Native versus banners -
Comparing exposure effects across popular online ad formats
by Andreas Storz
10967982
A thesis submitted in fulfillment of the requirements for the Master of Science degree of Research Master: Communication Science
Graduate School of Communication Supervisor: Dr. Theo Araujo
February 17th 2017
Native versus banners -
Comparing exposure effects across popular online ad formats
Abstract
‘Native advertising’ has been hailed as a promising solution to mitigate consumers’ ad avoidance and sustain online business models dependent on ad revenues. This study was conducted to assess the performance of this relatively new advertising format in
comparison to traditional banner ads across common ad effectiveness measures. A between-subjects experiment (N = 366) was designed to test for differences across explicit and implicit memory, brand attitude, as well as spontaneous brand awareness. In addition, the level of advertising clutter was varied in order to test for the negative impact of a web saturated with commercial messages on exposure effects of individual ads. Against the expectations suggested by existing theory, the results show that banner ads lead to a higher increase of spontaneous brand awareness than the presumably better liked native ads. Individuals who were exposed to a banner ad mentioned the target brand significantly more frequently (50%) for the relevant product category compared to
individuals who did not see any ads (37%), while native ads saw a smaller,
non-significant increase (42%). This finding suggests that while native advertising might be suitable to communicate more cognitively demanding messages to a particular audience, banner ads may be better able to convey simple reminders by priming the brand.
Spontaneous brand awareness represents a suitable measure to assess exposure effects across familiar brands. Moreover, while recognition was higher for native advertising, recall rates were higher for banners, suggesting that users do pay attention to native ads,
but a considerable share does not recognize it as advertising. A higher level of advertising clutter had a negative impact on recall but not on recognition of ads. Finally, exposure to advertising did not impact participants’ brand attitude, suggesting that a single exposure may often not be sufficient to impact people’s evaluation of a brand.
INTRODUCTION
Along with the rise of the Internet over the past quarter-century, advertising budgets have gradually shifted from traditional offline toward online media. Today, the web is a key arena for advertisers to reach consumers with their messages. However, there is also widespread skepticism as to whether online media are effective advertising channels. A key goal of marketers is to generate brand awareness. If advertising is to contribute to a brand’s success and help raise awareness, consumers need to pay attention to ads and form memory. However, various studies have shown that online audiences dislike, and consequently avoid, ads and develop ‘banner blindness’ (Benway 1998; Cho & Cheon, 2004; Resnick & Albert, 2014).
In reaction to these consumer habits, media professionals have developed new ad formats (Campbell, Cohen, & Ma, 2014) – including interstitials, floating ads, and skyscrapers (Burns & Lutz, 2006); rich media (Li & Leckenby, 2007); advergames (de Pelsmacker & Neijens, 2012); and mobile advertising (Rosenkrans & Myers, 2012) – that seek to gain consumers’ attention. In particular, ‘native advertising’ has been touted by advertisers and publishers as a promising solution (Levi, 2015; Carlson, 2015). Native advertising is “the practice by which a marketer borrows from the credibility of a content publisher by presenting paid content with a format and location that matches the
publisher’s original content” (Wojdynski & Golan, 2016, p. 1403). Therefore, rather than trying to attract users’ attention by standing out more, native ads attempt to blend in to regular site content, becoming “so cohesive with the page content, assimilated into the design, and consistent with the platform behavior that the viewer simply feels that they belong” (Interactive Advertising Bureau, 2013, p. 3).
Advocates claim that seamless integration of advertisers’ messages into a site’s main content allows advertisers to more effectively reach audiences, as native ads are less disruptive to the online experience and provide informational and entertainment value to users, while at the same time enabling the placement of embedded persuasive messages (Sullivan, 2015; Long, 2016). This seamless integration may also help a brand “break through” excessive amounts of online advertising. Given fierce competition for users’ attention, the online space is often described as cluttered and chaotic (Rotfeld, 2006), and consumers frequently perceive the overcrowding of media space with commercial
messages as disruptive and detrimental to the user experience, making them less receptive to traditional forms of advertising, particularly obtrusive banner ads (Elliot & Speck, 1998; Ha & McCann, 2008).
Yet the degree to which native advertising can actually deliver on each of these promises compared to conventional banner ads remains an open question. Indeed, we still know relatively little about the effectiveness of native advertising because, when it comes to measuring ad exposure effects, practitioners primarily focus on direct response metrics – mainly click-through rates and conversions. The former represents the percentage of users who click on a specific link; while the latter indicates the percentage of users who complete a desired action, such as purchasing an advertised product. However, research
on banner ads has demonstrated that these simple statistics do not capture the entire value of web advertising, as unclicked web ads also have brand-building potential (Briggs & Hollis, 1997; Chatterjee, 2008) due to their memory effects. Moreover, Chan Yun Yoo (2007; 2008; 2009) found that, even if web users cannot consciously remember being exposed to web advertising, i.e., they do not have explicit memory, they can still form implicit memory of online ads, or memory formed and retrieved unintentionally and without the conscious awareness of the person (Coates, Butler, & Berry, 2006). Implicit memory by itself can be sufficient to impact other brand metrics, including attitude toward the ad, click-through intention, and inclusion of a brand when a consumer later considers the product under which the brand falls.
Given the growth of native advertising, it is important to extend these studies’ examination of explicit and implicit memory effects to better understand the impact of exposure to this new advertising format. Both scholars and practitioners need a better grasp on how this type of advertising compares to traditional web banners across the more expansive, but still highly relevant, set of branding metrics. With this in mind, then, this study offers the following research question:
How do online native advertisements perform in comparison to traditional banner ads with respect to the formation of explicit and implicit memory in consumers and, in turn, to their awareness of and attitude toward a brand?
In order to answer this question, this thesis will first review relevant theoretical contributions, define core concepts, and derive testable hypotheses. Then, the
experimental setup and analysis is presented before turning to a discussion of the main results and their theoretical and practical implications.
THEORETICAL BACKGROUND
Memory and attention
Most advertising effectiveness models assume that memory is a necessary condition for ads to have any effect on consumers (Bettmann, 1979). Broadly defined, memory refers to the process by which the human mind encodes, stores, and retrieves information (Atkinson & Shiffrin, 1968). After all, people need to retain at least some amount of brand or product information for ads to have any impact on, say, their knowledge,
opinions, feelings, or purchase decisions. Moreover, given that memory and attention are interdependent (Chun & Turk-Browne, 2007), people will need to pay at least some level of attention to ads in order to form memory. Here, attention refers to the allocation of limited processing resources and the selection of object(s) on which to focus these resources (Chun & Turk-Browne, 2007; Lang, 2000). As a result, getting users’ attention is paramount for marketers (Teixeira, 2014).
However, numerous studies have shown that ad avoidance is widespread, meaning consumers tend to pay little attention to ads (Benway 1998; Cho & Cheon, 2004; Resnick & Albert, 2014). Typically, only a small share of the individuals who are exposed to any given ad actually interact with it and remember its content – including the brand (Shankar & Hollinger, 2007).
Users tend to avoid online ads for two main reasons. First, ads are perceived as a goal impediment (Cho & Cheon, 2004). In comparison to traditional media, the web is
considered a more goal and task-oriented medium (Chen & Wells, 1999; Eighmey, 1997). People primarily spend time online in order to satisfy a particular need, such as searching for information, communicating with others, or seeking entertainment (Yoo, 2007). Ads are frequently perceived as uninteresting and irrelevant to users’ tasks and goals (Pagendarm & Schaumburg, 2001), if not outright disruptive, especially if they are seen as intrusive (Edwards, Li, & Lee, 2002). Second, people avoid ads because of prior negative experiences (Cho & Cheon, 2004), where irritation over, for example, deceiving ad copy or spam leads to a learned response (Walsh, 2010).
There are also two main forms of online ad avoidance. First, users avoid ads cognitively, devoting selective attention to the web content deemed most relevant
(Fransen, Verlegh, Kirmani, & Smit 2015). This form of avoidance is said to dominate on the web and is considered an automatic response. Users learn to intuitively identify and screen out ads by adjusting their focus and gaze (Chatterjee, 2008; Kuisma, Simola, Uusitalo, & Öörni, 2010). Second, people show behavioral avoidance by, for example, actively avoiding ads or clicking away pop-ups, and by using tools such as ad blocking software (Fransen, Verlegh, Kirmani, & Smit 2015).
Explicit and implicit memory
While ad avoidance poses a serious challenge to advertisers, such avoidance is far from complete. Previous experimental research has demonstrated that “most participants fixate on ads at least once during their website visit” (Hervet, Guérard, Tremblay, & Chtourou, 2011, p. 708), making memory effects still plausible. As such, is important to distinguish two distinct forms of memory. Explicit memory, on the one hand, “refers to the conscious recollection of past experiences” while implicit memory “is defined as the
non-intentional, non-conscious retrieval of previously acquired information” (Coates, Butler, & Berry, 2004, p. 1195). Implicit memory effects are based on the priming mechanism, where a prior exposure event leads to enhanced perceptual fluency, which in turn results in greater salience and accessibility of the primed concept (Roediger & McDermott, 1993). This leads to better performance in an associated, yet seemingly unrelated task, without requiring conscious awareness by the individual. While explicit memory is retrieved deliberately, implicit memory retrieval is unintentional and automatic (Shapiro & Krishnan, 2001). For specific tasks, such as opinion-formation or decision-making, people can rely on explicit memory, implicit memory, or both (Shapiro & Krishnan, 2001; Lee, 2002).
Despite the relevance of implicit memory effects, explicit memory - usually measured by metrics such as recall and recognition of ad content and source (e.g., Heinz, Hug, Nugaeva, & Opwis, 2013) – has been dominant in advertising effects research (Shapiro & Krishnan, 2001). This results in an incomplete picture of potential memory effects (Duke & Carlson, 1993; 1994). In particular, previous studies have shown that as attention levels drop, explicit memory decreases more quickly than implicit memory (Richardson-Klavehn & Bjork, 1988; Schacter, 1987). For instance, under situations where individuals focus their attention on more than one object or task, experimental research on advertising effects has shown that implicit memory is often preserved, while explicit memory retrieval is adversely affected.
This may be because lower attention leads to lower levels of ad processing
(Shapiro & Krishnan, 2001). Cognitive psychologists assume that individuals can allocate more or less focus, time, and cognitive resources to different stimuli. The greater the
amount of these limited resources devoted to a specific stimulus, the higher, or ‘deeper’, the resulting level of processing will be (Craik & Lockhart, 1972). Explicit memory requires relatively high levels of attention and more active, elaborate processing. Implicit memory requires less attention and is formed automatically through so-called ‘low level processing’ (Heath, 2000).
The importance of implicit memory and its effects for advertising have been demonstrated across the spectrum of channels and tactics (cf. Vandeberg, Wennekers, Murre, & Smit, 2016; Krishnan & Chakravarti, 1999; Duke & Carlson 1993). For banner ads specifically, Chan Yun Yoo found that while explicit memory performance decreases under conditions of divided attention, implicit memory remains stable (Yoo, 2007). Moreover, field studies often report a low correlation between ad recall and sales, despite a strong relationship between advertising spending and sales, suggesting a significant influence of implicit memory on brands’ commercial success (Lee, 2002).
Indeed, several scholars argue that implicit memory is actually more relevant to brands for this and other reasons. For one, “implicit memory is closer to the behavioral predispositions of a consumer and is a form of memory used in everyday situations” (Krishnan & Chakravarti, 2001, p. 2). For another, respondents can more easily misrepresent their responses in test situations assessing explicit memory, where, for example, social desirability bias might impact responses in recognition tasks. Given the need for additional research into implicit memory effects, the following hypotheses are proposed:
H1a: Implicit memory of target words will be higher for individuals exposed to advertising containing these words compared to individuals who are not exposed. H1b: Implicit memory of distractor words will not differ across individuals.
Memory across ad formats
While specific knowledge about how individuals process different ad formats remains limited (Tutaj & van Reimersdal, 2012), previous studies have demonstrated that, in fact, there are relevant differences (Burns & Lutz, 2006). Moreover, there is some initial evidence that different formats are processed differently (Becker-Olsen, 2003). However, memory across different ad formats – native and banner, in particular – requires much greater exploration.
There are at least three reasons why one might expect native advertising to better mitigate ad avoidance and generate higher levels of memory than banner ads. First, several studies have shown that users generally prefer native advertising over conventional banner ads, as they find it “more informative, more amusing, and less irritating” (Tutaj & van Reijmerdsdal, 2012, p. 5). Because native advertising attempts to mimic regular site content, consumers likely perceive it as more relevant and less of an impediment to their goals. Second, given its relative novelty as an online advertising format, users will have had fewer prior negative experiences. Finally, research shows that people strongly prefer reading editorial over commercial content (Van Reijmersdal, Neijens, & Smit, 2005), and due to its resemblance to editorial content, it is more difficult for consumers to recognize native ads as advertising in the first place (Wojdynski, 2016).
If indeed consumers are less likely to avoid native advertising, it should receive relatively higher levels of attention and processing, which in turn should result in higher
explicit and implicit memory. These relationships are captured by the following hypotheses:
H2a: Recall will be higher for native advertising compared to banner ads. H2b: Recognition will be higher for native advertising compared to banner ads. H2c: Implicit memory performance will be higher for native advertising compared to
banner ads.
Clutter
However, the amount of advertising on a given website is likely to condition these effects (Speck & Elliot, 1997; Ha & McCann, 2008). Greater levels of advertising clutter – i.e., the number of ads a user is exposed to on a given webpage (Cho & Cheon, 2004) – are expected to have a negative impact on memory, as increased clutter means users have fewer cognitive resources available to process each individual ad (Teixeira, 2014). In addition, research suggests that the greater users perceive clutter to be on a given website, the less likely they are to attend to any of the ads (Cho & Cheon, 2004).
As banner ads are more easily recognized as advertising, and thus avoided, high clutter conditions should affect memory of banner ads more strongly compared to native advertising. That is, ad clutter should moderate the relationship between ad format and memory. These expected relationships yield the following hypotheses:
H3a: Recall will be higher in the condition of low clutter compared to high clutter. H3b: Recognition will be higher in the condition of low clutter compared to high clutter.
H3c: The clutter effect on recall (H3a) will be more pronounced for banner ads compared to native advertising.
H3d: The clutter effect on recognition (H3b) will be more pronounced for banner ads compared to native advertising.
H4a: Implicit memory performance will higher in the condition of low clutter compared to high clutter.
H4b: This clutter effect (H4a) will be more pronounced for banner ads compared to native advertising.
Brand attitude
As mentioned above, the literature on implicit memory effects suggests that implicit memory has a positive relationship with one’s attitude toward an advertised product or brand (Briggs & Hollis, 1997; Yoo, 2008), supporting the following hypotheses:
H5a: Brand attitude will be more favorable for individuals exposed to web advertising compared to individuals who are not exposed to web advertising.
H5b: Brand attitude will have a positive relationship with implicit memory performance.
The underlying mechanism for this effect is referred to as the ‘mere exposure effect’ (Zajonc, 1968; Kunst-Wilson & Zajonc, 1980). It draws on a core finding of cognitive psychology, which suggests that people show a preference for familiar items over unfamiliar ones and that familiarity is gained through simple prior exposure. This mechanism has been demonstrated in advertising (Fang, Singh, & Ahluwalia, 2007). For example, Matthes, Schemer, and Wirth (2007) find that consumers develop more
favorable attitudes towards brands after being exposed to film material containing brand placement.
Yet, there are different accounts with regard to the role of stimulus recognition. Most scholars claim that the effect hinges on implicit memory without conscious recognition – that is, without explicit memory (Bornstein & D’Agostino, 1992).
According to this account, individuals who do not remember having been exposed to the stimulus will perceive it as more familiar after a subsequent exposure on the basis of their implicit memory, which will increase their affect for it. However, this effect is
diminished if people consciously recognize the stimulus (Bornstein & Craver-Lemley, 2004). In the case of advertising, conscious recognition of an advertised brand can even lead to a backlash, as individuals recognize the covert persuasive intent and show resistance (Russell, 2002; Miller, 1976).
However, others claim that rather than inhibiting, conscious recognition of the stimulus can in fact enhance the mere exposure effect (Stafford & Grimes, 2012; Wang & Chang, 2004). Based on this account, recognition by itself reduces uncertainty and
increases familiarity, which leads to positive affect; in addition, these scholars argue that the recognition effect also positively interacts with implicit memory, thereby enhancing the mere exposure effect.
The following two competing hypotheses are proposed to capture the two opposing views on the mere exposure effect:
Competing hypotheses:
H6a: Brand attitude will increase with higher explicit memory performance. H6b: Brand attitude will decrease with higher explicit memory performance.
Spontaneous brand awareness
Finally, spontaneous or unaided brand awareness, i.e., the likelihood that respondents will mention a brand when given a product category cue, is an important metric and goal for marketers (Laurent, Kapferer, & Roussel, 1995; Keller, 2003). While it does not
guarantee commercial success, increased cognitive accessibility of a brand has been shown to increase its chances of inclusion in a consideration set (Nedungadi, 1990) and ultimately being chosen in a purchase decision (Shocker, Ben-Akiva, Boccara, & Nedungadi, 1991; Koniewski, 2012). Many experimental studies have used fictitious brands in order to control for potential confounding effects of prior knowledge across brands. While this certainly strengthens internal validity, it less suitable in order to understand the direction and magnitude of real-world effects of advertising. This study therefore heeds the call to extend the investigation of advertising effects to real, familiar brands (Yoo, 2008).
If the relationships hold as hypothesized, then individuals should form memory in reaction to exposure to advertising. Moreover, the level of memory formed is a function of the amount of attention paid and the level of ad processing. As ad avoidance is presumed to be lower for native advertising compared to banner ads, individuals are expected to form higher levels of memory when they are exposed to native ads compared to banner ads. Individuals who are not exposed to any advertising do not form any ad memory.
Advertising has been shown to increase brand awareness (Aaker, 1991, Huang & Sarigöllü, 2014). For the present case of a real, familiar target brand, exposure to
memory. Therefore, spontaneous brand awareness should be higher for individuals who are exposed to advertising by the target brand compared to people not exposed to
advertising. This effect should be strongest for individuals who are presented with native advertising, as it is expected to receive lower levels of ad avoidance compared to banner ads. These relationships are captured by the following hypothesis:
H7: Spontaneous brand awareness will be highest for native advertising, followed by banner advertising, and lowest for the group without advertising.
Sub research question – spontaneous brand awareness and implicit memory
Given that spontaneous brand awareness is tested without a direct reference to a previous ad exposure episode, it satisfies the conditions for an indirect memory test (Richardson-Klavehn & Bjork, 1988; Schacter, 1987). If ad exposure successfully primes the brand and enhances conceptual fluency of the stimulus, then it could be a suitable complement to assess implicit memory for real, familiar brands. However, whether this is the case has not been studied thoroughly by previous research. As a result, the following sub-research question is proposed to further investigate the relationship between spontaneous brand awareness and implicit memory performance:
RQ1: What is the relationship between spontaneous brand awareness and implicit memory performance?
METHODS
Participants
An online experiment was designed to test the proposed hypotheses. A total of 381 participants were recruited via the Amazon Mechanical Turk (MTurk) platform. MTurk is a crowdsourcing platform allowing requesters to find workers who perform so-called human intelligence tasks online. Requesters determine the compensation workers receive for completed tasks. Workers are free to choose which tasks they are willing to work on. As part of the MTurk policy, participants’ identities remain anonymous to the requester. MTurk is used frequently as a recruitment tool for data collection in the social and behavioral sciences (Paolacci & Chandler, 2014). Regarding data quality, prior research has found it to be at least on par with conventional methods (Buhrmester, Kwang, & Gosling, 2011). Moreover, MTurk samples have been shown to be more representative of the U.S. population than commonly used convenience samples (Berinsky, Huber, & Lenz, 2012).
Of the 381 individuals who completed the survey, 15 participants failed a simple attention check. These individuals were consequently removed from the sample, which ultimately comprised 366 participants. The age of the participants in the final sample ranged from 20 to 75 years (M = 37.96, SD = 10.83). With respect to gender, 169 (46%) identified as female while 196 (54%) reported being male; one individual identified as “non-binary”. The participants’ location was restricted to the United States in order to control for language skills as well as the presence and popularity of brands across different markets.
Pretest and target brand
The target brand was selected based on the results of a pretest (N = 43), in which participants were asked to name as many brands as they could think of for a given product category. As recruitment for the final experiment would not be restricted to a particular demographic or group with a shared interest, it was deemed important to avoid niche product categories, and to identify a segment with which most people would have a basic level of familiarity. This meant participants needed to have a somewhat equal chance of being familiar with relevant brands irrespective of gender, age, education, or socio-economic status. Three suitable product categories identified were “sneakers or running shoes”, “U.S. domestic airlines”, and “computer manufacturers”. Respondents were able to successfully recall relevant brands for all three categories. However, the range of products falling under the term ‘computer’, including desktops/laptops, tablets, wearable electronics, control systems, etc., was deemed too broad as a cue. Likewise, U.S. domestic airlines was considered somewhat ambiguous with respect to whether or not international carriers that offer many domestic connections or regional airlines would be included. Moreover, the use of computers and air travel, and consequently familiarity with the segment, might still vary too much across different demographics.
The product category of “sneakers or running shoes” was ultimately determined to be an appropriate target segment, as there are several competing brands with which most individuals have at least some level of familiarity. Moreover, sneakers are used both for sports activities and as casual footwear. As sneakers come in various price ranges, people across different social strata should be familiar with this product. Moreover, this category features a range of popular brands that compete for market share. Within the
sneakers category, it was important to select a target brand with a medium level of popularity, so that it would be unlikely to be the most frequently named brand, yet
common enough to be considered by a subset of the respondents. The final choice was for the sports manufacturer “Puma”, which was mentioned 11 times (28%) in the pretest and was the sixth most-frequently mentioned brand.
Stimulus materials and design
Five versions of sample web content were created to serve as stimuli for the different experimental conditions. Given the choice of brand and product category, an online health and fitness magazine was selected as an appropriate type of background website, on which the stimulus ads would appear. Five different versions of a sample web page were developed featuring short articles on healthy lifestyle and fitness. The “native” versions contained a native ad for a Puma fitness campaign, which was adapted based on an actual native ad by Puma. The native ad resembled the regular articles in appearance, but it contained the disclaimer “Sponsored Content” (Campbell, Cohen, & Ma, 2014) as well as a brand logo on top of the content. The “banner” versions, in turn, contained a vertical banner variant of the same Puma campaign. Both the native and the banner conditions had two variants with either a low or high level of ad clutter, respectively. The “high clutter” versions had two additional banner ads placed on the web page, while the “low clutter” versions only contained the target ads. The control version had no
advertising (cf. Appendix for examples of the stimuli). Therefore, there were five
experimental conditions in total (cf. Table 1 for an overview of the conditions and group sizes).
Table 1
Overview of experimental conditions and group sizes
Ad format High clutter Low clutter
Banner 61 59
Native 62 62
No ad (control) 122
Procedure
Participants were recruited to complete an academic survey on online media and user experiences in order to avoid focusing their attention on the advertisements. After
providing informed consent and receiving general instructions, they were first exposed to the sample web page based on the condition they were randomly assigned to. Exposure lasted 90 seconds in order to provide participants with an equal opportunity to process the stimuli. They were instructed to approach the page the same way as they do in their everyday use of the web. Moreover, participants were told to focus their attention on the editorial content based on which they would answer some follow-up questions. Following the exposure sequence, they answered a set of distractor questions about how they
approached the web page as well as their impressions of the content. Next, they
completed dependent measures, demographics, and other control measures, followed by manipulation and attention checks. Finally, participants were thanked and debriefed.
Measures
successfully complete it as “FINESSE”. Word fragments were carefully selected to represent a medium level of difficulty, so as to not generate too many or too few completions (Yoo, 2007; Duke & Carlson, 1994). Participants were given 8 target fragments, which occurred in the target ads embedded in the stimulus materials, as well as 8 distractor words. Their order was rearranged, and participants completed this task in two steps, for which they had 75 seconds each. They were instructed to complete the word stems with the first (matching) word that came to mind and to complete as many words as possible. Afterwards, successful completions for target words (M = 3.55, SD = 1.57) and distractor words (M = 2.78, SD = 1.26) were added up.
Explicit memory. Both recall and recognition of advertising material were measured to gauge respondents’ explicit memory. For the recall task, participants were asked to name any brands they could remember having seen in the advertisements during the exposure sequence. For the recognition task, participants had to indicate whether they remembered having seen a particular ad. They were shown 3 screenshots in random order, including the target ad along with 2 distractor ads. Answer options were “Yes”, “Maybe”, and “No”. Only a “Yes” answer for the target brand was counted as
recognition. Both measures are dichotomous and were coded either as 1 (did recall/recognize target ad) or 0 (did not recall/recognize target ad). Target ads were successfully recalled by 37 percent of the participants, while the overall recognition rate was 73 percent.
Brand attitude. Attitude towards the brand was measured with 5 items, using a 7-point semantic differential scale (Spears & Singh, 2004) ranging from 1-7. Items were anchored by “unappealing”/”appealing”, “bad”/”good”, “unpleasant”/”pleasant”,
“unfavorable”/”favorable”, and “unlikeable”/”likeable”. The mean score of the items was used as the overall index for brand attitude (α = .98, M = 5.3, SD = 1.20).
Spontaneous brand awareness (SBA). Participants were given 60 seconds to name up to 10 brands they could think of in a specific product category. They were instructed to enter the brand names in the order they came to mind. SBA for the target brand is a dichotomous measure indicating whether the respondent mentioned “Puma” as one of the brands in the sneakers category (1=yes, 0=no). Overall, 42.9 percent of the participants named “Puma” in their answers.
Controls. Control measures included basic demographics, namely participants’ age, gender, and level of education. Participants’ reported level of education ranged from “Did not complete high school” (n = 1) to “Doctoral degree” (n = 7). The most common category was “College degree” (n = 136) followed by “Some college or postsecondary training” (n = 100). In addition to that, participants’ level of involvement with the target segment was assessed via a single item, asking them to rate their level of interest in sportswear on a 7-point scale (M = 4.42, SD = 1.77), anchored by 1 (“not at all interested”) and 7 (“very interested”).
Manipulation check
The experimental groups were shown a screenshot of the low clutter and the high clutter version of their group’s ad format in randomized order. In both cases, they were asked to rate how they perceived the level of ad clutter on a 7-point scale ranging from 1 (“not cluttered at all”) to 7 (“extremely cluttered”). To see whether the manipulation of the ad clutter conditions was successful, a one-way ANOVA with repeated measures was conducted. The repeated measures were the conditions for ‘ad clutter’ on the dependent
variable measuring the perceived level of clutter on the web page. This test showed a significant result (F(1, 243) = 456.5, p < .001), indicating that participants indeed perceived the high ad clutter condition (M = 4.66, SD = 1.43) as more cluttered than the low clutter condition (M = 2.61, SD = 1.52).
Confound check
In order to verify whether the randomization procedure worked as intended, analyses of variance were conducted using the control variables as dependent measures and the experimental conditions as predictors. These tests returned non-significant results for the variables age (F(5, 360) = 0.54, p = .75), level of education (F(5, 360) = 0.64, p = .67), as well as participants’ stated level of involvement with the target product segment (F(5, 360) = 1.33, p = .25). For the variable gender, a chi-square test of independence against the experimental conditions was also non-significant, χ2
(5) = 6.24, p = .28. Together, these confound checks indicate that the randomization procedure was successful. Consequently, the control variables were excluded from subsequent analyses.
RESULTS
Ad format and memory
According to Hypothesis 1, implicit memory performance for target words was expected to be higher for individuals exposed to advertising compared to individuals who did not see any ads (H1a), while it was assumed that there would be no significant differences across these groups for the distractor words (H1b). These relationships were tested with two ANOVAs, using individuals’ implicit memory scores for target and distractor words
as the dependent variable and the three main experimental conditions as a between-subjects factor. The analysis for the target words yielded an overall non-significant result (F(2, 363) = .88, p = . 42). Moreover, planned contrasts comparing the groups with advertising against the control condition revealed no significant differences across conditions, t(363) = -0.788, p = .22 (one-tailed), thus H1a was not supported. The second analysis also returned a non-significant result, F(2, 363) = 0.38, p = .68, confirming H1b. Therefore, implicit memory performance did not differ across conditions for target as well as distractor words.
Hypothesis 2 stated that both explicit (H2a and H2b) and implicit (H2c) memory performance would be higher in the native condition compared to the banner ad
condition. In order to explore the relationships for explicit memory, two chi-square tests of independence were performed between recall and recognition, respectively, as well as the two ad formats (native and banner). Both chi-square tests for recall (χ2
(1) = 26.07, p < .001) and recognition (χ2
(1) = 12.05, p < .001) yielded significant results. However, only the difference for recognition was in the expected direction, as more participants in the native condition (83%) recognized the ad than in the banner condition (63%). In the case of recall, fewer participants in the native condition (21%) recalled the advertising
compared to participants exposed to the banner ad (53%). Therefore, H2a was not supported, while H2b was confirmed. To test the proposed relationship for implicit memory, an independent t-test was conducted comparing the mean scores for implicit memory across the native (M = 3.40, SD = 1.53) and banner (M = 3.62, SD = 1.66) groups. This test yielded a non-significant result, t(238.89) = 1.04, p = .30, therefore H2c was not supported.
Ad clutter and memory
Hypothesis 3 suggested that ad clutter would have a negative impact on explicit memory performance (H3a and H3b). Moreover, ad clutter was expected to moderate the
relationship between ad format and explicit memory performance (H3c and H3d). H3a and H3b were tested via two chi-square tests of independence testing both recall and recognition against the level of ad clutter. In the case of recall, a significant test result was obtained, χ2
(1) = 4.36, p < .05. Moreover, this relationship was in the expected direction as more participants in the low clutter condition (44%) were able to recall the advertising compared to those in the high clutter condition (30%). However, the chi-square test for recognition across clutter levels returned a non-significant result χ2
(1) = .41, p = .52, indicating that there was no statistical evidence for a difference in
recognition rates across the two levels of ad clutter. Thus, H3a was confirmed while H3b was not supported.
In order to test for a moderating influence of ad clutter on the relationship of advertising format on explicit memory performance, two logistic regressions were performed using recall and recognition as dependent variables and ad format, level of clutter, as well as their interaction as predictors. Neither in the test for recall (b = 0.90, p = .14) nor in the test for recognition (b = 0.95, p = .13) was the interaction between the level of ad clutter and the ad format statistically significant, suggesting that there is no evidence for a moderating effect of ad clutter on the impact of ad format on explicit memory performance. Therefore, H3c and H3d were not supported.
In analogous fashion, according to Hypothesis 4 a negative relationship between the level of ad clutter and implicit memory performance was expected (H4a). In addition
to that, the clutter level was supposed to have a moderating influence on the relationship between ad format and implicit memory (H4b). A one-way ANOVA was performed using participants’ implicit memory score as the dependent variable and ad clutter as the predictor. This test produced a non-significant result, F(1, 242) = 0.27, p = .61,
suggesting no evidence for an impact of ad clutter on implicit memory. Thus, H4a was not supported.
In order to test for a moderating influence of ad clutter, a two-way ANOVA was conducted with implicit memory as the dependent variable, while ad clutter and ad format were used as between-subjects factors. There was no significant interaction effect
between ad format and clutter level, F(1, 240) = 0.70, p = .41. Thus, H5b was not supported either.
Brand attitude
According to Hypothesis 5, target brand attitude is expected to be higher for groups exposed to advertising compared to the group that did not see any advertising (H5a). In order to test this relationship, a one-way ANOVA was performed with (target) brand attitude as the dependent variable and ad format as the predictor. This test showed a non-significant overall result, F(2, 363) = 0.82, p = .44. In addition to that, planned contrasts showed no evidence for a statistically significant difference in brand attitude between the groups exposed to advertising and the control condition, t(363) = -0.870, p = .39. Thus, H5a was not supported.
Furthermore, brand attitude was expected to have a positive relationship with implicit memory performance (H5b). To test this proposed relationship, a linear
the independent variable. This test showed a significant result for the effect of implicit memory (b = -0.09, p < .05). However, the magnitude of the estimated coefficient was fairly small, and in addition to that it was in the opposite direction. Therefore, there was no support for H5b.
Hypothesis 6 offered competing hypotheses, suggesting that brand attitude will either increase (H6a) or decrease (H6b) with explicit memory. In order to investigate these propositions, two one-way ANOVAs were conducted using brand attitude as the dependent variable as well as implicit memory along with recall and recognition, respectively, as between-subjects factors. Neither the test result for recall (F(1, 242) = 1.26, p = .26), nor for recognition (F(1, 242) = 0.05, p = .82) was statistically significant. Thus, there was no support for either hypothesis as neither recognition nor recall
appeared to impact brand attitude.
Spontaneous brand awareness
Hypothesis 7 suggested that spontaneous brand awareness should be highest for native advertising, followed by banner advertising, and lowest for the group without advertising. In order to test these propositions, a chi-square test of independence was first conducted, which yielded a non-significant result (χ2
(1) = 2.34, p = .13). Next, a logistic regression was performed, using spontaneous brand awareness as the outcome variable and the experimental conditions as the predictor. With the control condition serving as the baseline category, while the coefficient for native advertising was not statistically significant (b = 0.21, p = .42), the estimate for banner advertising was significant (b = 0.54, p < .05) and in the expected direction. Therefore, H7 was not confirmed.
However, there were considerable differences in spontaneous brand awareness across the experimental conditions. While 37 percent of the participants in the control condition mentioned the target brand, the rate was 42 percent for native advertising and 50 percent for the group exposed to a banner ad. In terms of the effect size, the odds ratio suggests that for an individual exposed to a banner ad, the odds of mentioning the target brand were 1.7 times larger than for someone who was not exposed to any advertising.
Finally, RQ1 was raised to explore a potential relationship between spontaneous brand awareness and implicit memory. A logistic regression was performed using brand awareness as the dependent variable and implicit memory as the predictor. The estimate for the coefficient of the main effect was marginally significant (b = 0.14, p = .09). Moreover, the predicted probability for spontaneous brand awareness (i.e., an individual mentioning the target brand) increases monotonically across implicit memory scores, from 0 (.34) to 8 (.61). Thus, there is some evidence for a positive relationship between implicit memory and spontaneous brand awareness.
CONCLUSION & DISCUSSION
Main findings and implications
The main aim of this study was to compare exposure effects across two different types of online advertising formats, namely native and banner ads. The experiment led to several important findings with relevant implications for communication theory. First, ad exposure had a considerable impact on spontaneous brand awareness. However, unlike hypothesized, the banner ad saw a higher lift in brand awareness than the native ad, despite the fact that the native ad likely received more attention as its recognition rate
was higher. This poses an interesting challenge to existing theory, based on which it was assumed that native ads, which are presumably better liked and less likely to be avoided than banner ads (Tutaj & van Reijmersdal, 2012), should lead to higher exposure effects.
Conclusions regarding the underlying reasons why banner ads seemed to be more effective with regard to brand awareness have to remain somewhat speculative and preliminary. However, one potential explanation might be that, as suggested by previous studies, different ad formats are indeed processed differently (Burns & Lutz, 2006; Becker-Olsen, 2003). Specifically, it appears that banners prime the brand while native ads prime the content of the sponsored article. While the brand successfully blends into the post, which resembles regular editorial content in form and function, it also seems to lose in salience as attention is directed more toward the content. This is despite the fact that the native ad contained a small brand logo above the post, identifying the sponsor. However, prior research has in fact shown that consumers may misperceive such logos as stand-alone ads (Wojdynski, 2016).
Moreover, regarding ad processing, it is important to point out that respondents had to make a cognitive leap on their own from brand to product. While they could draw on their memory formed in reaction to the ad exposure, the ads featured the target brand in the context of a sponsored fitness campaign without any cue to the target product category. Therefore, using new memory to help spontaneously recalling the target brand for the ‘sneakers’ category required a cognitive effort, which apparently was more easily accomplished by the group exposed to banner advertising. This further underscores the claim that different ad formats are processed differently (Li & Leckenby, 2007; Kuisma, Simola, Uusitalo, & Öörni, 2010; Rodgers & Thorson, 2000).
An additional practical implication is that spontaneous brand awareness is a valuable implicit measure for ad effects of familiar brands, which should be incorporated in future research if possible to complement direct measures such as recall and
recognition (Vandeberg, Smit, & Murre, 2015). Moreover, it can be a more empirically meaningful measure as it does not direct individuals to prior ad exposure episodes, which should better approximate consumers’ real-world behavior (Krishnan & Chakravarti, 2001).
Second, the results contribute to our understanding of native advertising in particular. While recognition rates were higher for the native ad, recall rates were higher for the web banner. This validates the finding by previous studies that many users do not think of native ads as advertising (Wojdynski & Evans, 2016; Hoofnagle & Meleshinsky, 2015). Therefore, one reason for lower levels of ad avoidance may indeed be that it is more difficult for users to recognize native advertising’s persuasive intent (Tutaj & van Reijmersdal, 2012), which would substantiate the claim by critics that native ads
constitute a form of deceptive marketing (Wojdynski, 2016; Levi, 2015; Gottfried, 2014). However, the higher recognition rate also provides some evidence that native advertising may indeed be able to better overcome consumers’ ad avoidance. Furthermore,
concerning effectiveness measures, an additional implication is the fact that recall is not necessarily a suitable metric to capture exposure effects of native advertising if many users do not recognize it as advertising. Therefore, additional measures are needed to accurately assess its full effect, including spontaneous brand awareness.
Third, ad exposure did not have any discernible impact on users’ brand attitude. This suggests that a single exposure is unlikely to change consumers’ affect toward a
familiar brand, which is in line with previous research (Schmidt & Eisend, 2015). Moreover, this finding implies that consumers’ evaluation of a brand is a fairly stable quantity in the case of established brands, which is unlikely to change much unless new information or a shift in evaluative standards causes consumers to revisit their
assessment.
Fourth, there is evidence that ad clutter indeed has a negative impact on memory, as suggested by previous research (Ha & McCann, 2008). While recognition was
unaffected, recall rates for the target ad were lower in the high clutter condition, irrespective of the ad format. Moreover, as described in the manipulation check, participants perceived the webpage with more ads as more cluttered. This finding confirms the claim by previous studies that ad clutter has a negative impact on the
effectiveness of individual ads (Rotfeld, 2006) and is a driver of consumers’ ad avoidance (Cho & Cheon, 2004).
Managerial implications
From the perspective of practitioners, the most central finding of this study is that even a single exposure to advertising can raise a brand’s level of awareness in the minds of consumers, at least in the short term. While there was no statistically significant increase for native advertising, the 13-point lift for the banner ad compared to the group with no ads is both statistically significant and empirically meaningful. Although brand
awareness by itself is no guarantee for commercial success, previous research has
demonstrated that higher levels substantially increase the odds that a consumer chooses a particular brand in a purchase decision (Hoyer & Brown, 1990; Macdonald & Sharp, 2000).
Regarding the functions and uses of the different ad formats, the results suggest that banner ads may be most suitable to convey simple messages for known products or brands, serving as succinct reminders to build or sustain awareness. Given native advertising’s greater popularity and the higher attention level it tends to garner, this format may be more appropriate for situations where advertisers need to convey more complex messages. This is in line with previous research by Martin Eisend and Franziska Küster (2015), who found in a meta analysis that publicity – securing space in regular media coverage – tends to be more effective to promote products for which consumers lack prior knowledge, while advertising is better suited to communicate about established products. They also found that advertorials – the offline equivalent to native advertising – are an appropriate alternative to publicity, as they give advertisers control over the
message while also retaining the relatively higher levels of credibility of regular publicity and evoking less skepticism than conventional advertising.
Furthermore, the results for ad clutter should be a warning to both advertisers and publishers. It seems that, as ad clutter increases, so does ad avoidance. Lower attention for ads decreases not only their effectiveness, but eventually also lowers the value of ad space. Therefore, overcrowding webpages with advertising poses a risk to online ad revenues (Rotfeld, 2006).
Limitations and future research
While this study produced several relevant results, it is important to acknowledge a number of important limitations. First, concerning participants’ implicit memory performance, none of the conditions showed statistically significant differences. While the average for the target words was somewhat higher than for the distractor words, this
general pattern was stable across all groups, with no significant differences across conditions. As a result, it is questionable whether the test procedure used, i.e., the word completion task, was able to accurately measure priming effects due to a prior exposure to advertising. Given that all participants were exposed to the stimulus materials for 90 seconds, it is rather unlikely that no implicit memory was generated at all. Failure to detect meaningful differences could either mean that (1) the implementation of the test procedure failed, or (2) the procedure itself was not a valid instrument for the purposes of this experiment.
It is possible that a different set of target word-stems might have provided a higher sensitivity to tap into priming effects. For example, word-stems with multiple completion options and a medium level of difficulty (such as “B _ O _”, which could be completed as “BOOK”, but also as “BLOG”) might have increased the likelihood that participants would choose the option they were exposed to, given that it would be more accessible in working memory. Alternatively, it is possible that the delay between exposure and memory test was too long, so that priming effects had already worn off. However, previous research has shown that priming effects can be stable and last for at least 7 days (Tulving, Schacter, & Stark, 1982). Either way, it is not certain that a short exposure episode allowed for enough low-level attention for the target words to generate a sufficiently large priming effect. This is possible in particular given that all target stimuli were words embedded in a larger stimulus page which itself primarily consisted of textual content. In either case, it seems likely that most implicit memory was simply not picked up by the measure.
Second, the experiment contained only a single exposure to advertising stimuli, therefore effects requiring repetition – such as possibly a change in brand attitude – could not be assessed here. More importantly, the experiment used only a single target brand and product category. Whether the conclusions are stable and extend to other brand and segments remains an open question. Finally, given that it was an experimental study, the findings’ level of external validity remains unclear as the test setting was necessarily somewhat artificial. Therefore, it should be assessed whether the main insights can be replicated in a more natural setting that involves individuals’ in their everyday use of the web.
There are several other potential paths for future studies. First further research is needed to verify whether the effect of ad exposure on spontaneous brand awareness is robust. This would require the use of multiple target brands and product categories. Furthermore, the present study showed only a short term effect, but it would be relevant to know how long the effect on brand awareness lasts based on a single or multiple ad impressions. Second, while there is some evidence that native advertising is avoided less than banner ads, we currently do not know who pays attention to native ads and how much, i.e., what individual-level factors drive people’s interest. Future studies should incorporate factors such as demographics, media behavior, and interest in brands and products to better understand individual-level drivers of attention. Finally, eye-tracking technology could be used to investigate whether people indeed treat native ads differently than banner ads. This would help corroborate the preliminary findings of this study.
This study has only been able to scratch at the surface of these important issues. Nonetheless, it opens the way for further investigations in a number of ways. For the time
being, the most important insight generated is that there is empirical evidence that a single ad exposure can impact web users level of spontaneous brand awareness, which should be a relevant finding to both academics and practitioners.
REFERENCES
Aaker, D. A. (1991). Managing brand equity: Capitalizing on the value of a brand name. New York, N.Y.: The Free Press.
Atkinson, R. C., & Shiffrin, R. M. (1968). Human memory: A proposed system and its control processes. Psychology of learning and motivation, 2, 89-195.
Becker-Olsen, K. L. (2003). And now, a word from our sponsor—a look at the effects of sponsored content and banner advertising. Journal of Advertising, 32(2), 17-32. Benway, J. P. (1998, October). Banner blindness: The irony of attention grabbing on the
World Wide Web. In Proceedings of the Human Factors and Ergonomics Society Annual Meeting (Vol. 42, No. 5, pp. 463-467). Sage CA: Los Angeles, CA: SAGE Publications.
Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012). Evaluating online labor markets for experimental research: Amazon. com's Mechanical Turk. Political
Analysis, 20(3), 351-368.
Bettman, J. R. (1979). Memory factors in consumer choice: A review. The Journal of Marketing, 43(Spring), 37-53.
Bornstein, R. F., & Craver-Lemley, C. (2004). Mere exposure effect. In R. F. Pohl (Ed.), Cognitive illusions: A handbook on fallacies and biases in thinking, judgment and memory (pp. 215–233), Hove, UK: Psychology Press.
Bornstein, R. F., & D'Agostino, P. R. (1992). Stimulus recognition and the mere exposure effect. Journal of personality and social psychology, 63(4), 545.
Briggs, R., & Hollis, N. (1997). Advertising on the Web: Is there response before click-through?. Journal of Advertising research, 37(2), 33-46.
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's Mechanical Turk a new source of inexpensive, yet high-quality, data?. Perspectives on psychological science, 6(1), 3-5.
Burns, K. S., & Lutz, R. J. (2006). The function of format: Consumer responses to six on-line advertising formats. Journal of Advertising, 35(1), 53-63.
Campbell, C., Cohen, J., & Ma, J. (2014). Advertisements just aren't advertisements anymore: a new typology for evolving forms of online" advertising". Journal of Advertising Research, 54(1), 7-10.
Carlson, M. (2015). When news sites go native: Redefining the advertising–editorial divide in response to native advertising. Journalism, 16(7), 849-865.
Chatterjee, P. (2008). Are unclicked ads wasted? Enduring effects of banner and pop-up ad exposures on brand memory and attitudes. Journal of electronic commerce Research, 9(1), 51-61.
Chen, Q., & Wells, W. D. (1999). Attitude toward the site. Journal of advertising research, 39(5), 27-38.
Cho, C. H., & Cheon, H. J. (2004). Why do people avoid advertising on the internet?. Journal of advertising, 33(4), 89-97.
Chun, M. M., & Turk-Browne, N. B. (2007). Interactions between attention and memory. Current opinion in neurobiology, 17(2), 177-184.
Coates, S. L., Butler, L. T., & Berry, D. C. (2004). Implicit memory: A prime example for brand consideration and choice. Applied Cognitive Psychology, 18(9), 1195-1211.
Coates, S. L., Butler, L. T., & Berry, D. C. (2006). Implicit memory and consumer choice: The mediating role of brand familiarity. Applied Cognitive
Psychology, 20(8), 1101-1116.
Craik, F. I., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of verbal learning and verbal behavior, 11(6), 671-684. de Pelsmacker, P., & Neijens, P. C. (2012). New advertising formats: How persuasion
knowledge affects consumer responses. Journal of Marketing Communications, 18(1), 1-4.
Duke, C. R., & Carlson, L. (1993). A conceptual approach to alternative memory measures for advertising effectiveness. Journal of Current Issues & Research in Advertising, 15(2), 1-14.
Duke, C. R., & Carlson, L. (1994). Applying implicit memory measures: Word fragment completion in advertising tests. Journal of Current Issues & Research in
Advertising, 16(2), 29-39.
Edwards, S. M., Li, H., & Lee, J. H. (2002). Forced exposure and psychological reactance: Antecedents and consequences of the perceived intrusiveness of pop-up ads. Journal of Advertising, 31(3), 83-95.
Eighmey, J. (1997). Profiling user responses to commercial web sites. Journal of advertising research, 37(3), 59-67.
Eisend, M., & Küster, F. (2011). The effectiveness of publicity versus advertising: A meta-analytic investigation of its moderators. Journal of the Academy of Marketing Science, 39(6), 906-921.
Elliott, M. T., & Speck, P. S. (1998). Consumer perceptions of advertising clutter and its impact across various media. Journal of advertising research, 38(1), 29-30. Fang, X., Singh, S., & Ahluwalia, R. (2007). An examination of different explanations
for the mere exposure effect. Journal of consumer research, 34(1), 97-103. Fransen, M. L., Verlegh, P. W., Kirmani, A., & Smit, E. G. (2015). A typology of
consumer strategies for resisting advertising, and a review of mechanisms for countering them. International Journal of Advertising, 34(1), 6-16.
Gottfried, R. A. (2015). Six Ways This Article is Most Definitely Not an Ad: Deceptive Marketing and the Need for Clearly-Defined Disclosure Rules in Online Native Advertising. Loyola Consumer Law Review, 27(3), 399.
Ha, L., & McCann, K. (2008). An integrated model of advertising clutter in offline and online media. International Journal of Advertising, 27(4), 569-592.
Heath, R. (2000). Low involvement processing—a new model of brands and advertising. International Journal of Advertising, 19(3), 287-298.
Heinz, S., Hug, M., Nugaeva, C., & Opwis, K. (2013, April). Online ad banners: the effects of goal orientation and content congruence on memory. In CHI'13 Extended Abstracts on Human Factors in Computing Systems (pp. 1875-1880). ACM.
Hervet, G., Guérard, K., Tremblay, S., & Chtourou, M. S. (2011). Is banner blindness genuine? Eye tracking internet text advertising. Applied cognitive
psychology, 25(5), 708-716.
Hoofnagle, C. J., & Meleshinsky, E. (2015). Native advertising and endorsement:
Schema, source-based misleadingness, and omission of material facts. Technology Science. Retrieved from http://techscience.org/a/2015121503/
Hoyer, W. D., & Brown, S. P. (1990). Effects of brand awareness on choice for a
common, repeat-purchase product. Journal of consumer research, 17(2), 141-148. Huang, R., & Sarigöllü, E. (2014). How brand awareness relates to market outcome,
brand equity, and the marketing mix. In Fashion Branding and Consumer Behaviors (pp. 113-132). Springer New York.
Interactive Advertising Bureau (2013), The Native Advertising Playbook: Six Native Ad Categories, Six Marketplace Consideration, and IAB Recommended Disclosure Principles. Retrieved from Interactive Advertising Bureau:
http://www.iab.com/wp-content/uploads/2015/06/IAB-Native-Advertising-Playbook2.pdf.
Keller, K. L. (2003). Brand synthesis: The multidimensionality of brand knowledge. Journal of consumer research, 29(4), 595-600.
Koniewski, M. (2012). Brand awareness and customer loyalty. Retrieved from PMR Research: http://www.research-pmr.com/userfiles/file/wp/Brand-Awareness-and-Customer-Loyalty.pdf.
Krishnan, H. S., & Chakravarti, D. (1999). Memory measures for pretesting advertisements: An integrative conceptual framework and a diagnostic template. Journal of Consumer Psychology, 8(1), 1-37.
Kuisma, J., Simola, J., Uusitalo, L., & Öörni, A. (2010). The effects of animation and format on the perception and memory of online advertising. Journal of Interactive Marketing, 24(4), 269-282.
Kunst-Wilson, W.R., & Zajonc, R.B. (1980). Affective discrimination of stimuli that cannot be recognized. Science, 207(4430), 557–558.
Lang, A. (2000). The limited capacity model of mediated message processing. Journal of communication, 50(1), 46-70.
Laurent, G., Kapferer, J. N., & Roussel, F. (1995). The underlying structure of brand awareness scores. Marketing Science, 14(3_supplement), G170-G179.
Lee, A. Y. (2002). Effects of implicit memory on memory-based versus stimulus-based brand choice. Journal of Marketing Research, 39(4), 440-454.
Levi, L. (2015). A faustian pact: native advertising and the future of the press. Ariz. L. Rev., 57, 647.
Li, H., & Leckenby. J. D. (2007). Examining the effectiveness of Internet advertising formats. In D.W. Schumann & E. Thorson (Eds.), Internet advertising: Theory and research (pp. 203–24), New Jersey: Lawrence Erlbaum Associates, Inc. Long, J. (2016, November 4). How Native Ads Can Benefit Both Advertisers And
Publishers. Forbes. Retrieved from
http://www.forbes.com/sites/theyec/2016/11/04/how-native-ads-can-benefit-both-advertisers-and-publishers/#2caef7f84c85.
Macdonald, E. K., & Sharp, B. M. (2000). Brand awareness effects on consumer decision making for a common, repeat purchase product:: A replication. Journal of
Business Research, 48(1), 5-15.
Matthes, J., Schemer, C., & Wirth, W. (2007). More than meets the eye: Investigating the hidden impact of brand placements in television magazines. International Journal of Advertising, 26(4), 477-503.
Miller, R. L. (1976). Mere exposure, psychological reactance and attitude change. Public Opinion Quarterly, 40(2), 229-233.
Nedungadi, P. (1990). Recall and consumer consideration sets: Influencing choice without altering brand evaluations. Journal of Consumer Research, 17(3), 263-276.
Pagendarm, M., & Schaumburg, H. (2001). Why are users banner-blind? The impact of navigation style on the perception of web banners. Journal of Digital
Information, 2(1). Retrieved from http://journals.tdl.org/article/view/jodi-37/38. Paolacci, G., & Chandler, J. (2014). Inside the Turk: Understanding Mechanical Turk as
a participant pool. Current Directions in Psychological Science, 23(3), 184-188. Resnick, M., & Albert, W. (2014). The impact of advertising location and user task on
the emergence of banner ad blindness: An eye-tracking study. International Journal of Human-Computer Interaction, 30(3), 206-219.
Richardson-Klavehn, A., & Bjork, R. A. (1988). Measures of memory. Annual review of psychology, 39(1), 475-543.
Rodgers, S., & Thorson, E. (2000). The interactive advertising model: How users perceive and process online ads. Journal of interactive advertising, 1(1), 41-60.
Roediger, H. L., & McDermott K. B. (1993). Implicit memory in normal human subjects. In F. Boller & J. Grafman (Eds.), Handbook of Neuropsychology (Vol. 8, pp. 63– 131), Amsterdam: Elsevier.
Rosenkrans, G., & Myers, K. (2012). Mobile advertising effectiveness. International Journal of Mobile Marketing, 7(3), 5-21.
Rotfeld, H. J. (2006). Understanding advertising clutter and the real solution to declining audience attention to mass media commercial messages. Journal of Consumer Marketing, 23(4), 180-181.
Russell, C. A. (2002). Investigating the effectiveness of product placements in television shows: The role of modality and plot connection congruence on brand memory and attitude. Journal of consumer research, 29(3), 306-318.
Schacter, D. L. (1987). Implicit memory: History and current status. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13(3), 501-508. Schacter, D. L. (2008). Searching for memory: The brain, the mind, and the past. Basic
Books.
Schmidt, S., & Eisend, M. (2015). Advertising repetition: a Meta-analysis on effective frequency in advertising. Journal of Advertising, 44(4), 415-428.
Shankar, V., & Hollinger, M. (2007). Online and mobile advertising: current scenario, emerging trends, and future directions. Marketing Science Institute, 31(3), 207-206.
Shapiro, S., & Krishnan, H. S. (2001). Memory-based measures for assessing advertising effects: A comparison of explicit and implicit memory effects. Journal of
Shocker, A. D., Ben-Akiva, M., Boccara, B., & Nedungadi, P. (1991). Consideration set influences on consumer decision-making and choice: Issues, models, and
suggestions. Marketing Letters, 2(3), 181-197.
Spears, N., & Singh, S. N. (2004). Measuring attitude toward the brand and purchase intentions. Journal of Current Issues & Research in Advertising, 26(2), 53-66. Speck, P. S., & Elliott, M. T. (1997). The antecedents and consequences of perceived
advertising clutter. Journal of Current Issues & Research in Advertising, 19(2), 39-54.
Stafford, T., & Grimes, A. (2012). Memory enhances the mere exposure effect. Psychology & Marketing, 29(12), 995-1003.
Sullivan, M. (2015, November 12). As Print Fades, Part 4: Native Advertising on The Rise. The New York Times. Retrieved from
https://publiceditor.blogs.nytimes.com/2015/11/12/as-print-fades-part-4-native-advertising-on-the-rise/?_r=5.
Teixeira, T. S. (2014). The rising cost of consumer attention: why you should care, and what you can do about it. (Harvard Business School Working Paper, No. 14-055). Retrieved from Harvard Business School website:
http://www.hbs.edu/faculty/Pages/item.aspx?num=46132.
Tulving, E., Schacter, D. L., & Stark, H. A. (1982). Priming effects in word-fragment completion are independent of recognition memory. Journal of experimental psychology: learning, memory, and cognition, 8(4), 336-342.
Tutaj, K., & van Reijmersdal, E. A. (2012). Effects of online advertising format and persuasion knowledge on audience reactions. Journal of Marketing
Communications, 18(1), 5-18.
Vakratsas, D., & Ambler, T. (1999). How advertising works: what do we really know?. The Journal of Marketing, 63(1), 26-43.
Van Reijmersdal, E., Neijens, P., & Smit, E. (2005). Readers' reactions to mixtures of advertising and editorial content in magazines. Journal of Current Issues & Research in Advertising, 27(2), 39-53.
Vandeberg, L., Smit, E. G., & Murre, J. M. (2015). Implicit measurement. Psychological Review, 84, 231-259.
Vandeberg, L., Wennekers, A. M., Murre, J. M., & Smit, E. G. (2016). Implicit and Explicit Measures: What Their Dissociations Reveal About the Workings of Advertising. In Advances in Advertising Research (Vol. VI, pp. 269-279). Springer Fachmedien Wiesbaden.
Walsh, M. F. (2010). New insights into what drives Internet advertising avoidance behaviour: The role of locus of control. International Journal of Internet Marketing and Advertising, 6(2), 127-141.
Wang, M. Y., & Chang, H. C. (2004). The mere exposure effect and recognition memory. Cognition & Emotion, 18(8), 1055-1078.
Wojdynski, B. W. (2016). The deceptiveness of sponsored news articles: How readers recognize and perceive native advertising. American Behavioral Scientist, 60(12), 1475-1491.
