increase positive affect as a function when an individual engages in central processing and that ad cred-
of availability of stored representation in memory representation. ibility is affected by the advertiser credibility under peripheral pro-
According to Janiszewski (1993), the effect of mere exposure is cessing. Finally, in the classic attitude toward ad framework
influenced by how much the incoming message matches the stored (MacKenzie & Lutz, 1989; Spears & Singh, 2004), attitude toward
memory representation formed in previous incidental exposure, the brand has a direct impact on purchase intent.
which is called the match-activation hypothesis. As the traditional
cognitive response theory posits, such positive affect, including 2.4. Multiple-source effect under different levels of involvement
positive thoughts, wields influence over attitude toward brand
(MacKenzie, Lutz, & Belch, 1986; Maclnnis & Jaworski, 1989). An important factor that can affect the overall impact of the
Thus, we proposed the following hypothesis: cross-media advertising on persuasion is consumers’ product
involvement (Voorveld, Neijens, & Smit, 2012). As demonstrated
H6. Participants exposed to an ad in multiple media will have by numerous studies, product involvement tends to moderate
more positive attitude toward the brand than those who viewed the influences of advertising (Petty et al., 1983).
the same ad in a single medium. An emerging question is whether the level of involvement
From a perspective of IMC approach, the combined effects of would moderate the multiple-source effect. Voorveld et al.
multimedia activities can exceed the sum total of their individual (2012) found that a TV–Internet commercial sequence was effec-
contributions (Naik & Raman, 2003). Thus, such a multiple source tive for inducing persuasion for both high- and low-involvement
effect is able to exercise influence on brand credibility and pur- products while Internet–TV sequence was only effective for high-
chase intent. involvement products. Their findings indicate that TV commercials
In ELM, repeated ad exposure effect was explained in terms of are most effective for a low-involvement product, while they can
peripheral processing. In other words, those who are exposed to also exert influence on consumers who process a high-involvement
repetitive ads focus and elaborate on peripheral signals such as product. In contrast, Internet advertising requires consumers to be
advertiser credibility (Petty & Cacioppo, 1996). Ad credibility that involved in a certain level of ‘‘active, conscious, and cognitive infor-
influences brand credibility is determined by advertiser credibility mation process’’ (Cho, 1999, p. 36) such that it may limit the effect
(Chang & Thorson, 2004; MacKenzie & Lutz, 1989). Higher ad cred- for a high-involvement product.
ibility leads to more positive thoughts through a lower level of We assume that the way that DMB advertising works for induc-
information processing such as the pure affect transfer or heuristic ing persuasion would be similar to how TV advertising works since
evaluation (Maclnnis & Jaworski, 1989). Together with ad credibil- it allows individuals to consume the message on the go and spend
ity, positive thoughts induce more positive brand credibility less cognitive effort than they do to process Internet advertising.
(Fishbein & Ajzen, 1975) and increase purchase intention. So, we can predict that the DMB–Internet sequence might be
effective for either high- or low-involvement products as the TV–
H7. Participants exposed to an ad in multiple media will have a Internet sequence has proven to be effective for both high- and
higher purchase intention than those who viewed the same ad in a low-involvement products (Voorveld et al., 2012). Following the
single medium. study from Voorveld et al. (2012), we also assume that the Internet–
DMB sequence will be effective only for a high-involvement
product.
2.3. Summary of the constructs and relationships among them Although DMB and TV have considerable common denomina-
tors, the advertising effectiveness between the DMB–TV sequence
The aforementioned hypotheses aimed to examine general out- and the TV–DMB sequence can be different. Equipped on a smart-
come measures of the synergetic advertising repetition that have phone, DMB enables users to personalize the content on the go,
been examined in various empirical studies. Table 1 showcases which inherently increases individuals’ cognitive effort. Therefore,
the selected research results that tested the synergistic repetition the DMB–TV sequence can be effective for the high-involvement
effects on various outcomes for advertising effectiveness. product because advertising on DMB devices can enhance advertis-
The measures can be identified in three types of responses: ing effectiveness by reducing ‘‘the ad clutters or consumer ad
credibility measures, cognitive responses, and attitudinal mea- avoidance’’ (Kim & Jun, 2008). On the basis of key attributes of
sures. Credibility measures include ad credibility, message cred- DMB devices, we assume that a DMB commercial may exert less
ibility, brand credibility and advertiser credibility. When influence than a TV commercial for a low-involvement product.
 J.S. Lim et al. / Computers in Human Behavior 48 (2015) 463–472 467

Table 1
Selected studies on the cross-media synergy effect.

Author(s) (year), [Method] Dependent Media Major findings

variables
Edell and Keller (1989), [Experiment] CR, AB, PI T+R The T–R and R–T elicited significantly fewer cognitive responses than did the T–T but
more responses than the R–R condition
AB and PI: only significant main effect of media exposure. T yielded greater effect than
did R. No cross-media exposure effect was found
Confer and McGlathery (1991), [Experiment] AB, PI T+P The combination of T + P produces greater attitudinal change than a single-medium
repetition
Campbell and Keller (2003), [Experiment] CR, AB T+W Brand familiarity moderated the effect of repetition on AB
Chang and Thorson (2004) [Experiment] AC, AB, PI, BC, T+W Synergy effects on MC, CR
MC, CR No synergy effects on AC, AB, PI, BC, and ATC
Dijkstra, Buijtels, and van Raaij (2005), CR T + P+ CR: T was superior in evoking CR
[Experiment] AB W AB and PI: No significant main effect of media type. Only the main effect of
involvement was witnessed
PI
Stammerjohan et al. (2005), [Experiment] AB P+R The main effect of media on AB was found only in one condition of two experiments
Tang et al. (2007), [Experiment] MC, AB T+P The coordinated T and P repetition led to higher perceived MC and more positive AB
compared to a single-medium repetition
Voorveld et al. (2011) [Experiment] AB, PI T+W Cross-media repetition yielded a more positive AB and a higher purchase intention
than a repetition in the Internet
Voorveld (2011), [Experiment] AB, PI W+R Combining W and R ads resulted in more positive AB and PI than using only one
medium
Voorveld et al. (2012), [Experiment] MC T+W Significant interaction effect of the cross-media repetition and involvement on MC
This study, [Experiment] AC, AB, PI, BC, T+W+M Synergistic repetitions from paired media of T, W, and M yielded greater scores for all
MC, CR, ATC dependent variables than single medium repetitions

Note: AC = Ad Credibility, AB = Attitude toward the Brand, PI = Purchase Intention, BC = Brand Credibility, MC = Message Credibility, CR = Cognitive Response, ATC = Advertiser
Credibility.
T = Television, R = Radio, P = Print, W = Internet, M = Mobile.

When consumers are first exposed to a TV commercial, however, Chang and Thorson’s (2004) experiment on television and Internet
the TV–DMB sequence can also be effective for both high- and advertising synergies. The design was a 3 (paired media condition
low-involvement products since the television ad first triggers for ad repetition: repetition on a single medium vs. repetition on
interest and enhances attention as evidenced by previous research multiple media vs. reversed repetition on multiple media)  2
(Voorveld et al., 2013). (level of product involvement: high vs. low) mixed factorial design.
In the current study, we replicated the previous research on the The paired media condition was a between-subjects factor with
synergy effect by varied repetition by adding mobile TV advertis- exposure to a commercial on a single medium [TT (television–tele-
ing. While adding DMB in cross-media advertising for examining vision), DD (DMB–DMB), and WW (Internet–Internet)] vs. multiple
the synergy effect along with television and the Internet, we treat- media [TD (television–DMB), DW (DMB–Internet), and TW (televi-
ed involvement as a within-subject factor. Though Voorveld et al. sion–Internet)]. The exposure to multiple devices was replicated
(2012) tested the interacting role of media sequence and product by changing the order of a multiple repetition condition [DT–
involvement in cross-media campaigns, they only used two multi- WD–WT]. The level of product involvement was a within-subject
channel (exposure to multiple media) conditions of TV–Internet factor by which each subject was exposed to each commercial in
and Internet–TV sequences. In the current study, we also added both high- and low-level of product involvement.
another condition that was not considered in Voorveld et al.’s Two hundred eighty-two undergraduate students in a large pri-
(2012) study, which is to examine the varied multiple-repetition vate university in South Korea who had experience using TV, DMB,
effect when the order of the ad presentations was reversed. and Internet were recruited for the current study in return for
Since there is scant empirical evidence of the interaction effect research participation credit for an introductory marketing class.
of varied multiple repetitions and involvement on persuasion, we Participants were randomly assigned to one of three paired media
propose a research question: conditions in which they were simultaneously asked to watch two
30-s commercials that were high vs. low in product involvement
RQ1. Will the persuasive effect of varied multiple repetition for about 1 min. Then participants filled in a paper-and-pencil
predicted in H1–H7 be different for a high- vs. a low-involvement questionnaire.
product? In other words, will there be differences in the number of Ninety-eight participants took an experiment under the repeti-
positive thoughts, message/source/ad/brand credibility, attitude tive ad condition (TT = 37, DD = 31 and WW = 30), 91 participants
toward brand, and purchase intention between repetition on a under the multiple-device condition (TD = 31, DW = 30, TW = 30)
single device and varied repetition under different level of and 93 under the multiple-device condition in a reversed order
involvement? of media (WT = 32, DT = 31 and WD = 30). Two hundred eighty-
two participants’ mean age was 23 years old and 53.9% of par-
ticipants were male.
3. Method
3.2. Experimental stimuli
3.1. Design, participants, and procedure
One news story and two television commercials were presented
An experiment was conducted to analyze how media conditions to participants in each of the three experimental conditions. In
and involvement of advertised products impacted on consumers’ order to create a natural viewing atmosphere for the commercials,
thinking and attitude. In the first experiment, we replicated a news story of YTN, Korea’s news channel, was included. The story
468 J.S. Lim et al. / Computers in Human Behavior 48 (2015) 463–472

was a tidbit of the Ukrainian couple who did not wash for sixteen unidimensional summary evaluation of the brand that presumably
years. The news story and commercials for Internet and DMB were energizes behavior’’ (p. 55).
identical to the original television commercial. A pretest was con-
ducted to select a commercial of a high-involvement product and a
3.3.3.2. Purchase intention. The purchase intention in this study is
commercial of a low-involvement product.
referred to as ‘‘an individual’s conscious plan to make an effort to
For the commercials to be presented on the Internet, a website
purchase a brand.’’
of YTN was professionally simulated from an existing YTN site. Fol-
lowing Chang and Thorson (2004), participants in the TW condi-
tion first watched the news and commercials on a 42-inch 4. Results
television and then they were presented with the stimuli on a
19-inch computer screen. Participants in the Internet–DMB condi- 4.1. Manipulation check
tion first viewed the stimuli on the Internet through a 19-inch
computer monitor, followed by viewing on their 3.5-inch DMB-e- To check participants’ involvement in two products promoted
quipped smart phone. in two commercials, four statements adapted from Petty and
Cacioppo’s (1996) study were asked on a Likert scale anchored
3.3. Measures by ‘‘1’’ = ‘‘strongly disagree’’ and ‘‘7’’ = ‘‘strongly agree’’. The four
statements were: (1) I carefully evaluated the product’s advantages
Table 2 presents key constructs, scale items and reliability of and disadvantages; (2) The product is important to me; (3) I am
the measured items in different involvement levels. The constructs interested in the product; (4) The product means a lot to me. Par-
of the study can be identified in ad- and brand-related credibility, ticipants rated a laptop as a high-involvement product and a pain-
cognitive responses and attitudes. The following sections summa- killer as a low-involvement product. The mean difference of a
rize the operational definitions for each construct. summed involvement index for two products was statistically sig-
nificant. Mean for a high involvement product was 3.92 (SD = .79),
3.3.1. Ad- and brand-related credibility measures and mean for a low-involvement product was 2.17 (SD = .88),
3.3.1.1. Ad message credibility. The measures for ad-related percep- t = 26.70, df = 281.
tions include ad message credibility, advertiser credibility and ad
credibility. Ad message credibility is the perceived believability 4.2. Hypothesis testing
and trustworthiness of an advertisement’s message (Brennan &
Bahn, 2006; Chang & Thorson, 2004). 4.2.1. Ad message credibility
A one-way analysis of variance (ANOVA) was performed for
3.3.1.2. Advertiser credibility. Advertiser credibility is defined as assessing message credibility of the three experimental conditions.
‘‘the perceived truthfulness or honesty of the sponsor of the ad’’ The analysis was performed by splitting the results for commer-
(MacKenzie & Lutz, 1989, p. 51). cials for the high and low-involvement products. Table 3 presents
the overall results for the hypothesis tests using a one-way analy-
3.3.1.3. Ad credibility. Ad credibility refers to ‘‘the extent to which sis of variance (ANOVA).
the consumer perceives the message in the ad to be believable’’ The ad repetition in the multiple media conditions yielded high-
(MacKenzie & Lutz, 1989, p. 51). er message credibility than in the repetition condition on a single
device for both high- and low-involvement products. In a high-in-
3.3.1.4. Brand credibility. Brand credibility was defined as ‘‘the volvement condition, the message credibility score was higher for
extent to which the consumer perceives claims made about the repetition on multiple devices (M = 3.11, SD = .79) and reversed
brand in ads to be truthful and believable’’ (MacKenzie & Lutz, repetition on multiple media (M = 2.96, SD = .72) than for a single
1989, p. 51). According to MacKenzie and Lutz (1989), the ad-relat- medium (M = 2.63, SD = .74), F (2/279) = 10.43, p < .001. For a
ed credibility influences brand-related perception (i.e., brand low-involvement product, participants also rated higher message
credibility). credibility in repetition from multiple devices (M = 3.04, SD = .87)
and a reversed repetition from multiple devices (M = 2.77,
3.3.2. Cognitive response SD = .95) than a repetition from a single device (M = 2.42,
As reviewed earlier, cognitive response is an important measure SD = 0.78), F (2/279) = 12.28, p < .001. A post hoc analysis using
to understand the repetition-induced persuasion in advertising Tukey’s b further demonstrated that there was no mean difference
research (Cacioppo & Petty, 1979). We adopted the same technique between two multiple-media repetitions, whereas there were sig-
that Chang and Thorson (2004) used to elicit cognitive thoughts as nificant mean differences between multiple-media conditions and
a response to ad exposures. Chang and Thorson’s technique to single-medium conditions. Therefore, H1 was supported.
measure cognitive responses is composed of two stages: first, par-
ticipants were asked to list all the thoughts that came to mind 4.2.2. Advertiser credibility
while watching the commercials of a laptop and a headache pill; To test H2, a one-way ANOVA was performed. As predicted in
then they rated each of the elicited thoughts as positive, negative, H2, advertiser credibility was perceived higher in multiple-media
or neutral. repetition conditions than single-medium repetition conditions
for both high- and low-involvement products. A post hoc
3.3.3. Attitudinal measures Tukey’s b pairwise comparison revealed that there was slight
Spears and Singh (2004) reviewed that two attitudinal con- difference of means between two multiple-media conditions
structs of attitude toward the brand and purchase intention are [MMultiple-media repetition = 3.32 (SD = .65), MReversed multiple-media repetition =
most popular and pivotal in the marketing research based on the 3.05 (SD = .79)] for a low-involvement product, while there was
hierarchy of effect model rooted in attitudes. also a significant difference between multiple-media repetition
and single-medium repetition [MSingle-medium repetition = 2.78
3.3.3.1. Attitude toward the brand. Spears and Singh (2004) (SD = .82)], F (2/279) = 11.89, p < .001. For a high-involvement pro-
define attitude toward the brand as ‘‘a relatively enduring, duct, both multiple-media repetitions yielded higher advertiser
 J.S. Lim et al. / Computers in Human Behavior 48 (2015) 463–472 469

Table 2
Key constructs, measurement items, and item reliability for composite index.

Construct Question Items Cronbach’s a

Message credibility (Chang and Thorson, 2004) How reliable [. . .] was the information of the ad you Not reliable/reliable High Involvement .836
watched? Not credible/credible Low Involvement .917
Not believable/believable
Advertiser credibility (MacKenzie and Lutz, How convincing [. . .] did you think the advertiser of the Not convincing/convincing High Involvement .840
1989) ad you watched? Biased/unbiased
Not believable/believable Low Involvement .880
Ad credibility (MacKenzie and Lutz, 1989) How convincing [. . .] did you feel that the ad was? Not convincing/convincing High Involvement .848
Not believable/believable Low Involvement .872
Biased/unbiased
Brand credibility (Chang and Thorson, 2004) How would you rate advertised brand along these Not reliable/reliable High Involvement .901
scales? Not credible/credible Low Involvement .897
Not believable/believable
Attitude toward the brand (Chang and Thorson, Not likable/likable High Involvement .880
2004; Spears and Singh, 2004) Please describe your overall feelings about the brand Unpleasant/pleasant
featured in the ad you just watched
Bad/good Low Involvement .867
Not appealing/appealing
Purchase Intention (MacKenzie et al., 1986) How likely is it that you will try the advertised product if Not likely/likely High Involvement .859
it becomes available in your area? Improbable/probable
Impossible/possible Low Involvement .847
Unwilling/willing

Note: All measures are based on a 5-point semantic-differential scale: the higher the composite score is, the more positive is the assessment for each measure. Each
individual’s responses to the above scales were averaged to produce the single index of measured construct.

Table 3
Tests of proposed hypotheses: results of one-ways ANOVAs for different levels of product involvement.

Variable (Hypothesis) Involvement Repetition on a Repetition on Reversed repetition F (2, 279)

single device multiple devices on multiple devices
Message credibility (H1) High 2.63a (.74) 3.11b (.79) 2.96b (.72) 10.43***
Low 2.42a (.78) 3.04b (.87) 2.77b (.95) 12.28***
Advertiser credibility (H2) High 2.89a (.85) 3.25b (.70) 3.16b (.68) 5.82**
Low 2.78a (.85) 3.32c (.65) 3.05b (.79) 11.89***
Ad credibility (H3) High 2.60a (.66) 3.11b (.77) 2.97b (.75) 12.62***
Low 2.42a (.74) 2.95b (.85) 2.80b (.79) 11.35***
Brand credibility (H4) High 2.29a (.74) 2.85b (.79) 2.80b (.95) 13.43***
Low 2.31a (.82) 2.82b (.92) 2.76b (.81) 10.55***
Number of positive thoughts (H5) High .72a (1.08) 1.93b (1.53) 1.73b (1.58) 20.23***
Low .42a (.82) 1.32b (1.32) 1.16b (1.51) 14.39***
Attitude toward the brand (H6) High 2.30a (.84) 2.90b (.80) 2.70b (.86) 13.17***
Low 2.22a (.79) 2.60b (.91) 2.55b (.87) 5.73**
Purchase intention (H7) High 1.88a (.80) 2.39b (.77) 2.36b (.99) 10.56***
Low 2.00a (.86) 2.50b (.91) 2.48b (.91) 9.60***

Note: Repetition on a single device indicates repeated exposure of the same ad on TT (TV–TV), DD (DMB–DMB), and WW (Web–Web). Repetition on multiple devices indicates
exposure to the same ad in cross-media in sequence of TD, DW, and WT. Reversed Repetition on multiple devices is the replication of the ‘‘Repetition on multiple devices’’
condition, by changing the presentations order (e.g., DT, WD, and TW). For each condition, N = 93. Means with different superscript letters are significantly different according
to Tukey’s b post hoc test (p < .05). The numbers in parentheses are standard deviation.
⁄
p < 0.05.
**
p < 0.01.
***
p < 0.001.

credibility than a single-medium repetition, F (2/279) = 5.82, single-medium repetition regardless of product involvement for
p < .01. These results corroborate H2. the ad. Therefore, H3 was also supported.

4.2.3. Ad credibility 4.2.4. Brand credibility

When it comes to ad credibility, both multiple-media conditions To test H4, a one-way ANOVA was conducted for the brand
yielded significantly higher ad credibility than the single-medium credibility index. As a result, brand credibility was higher for mul-
repetition condition for both high- and low-involvement products. tiple-media repetitions than for a single-medium repetition for
For the ad exposure to high-involvement product, MMultiple-media repetition both high- and low-involvement products. The mean of multiple-
was 3.11 (SD = .77), MReversed multiple-media repetition was 2.97 (SD = .75), media repetitions and reverse multiple-media repetition was
and MSingle-medium repetition = 2.60 (SD = .66), F (2/279) = 12.62, 2.85 and 2.80 respectively for a high-involvement product, F (2/
p < .001. For the ad exposure to a low-involvement product, 279) = 13.43, p < .001. The mean of multiple-media repetition and
MMultiple-media repetition was 2.95 (SD = .85), MReversed multiple-media repetition reverse multiple media repetition was 2.82 and 2.76 respectively,
was 2.80 (SD = .80), and MSingle-medium repetition = 2.42 (SD = .74), F (2/ and the mean of single-media repetitions was 2.31 for a low-in-
279) = 11.35, p < .001. These results indicate that ad credibility volvement product, F (2/279) = 10.55, p < .001. Consequently, H4
is perceived higher for cross-media repetition than for a was also corroborated.
470 J.S. Lim et al. / Computers in Human Behavior 48 (2015) 463–472

4.2.5. Cognitive response superior to single-medium repetitions for the most important out-
As for the cognitive response measure by ad repetition, both comes measures of advertising effectiveness.
multiple-media conditions resulted in a significantly greater number
of positive thoughts than single-medium repetitions. For a high-in- 5.1. Theoretical implications
volvement product commercial, MMultiple-media repetition was 1.93
(SD = 1.53), MReversed multiple-media repetition was 1.73 (SD = 1.59) where- There is little research on the effects of mobile video advertis-
as MSingle-medium repetition was 0.72 (SD = 1.08), F (2/279) = 20.23, ing. This research suggests that the mobile ad environment oper-
p < .001. For a low-involvement product, MMultiple-media repetition was ates similarly to the much studied Internet environment.
1.32 (SD = 1.32), MReversed multiple-media repetition was 1.16 (SD = 1.51) The robust effects of synergistic ad repetition on general out-
whereas MSingle-medium repetition was 0.42 (SD = .82), F (2/279) = comes of perceived credibility, cognitive responses, and attitude
14.39, p < .001. A post hoc using Tukey’s b method also revealed shed light on the cross-media synergy effects for digital advertis-
that the mean differences between multiple-media repetition ing. In this study, the original television commercial was digitized
and single-media repetition were statistically significant under and presented in three different media of different screen sizes.
the p < .05 level. Therefore, H5 was also supported. Any combinations of cross-media presentation yielded greater
and more positive results for intended measures than simple
4.2.6. Attitude toward the brand repetitions on a single media. In proposing the hypotheses, we rea-
H6 was tested by performing a one-way ANOVA for the index of soned four theoretical underpinnings for the cross-media synergy
attitude toward the brand. As predicted, both multiple-media effect. The greater perceived credibility was postulated based on
conditions showed higher attitude toward brand scores than the multiple-source effect. The more positive cognitive responses
single-medium repetitions for both product commercials in differ- were predicted based on the differential attention hypothesis and
ent levels of product involvement. For high-involvement products, forward encoding hypothesis. Finally, attitudinal change was
participants showed more positive attitude toward the grounded on repetition-variation theory. Therefore, findings of
brand when exposed to the commercial in multiple media the current study can provide useful theoretical accounts for the
[MMultiple-media repetition = 2.90 (SD = .79), MReversed multiple-media repetition = cross-media synergy effect.
2.70 (SD = .86)] than in single medium [(MSingle-medium repetition = Findings of this study are the first empirical results that exhib-
2.30 (SD = .84)], F (2/279) = 13.17. For a low-involvement product, ited the synergy effect of the Internet–television–mobile advertis-
multiple-media repetition conditions also resulted in more posi- ing on inducing positive perceptions, cognition, and affects. One
tive attitude toward the brand than single-media repetition, notable thing is that the varied ad repetition in our study exhibited
F = (2/279) = 5.73. Therefore, H6 was also supported. positive effects on ad credibility, brand credibility, attitude toward
the brand and purchase intentions, which was not found in Chang
4.2.7. Purchase intention and Thorson’s (2004) study. We note that participants in this study
In H7, we also predicted that multiple-media repetitions of a were exposed to digitized commercials that were presented in
commercial would yield higher purchase intention than single- starkly different screen sizes (42 vs. 19 vs. 3.5 inches respectively).
medium repetitions. The result of the ANOVA test revealed that Reeves et al. (1999) revealed the display size effect on the audi-
two multiple-media conditions yielded higher purchase intention ence’s arousal. Although little evidence has been available regard-
than single-media repetition for both a high-involvement product ing the repetition effects through varied screen sizes, the varied
[F (2/279) = 10.56, p < .001] and a low-involvement product [F (2/ repetition in different screen sizes from this study could have gen-
279) = 9.60, p < .001]. Thus, H7 was supported. erated more striking variation. We presume that this helped to
In RQ1, we asked a research question (RQ1) regarding whether yield the effects that the forward encoding hypothesis and the
the predicted advantage of multiple-media repetitions over single- repetition variation theory postulate. In other words, it is probable
medium repetition could be different for a high- vs. a low-involve- that participants in our study may have formed more positive cred-
ment product. As reported for hypothesis testing, the advantage of ibility perceptions and attitudes due to much greater variations in
multiple-media repetitions over single-media repetitions was synergistic repetition than Chang and Thorson’s experimental con-
robust across different dependent measures. Therefore, we conclude dition (similar size of television and computer screen).
that the multiple-media repetitions are more effective than single- Because the higher level of encoding affected the participants’
medium repetitions in achieving intended goals of advertising. attitude toward the brand and purchase intention (Voorveld
et al., 2011), more positive attitude toward the brand and higher
5. Discussion purchase intention in multiple-source repetition than in a single
media repetition can be explained. Also, screen size can increase
This study expanded the knowledge on the synergy effect that attention and arousal, which provide variations (Reeves et al.,
came from the sequential presentation on multiple devices of differ- 1999). Repetition with such variations connects to a significant
ent screen sizes. Consistent with findings from the previous research, positive effect on the viewer’s cognitive response activity including
we evidenced that ad repetition on multiple devices had a greater ad credibility and brand credibility (McCullough & Ostrom, 1974).
advantage in achieving intended ad effects than the repetition on a Consequently, this research found intriguing results that were dif-
single device. In summary, the repetition of ads on multiple media ferent from Chang and Thorson’s research. An explanation for the
induced significantly higher credibility perception, cognitive different findings is that the ad repetitions on multiple-media of
responses, attitude toward the brand, and purchase intention than distinctively different screen sizes are likely to invoke higher ad/
repetitions on single devices of television, DMB, and the Internet. brand credibility, more positive attitude toward the brand, and
Extending the earlier study by Chang and Thorson (2004), this higher purchase intention than the repetitions on two screens
study also considered the product involvement as a potential mod- (i.e., television and computer) of the same size.
erator. Unlike Voorveld et al.’s (2012) study in which product
involvement was treated as a between-subject factor, this study 5.2. Managerial implications
treated it as a within-subject factor. It is noteworthy that the pro-
duct involvement condition did not make any impact on the main The current study provides managerial implications in regards
effect of cross-media advertising on the ad effectiveness. In to the synergistic repetition of digital advertising. First, synergistic
conclusion, it was confirmed that multiple-media conditions were repetition through variations can help increase more positive
 J.S. Lim et al. / Computers in Human Behavior 48 (2015) 463–472 471

advertising effects than a single-media based repetition. In doing tising is adapting to three-screen based advertising. Further
so, professionals need to consider the so-called three-screen (tele- research is needed to analyze multiple-media repetitions involving
vision–Internet–mobile) variation strategy (Truong, McColl, & more than two media.
Kitchen, 2010) to maximize the influence of distinct types of varia- Finally, the commercials presented to the participants were
tion (Haugtvedt, Schumann, Schneier, & Warren, 1994; Schumann identical in this study. However, future research may consider pre-
et al., 1990). senting different versions of the same message that are customized
Secondly, advertisers, in an interactive media age, need to adapt to each medium’s screen size. Since the goal of the current study
the non-interactive nature of television spots to address the syner- was to see the effect of the same exact commercial that was
gistic effects of digital media platforms. In this regard, Assael aptly originally created for television, we did not consider such cus-
states: ‘‘As applied to cross-media, an IMC approach would require tomization tailored to each medium of the Internet and DMB. As
an integrative media plan capable of measuring the interactive previously mentioned, the ad message is now tailored to each
effects of media components on exposure, attention, and behavior’’ device’s screen size so that voters who view a political attack ad
(Assael, 2011, p. 7). Political campaigns in the U.S. have taken on television and YouTube later view the attack message while
unique advantage of the cross-media opportunities by chasing surfing the Internet and through their smartphones when com-
the potential voters at home and on the go. As showcased in a muting by subway (Corasaniti & Parker, 2014).
recent New York Times report of the 2014 election (Corasaniti & We anticipate that the future research will not only replicate
Parker, 2014; Corrigan, Powell, & Michaels, 2013), implementing the findings of the current study but will also expand the knowl-
advertising on multiple devices including mobile devices is not edge of the synergy effect in the rapidly changing, newer, and
merely an option but an integral part of today’s media planning. mobile media platforms.
Thirdly, airing the ad on mobile TV may not make any difference
in the synergy effect of cross-media advertising. In other words,
References
the screen size may not yield a difference in the ad synergy effect
since the ad presentation on DMB in different sequential order did Abernethy, A., Cannon, H. M., & Leckenby, J. D. (2002). Beyond effective frequency:
not result in any statistical difference for the dependent variables Evaluating media schedules using frequency value planning. Journal of
of the current study. This does not mean that the screen size does Advertising Research, 42(6), 33–47.
Anand, P., & Sternthal, B. (1990). Ease of message processing as a moderator of
not entirely matter. When examined separately in a post hoc ana- repetition effects in advertising. Journal of Marketing Research, 27(3), 345–353.
lysis of the comparison of multiple repetitions from a single device, Assael, H. (2011). From silos to synergy: A fifty-year review of cross-media research
an ad delivered on a DMB device yielded the lowest score across shows synergy has yet to achieve its full potential. Journal of Advertising
Research, 51(1), 1–17.
different measures.
Bart, Y., Stephen, A. T., & Sarvary, M. (2014). Which products are best suited to
mobile advertising? A field study of mobile display advertising effects on
5.3. Limitations and suggestions for future research consumer attitudes and intentions. Journal of Marketing Research, 51(3),
270–285.
Batra, R., & Ray, M. L. (1986). Situational effects of advertising repetition: The
While this study found the effectiveness of cross-media adver- moderating influence of motivation, ability, and opportunity to respond. Journal
tising over repetition of advertising in the same media, some of Consumer Research, 12(4), 432–445.
limitations should be applied to make a generalization of the Belch, G. E. (1982). The effects of television commercial repetition on cognitive
response and message acceptance. Journal of Consumer Research, 9(1), 56–65.
repetition effect. Firstly, the condition in which participants were Berlyne, D. E. (1970). Novelty, complexity, and hedonic value. Perception and
exposed to advertising in repetition may be slightly different from Psychophysics, 8(5), 279–286.
the real-life situation since participants’ viewing behavior was con- Brennan, I., & Bahn, K. D. (2006). Literal versus extended symbolic messages and
advertising effectiveness: The moderating role of need for cognition. Psychology
trolled in a laboratory experimental setting. For instance, con- and Marketing, 23(4), 273–295.
sumers may watch the mobile advertising typically on the move, Cacioppo, J. T., & Petty, R. E. (1979). Effects of message repetition and position on
which may affect their attention and make their information pro- cognitive response, recall, and persuasion. Journal of Personality and Social
Psychology, 37(1), 97–109.
cessing less rich than when they watch it on a larger screen. Future Campbell, Margaret C., & Keller, K. L. (2003). Brand familiarity and advertising
researchers for cross-media advertising may consider the condi- repetition effects. Journal of Consumer Research, 30(2), 292–304.
tion that can simulate more realistic viewing behavior. Future Chang, Y., & Thorson, E. (2004). Television and Web advertising synergies. Journal of
Advertising, 33(2), 75–84.
research also needs to replicate the results obtained from the cur-
Cho, C.-H. (1999). How advertising works on the WWW: Modified elaboration
rent study, possibly adding more conditions for screen size. In this likelihood model. Journal of Current Issues and Research in Advertising, 21(1),
regard, Edell and Keller (1989) once pointed out that the screen 34–50.
size effect could be affected by the cognitive capacity to process Confer, M. G., & McGlathery, D. (1991). The research study: The advertising impact
of magazines in conjunction with television. Journal of Advertising Research,
ad messages. Researchers who aim to incorporate the cognitive 31(1). 64–64.
capacity condition into the synergy effect by different screen size Corasaniti, N., & Parker, A. (2014). G.O.P. ads chase voters at home and on the Go. The
can also consult with Janiszewski’s (1993) previous study. New York Times (pp. A1). .
Corrigan, P. W., Powell, K. J., & Michaels, P. J. (2013). The effects of news stories on
Secondly, the findings of the current study indicated that the the stigma of mental illness. Journal of Nervous and Mental Disease, 201(3),
adverse effect of additional ad exposures on attitude could be mod- 179–182.
erated by the variation in ad repetition through cross-media Cox, D. S., & Cox, A. D. (1988). What does familiarity breed? Complexity as a
moderator of repetition effects in advertisement evaluation. Journal of Consumer
sequential presentation. As the two-factor theory postulates, how- Research, 15(1), 111–116.
ever, the frequency of repetitions would eventually reduce the Dijkstra, M., Buijtels, H. E. J. J. M., & van Raaij, W. F. (2005). Separate and joint effects
positive affect of varied repetition as the number of repetition of medium type on consumer responses: A comparison of television, print, and
the Internet. Journal of Business Research, 58(3), 377–386.
increases (Rau, Zhou, Chen, & Lu, 2014; Rethans et al., 1986). Edell, J. A., & Keller, K. L. (1989). The information processing of coordinated media
Future research needs to examine how many varied repetitions campaigns. Journal of Marketing Research, 26(2), 149–163.
would start to generate consumer wearout with each additional Fishbein, M., & Ajzen, I. (1975). Beliefs, attitude, intention, and behavior: An
introduction to theory and research. Reading, MA: Addison-Wesley.
repetition diminishing positive affect. It may be of some benefit
Grass, R. C., & Wallace, W. H. (1969). Satiation effects of TV commercials. Journal of
to the industry if future research can formulate repetition wearout Advertising Research, 9(3), 3–8.
rates under differential synergistic combination. Harkins, S. G., & Petty, R. E. (1987). Information utility and the multiple source
Thirdly, the synergistic repetition in this study is based on only effect. Journal of Personality and Social Psychology, 52(2), 260–268.
Haugtvedt, C. P., Schumann, D. W., Schneier, W. L., & Warren, W. L. (1994).
two-paired media. As suggested in Truong et al.’s (2010) research, Advertising repetition and variation strategies: Implications for understanding
however, the future trend of cross-media synergy for digital adver- attitude strength. Journal of Consumer Research, 21(1), 176–189.
472 J.S. Lim et al. / Computers in Human Behavior 48 (2015) 463–472

Havlena, W., Cardarelli, R., & De Montigny, M. (2007). Quantifying the isolated and Petty, R. E., Cacioppo, J. T., & Schumann, D. (1983). Central and peripheral routes to
synergistic effects of exposure frequency for TV, print, and Internet advertising. advertising effectiveness: The moderating role of involvement. Journal of
Journal of Advertising Research, 47(3), 215–221. Consumer Research, 10(2), 135–146.
Hawkins, S. A., & Hoch, S. J. (1992). Low-involvement learning: Memory without Rau, P.-L. P., Zhou, J., Chen, D., & Lu, T.-P. (2014). The influence of repetition and time
evaluation. Journal of Consumer Research, 19(2), 212–225. pressure on effectiveness of mobile advertising messages. Telematics and
Janiszewski, C. (1993). Preattentive mere exposure effects. Journal of Consumer Informatics, 31(3), 463–476.
Research, 20(3), 376–392. Reeves, B., Lang, A., Kim, E. Y., & Tatar, D. (1999). The effects of screen size and
Jessen, I. B., & Graakjær, N. J. (2013). Cross-media communication in advertising: message content on attention and arousal. Media Psychology, 1(1), 49–67.
Exploring multimodal connections between television commercials and Rethans, A. J., Swasy, J. L., & Marks, L. J. (1986). Effects of television commercial
websites. Visual Communication, 12(4), 437–458. repetition, receiver knowledge, and commercial length: A test of the two-factor
Jin, H. S., Suh, J., & Donavan, D. T. (2008). Salient effects of publicity in advertised model. Journal of Marketing Research, 23(1), 50–61.
brand recall and recognition: The list-strength paradigm. Journal of Advertising, Schultz, D. E., Block, M. P., & Raman, K. (2011). Understanding consumer-created
37(1), 45–57. media synergy. Journal of Marketing Communications, 18(3), 173–187.
Jung, Y., Perez-Mira, B., & Wiley-Patton, S. (2009). Consumer adoption of mobile TV: Schumann, D. W., Petty, R. E., & Clemons, D. S. (1990). Predicting the effectiveness of
Examining psychological flow and media content. Computers in Human different strategies of advertising variation: A test of the repetition-variation
Behavior, 25(1), 123–129. hypotheses. Journal of Consumer Research, 17(2), 192–202.
Keller, K. L. (1987). Memory factors in advertising: The effect of advertising retrieval Sheehan, K. B., & Doherty, C. (2001). Re-weaving the web: Integrating print and
cues on brand evaluations. Journal of Consumer Research, 14(3), 316–333. online communications. Journal of Interactive Marketing, 15(2), 47–59.
Kim, M. J., & Jun, J. w. (2008). A case study of mobile advertising in South Korea: Shin, D. H. (2009). Understanding user acceptance of DMB in South Korea using the
Personalisation and digital multimedia broadcasting (DMB). Journal of Targeting, modified technology acceptance model. International Journal of Human–
Measurement and Analysis for Marketing, 16(2), 129–138. Computer Interaction, 25(3), 173–198.
Krugman, H. E. (1965). The impact of television advertising: Learning without Spears, N., & Singh, S. N. (2004). Measuring attitude toward the brand and purchase
involvement. Public Opinion Quarterly, 29(3), 349–394. Intentions. Journal of Current Issues and Research in Advertising, 26(2), 53–66.
Lin, C., Venkataraman, S., & Jap, S. D. (2013). Media multiplexing behavior: Stammerjohan, C., Wood, C. M., Chang, Y., & Thorson, E. (2005). An empirical
Implications for targeting and media planning. Marketing Science, 32(2), investigation of the interaction between publicity, advertising, and previous
310–324. brand attitudes and knowledge. Journal of Advertising, 34(4), 55–67.
Machleit, K. A., & Wilson, R. D. (1988). Emotional feelings and attitude toward the Stolyarova, E., & Rialp, J. (2014). Synergies among advertising channels: An
advertisement: The roles of brand familarity and repetition. Journal of efficiency analysis. Journal of Promotion Management, 20(2), 200–218.
Advertising, 17(3), 27–35. Tang, T., Newton, G. D., & Wang, X. (2007). Does synergy work? An examination of
MacKenzie, S. B., & Lutz, R. J. (1989). An empirical examination of the structural cross-promotion effects. International Journal on Media Management, 9(4),
antecedents of attitudes toward the ad in an advertising pretesting context. 127–134.
Journal of Marketing, 53(2), 48–65. Truong, Y., McColl, R., & Kitchen, P. (2010). Practitioners’ perceptions of advertising
MacKenzie, S. B., Lutz, R. J., & Belch, G. E. (1986). The role of attitude toward the ad strategies for digital media. International Journal of Advertising, 29(5), 709–725.
as a mediator of advertising effectiveness: A test of competing explanations. Unnava, H. R., & Burnkrant, R. E. (1991). Effects of repeating varied ad executions or
Journal of Marketing Research, 23(2), 130–143. brand name memory. Journal of Marketing Research, 28(4), 406–416.
Maclnnis, D. J., & Jaworski, B. J. (1989). Information processing from Varan, D., Murphy, J., Hofacker, C. F., Robinson, J. A., Potter, R. F., & Bellman, S.
advertisements: Toward an integrative framework. The Journal of Marketing, (2013). What works best when combining television sets, PCs, tablets, or mobile
53(4), 1–23. phones? How synergies across devices result from cross-device effects and
Malaviya, P. (2007). The moderating influence of advertising context on ad cross-format synergies. Journal of Advertising Research, 53(2), 212–220.
repetition effects: The role of amount and type of elaboration. Journal of Voorveld, H. A. M. (2011). Media multitasking and the effectiveness of combining
Consumer Research, 34(1), 32–40. online and radio advertising. Computers in Human Behavior, 27(6), 2200–2206.
McCullough, J. L., & Ostrom, T. M. (1974). Repetition of highly similar messages and Voorveld, H. A. M., Neijens, P. C., & Smit, E. G. (2011). Opening the black box:
attitude change. Journal of Applied Psychology, 59(3), 395–397. Understanding cross-media effects. Journal of Marketing Communications, 17(2),
Naik, P. A., Mantrala, M. K., & Sawyer, A. G. (1998). Planning media schedules 69–85.
in the presence of dynamic advertising quality. Marketing Science, 17(3), Voorveld, H. A. M., Neijens, P. C., & Smit, E. G. (2012). The interacting role of media
214–235. sequence and product involvement in cross-media campaigns. Journal of
Naik, P. A., & Peters, K. (2009). A hierarchical marketing communications model of Marketing Communications, 18(3), 203–216.
online and offline media synergies. Journal of Interactive Marketing, 23(4), Voorveld, H. A. M., Smit, E., & Neijens, P. (2013). Cross-media advertising: Brand
288–299. promotion in an age of media convergence media and convergence
Naik, P. A., & Raman, K. (2003). Understanding the impact of synergy in multimedia management. Springer, 117–133.
communications. Journal of Marketing Research, 40(4), 375–388. Voorveld, H. A. M., & Valkenburg, S. M. F. (2014). The fit factor: The role of fit
Nielsen (2014). Shifts in viewing: The cross-platform report Q2 2014. <http://www. between ads in understanding cross-media synergy. Journal of Advertising, 1–11.
nielsen.com/us/en/insights/reports/2014/shifts-in-viewing-the-cross-platform- Wang, A. (2011). Branding over internet and TV advertising. Journal of Promotion
report-q2-2014.html>. Management, 17(3), 275–290.
Petty, R. E., & Cacioppo, J. T. (1996). Attitudes and persuasion: Classic and Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and
contemporary approaches. Boulder, CO: Westview Press. Social Psychology, 9(2), 1–27.