A study of media multitasking effects on implicit and explicit memory of advertisements

0

A Study of Media Multitasking Effects on

Implicit and Explicit Memory of Advertisements

Rantmila A. Tanasi

10494634

Master’s Thesis

University of Amsterdam

Graduate School of Communication

Persuasive Communication 2013-2014

UvA

Supervisor

Hilde Voorveld

1

Table of Contents

Abstract ... 2

Introduction ... 2

Theoretical Framework ... 5

Multitasking Effects on Memory ... 5

Theoretical Explanation of Multitasking Effects ... 6

Limited Capacity Model of Information Processing ... 6

Elaboration Likelihood Model ... 6

Hypotheses about Multitasking Effect on Memory of Ads ... 6

Explicit and Implicit Memory ... 7

Theoretical Explanation of the Disassociation of Explicit and Implicit Memory ... 8

Dual-processing Theories ... 8

Cognitive processing approaches ... 9

Hypotheses about Multitasking Effects on Implicit Memory of Ads ... 9

Method ... 11

Design and Sample ... 11

Pre-test for Material Selection ... 12

Experimental Material ... 13

Procedure ... 13

Measures ... 14

Results ... 18

Multitasking Effects on Explicit Memory of Ads ... 18

Multitasking Effects on Implicit Memory of Ads ... 19

Conclusion ... 20

References ... 24

2 Abstract

The current study examined the effects of multitasking with television and tablet on explicit and implicit memory of ads, as well as the role of the focus of attention. An experiment with a mixed-level design was conducted. Explicit and implicit memory of television ads was reduced by multitasking, whereas the memory of banner ads remained unaffected. The results are discussed minutely.

Introduction

In recent years, technological evolution has introduced a variety of new media that combine increased speed and decreased cost of access. These changes in the media field favored phenomena such as media multitasking (Minear, Brasher, McCurdy, Lewis, & Younggren, 2013). Media multitasking is defined as the exposure to multiple media simultaneously (Bardhi, Rohm, & Sultan, 2010), a trend that has been found to be relevant to all ages (Voorveld & van der Goot, 2013).

Internet seems to play an important role in multitasking, as it has been found that more than half of internet usage is combined with other media (Papper, Holmes, & Popovich, 2004). The majority of multitasking combinations is either between two internet activities, or between an internet activity and another medium, such as television or radio (Voorveld & van der Goot, 2013). Moreover, new portable media have been found to increase this multitasking tendency (McClard & Somers, 2000). About 85% of tablet and smartphone users combine these media with TV watching, with half of them doing it on a daily basis (Nielsen, 2012). Though usage of portable media presents a tremendous rise, there are not many studies concerning the multitasking with tablet or smartphones. Therefore, this study will focus on the simultaneous usage of tablet and television.

The aforementioned increase of media multitasking has created concerns- among advertisers and marketers- regarding its impact on advertising effects. Previous communication studies investigating multitasking indicate that conditions of divided attention decrease memory (Armstrong & Chung, 2000; Armstrong & Greenberg, 1990; Armstrong & Sopory, 1997, Srivastava, 2013; Zhang, Jeong, & Fishbein, 2010). Divided attention is when someone performs two tasks simultaneously, in contrast with full attention- which is characterized by one task (Lozito & Mulligan, 2010). In the case of media multitasking, the exposure to multiple media divides the

3

attention, which has been found to result in weaker memory of the advertisements (Voorveld, 2011). This negative effect relies on theoretical justifications, such as the Elaboration Likelihood Model (ELM; Petty & Cacioppo, 1984; Petty, Gleicher, & Baker, 1991) and the Limited Capacity Model of information processing (Lang, 2000), both explained minutely in the next chapter.

Although studies using explicit memory measures conclude that multitasking has a negative effect on advertising effectiveness, a recent stream of academic studies focusing on brand placement uses implicit memory measures and reveals promising insights to the research of media multitasking effects (Choi, Lee, & Li, 2013; Yang, Roskos-Ewoldsen, Dinu, & Arpan, 2006; Yang & Roskos‐Ewoldsen, 2007; Yeu, Yoon, Taylor, & Lee, 2013). Despite the different nature of media multitasking and brand placement, both have one mutual characteristic- that is divided attention.

Earlier cognitive psychology research has shown that divided attention harms explicit memory but not implicit memory (Isingrini, Vazou, & Leroy, 1995; Jennings & Jacoby, 1993; Lozito & Mulligan, 2010; Mulligan, 1998; Parkin & Russo, 1990). Hence, it is important to distinguish between explicit and implicit memory when studying media multitasking. Explicit memory is the conscious recollection of past information encoded in memory, whereas implicit memory does not necessitate intentional retrieval due to the occurrence of an automatic process of information recollection (Duke & Carlson, 1993). Therefore, the first aim of this study is to provide insights regarding the implicit and explicit memory of ads in media multitasking by answering the below question:

RQ1: To what extent does multitasking with tablet and television affect

implicit and explicit memory of advertised brands on both media?

Furthermore, the effects of multitasking on implicit memory may differ depending on the focus of attention. Even though the majority of previous studies argue that implicit memory remains unaffected by conditions of divided attention (Isingrini, et al., 1995; Jenning & Jacoby, 1993; Lozito & Mulligan, 2010; Mulligan, 1998; Parkin & Russo, 1990; Shapiro & Krishnan, 2001), there is evidence about the potential role of the focus of attention. Some studies indicate that implicit memory is affected by divided attention (Mulligan, 2002; Sparato, Cestari, & Rossi-Arnaud, 2011; Wolters & Prinsen, 1997), whereas one presents implicit memory being boosted (Spataro, Mulligan & Rossi-Arnaud, 2013). Mulligan (2002) argues that this effect is dependent on attention. Based on this evidence, it is expected that focus of attention

4

will influence the media multitasking effect on implicit memory. In the case of media multitasking, focus of attention is interpreted as the distinction of the primary and the secondary medium, with the primary being on the main focus of attention and the secondary receiving less cognitive resources (Jeong & Hwang, 2012). Therefore, the second aim of this study is to examine the role of the focus of attention regarding the multitasking effect on implicit memory. A second question is as follows:

RQ2: To what extent does focus of attention influence the effect of media

multitasking on the implicit memory of brands?

Due to the evidence that applying also implicit memory measures provides a more integrative comprehension of media effects (Hefner, Rothmund, Klimmt & Gollwitzer, 2011), this study contributes to the interpretation of multitasking effects on advertising. Further, by investigating the role of attention focus, this study attempts to explain potential differences of media multitasking effects on implicit memory. Finally, even though previous research has found that internet with television is the second most common combination (Voorveld & van der Goot, 2013), there are not many studies focusing on multitasking with new media, such as tablets. Hence, by examining the combination of tablets and television, this paper adds contemporary value to the media multitasking literature.

It is not only scholars who are interested in understanding media multitasking impacts, but also marketers and advertisers seeking to adjust their strategies according to the changes in the media field. Many years now, tablet devices have entered in our life and favored the rise of multitasking (McClard & Somers, 2000). As mentioned previously, about 41% of the tablet users combine this medium with watching television at least once per day (Nielsen, 2012). However, it is not clear what the implications are, when tablet is combined with television. Moreover, it is supported that implicit memory is highly correlated with purchase intention (Kardes, 1986). Thus, by using implicit memory measures, this study allows marketers to comprehend the effects of multitasking and utilize these insights in their strategies accordingly. In sum, this research aims to guide practitioners in choosing a suitable medium for their advertisements by taking into consideration potential differences in multitasking effects.

5

Theoretical Framework Multitasking Effects on Memory

The ability to multitask with ease and efficiency has traditionally been considered a positive characteristic (Judd, 2013). However, this view has been outdated due to earlier empirical findings in communication studies, which support that media multitasking reduces memory of information (Armstrong & Chung, 2000; Armstrong & Greenberg, 1990; Armstrong & Soropy, 1997; Srivastra, 2013; Zhang, et al., 2010). Similarly, engaging with multiple media harms the memory of advertisements (Voorveld, 2011). This decrease in cognitive response occurs because multitasking inhibits the processing of information to short- and long-term memory (Edwards & Gronlund, 1998).

Communication studies focusing on media multitasking are in line with earlier empirical findings of psychology research, which support that divided attention affects memory negatively (Isingrini et al., 1995; Jennings & Jacoby, 1993; Lozito & Mulligan, 2010; Mulligan 1997; Parkin & Russo, 1990; Wolters et al., 1997). These studies argue that, in contrast to full attention, divided attention impairs the recall or recognition of information. This happens because attention is a prerequisite of a mindful processing of information, which can lead to stronger memory. When attention is divided to two or more sources of information, it is not possible to process mindfully the content from all the sources simultaneously or to memorize the information as good as in a single-tasking circumstance.

A distinction between primary and secondary medium is essential in order to comprehend the effects of media multitasking (Jeong & Hwang, 2012). Primary is the medium that receives more cognitive resources (e.g. print), whereas the secondary medium can be consumed with less attention (e.g. radio). The negative effects of multitasking occur because the secondary medium functions as an interruption to the primary and, according to previous findings, interruption harms memory (Edwards & Gronlund, 1998). On the other hand, reduced information processing is particularly related to the content disseminated by the secondary medium (Jeong & Hwang, 2012). This is caused by the fact that primary medium receives more attention and limits the processing of the secondary medium. Thus, primary medium harm the memory of the secondary, and vice versa.

6

Theoretical Explanation of Multitasking Effects

Both streams of academic research can be justified by the same theoretical foundations, such as the Limited Capacity Model of information processing (Lang, 2000) and the Elaboration Likelihood Model (Petty & Cacioppo, 1984).

Limited Capacity Model of information processing. First, the Limited Capacity Model of information processing (Lang, 2000) can explain the negative effects of multitasking and divided attention, in general. This model maintains that people have limited cognitive resources, which implies limited message processing during conditions of distraction. Consequently, limited message processing results in weaker memory and reduced comprehension of the content of a message. In other words, when people are exposed to two media simultaneously, they face difficulties processing information from both media due to limited cognitive resources.

Elaboration Likelihood Model. A second theoretical foundation that expounds the aforementioned negative effects of multitasking is the ELM developed by Petty and Cacioppo (1984). This theory argues that information can be processed in two different ways. On the one hand, the peripheral processing of information is affected by unrelated cues and implies unstable outcomes, such as weak memory of the exposure. On the other hand, people process information centrally when they have the motivation and ability for the processing. This occurs after deliberate encoding and results in stronger memory of the information. In the case of multitasking, the existence of two media restricts the ability to process information centrally. The interruption caused by the secondary medium inhibits thoughtful processing of the information and, thus, leads to weaker memory.

Hypotheses about Multitasking Effects on Memory of Ads

According to the previous studies and the theoretical justification, it is expected that multitasking will affect negatively the memory of the ads presented on both media: television and tablet. When a television is used alone, all the cognitive resources are available in order to process the information centrally and encode that information in the memory. However, when television is used in combination with a tablet, then it becomes a secondary medium, as multitasking with television and computer devices leads the attention primarily to the latter ones (Brasel & Gips, 2011). This is expected to reduce the memory of television advertisements. Hence, it is hypothesized that:

7

H1: When people are simultaneously exposed to television and tablet,

memory of advertised brands on television is weaker than memory of the same ads when television is used alone.

Similarly, it is expected that memory of banner ads of the tablet will be weaker when this new medium is combined with exposure to television. Although a computer device is used as primary medium, when someone is engaged in multitasking with television (Brasel & Gips, 2011), the secondary medium distracts attention and reduces the encoding in memory (Edwards & Gronlund, 1998). Armstrong and his colleagues (1990; 1997; 2000) distinguished the interruptive role of television in their studies by calling it “background television”. Based on this evidence, it is expected that multitasking with television will affect the memory of banner ads presented through tablet. Hence, it is hypothesized that:

H2: Compared to engagement with tablet solely, memory of brands

presented via banner ads on a tablet is weaker when someone multitasks with television.

Explicit and Implicit Memory

Earlier research of multitasking has focused mainly on explicit memory of ads, though there is evidence that implicit memory should also be considered when studying the effects of advertisements (Duke & Carlson, 1993; Hefner et al., 2011). It was defined previously that explicit memory is the conscious recollection of information from memory, whereas the implicit memory is the unconscious retrieval from memory (Duke & Carlson, 1994). Although researchers in the communication and advertising field neglect studying both memories, there are some scholars who focus on implicit memory as well.

Research of brand placement is a good example of studying the effects of divided attention and implicit memory within the field of advertising. When people are exposed to brand placement, their attention is divided between the main content and the advertisements. Earlier research shows that implicit memory of in-game ads is higher than explicit memory (Choi et al., 2013; Yang et al., 2006). Similarly, a study of brand placement in movies concluded that implicit memory of ads is unaffected by the centrality of the placement compared to explicit memory, which is reduced for ads in peripheral location (Yang & Roskos-Ewoldsen, 2007). Additionally, Shapiro and Krishnan (2001) argue that implicit memory of ads is not affected by divided attention,

8

whereas explicit memory is decreased. These findings support the existence of a disassociation between the implicit and explicit memory of ads, when someone is engaged in multitasking.

The assumption of the explicit and implicit memory disassociation is enhanced also through findings of cognitive psychology. It was discussed earlier that studies of divided attention indicate negative effects on explicit memory of information (Isingrini et al., 1995; Jennings & Jacoby, 1993; Lozito & Mulligan, 2010; Mulligan, 1998; Parkin & Russo, 1990). However, the same studies show that implicit memory is not affected by divided attention. The differences between implicit and explicit memory are based on the different encoding of information (Hefner, et al., 2011). The theoretical justification of implicit and explicit memory differences are provided minutely in the next chapter.

Theoretical Explanation of the Disassociation of Explicit and Implicit Memory In order to explain the differences between the explicit and the implicit memory in conditions of divided attention, it is useful to examine some related theories of information processing, such as dual-processing theories and the Transfer Appropriate Processing theory.

Dual-processing theories. Fazio and Olson (2003) criticized the lack of theoretical framework regarding implicit memory and proposed the Motivation and Opportunity as Determinants model (MODE; Chaiken & Trope, 1999). This model is a dual-processing model, such as the ELM (Petty & Cacioppo, 1984). According to the dual-process theories, processing of information can occur either in a deliberate way or through a spontaneous, automatic way (Strack & Deutsch, 2004). As aforementioned, the former leads to a predictable and stable outcome, while the latter leads to a less stable response.

Smith and DeCoster (2000) expanded the dual-process theory and argued about the existence of a dual memory system. Particularly, the authors distinguished between two processing modes: the rule-based and the associative processing. The rule-based

processing is a deliberate and conscious learning of regularities, a condition that

requires motivation and capacity. These regularities can be learned in just one or a few exposures and can be retrieved from memory intentionally. On the other hand,

associative processing is a less effortful process that occurs automatically after many

9

information in the memory. In contrast to the rule-based retrieval, this underlying memory is accessed unintentionally and unconsciously. Scholars not only support the distinction of explicit and implicit memory, but they also claim that the two memories work simultaneously, rather than in sequence (Smith & DeCoster, 2000).

Thus, when attention is divided between multiple media, the information is processed, and retrieved, based on this dual-processing memory construct. The primary medium is processed centrally, whereas the secondary medium is processed peripherally. Hence, the information delivered by the primary medium can be retrieved consciously, while the content disseminated by the secondary medium can be retrieved unconsciously.

Cognitive processing approaches. A further explanation to the differences of explicit and implicit memories is also based on the assumption of the existence of two different processes. The major argument is that disassociations in explicit and implicit memory are caused by differences regarding two processes of information: perceptual and conceptual (Duke & Carlson, 1993). The Transfer Appropriate Processing (TAP) theory distinguishes between data-driven and conceptually driven tasks, with the first being based on perceptual processes characterized by analysis of the surface properties of a stimulus and the latter being based on conceptual processes of analyzing the meaning of the stimulus (Spatar et al., 2011).

As Mulligan and Hartman (1996) mention, during conditions of full attention, semantic or conceptual processes are more prominent due to the elaborative encoding that necessitates more capacity. Thus, a decrease in attention affects only the memory that is related to these effortful processes, leaving perceptual memory unaffected. In other words, when someone is engaged with one medium only, the information is processed conceptually due the ability to analyze the meaning of the disseminated information. In contrast, the engagement with multiple media inhibits the conceptual processing of information from all the sources and the content is analyzed superficially.

Hypotheses about Multitasking Effects on Implicit Memory of Ads

Even though the majority of studies using implicit memory measures argue that implicit memory is not affected by divided attention (Isingrini et al., 1995; Jennings & Jacoby, 1993; Lozito & Mulligan, 2010; Mulligan, 1998; Parkin & Russo, 1990; Shapiro & Krishnan, 2001), there are some findings that suggest taking into

10

consideration the role of attention. For instance, Wolters and Prinsen (1997) found that strong division of attention is harmful for explicit and implicit memory, whereas mediocre division of attention affects negatively only the explicit memory. A further argument is that implicit memory is decreased only in cases of attention being divided on two objects (Mulligan, 2002). These negative results are further supported by a meta-analysis of implicit memory in divided attention conditions which showed implicit memory being affected slightly negatively by divided attention (Sparato et al., 2011).

The above findings enhance the importance of attention for the existence of implicit memory (Mulligan, 2002). When someone multitasks with tablet and television, attention is divided across two media and, a distinction of primary and secondary focus of attention is created automatically. Due to the different attentional load that each medium receives, different expectations can be formed for implicit memory of television and banner ads.

As aforementioned, when people multitask with tablet, television instantly becomes the secondary medium, as the attention spent on this medium is decreased. Hence, it is expected that multitasking will affect negatively the implicit memory of television ads, because the attention is reduced due to the transition from the primary to secondary medium. Hence, it is predicted that:

H3: Compared to single-tasking, implicit memory of advertised brands on

television is reduced by multitasking with tablet.

On the other hand, with regard to the tablet, it is important to understand that- in multitasking with tablet and television- there are two divisions of attention, one within the tablet (main content and banner ads) and one between two media (television ad tablet). Despite this double division, however, computer devices remain the primary medium in multitasking with television (Brasel & Gips, 2011) and, as already noticed, limited processing is mostly related to the information of the secondary medium (Jeong & Hwang, 2012). Therefore, it is expected that implicit memory of banner ads will not be affected by multitasking, as tablet remains on the main focus of attention. Hence, it is hypothesized that:

H4: Compared to single-tasking, implicit memory of advertised brands on the

tablet is not affected by multitasking with television.

Despite the potential reduction of implicit memory, overall implicit memory of advertisements is higher than explicit memory (Choi et al., 2013; Yan &

Roskos-11

Ewoldsen, 2007; Yang et al., 2006). Also, a cognitive psychology meta-analysis conducted by Sparato et al. (2011) concluded that divided attention had smaller effect on implicit memory than on explicit memory of information. Hence, it is expected that implicit memory of ads presented on tablet and television will be affected less by multitasking than the explicit memory of the same ads. According to this evidence, it is hypothesized that:

H5: Implicit memory of advertised brands on tablet and television is affected

less by media multitasking than explicit memory of the same ads.

For better interpretation of the expected effects, the conceptualization of the multitasking effects on explicit and implicit memory is provided in the Figure 1.

Figure 1

Multitasking effects on explicit and implicit memory. The role of focus of attention

Method Design and Sample

A mixed-level factorial experimental design was conducted, with 3 between-group conditions (multitasking vs. television-tasking vs. tablet-tasking) and 2 within-subject conditions (multitasking primary attention vs. multitasking secondary attention). Participants were assigned randomly in one of the three between-group conditions, with the first subject being assigned to the tablet-tasking condition, the second to the tv-tasking, the third to the multitasking, and so on.

The sample was recruited through convenience and snowball sampling, and in total, 72 people took part in this experiment. One person did not complete the procedure, and three subjects declared awareness of the purpose of the implicit memory test. By excluding the participants, the sample was N = 68. About 53% of the

12

sample was female and 47% male. Participants ranged from 18 to 58 years old, and the mean age was 28.65 (SD = 8.21, Mdn = 27).

It is important to mention that the nationality of the sample was homogeneous, as all the participants were Greek. This was chosen mainly for two reasons. First, by choosing a nationally homogeneous sample, it was intended to avoid disassociations in the levels of brand knowledge and brand familiarity among the sample. Second, it was expected that language plays an important role, thus by using the native language of the homogeneous sample, it was anticipated that everyone would be able to comprehend the tasks at the same level.

Pre-test for Material Selection

Initially, the researcher made a selection of twenty television ads and six banner ads, all of them expected to be familiar and low or medium involvement. Familiarity and involvement were taken into consideration, because the former has been found to increase implicit memorizing (Pieters, Warlop, & Wedel, 2002), and the latter one influences information processing (Greenwald & Leavitt, 1984). An online pre-test was conducted in order to measure the brand and ad attitude as well as the familiarity with the chosen brands. Respondents were asked to watch the ads, and after each advertisement, they had to answer three questions. The whole procedure of the pre-test study lasted 25 to 30 minutes. A sample of 25 people was reached, and the questionnaire was completed by N= 17. The final sample was 82% female, and the average age was 29.59 (SD = 8.32, Mdn = 28). All the people who took part in the pre-test study were excluded from the sampling of the main experiment.

The questionnaire focused on three measures. Brand attitude was derived from the study of Campbell and Keller (2003), and was measured with four items (bad-good, low quality-high quality, unappealing-appealing, unpleasant-pleasant) on a seven-point differential scale with 1 for positive and 7 for negative attitude (Cronbach’s alpha = 0.88, M = 2.67, SD = 0.58). Ad attitude (Campell & Keller, 2003) was measured with the same four-item seven-point differential scale (Cronbach’s alpha = 0.91, M = 3.08, SD = 0.80). Familiarity (Kent & Allen, 1994) was assessed with three items (familiar-unfamiliar, inexperienced-experienced, knowledgeable-not knowledgeable), on a seven-point differential scale with 1 for familiar and 7 for unfamiliar (Cronbach’s alpha = 0.91, M = 2.62, SD = 0.85). A preview of the pre-test questionnaire can be found in the Appendix 1 at the end of this study.

13 Experimental material

In order to select the advertisements, television and banner ads were placed in ascending order for both measures: ad and brand attitude. By ordering the ads from the most positive to the most negative (Table 1-2 in Appendix 2), it was possible to exclude the ads and brands that were characterized by extreme scores. This selection was based on the fact that extremely positive or negative ads may attract more the attention and be memorized better. Thus, eight television ads and three banner ads were selected for the experiment. The number of ads was chosen according to observations of the number of advertisements on websites and television channels. By using a number of ads that corresponds to the real life, it was possible to naturalize the settings of the exposure. The average familiarity, brand and ad attitude for all the selected ads can be found in the Table 3 of the Appendix 2.

The eight selected television ads had about a 30 second duration each. A montaged video was created for the needs of the study, with total duration about four minutes. The video was starting and ending with ads about TV shows of a specific channel, in order to make the video more natural. On the other hand, a blog was created for the experiment with an article accompanied by the three selected banner ads. The professional assistance of an IT developer was necessary in order to create a compatible website version, with the ads being visible through any device. The blog article was about an exhibition of physics which was held in a different city than the one where the experiment was conducted. This article was chosen in order to keep the involvement with the article low, as it was expected that people will not be interested in an exhibition that was mostly for children and held in a different city. Moreover, the duration of the video was similar with the duration of reading the blog article (about four minutes). A link of the montaged video and a preview of the website are provided in the Appendix 3.

Procedure

The experiment was held in an office, where a couch was placed about four meters across a television of 32 inches. Television was switched off for participants in the tablet-tasking condition, as those subjects were provided solely with a tablet of seven inches. Participants were allowed to take the test only individually. Before starting the procedure, they were informed with a cover story about the nature and the topic of the study, they also signed an informed consent regarding their participation.

14

People were informed that they were going to participate in a study concerning “Media and Personality” and no reference to multitasking or memory occurred, as this would create reaction effects to the experiment.

Initially, participants followed the instructions for the exposure phase. The researcher asked from them either to “watch the video on the TV”, to “take a look at the blog”, or to “take a look at blog while a video on the TV plays simultaneously”. There was no any reference to the advertisements or the following memory tests. By giving the above guidance, the researcher aimed to avoid creating awareness concerning the ads and to keep the settings of the experiment as naturalized as possible. While the subjects were involved in the exposure part, the researcher observed and recorded multitasking that was out of the control of the researcher, such as engagement with the mobile phone.

Afterwards, a questionnaire was provided to measure the effects on explicit and implicit memory, as well as potential covariance. The total time of the exposure and questionnaire phase lasted about 15-25 minutes. At the end of the procedure, the researcher conversed with the participants in order to identify whether or not they realized the purpose of this study. A brief description of the aim of the research was provided. No reward was given for the participation in the experiment.

Measures

Independent variables. The current research focuses on the role of two independent variables, multitasking and focus of attention.

Multitasking. The manipulation of this experiment was concerning the number of

media that people become engaged with. Participants were exposed either to two media (tablet and television), or only to one of the two media. Thus, three groups of media exposure were created which were coded with 1 for people exposed solely to tablet (ntablet = 23), 2 for those who were engaged only with television (ntv = 22), and 3 for participants who were exposed to both media (nmt = 23).

Focus of attention. It is expected that attention influences the effect of

multitasking on implicit memory. Due to the fact that attention is difficult to be measured without technical tools, such as eye-tracking, this study operationalizes the attention based on the distinction of the primary-secondary media usage (Jeong & Hwang, 2012). In the case of multitasking with computer devices and television, tablet is the primary medium and television is the secondary, as the former receives more

15

attention than the latter (Brasel & Gips, 2011). There was no coding, because the television ads represented the memory of the secondary medium and the banner ads represented the primary medium.

Dependent variables. This study focuses on the effects of multitasking on memory, in particular explicit and implicit memory. Thus, the measure phase of the experiment began with an implicit memory test. The implicit memory test, along with the other measures of the questionnaire, can be found in the Appendix 4.

Implicit memory. There are several different implicit memory tests, but the more

accurate for comparison with explicit memory tests are the word stem and the word fragment completion tests (Duke & Carlson, 1993). Both word completion tests aim to measure the performance of subjects on stimuli they have seen previously, compared to stimuli that have not been attended. The difference between the tests is that word stem completion test necessitates the completion of the ending of a word (e.g. NES_ _ _ for Nestle), whereas, in the word fragment completion test, participants have to complete random letters to create a word (e.g. N_S_L_ for Nestle).

For the needs of this study, the word fragment completion test was used in order to assess the implicit memory of the ads. This selection was based on previous findings that support the suitability of the specific test for studies focusing on divided attention (Mulligan & Hartman, 1996) and for studies of advertising effects (Duke & Carlson, 1994). Hence, for each brand name, a fragmented word was created by excluding random letters. Distractor words of brand names that were not contained in the exposure phase were also added. Two Wilixon signed-rank tests were performed in order to compare the performance of stimuli and distractors on television and banner ads. The results showed that people performed significantly higher on stimuli than on distractors (p < .01). The initial variable of the implicit memory for each ad was dichotomous (0 for no memory, 1 for memory), and two variables were computed, one for implicit memory of television ads (M = 0.38, SD = 0.25) and one for implicit memory of banner ads (M = 0.14, SD = 0.22).

Finally, it is important to mention that all the selected brand names were composed of letters that exist also in the Cyrillic alphabet. This initial choice was based on the fact that by presenting Latin letters to a native sample that does not use the Latin alphabet would create bias and awareness concerning the aim of the procedure.

16

Explicit memory. Explicit memory is usually measured with recall or recognition.

Both measures aim to elicit information from memory through an intentional and conscious way (Lozito & Mulligan, 2010). The current research used both explicit measures based on the argument that recall and recognition represent different types of brand awareness (Percy & Rossiter, 1992).

Recall is defined as the intentional reconstruction of a stimulus based on a

previous exposure (Duke & Carlson, 1993). This study used unaided brand recall, and thus, after the implicit test, participants were asked to recall brand names from the exposure phase. This was also a dichotomous variable with 0 for absence of recall and 1 for recall. Two new variables were computed in order to represent the average recall of television (M = 0.25, SD = 0.22) and banner (M = 0.15, SD = 0.24) advertisements.

Recognition task was performed directly after the recall test in order to avoid

priming the previous memory test. Recognition is defined as the conscious selection of a stimulus from a previous exposure, in combination with rejection of a non-seen stimulus (Duke & Carlson, 1993). In this study, for each ad, three brand names of the same product category were given, and subjects were asked to recognize the brand that was advertised previously. Moreover, they were asked to avoid replying based on luck, as there was a 33% chance of giving the correct answer. Similarly to the other memory tests, this variable was on a dichotomous scale (0 for no recognition, 1 for recognition), based on which two new variables were computed for recognition of television (M = 0.48, SD = 0.31) and banner (M = 0.29, SD = 0.32) ads.

Covariates. In order to control for potential covariates with other factors, some additional questions were included in the questionnaire. After the memory tests, participants had to indicate their familiarity and involvement with each brand. In the end, they were asked about their gender and age.

Brand familiarity. Familiarity is the reflection of the direct and indirect

knowledge concerning a brand, a structure with associations that exist in the memory (Campbell & Keller, 2003). It is expected that familiarity plays an important role in memorizing a brand implicitly (Pieters et al., 2002), thus it seemed important to include this factor in the study. Brand familiarity was measured on the same three-item (experienced-inexperienced, knowledgeable-unknowledgeable, familiar-unfamiliar) seven-point differential scale which was used in the pre-test study (Kent & Allen, 1994), but this time reversed: with 1 for unfamiliar brand and 7 for familiar. Reliability analyses showed that the measure had acceptable reliability (Cronbach’s

17

alpha > .70). For the needs of this study, a variable of familiarity was created for each ad separately. Afterwards, a variable was computed to represent familiarity with television ads (M = 4.79, SD = 0.99) and another one for the familiarity with banner ads (M = 4.33, SD = 0.97), with both being on an interval scale.

Involvement. This study interprets involvement not as personal relevance to a

brand or a product (Zaichkowsky, 1985) but as a characteristic of the brand/product itself. Despite the fact that involvement is experienced by the individual, there is evidence that it is more accurate to talk about low or high involvement products and brands (McWilliam, 1997). The level of the involvement plays an important role in decision making, as different levels of involvement signify different levels of processing information (Greenwald & Leavitt, 1984). Thus, the current study used the involvement measure of the McWilliam (1997) study which consists of three items on a seven-point differential scale (decision is/not very important, decision requires a lot of thought/decision requires little thought, a lot to lose if you choose the wrong brand/little to lose if you choose the wrong brand) with 1 for low involvement and 7 for high involvement. After computing a variable for each ad, two variables were created in order to represent the involvement of banner ads (M = 3.80, SD = 1.24) and involvement of television ads (M = 3.34, SD = 0.92).

Age. It is expected that age has negative effect on explicit memory (Isingrini et al.,

1995), thus it seemed necessary to examine for potential covariance of this factor. The initial ordinal variable of age was recoded according to its median (Mdn = 27) in order to create a dichotomous variable, with 0 for younger and 1 for older participants.

Gender. Participants were asked to indicate their gender, which was a

dichotomous variable (1= Male, 2= Female).

Additional multitasking. The researcher observed the participants and kept data

about any further tasks, such as engagement with mobile phone or discussion with the researcher. This variable was also on a binary scale, with 0 for no additional tasks and 1 for additional multitasking.

Finally, the average implicit memory, recall, recognition, familiarity and involvement for all the ads is provided in the Appendix 5.

Results

Before starting the analyses, it was necessary to examine the distribution of the sample. Two chi-square goodness-of-fit tests were performed for age and gender in

18

order to determine whether the sample was equally distributed to older-younger and male-female respectively. The results indicated that age was equally distributed, χ2 (1,

Ν = 68) = 0.235, p = .628, as well as gender, χ2

(1, Ν = 68) = 0.235, p = .628. Additionally, two chi-square analyses were applied to test the distribution of the population within the experimental conditions. Older and younger participants were distributed equally in the three conditions, χ2 (2, Ν = 68) = 1.98, p = .372 (2-tailed), as well as male and female participants, χ2 (2, Ν = 68) = 2.04, p = .204 (2-tailed). Thus, the randomization of the sample was successful.

Moreover, assumption of normality was tested with Kolmogorov-Smirnov (K-S) tests to clarify which statistical technique must be chosen. Results indicated that the assumption of normality was violated for all memory measures (p < .001). By splitting the data for single-tasking and multitasking, the results of the K-S tests remained significant. Therefore, nonparametric statistical techniques are more appropriate for the analyses of the data.

In order to test whether control variables need to be incorporated in the analyses of data, multiple nonparametric tests (such as chi-square test and Spearman’s rho correlation) were run. Familiarity, involvement, age, gender and additional multitasking were examined. The results indicated that there was no relationship among the variables, except for three cases. According to nonparametric correlational analysis, involvement increased recall of television ads, but this effect was weak, rs = .26, p = .040. However, the opposite findings revealed for banner ads, as involvement decreased the recall of banner ads, but this effect was also weak, rs = -.285, p = .028. Additional multitasking was found being significantly related to the recall of banner ads, χ2 (2, Ν = 68) = 6.09, p = .024 (1-tailed). Due to the weakness of the involvement and additional multitasking effects, these variables were not taken into consideration for covariance.

Multitasking Effects on Explicit Memory of Ads

In H1, it was predicted that, compared to single-tasking, explicit memory of television ads is weaker when someone multitasks with tablet. Two Man-Whitney exact tests were performed for recall and recognition of television ads as dependent variables and experimental conditions as independent. Recall of TV ads in multitasking (Mdn = 0.12) was significantly weaker than in single-tasking condition (Mdn = 0.50), U = 48.50, z = -4.75, p < .001 (1-tailed). Similarly, participants

19

engaged with two media (Mdn = 0.12) recognized significantly less ads than participants who were engaged only with TV (Mdn = 0.75), U = 29.00, z = -5.13, p < .001 (1-tailed). Hence, the first hypothesis is confirmed because, in both measures, multitasking affected negatively the explicit memory of television ads.

According to H2, it was expected that when someone is engaged with tablet and television explicit memory of banner ads is weaker than in engagement with tablet only. Identically to the previous hypotheses, two Mann-Whitney exact tests were performed, with recall and recognition of banner ads as outcome variables and experimental conditions as the predictor. Results showed that participants in single-tasking condition (Mdn = 0) did not recall significantly more ads than multisingle-tasking participants (Mdn = 0), U = 231.50, z = -0.86, p = .246 (1-tailed). Concerning the recognition of banner ads, participants in multitasking condition (Mdn = 0.33) did not differ significantly from participants in single-tasking condition (Mdn = 0.33), U = 243.00, z = -0.51, p = .307 (1-tailed). Therefore, H2 is rejected.

Multitasking Effects on Implicit Memory of Ads

According to H3, implicit memory of brands on television is affected negatively by multitasking with tablet. Identically to the previous hypotheses, a Mann-Whitney exact test was conducted in order to test for differences between single- and multitasking. Results showed that participants in the single-tasking condition (Mdn = 0.50) had significantly higher implicit memory than participants in the multitasking condition (Mdn = 0.25), U = 157.50, z = -2.20, p = .014 (1-tailed). Hence, H3 is confirmed, as the results show that, compared to single-tasking, implicit memory of television ads is weaker when someone multitasks.

In the H4 it was predicted that implicit memory of banner ads is not affected by multitasking. According to the Mann-Whitney exact test, implicit memory of banner ads did not differ significantly in single-tasking (Mdn = 0) and multitasking (Mdn = 0) condition, U = 264.00 , z = -0.01, p = .500 (1-tailed). Hence, H4 is also confirmed.

Finally, according to H5, implicit memory of ads is expected to be affected less negatively by multitasking than explicit memory for both television and tablet. Multiple Mann-Whitney exact tests were performed in order to compare the effect on implicit and explicit memory. For tablet ads, implicit memory of single-taskers (Mdn = 0) did not differ significantly from implicit memory of multitaskers (Mdn = 0), U = 264.00, z = -0.01, p = .500 (1-tailed). Concerning the recognition of banner ads,

20

participants in the multitasking condition (Mdn = 0.33) did not differ from those in the single-tasking condition (Mdn = 0.33), U = 243.00, z = -0.51, p = .307 (1-tailed). Finally, participants engaged with two media (Mdn = 0) did not recall significantly more banner ads than participants engaged with tablet only (Mdn = 0), U = 231.00, z = -0.86, p = .246 (1-tailed). Thus, the hypothesis is not accepted for the memory of banner advertisements, as there are no differences between single- and multitaskers. On the other hand, with regard to the television ads, participants engaged with TV only (Mdn = 0.50) had significantly higher implicit memory than participants engaged with both media (Mdn = 0.25), U = 157.50, z = -2.20, p = .014 (1-tailed). With regard to the recall of television ads, participants who engaged with television solely (Mdn = 0.50) had significantly higher memory than participants in the multitasking condition (Mdn = 0.12), U = 48.50, z = -4.75, p < .001 (1-tailed). Similarly, participants in the single-tasking condition (Mdn = 0.75) recognized significantly more television ads than participants in the multitasking condition (Mdn = 0.12), U = 29.00, z = -5.13, p < .001 (1-tailed). In order to provide answer to the hypothesis, it was necessary to calculate the effect size as follows (Field, 2009):

N z

r  (1)

By using the z-scores that the Mann-Whitney tests generated and the root of the sample (n = 45), the effect size was calculated manually. Based on the results, implicit memory (r = .33) of television ads was reduced less by multitasking than recall (r = -.71) and recognition (r = -.76) of the same ads. Thus, H5 is accepted partially because the hypothesis was confirmed only for the advertisements presented on television but not for banner ads.

Conclusion

The aim of this research was to examine the effects of multitasking- with television and tablet- regarding explicit and implicit memory of advertisements presented on both media. A second aim was to identify the role of the focus of attention concerning the implicit memory of ads. The analyses showed that multitasking had strong negative effect on the explicit memory of television ads, whereas implicit memory of these ads was affected weakly. With regard to the memory of advertisements presented on tablet, explicit and implicit memory were unaffected by multitasking with television. Focus of attentions plays an important role

21

for implicit memory of ads, as advertisements of the secondary medium were memorized less due to multitasking, whereas implicit memory of ads presented on the primary medium remained unaffected.

The analyses revealed conflicting results, with the expectations being confirmed partially. Explicit memory of television ads was reduced by engagement with tablet, with the findings being in line with previous studies of multitasking effect on explicit memory (Srivastava, 2013; Voorveld, 2011; Zhang et al., 2010). However, explicit memory of banner ads was not reduced by simultaneous engagement with television. This was contrary to the expectations and earlier research that has shown that explicit memory is reduced due to background television (Armstrong & Chung, 2000; Armstrong & Greenberg, 1990; Armstrong & Soropy, 1997). The findings with regard to the implicit memory of ads and the focus of attention supported the argument that attention plays role in memorizing implicitly (Mulligan, 2002, Sparato et al., 2011), as the implicit memory of television ads was reduced due to the reduction of the attentional load when someone multitasks with tablet.

There are two possible explanations for these results. First, potential reasoning for the low memory of banner ads is a fact that was observed during the procedure by the researcher. Some of the participants who used the tablet expressed ad avoidance by using the “zoom” function with their hand, focusing only on the article. Even though ad avoidance might be the explanation for the low memory of banner ads, it does not explain why there was no effect of multitasking. Previous studies have shown that television impairs the memory of the primary task (Armstrong & Chung, 2000; Armstrong & Soropy, 1997). However, the difference among these studies and the current is that previous studies examined the memory of information that was processed deliberately, whereas this study does not. Banner ads are not the main focus when someone is engaged with a website due to the inner division of attention. It might be the case that a second division of attention does not impair the memory of the information that was processed peripherally in the initial divided attention. In order to clarify whether ad avoidance is the explanation, a replication with a desktop computer would be suggested, as it has a larger screen and users do not zoom.

A more integrative explanation is provided by the interpretation of primary-secondary medium distinction in combination with the double division of attention. In other words, low memory of banner ads is caused by the fact that attention is already divided within the tablet because of the simultaneous existence of multiple stimuli

22

(title, article, ads, et cetera). Thus, by adding an additional division of attention, such as television, the already low memory of ads is not reduced further due to the fact that the computer devices remain the primary medium in both conditions (Brasel & Gips, 2011). This absence of transmission from primary to secondary medium might be the cause of the absence of the multitasking effect for banner ads. Therefore, focus of attention seems to play an important role not only for the implicit memory but also for the explicit memory of the advertisements.

As with all the researches, there are some limitations in this study. First, the sample size is slightly small, and it is believed that by obtaining bigger sample (N > 30), the results might differ. In order to overcome this limitation, the main analyses (Mann-Whitney tests) were accompanied by exact probability tests. According to Filed (2009), exact probability tests are more suitable to generate reliable statistical results for small samples than approximate probability tests. Second, the experimental groups of tablet, television and multitasking differed concerning the number of ads (three, eight, and eleven, respectively). The different number of ads would necessitate a different amount of cognitive resources. This limitation would be expected to prime the memory of the banner ads in the single-tasking condition, as those participants had to remember only three ads. However, as mentioned earlier, memory of banner ads was extremely low. Thus, this limitation had no impact on the study but, instead, provided a more naturalized environment for the experimental settings. Finally, an important limitation was the fact that the pre-test sample was 82% female. This has probably created gender bias in the selection of the experimental material, as women are expected to have different predisposition towards ads than men.

Despite some limitations, it is believed that the current research has contributed to the academic fields of media and advertising studies. Although the majority of the studies focus on explicit memory, there is evidence that implicit memory measures should also be used (Duke & Carlson, 1993; Grimes & Kitchen, 2007; Yoo, 2007). It was found that implicit memory of television advertisements also exists, and it is not affected by multitasking as negatively as explicit memory. Thus, future researches are advised to take into consideration implicit memory as well, by adding this measure a more integral examination of advertising effects on memory may be performed. Moreover, this study contributes to the media multitasking research by distinguishing between primary and secondary focus of attention. This distinction allows a better interpretation of what changes occurs due to multitasking,

23

and how a transition from a primary to secondary medium affects the memory of advertisements. It was expected that by remaining on the primary focus, implicit memory of ads presented on a tablet is not reduced. Not only the results supported this argument, but also the same pattern revealed for the explicit memory of banner ads. Therefore, further research should elaborate on the role of attention when studying multitasking, by distinguishing between primary and secondary medium.

The findings of this study have also some practical implications that seem to be helpful for advertisers and marketers. Considering the fact that multitasking with television increases (Nielsen, 2010), professionals should not be concerned about the effects on advertising, as this tendency harms mostly explicit memory. Although explicit memory is related to decision-making prior to purchase and at the point of purchase (Percy & Rossiter, 1992), this lack of the explicit memory can be the reason for low counterargument towards ads consumed during engagement with multiple media (Jeong & Hwang, 2012). It is also promising that implicit memory of television ads is reduced less than explicit memory, as previous evidence supports that implicit memory is positively related with judgments and purchase intention (Kardes, 1986). Second, even though the memory of banner ads was low, it was not reduced due to multitasking with television. Thus, advertisers should not worry about the impact of multitasking on online advertising, as this medium remains on the primary focus of attention, and the memory of ads is not affected by the existence of a second medium. The only thing that must be handled with caution is the compatibility of the website with the new devices. In other words, marketers should pay heed on whether their website is adjustable to different screen sizes or not.

To conclude, hopefully, this study provided valuable insights to the research of multitasking with new media and its effect on the memory of ads. In particular, the current research contributed by highlighting the importance of utilizing also implicit memory measures when studying advertising.

24 References

Armstrong, G. B., & Chung, L. (2000). Background television and reading memory in context assessing TV interference and facilitative context effects on encoding versus retrieval processes. Communication Research, 27(3), 327-352.

Armstrong, G. B., & Greenberg, B. S. (1990). Background television as an inhibitor of cognitive processing. Human Communication Research, 16(3), 355-386. Armstrong, G. B., & Sopory, P. (1997). Effects of background television on

phonological and visuo-spatial working memory. Communication

Research,24(5), 459-480.

Bardhi, F., Rohm, A. J., & Sultan, F. (2010). Tuning in and tuning out: media multitasking among young consumers. Journal of Consumer Behaviour, 9(4), 316-332.

Brasel, S. A., & Gips, J. (2011). Media multitasking behavior: Concurrent television and computer usage. Cyberpsychology, Behavior, and Social Networking, 14(9), 527-534.

Campbell, M. C., & Keller, K. L. (2003). Brand familiarity and advertising repetition effects. Journal of Consumer Research, 30(2), 292-304.

Chaiken, S., & Trope, Y. (Eds.). (1999). Dual-process theories in social psychology. Guilford Press.

Choi, Y. K., Lee, S. M., & Li, H. (2013). Audio and visual distractions and implicit brand memory: A study of video game players. Journal of Advertising, 42(2-3), 219-227.

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, B. M., & Gronlund, S. D. (1998). Task interruption and its effects on memory. Memory, 6(6), 665-687.

Fazio, R. H., & Olson, M. A. (2003). Implicit measures in social cognition research: Their meaning and use. Annual Review of Psychology, 54(1), 297-327.

25

Greenwald, A. G., & Leavitt, C. (1984). Audience involvement in advertising: Four levels. Journal of Consumer Research, 11, 581-592.

Grimes, A., & Kitchen, P. J. (2007). Researching mere exposure effects to advertising: Theoretical foundations and methodological implications.

International Journal of Market Research, 49(2), 191-219.

Hefner, D., Rothmund, T., Klimmt, C., & Gollwitzer, M. (2011). Implicit measures and media effects research: Challenges and opportunities. Communication

Methods and Measures, 5(3), 181-202.

Isingrini, M., Vazou, F., & Leroy, P. (1995). Dissociation of implicit and explicit memory tests: Effect of age and divided attention on category exemplar generation and cued recall. Memory & Cognition, 23(4), 462-467.

Jeong, S. H., & Hwang, Y. (2012). Does multitasking increase or decrease persuasion? Effects of multitasking on comprehension and counterarguing.

Journal of Communication, 62(4), 571-587.

Jennings, J. M., & Jacoby, L. L. (1993). Automatic versus intentional uses of memory: aging, attention, and control. Psychology and Aging, 8(2), 283.

Judd, T. (2013). Making sense of multitasking: Key behaviours. Computers &

Education, 63, 358-367.

Kardes, F. R. (1986). Effects of initial product judgments on subsequent memory-based judgments. Journal of Consumer Research, 13(1), 1-11.

Kent, R. J., & Allen, C. T. (1994). Competitive interference effects in consumer memory for advertising: The role of brand familiarity. Journal of

Marketing, 58(3).

Lang, A. (2000). The limited capacity model of mediated message processing.

Journal of Communication, 50(1), 46-70.

Lozito, J. P., & Mulligan, N. W. (2010). Exploring the role of attention during implicit memory retrieval. Journal of Memory and Language, 63(3), 387-399. McClard, A., & Somers, P. (2000, April). Unleashed: Web tablet integration into the

home. In Proceedings of the SIGCHI conference on Human factors in

computing systems (pp. 1-8). ACM.

McWilliam, G. (1997). Low involvement brands: is the brand manager to blame?. Marketing Intelligence & Planning, 15(2), 60-70.

26

Minear, M., Brasher, F., McCurdy, M., Lewis, J., & Younggren, A. (2013). Working memory, fluid intelligence, and impulsiveness in heavy media multitaskers. Psychonomic Bulletin & Review, 20(6), 1274-1281

Mulligan, N. W. (1998). The role of attention during encoding in implicit and explicit memory. Journal of Experimental Psychology: Learning, Memory, and

Cognition, 24(1), 27.

Mulligan, N. W. (2002). Attention and perceptual implicit memory: effects of selective versus divided attention and number of visual objects. Psychological

Research, 66(3), 157-165.

Mulligan, N. W., & Hartman, M. (1996). Divided attention and indirect memory tests. Memory & Cognition, 24(4), 453-465.

Nielsen. (2010). Americans Using TV and Internet Together 35% More Than a Year

Ago. Retrieved on June 21, 2014 from:

http://www.nielsen.com/us/en/newswire/2010/three-screen-report-q409.html Nielsen. (2012). State of the Media: The Cross-Platform Report. Retrieved on April 3,

http://www.nielsen.com/content/dam/corporate/us/en/reports-2014 from:

downloads/2012-Reports/Nielsen-Cross-Platform-Report-Q2-2012-final.pdf Papper, R. A., Holmes, M. E., & Popovich, M. N. (2004). Middletown media

studies. International Digital Media and Arts Association Journal, 1(1), 4-56.

http://idmaa.org/wp-Retrieved on April 3, 2014 from:

content/uploads/2013/07/V1N1-Middletown-Media-Studies.pdf

Parkin, A. J., & Russo, R. (1990). Implicit and explicit memory and the automatid effortful distinction. European Journal of Cognitive Psychology, 2(1), 71-80. Percy, L., & Rossiter, J. R. (1992). A model of brand awareness and brand attitude

advertising strategies. Psychology & Marketing, 9(4), 263-274.

Petty, R. E., & Cacioppo, J. T. (1984). The effects of involvement on responses to argument quantity and quality: Central and peripheral routes to persuasion.

Journal of Personality and Social Psychology, 46(1), 69.

Petty, R. E., Gleicher, F., & Baker, S. M. (1991). Multiple roles for affect in persuasion. In J. P. Forgas (Ed.), Emotion & Social Judgments (pp. 181-200). Oxford: Pergamon Press.

Pieters, R., Warlop, L., & Wedel, M. (2002). Breaking through the clutter: Benefits of advertisement originality and familiarity for brand attention and memory. Management Science, 48(6), 765-781.

27

Russo, R., & Parkin, A. J. (1993). Age differences in implicit memory: More apparent than real. Memory & Cognition, 21(1), 73-80.

Shapiro, S., & Krishnan, H. S. (2001). Memory-based measures for assessing advertising effects: a comparison of explicit and implicit memory effects. Journal of Advertising, 30(3), 1-13.

Smith, E. R., & DeCoster, J. (2000). Dual-process models in social and cognitive psychology: Conceptual integration and links to underlying memory systems. Personality and Social Psychology Review, 4(2), 108-131.

Spataro, P., Cestari, V., & Rossi-Arnaud, C. (2011). The relationship between divided attention and implicit memory: A meta-analysis. Acta Psychologica,136(3), 329-339.

Spataro, P., Mulligan, N. W., & Rossi-Arnaud, C. (2013). Divided attention can enhance memory encoding: The attentional boost effect in implicit memory.

Journal of Experimental Psychology: Learning, Memory, and Cognition, 39(4),

1223.

Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8(3), 220-247.

Srivastava, J. (2013). Media multitasking performance: Role of message relevance and formatting cues in online environments. Computers in Human

Behavior, 29(3), 888-895.

Voorveld, H. A. (2011). Media multitasking and the effectiveness of combining online and radio advertising. Computers in Human Behavior, 27(6), 2200-2206. Voorveld, H. A., & van der Goot, M. (2013). Age differences in media multitasking:

A diary study. Journal of Broadcasting & Electronic Media, 57(3), 392-408. Wolters, G., & Prinsen, A. (1997). Full versus divided attention and implicit memory

performance. Memory & Cognition, 25(6), 764-771.

Yang, M., & Roskos‐Ewoldsen, D. R. (2007). The effectiveness of brand placements in the movies: Levels of placements, explicit and implicit memory, and brand‐ choice behavior. Journal of Communication, 57(3), 469-489.

Yang, M., Roskos-Ewoldsen, D. R., Dinu, L., & Arpan, L. M. (2006). The effectiveness of" in-game" advertising: Comparing college students' explicit and implicit memory for brand names. Journal of Advertising, 35(4), 143-152. Yeu, M., Yoon, H. S., Taylor, C. R., & Lee, D. H. (2013). Are banner advertisements

28

Yoo, C. Y. (2007). Implicit memory measures for web advertising effectiveness.

Journalism & Mass Communication Quarterly, 84(1), 7-23.

Zaichkowsky, J. L. (1985). Measuring the involvement construct. Journal of

Consumer Research, 12(3), 341.

Zhang, W., Jeong, S. H., & Fishbein, M. (2010). Situational factors competing for attention. Journal of Media Psychology: Theories, Methods, and Applications, 22(1), 2-13.

29

Appendix 1: Pre-test Questionnaire

Figure 1. Online questionnaire for material selection

Repeated for 28 ads

30

Appendix 2: Pre-test Results

Table 1

Ascending Ordering of Television Ads According to Average Ad and Brand Attitude

Ad attitude Brand attitude

M SD M SD Plomari 1.98 1.02 Ion 1.54 0.49 Allatini 2.18 0.94 Pavlidis 1.68 0.74 Nikas 2.19 1.01 Altis 1.88 0.93 Manna 2.21 0.98 Papadopoulou 1.93 0.83 Ion 2.23 1.40 Allatini 1.98 0.77 Altis 2.34 0.92 Manna 2.01 1.04 Melissa 2.37 1.09 Melissa 2.06 0.95 Epirus 2.56 1.42 Plomari 2.09 1.22 Fage 2.62 1.63 Loumidis 2.20 0.98 Pavlidis 2.68 1.34 Avra 2.29 1.07 Ifantis 2.88 1.77 Epirus 2.31 0.97 Jotis 2.89 1.78 Fage 2.38 1.16 Avra 2.91 1.39 Attiki 2.46 1.05 Nounou 3.03 0.98 Jotis 2.48 0.92 Attiki 3.25 1.39 Nikas 2.50 1.06 Alfa 3.48 1.71 Nounou 2.62 1.20 Agno 3.60 1.35 Alfa 2.91 1.11 Masoutis 3.72 1.30 Masoutis 3.20 1.03 Loumidis 3.75 1.63 Ifantis 3.35 1.45 Papadopoulou 5.09 1.50 Agno 3.48 1.38