1 The effect of meaning operation on consumer responses
To what extent do varying degrees of meaning operation richness have an effect on cognitive elaboration, ad attitude and product perception?
Koen de Bruijn ,10618341 Master’s Thesis
Graduate School of Communication
Master’s programme Communication Science
Ester de Waal
2 Abstract
Visual metaphors have been used for decades in advertising and over the years their use has grown. The typology of McQuarrie and Phillips (2004) states that visual metaphors are divided into two parts: complexity and ambiguity. The focus of this study lies on the ambiguity
dimensions of visual metaphors, also called meaning operation (three levels, increasing ambiguity: connection, similarity, opposition), and the effect on cognitive elaboration,
advertisement attitude and product perception. A convenience sample of 154 subjects partook in a three (connection vs. similarity vs. opposition) between participants design. The results show some similar patterns that were found in former literature. For cognitive elaboration an
increasing linear trend was observed: more ambiguity, more elaboration. For ad attitude an inverted u-curve was observed, where similarity scored higher than opposition but not higher than connection. Product perception was not affected by any level of meaning operation: it did not increase when ambiguity increased. For this variable an inverted u-curve was also observed. Gender seemed to influence the effect of meaning operation on product perception and brand familiarity seemed to influence the effect of meaning operation on cognitive elaboration and ad attitude.
Keywords: visual metaphor, advertising, meaning operation, richness, typology, weak implicatures
Introduction
“A picture is worth a thousand words”. A very well known adage, that through the years has gained its credibility. It refers to making a complex issue easily understandable in just a single picture/visual. Something that cannot be achieved in a verbal manner. In advertising it is often important to make the message easily understandable because you want people to comprehend what you would like to say in the most effective way possible. Often people do not have the time
3 nor the motivation to process a print ad, television commercial or internet banner. Using a
picture could be just the way to go in these situations. When pictures are used in ads they are often referred to as visual rhetoric (Scott, 1994; McQuarrie & Mick, 1999). Visual rhetoric could be described as the way images communicate, more precisely how they communicate meaning. To characterize a visual image as visual rhetoric (a way of communication like advertising) three criteria have to be checked off: the visual image has to be symbolic, be created by human intervention and has to be created to communicate to an audience (Poss, 2005). A traffic sign can be perceived as visual rhetoric. Its colors are symbolic for stopping or going, they are made by human intervention (these colors are not naturally connected with the actions they
communicate, so they are manipulated by human interaction) and it is targeted at an audience to communicate something (stop/go). This is a very basic idea of what visual rhetoric is.
Use of visual rhetoric in advertising has become more prevalent in advertising (Leiss, Kline & Jhally, 1986; Pollay, 1986; Philips & Mcquarrie, 2004; Leigh, 1994; Andrews, 2011; Madupu, Sen & Than, 2013) and especially visual metaphors have been increasingly used in advertising over the years (Enschot, Beckers & Mulkers, 2010; McQuarrie & Mick, 2003; , Phillips & McQuarrie, 2009; Kim et al., 2012). An explanation could be that people need less time to process and comprehend an ad containing primarily visual image(s) than an
advertisement that contains primarily verbal anchoring (copy text) (MacInnis, Moorman & Jaworski, 1991), which makes it more effective in communicating its intended message in these days of advertising clutter. Chang and Yen (2013) even call metaphors “the very heart of the modern communication form in advertising”. Internationally, in culturally alike societies, visual metaphorical advertising can be implemented widely because there is no language barrier, which gives it another advantage over verbal advertising. Evidence suggests a superior persuasive effect of visual metaphors over non-visual metaphors and verbal-anchored visual metaphors, by causing more positive consumer responses (Hitchon, 1997; Jeong, 2008; Kim et
4 al., 2012); McQuarrie & Mick, 1999,2003,2004,2005; Madapu et al., 2013; Mothersbaugh,
Hughman & George, 2002; Tom & Eves, 2012; Torn, 1999).
It is not just interesting to study the differences between non-visual metaphorical ads and visual metaphorical ads, but also between the visual metaphorical ads themselves. It was
mentioned by McQuarrie & Phillips (2004) that between visual metaphors there were also observed differences in how consumers responded to them. It was observed that some visual metaphors were more open to interpretation than others. Therefore it would be great to know how these differences translate to various consumer responses and how they aid to the persuasiveness of the different visual metaphors.
To effectively study VM’s researchers needed to come up with some kind of typology to discriminate between different types of metaphors by classifying them on certain characteristics. Rather soon two characteristics were distinguished by McQuarrie & Phillips (2004): complexity and ambiguity. Where other studies focused primarily on complexity, this study will focus on the ambiguity dimension of a visual metaphor because the working mechanisms behind these different levels are not yet completely revealed. Holding the complexity dimension constant, the opposite of what Madapu et al. (2013) did in their study. Ambiguity simply put, is how open to, or how rich in interpretations a visual metaphor is. In McQuarrie & Phillips (2004) typology
ambiguity is called meaning operation. Meaning operation has three levels, from not too many interpretations (connection) to moderate interpretations (similarity) to a lot of interpretations (opposition) caused by the VM ad (Mcquarrie & Phillips,2004; Madapu et al., 2013). These different levels of meaning operation could have consequences for how consumers respond to the advertisement.
It is hypothesized in earlier research that the richer the meaning operation gets, the more mental activity is provoked (McQuarrie & Phillips, 2004,2005). A lot of elaborative processing has been thought to be positively related to the persuasiveness of a message (Petty, Cacioppo
5 & Schumann, 1983). It is important to know whether cognitive elaboration increases when
meaning operation increases in richness of interpretations, like suggested by McQuarrie and Phillips (2004), or that the elaborative processing stops when people are exposed to meaning operations that pose too many interpretations. Persuasiveness could be influenced by the amount of cognitive elaboration (Petty, Cacioppo & Schumann, 1983) so it is important to find out to what extent meaning operation influences elaboration, because it could be that more evoked elaboration causes more acceptation of the claims the ad makes.
For an advertiser it would be preferable that consumers evaluated your advertisements positively. Although it is known that visual metaphorical ads are often evaluated more positively than non visual metaphorical ads (McQuarrie & Mick, 2003; Kim et al., 2012), it would be helpful to know which visual metaphor (varying in levels of meaning operation) should be the most positively evaluated (more positive ad attitude). In former literature there is some support for an increasing positive evaluation when the richness of meaning operation also increases
(McQuarrie & Phillips, 2004, 2005), but there is also some support for an inverted u-curve assuming that to a visual metaphor too rich in meaning operation causes a less positive
evaluation of than a moderate rich meaning operation (Lagerwerf, Meijer, 2008; van Hooft, van Mulken & Nederigt, 2011, 2013).
Next to a positive evaluation of the ad itself, it is important to know how perceptions of the product itself are influenced by varying degrees of meaning operation. Advertisers try to give you a certain perception of their products when they advertise them (product is reliable, cool, exciting, high status etc.). Often it is one of their main purpose to give consumers positive perceptions of the product. Richer meaning operations cause more mental activity and “may lead to a product of audience’s own construction of meaning” (Proctor, Proctor & Papasolomou, 2005). So meaning operation seems to influence product perception, although it is not clear how. Too much interpretations could for example cause diluted perceptions of the product, but a lot of
6 interpretations could also be beneficial because of the accumulation of positive perceptions (McQuarrie & Phillips, 2004).
RQ: To what extent do varying degrees of meaning operation richness have an effect on cognitive elaboration, ad attitude and product perception?
Theory
Visual rhetoric: visual metaphors in advertising
When defining what a rhetoric figure really is, the definition of McQuarrie and Mick (1996) is the most used in literature concerning this subject: an expression that in a artful manner diverges from what you would expect, but not in a way one would perceive it as wrong or mistaken. This study will be focusing on visual rhetoric and more specific on the visual metaphor in advertising. Visual metaphors compare two dissimilar things and make it look like they are similar (Sopory & Dillard, 2006). Lakoff and Johnson (1980) describe metaphors as “understanding and
experiencing one kind of thing in terms of another”. Sopory and Dillard (2002) describe a metaphor by saying it is a comparison between two unrelated objects where a certain aspect of one object is transported to another. Lakoff and Johnson (1999), describe metaphors as
combining two uncommon things and let the audience (consumers in advertising) infer what the two things have in common. Metaphors are often defined as figures that cause people to compare dissimilar things, as that one thing is seen similar to the other. Visual metaphors work just the same, only they picture a metaphor instead of verbally present it. A few advertising examples where visual metaphors are used: comparing a fabric softener with a teddy bear (implying softness), comparing toilet paper with a puppy (implying softness), pairing a cleaning liquid with a grenade (explosive stain remover).
7 Characteristics of visual metaphors: ambiguity and difficulty
By straying of the expected path (comparing things that are normally not directly related to each other), the visual metaphor has been known to cause a tension-release process (Jeong, 2008). A visual metaphor does something unexpected in the ad which causes tension. Because the consumer wants to get rid of this tension he/she tries to resolve the ‘puzzle’ (the visual metaphor) by trying to find a relationship between the two objects. The consumer is hereby doing two things: he is trying to resolve the incongruity between the objects in the ad and he is trying to infer and comprehend the meaning of the VM. By doing so (finding the relationship between the two and resolving the incongruity) the tension is released (inconsistency resolved) giving the consumer a pleasant feeling (Jeong, 2008). Visual metaphors can vary in their ambiguity (what is it trying to say) and how complex they are (McQuarrie & Phillips, 2004; Gzioukepas & Hogg, 2011; Forceville, 2006) .
It is assumed, that making a too difficult and ambiguous puzzle the tension will be too high and the release will have a greater chance of not coming at all. At some point the
complexity of the puzzle does not cause pleasure, but other negative feelings as suggested by McQuarrie and Phillips (2004). This process can be seen as a inverted u-curve (as mentioned before in the introduction), were the less complex puzzle (VM) does not cause too much tension and the release is in proportion with this tension (not many positive feelings evoked). But when the puzzle (VM) gets more difficult it creates more tension and when solved more release and thus more positive effects/feelings. Finally the puzzle (VM) gets too difficult and is hard to solve, or it cannot be solved by the consumer (not comprehended) which causes no tension relieve and thus negative feelings (Berlyne, 1971).
Although it is clear that VM’s are an effective tool in advertising, a long time it was not clear if there were differences between them, or more fundamentally if there were different types
8 of VM’s and if so how they could be classified. Although it was already mentioned that VM’s differed in dimensions of ambiguity and complexity, just recently classifications were made.
Visual metaphor typologies
Because of the increasing use of pictures in advertising and the increasing interest in visuals in advertising, the need to come up with some kind of theory/classification to categorize the most well-known visual rhetoric: different types of visual metaphors and their effects (depending on what kind of VM is used) was growing. The typology that is chosen in this study is that of Phillips and McQuarrie (2004). This typology has been chosen, because it is a clear typology that is easily measurable. Other typologies that were proposed in other articles do classify visual metaphors the same way, but just in different formulations.
Following the typology of Phillips and McQuarrie (2004), VM’s vary in two ways: their visual structure and their operation of meaning. Visual structure means how the object
(dissimilar from the product) and subject (product) are placed together in the advertisement. This can be either side-by-side (juxtaposition), object and subject are combined and form a new object (fusion) and only one object that represents the subject that is replaced by the object (replacement). With an increase in complexity from juxtaposition (less complex), fusion (more complex) and finally replacement (most complex). When complexity increases more cognitive resources are needed to process the VM, as suggested by McQuarrie and Phillips (2004). The second dimension is called operation of meaning: how the object and subject conceptually relate to each other and communicate meaning. Here are also three types: the object and subject are associated with each other (connection), the object is like the subject (similar) and the object is not like the subject (opposite). These dimensions vary in how ambiguous or rich they are: how many different interpretations or inferences are caused depending on the degree of
9 hanging on a washing rack with a wolf howling to it can be seen as an VM ad with the meaning operation connection. The shirt is associated with the moon, because it really white. An ad for VAIO laptops as described before can be seen as an ad with the meaning operation similarity. The VAIO laptops are seen as similar as paper (same weight). Another ad for fabric softener that pictures socks (one of normal socks, the other pair of socks like a cactus) is an example of opposition: dissimilar.
Figure 1 illustrates how the typology looks like.
Visual structure Operation of meaning Richness Less → more Complexity Less ↓ More
Connection Similarity Opposition
Juxtaposition
Fusion
Replacement
Figure 1: Visual metaphor typology by: McQuarrie & Phillips (2004)
Looking back the previous section it can be seen that earlier research already mentioned these two dimension (richness and complexity, before called ambiguity and complexity). With this classification it has become easier to empirically test different types of visual metaphors and their hypothesized outcomes (Lagerwerf & Meijer, 2012; Madupu et al., 2013). The crossed squares are the VM’s that are empirically tested in this study.
10 Richness of meaning operation
Visual metaphors often relate objects to each other in an unusual way. By doing this the consumer is invited to think about the relationship between the objects (how are they related, what is the advertiser trying to say?). VM’s can vary in how much thought is given to the VM’s. This is conceptualized in the richness of meaning operation. Richness is “the number of alternative responses allowed for by the various meaning operations indicated in the picture” (Lagerwerf & Meijer, 2008). Richness of meaning operation is, as defined previously, the varying degree of alternative consumer responses (interpretations, inferences) caused by the VM advertisement. This inferring is caused by the need to understand how the objects depicted in the VM are related to each other.
In former literature these alternative consumer response are called “implicatures” or more specific “weak implicatures” (Serber & Wilson, 1986; Forceville, 1996; Phillips, 1997; McQuarrie & Phillips, 2004,2005; Lagerwerf & Meijers, 2008). The richer a visual metaphor is, the more weak implicatures are made, as stated by former literature (McQuarrie & Phillips, 2005; Lagerwerf & Meijer, 2005). When the consumer is trying to comprehend what the advertiser is trying to say with the VM in the advertisement they generate various weak implicatures
(inferences) (McQuarrie & Phillips, 2005).
Philips and McQuarrie (2004) formulate three types varying in richness: connection, similarity and opposition. In this three types two major operations of meaning are present: association and comparison. Connection concerns association, a metaphor associates two objects with each other. For example when a hamburger is pictured together with a slim model, the association with beauty can be facilitated. Similarity and opposition both work with
comparison, similarity compares objects to cause the perception that the objects are similar and opposition compares objects to cause the perception that the objects are different. Similarity compares two objects in the metaphor and then makes it look that one is similar like the other.
11 An ad of VAIO (laptops) pictures a laptop besides floating paper, comparing the papers to the laptop: suggesting that the laptop is as light-weighted as the papers. Opposition compares things as well but makes people see that the objects in the metaphor are not similar but the opposite of each other. An ad for liquid detergent softener can depict two pictures of socks, the left picture contains socks that look like a cactus and the right picture contains socks that look very soft. The product is placed in the right picture, making the opposition clear: when you use this softener you won’t get socks like cacti.
In the framework of McQuarrie and Phillips (2004) connection is seen as the least rich (the one which is the least ‘rich’ when it comes to how many inferences and interpretations are caused by the VM when consumers try to understand how the objects in the VM relate to each other). They (McQuarrie & Phillips, 2004) argue that a visual metaphor that associates objects together causes less weak implicatures because “” how is X associated with Y?” only has a few answers and “how are X and Y similar/different?” causes more responses. Between the two comparisons (similarity and opposition) it is argued by McQuarrie and Phillips (2004) that opposition is richer than similarity because people have to think why the objects in the VM are similar and after that, how they are different. This could cause even more responses for this last type of meaning operation and is thus called the richest meaning operation of all three
(McQuarrie & Phillips, 2004).
Meaning operation and the effect on cognitive elaboration
This classification system could be very useful to test various forms of visual metaphors and their effects on various consumer responses. It was already mentioned that VM’s tend to encourage comparisons between two things and in this way are encouraged to look how they are similar (Stern, 1990). When there is not a quite straightforward association between the two objects, consumers are going to make multiple inferences about the VM (Sawyer & Howard,
12 1991). The ambiguity (VM can mean several things) is found to cause this (McQuarrie & Phillips 2005). Consumers get motivated to unravel the meaning of the VM (how the objects are
associated or similar/opposite from each other and what they try to say with this) and it is hypothesized that this causes them to cognitively elaborate more intensely on the ad to generate more weak implicatures (McQuarrie & Phillips, 2004).
In the literature cognitive elaboration on visual metaphors is defined as how intense people engage in a message containing a VM, how thoroughly they process an ad with VM (how much “cognitive resources” are spend on a message (Mick & McQuarrie, 1993). Cognitive elaboration “reflects the extent to which information in working memory is integrated with prior knowledge structures” (MacInnis & Price 1987). More thoughts (weak implicatures) devoted to the advertisement in this study means a higher cognitive elaboration. Generally increased cognitive elaboration is caused because VM’s are “artful deviations” (McQuarrie & Mick, 1996), or in other words incongruent figures where people need more time to look at (Heckler & Childers (1992).
As the meaning operation of VM’s tends to get richer, it is hypothesized that the amount of weak implicatures tends to increase as well. Increasing implicitness of a visual argument (here the meaning operation dimension of a VM) encourages consumers to elaborate longer and more intensely on this ad and to process this centrally (Bulmer & Buchanan-Oliver, 2004; Stern, 1989). Chang and Yen (2013) found that implicit VM ads (where the visual metaphor was more rich in inferences and interpretations, : http://www.coloribus.com/adsarchive/prints/ariel-liquid-washing-detergent-shark-3888905/) generated more cognitive elaboration than explicit VM ads (where the visual metaphor was less rich, more straightforward:
http://www.adsora.com/electronics/sony-vaio-ads). Also, Lagerwerf & Meijer (2008) found that richer visual metaphors evoked more “elaborative thoughts”. Based on former literature it could
13 be hypothesized that VM’s rich in interpretations are positively related to more cognitive
elaboration (McQuarrie & Phillips, 2004, 2005).
H1: Visual metaphor advertisements with the meaning operation opposition will cause more cognitive elaboration than similarity and visual metaphor advertisements with the meaning operation similarity will cause more cognitive elaboration than connection.
Meaning operation and the effect on ad attitude and product perception
Quite a few studies have found empirical evidence for the persuasive power of visual metaphor advertising over no visual metaphor advertising (Messaris, 1997; Torn, 1999; McQuarrie & Phillips, 2004; Jeong, 2008). The persuasiveness of advertisements including VM’s can be probably best explained by the Elaboration Likelihood Model of Petty and Cacioppo (1986). This model states that there is a relationship between how much thought is spend on an message (advertisement), the persuasive power of this message and attitude change/formation of this message. Two types of processing are identified in the model. Although visuals are often seen as cues that motivate peripheral processing (superficially processing of a message, because you are just looking at a picture)(Chang & Yen, 2012), visual metaphors make a strong argument by inviting consumers to make inferences about the visual metaphor themselves, thus encouraging and motivating consumers to engage in central processing of the message (Jeong, 2008).
VM’s hold the attention of consumers and force them to solve the ambiguous figure (Sperber & Wilson, 1986). When finding out what the VM is trying to say, consumers will feel rewarded for their effort and this causes more appreciation for the advertisement (Tanaka, 1992). The richer the VM, the more weak implicatures (inferences, interpretations) are evoked (McQuarrie & Philips, 2004,2005). More weak implicatures are thought to stimulate multiple interpretations and when resolved cause more pleasure and appreciation (Lagerwerf & Meijers, 2008; Perrachio & Meyers-Levy, 1994). The pleasure of the processing the VM will cause the
14 consumers to like the ad more (McQuarrie & Mick, 1999).
Researchers have proposed that a too rich VM (too much weak implicatures) has a negative effect. It is thought that pleasure and appreciation of the ad decreases when too many interpretations are possible (Booth, 1961; Lagerwerf, 2002; Lagerwerf & Meijers, 2008; Phillips, 2000). When the VM gets too ambiguous (lots of different interpretations) the resolving of the puzzle stays out. There are too many thoughts and interpretations generated about the VM and it gets unclear what the meaning of the VM is. This means the release (as previously mentioned) is absent and which in turn causes more negative consumer responses, like attitude towards the ad (unsolved incongruity) (Mick, 1992).
An inverted U-curve, as earlier mentioned, is suggested (Lagerwerf & Meijer, 2008; Van Hooft, Van Mulken & Nederigt, 2011, 2013). The inverted U-curve states that stimuli that are too weak or too strong are less effective than moderate ones (Yerkes & Dodson, 1908), which predicts that a moderate degree of weak implicatures could be the most effective way to influence attitude towards the ad. Following the reasoning of McQuarrie and Phillips (2004/2005), visual metaphors that are moderate rich evoke a moderate degree of weak implicatures. That is why it is hypothesized that a moderate rich VM (meaning operation: similarity) has the most positive influence on attitude towards the ad.
H2: Visual metaphor advertisements with the operation of meaning similarity will more positively influence attitude towards the ad than the other two VM ads with the operation of meaning connection and opposition.
Earlier research has found that pictures can influence perceptions (Miniard, Bhatla, Lord & Dickson, 1991). An old but relevant study of Mitchell and Olson (1981) found that visual rhetoric in advertising could cause consumers to make inferences about attributes of the product that was being advertised. Jeong (2008) also found that product beliefs were most positively
15 influenced when VM’s were used in advertisement in comparison with advertisements without VM’s. Visual metaphors that visualized how smooth a car was, were effective in making the consumer perceive the car as smooth (Jeong, 2008). McQuarrie and Mick (1992) found that consumers would rather seek positive meanings in ads with visual rhetoric than negative
meanings. McQuarrie and Phillips (2005) found that VM’s indeed caused consumers to generate more positive inferences about the product when they were exposed to ads with VM’S and also accept the implicit claims (weak implicatures) faster than reject them. More specifically, the claims the ads wanted to make with the VM’s were also inferred by the participants in their study (McQuarrie & Phillips, 2005) and easier accepted than rejected. Thereby creating positive perceptions of the advertised products.
People have learned how to respond to VM’s in advertising over the years (McQuarrie & Phillips, 2005), probably because VM are widely used in advertising and people got used to them and therefore learnt how to respond to them (Chang & Yen, 2013; Phillips & McQuarrie, 2009). Because of this they can be guided to infer what the VM really tries to say (O’Donohoe, 2001; McQuarrie & Phillips, 2004). These self generated inferences (or claims) are more accessible and less challenged by the same consumers (Lee & Olshavsky, 1995). This is why VM’s were often inferred by the participants in their study and were easier accepted.
It was already hypothesized that increasing richness of a visual metaphor causes more positive inferences. It was also found that these (the weak implicatures) are easier accepted and were often positive (McQuarrie & Phillips, 2005). Kim et al. (2012) found that more thoughts evoked by VM’s caused more positive ad perceptions. It is thought that consumers do not just generate thoughts about the ad but also about the product advertised (Jeong, 2008). We could support this thought with the balance theory (Heider, 1946). When the ad causes more positive ad perceptions, people feel they have to have positive perceptions about the advertised product as well to correct the balance. Although this sounds fair, it is not that odd to think of a similar
16 inverted u-curve concerning product perception. It would not be directly related to negative thoughts (because VM in general evoke more positive responses than negative ones (McQuarrie & Mick, 1999), but rather to having too many interpretations about the ad and perceptions about the product. With a meaning operation too rich in interpretation, the consumer could develop too many perceptions about the product causing the consumer to be confused of what the advertiser is trying to say with the advertisement (competing perceptions). Although it is not yet completely clear, this study has earlier literature as starting point which states that richer meaning
operations cause more positive perceptions (McQuarrie & Phillips, 2005). Opposition should be the most rich meaning operation and therefore cause the most positive product perceptions. Thus, it is hypothesized that positive product perceptions increase when richness increases. So that the least rich meaning operation (connection) causes the least positive product perceptions and the most rich meaning operation (opposition) causes the most positive product perceptions.
H3: Visual metaphor advertisements with the operation of meaning opposition will be more positively influencing product perception than similarity and the meaning operation similarity will be more positively influencing connection.
Method
Participants
Participants were approached through social media (Facebook), via e-mail and through face-to-face contact. The respondents participated voluntarily in the study. The sample of respondents was a convenience sample and consisted primary out of students. Respondents were randomy divided over three conditions. 170 participants (N=170) participated in this study. 154 of the 170 (91%) respondents completed the study completely and only the data of these participants has been used in this study. Having 50 participants per cell should make this sample powerful enough to make any predictions. Of the respondents 63,6% was female and 36.4% was male.
17 The average age was 23,75 (µ=23.6, SD= 5.61). The majority of the participants were originally from the Netherlands (89.6%).Highest accomplished education of the participants was : 2.6% high school or equivalent, 0.6% Vocational/technical school, 13% some college, 31.2% college graduate, 50.6% (finished bachelor HBO/University), master’s degree, 0.6% doctoral degree and 0.6% other education.
Design
A 3(connection vs. similarity vs. opposition) between participants design was used in the research. The dependent variables were cognitive elaboration, ad attitude and product perception.
Stimulus
Research evidence suggests that visual arguments may be more persuasive when there are no verbal explanations for the visual argument (Jeong, 2008). This is why the stimuli have as little verbal copy as possible (only the brand name). Also, this has been done to make sure that no other influences other than the picture have an effect on the consumers responses measured in this study. Print ads for existing brand products were chosen to increase external validity of the results (Thorson, 1990). Print ads were chosen because of the non-time compression character, meaning that respondents can spend as much or as little time looking at the ad an processing the ad (MacInnis, Moorman & Jaworski, 1991). Finally the print ads were not altered in anyway.
In this study visual structure has been held constant. This means one form of visual structure has been chosen to appear in every visual metaphor next to the three dimensions of meaning operation (visual structure: juxtaposition and meaning operations: connection, similarity and opposition). The visual structure that was chosen to be held constant was juxtaposition (two objects next to each other). This structure has been used in all the three stimuli. Madapu et al. (2013) found that this was the most positively evaluated visual structure and using this structure
18 help find the most effective visual metaphor (juxtaposition is found to be the most effective, this study adds the three levels of meaning operation and the results will hopefully show which combination has the most positive impact on consumers). The same product was used in all three advertisements. This was done to hold the effect of product type constant. To reduce the effect of brand associations on the outcome variables, washing detergent was used as the advertised product. It is assumed that a majority of people do not have strong brand associations with washing detergent brands. The print ads were found on the site:
http://adsoftheworld.com. Table 1 in appendix D describes the three different visual metaphors varying in meaning operation and holding visual structure constant.
Procedure
At the beginning of the survey participants were told the study was about print advertising and they were told how the study was build up: the first part a print ad and the second part a
questionnaire. They were also told how long participating would take (5-10 minutes) and that all the results would be kept anonymous. After this came an informed consent section to which they had to agree if they wanted to proceed. The next part was one of the three visual metaphor print ads, containing three different degrees of meaning operation. Participants were randomly assigned to one of the three VM’s: Juxtaposition-connection, juxtaposition-similarity or
juxtaposition-opposition. 51 participants were assigned to the juxtaposition-connection group (N = 51), 53 participants were assigned to the juxtaposition-similarity group (N = 53) and 50
participants were assigned to the juxtaposition-opposition group (N = 50).
After the manipulation/stimulus respondents answered questions that measured cognitive elaboration, ad attitude and product perception. Then came a question that asked if participants already were familiar with the brand advertised: “Before this study, were you already familiar with the brand displayed in the advertisement?”, with the answer categories: yes/no.
19 After this some demographical questions were asked: sex, age, highest education and country of birth. Finally respondents were thanked for their participation and this was the end of the online survey.
Measures
Cognitive elaboration
The first dependent variable in the survey was cognitive elaboration. Cognitive elaboration in this study is defined as how much thought or “cognitive resources” are used to process a message, in this study the ad which contains the VM. This variable was measured on a scale of Jeong (2008). The scale contained questions that asked to what extent respondents agreed with the statements: ‘‘I had many thoughts in response to the advertisement’’ and ‘‘The advertisement elicited lots of thinking’’. The answers were measured on a 7-points likert scale, (1) strongly disagree – (7) strongly agree. The scale (M = 4.4, SD = 1.48) had a cronbach’s alpha of α =.90. No items were removed from the scale.
Advertisement attitude
The second dependent variable in the survey was advertisement attitude. This variable was measured with “four sementic differential items” by MacKenzie, Lutz and Belch (1986):
bad/good, unfavorable/favorable, unpleasant/pleasant and unappealing/appealing. The answers were measured on a 7-points scale. The scale (M = 4.65, SD = 1.28) had a cronbach’s alpa of α =.90. No items were removed from the scale.
Product perception
The last dependent variable in the survey was product perception. This variable was measured on a modified scale of Keller (1997) to measure product perception. Three items measured to
20 what extent respondents perceived the product advertised as reliable, exciting and cool. “To what extent do you think the product advertised in the ad is reliable/exciting/cool?” The answers were measured on a 7-point scale, (1) none at all – (7) a lot. The scale (M = 4.38, SD = 1.49) had a cronbach’s alpha of α =.74. When looking if the scale could be improved it was noticed that when removing the item that measured the reliable item, the scale’s reliability increased from α =.74 to α =.91. The item “reliable” was thus removed from the scale.
Results
Main analysis
To assume homogeneity of covariance matrices the Box’s test was conducted. The Box’s Test was found non-significant, M = 12.44, F(12, 109,916) = 1.01, p>.05 (p= .439). This means homogeneity of covariance matrices can be assumed.
Using Pillai’s trace , there was a main significant effect of meaning operation richness on cognitive elaboration, ad attitude and product perception, V = 0.13, F(6, 296) = 3.42, p <.001. This means that the meaning operation richness of the visual metaphor print advertisement had an effect on the three outcome variables. The effect size was 0.065, = .065. This means 6.5% of the variance in the outcome variables was caused by the independent variable.
Brand familiarity was included in the model, because it was suspected that familiarity might have an effect on elaboration, by means of earlier associations which could stop the respondent from generating a lot of thoughts. After conducting bivariate correlations between brand familiarity and cognitive elaboration, it was found that brand familiarity correlated weak significantly with cognitive elaboration: r = .19, p = .016. First a main univariate one way ANOVA was done without this variable and the effect of meaning operation on cognitive elaboration was found significant, F(2, 151) = 5.34, p < .01, = .066. After doing the main univariate one way ANOVA with brand familiarity included, it became clear that the effect on cognitive elaboration
21 had a tendency towards significance , F(2, 151) = 2.33, p = .10 (p>.0.5), = .030.
Because ad attitude also correlated moderately significant with brand familiarity, r = .31, p<.001, two univariate one-way ANOVAs (included in the MANOVA) were done. The results of the univariate one-way ANOVA found that the main effect of meaning operation richness on ad attitude was significant, F(2, 151) = 13.42, p <.001, = .152. After including brand familiarity in the model, the main effect was still significant, F(2, 151) = 5.49, p <.01, = .068.
Proctor et al. (2005) stated that women’s product -and ad perceptions would be affected differently than those of men. Gender correlated weak negatively significant with product
perception, r = -.18, p = .021 (p<.05). Gender was coded male (1) and female (2). “Increase” in gender would cause the product perception score to go down. Product perception would be more negative for females than males. This is why gender was included in the model. The results of the first univariate one-way ANOVA (gender excluded) found that the main effect of meaning operation richness on product perception was not significant, F(2, 151) = 2.36, p = .098 (p>.05). The second univariate one-way ANOVA that was part of the MANCOVA found that the main effect of meaning operation richness on product perception was significant F(2, 151) = 2.91, p = .05 (p ≤ .05)
As the results show, significant univariate main effects were found for ad attitude, F(2, 149) = 5.75, p <.001, = .072 ,and product perception, F(2, 151) = 3.07, p = .05 (p≤.05), = .04, but not for cognitive elaboration, F(2, 151) = 2.34, p = .10 (p>.0.5), = .031.
The assumption of homogeneity of variance was tested next. For cognitive elaboration, variances were equal for all the three meaning operations, F(2,151) = 0.59, ns (p = .556). For product perception, variances were equal for all the three meaning operations, F(2, 151) = 2.26, ns (p = .107).The assumption for homogeneity for ad attitude across the three meaning
operations was violated, F(2, 151) = 4.12, p = .02 (p<.05). To double-check this outcome, Hartley’s Fmax was calculated (Pearson & Hartley, 1954). This is the variance ratio between the
22 dependent variable with the most variance and the least variance. Both the highest variance and the lowest variance were squared and the highest squared variance was then divided by the lowest squared variance (1.059901176² / 0.7330357² = 1.97). This outcome (Hartleys Fmax) was 1.97. When looking at “the critical values of Hartley’s maximum F” (Joglekar, 2003) it is seen that for three groups that each contain 30-60 respondents the critical value is 2.40. Because 1.97 is lower than this number, there is no need to worry about the unequal variance of ad attitude over the three meaning operations
Hypotheses testing
The first hypothesis stated that respondents would elaborate the most on the richest meaning operation (opposition). Juxtaposition-opposition had a higher elaboration score (M = 0.34, SD = 0.18) than juxtaposition-similarity (M = -0.11, SD = 0.15) and juxtaposition-similarity higher than juxtaposition-connection (M = -0.23, SD = 0.15).
To test if the hypothesis could be partly accepted, a post-hoc test was done (bonferroni test) to do multiple comparisons. The scores of the hypothesized least rich meaning operation and the hypothesized most rich meaning operation could still differ significantly. After the Bonferroni test it was found that there was no significant difference in the means between connection (hypothesized least rich meaning operation) and
juxtaposition-opposition (hypothesized most rich meaning operation) (Mdifference = 0.57, p = .098) nor was there
a significant difference in the means between juxtaposition-similarity (hypothesized moderate rich meaning operation) and juxtaposition-opposition (Mdifference = 0.45, p = .27). Hypothesis 1 is
thereby not accepted, although there exists a tendency towards the hypothesized direction as graph 1 illustrates.
23 Graph 1
The second hypothesis stated that similarity would produce the highest ad attitude of all the meaning operations. Similarity had a higher ad attitude score (M = -0.39, SD = 0.14) than connection (M = -0.01, SD = 0.14) and juxtaposition-opposition (M = -0.41, SD = 0.17). To test if juxtaposition-similarity causes a higher ad attitude than juxtaposition-connection and
juxtaposition-opposition a post-hoc test (Bonferroni test) was conducted to generate multiple comparisons. After the bonferroni test it was found that there was no significant difference in the means between juxtaposition-similarity and juxtaposition-connection (Mdifference = 0.40 , p = .093) ,
but there was a significant difference in the means between juxtaposition-similarity and juxtaposition-opposition (Mdifference = 0.80, p < .001). Hypothesis 2 is thereby partly accepted,
juxtaposition-similarity causes a higher ad attitude than juxtaposition-opposition but there is no significant difference between juxtaposition-similarity and opposition. This effect on ad attitude is illustrated in graph 2.
24 Graph 2
The third hypothesis stated that product perception would be the most positively influenced by the meaning operation opposition. Juxtapositionopposition had a higher perception score (M = -0.12, SD = 0.18) than juxtaposition-connection (M = -0.17, SD = 0.15) but not higher than juxtaposition-similarity (M = 0.28, SD = 0.15). A post-hoc comparison test (bonferroni test) was done to test the hypothesis in more depth and to gain more understanding of how product perception is affected by meaning operation. There was no significant difference in the means between opposition and similarity (Mdifference = -0.41, p = .37) nor was there a significant
difference in the means between opposition and connection (Mdifference = 0.05, p = 1.00).
Hypothesis 3 is thereby not accepted, opposition does not influence product perception more positively than similarity or connection. The effect on product perception is illustrated in graph 3.
25 Graph 3
Discussion & Conclusion
Summary of the results
In the introduction it was asked “To what extent do varying degrees of meaning operation richness have an effect on cognitive elaboration, ad attitude and product perception?” As main purpose this study tried to answer how different levels of meaning operation, varying from least rich to most rich (according to the typology of McQuarrie and Phillips, 2004), influenced cognitive elaboration, ad attitude and product perception. The first hypothesis stated that cognitive
elaboration scores would be the highest for opposition. This meaning operation did not however, elicit more thinking and people exposed to this meaning operation did not have more thoughts than people who were exposed to the other two meaning operations. This non-significant effect of meaning operation on cognitive elaboration could be explained by how familiar the respondent was with the brand. This control variable correlated highly with cognitive elaboration. When not
26 included in the model cognitive elaboration scores were significantly higher for opposition than for the other two meaning operations. Earlier research found that prior knowledge could have a negative effect on the impact of “metaphorical frames” (Barker, 2005). Although using existing brands is recommended (Madapu et al., 2013; Thorson, 1990), some scholars warn for the effect of established associations with the brands that could dilute the effectiveness of a visual metaphorical advertisement (Chang & Yen, 2013). It could be suggested that because people are already familiar with the brand advertised they are generating less thoughts about the advertisement. This could be because they already know the brand and thus not think about it that much, even though the most rich meaning operation is used. It may be wise for companies to use visual metaphorical ads for newer products/brands instead to use VM’s for established brands.
Although the hypothesis could not be accepted, the meaning operation most rich in interpretations and thought eliciting did cause more cognitive elaboration than the other two meaning operations, a common trend that also was observed in other studies. When looking at graph 1 we see an increasing line with connection causing the least cognitive elaboration, similarity causing more elaboration and finally opposition causing the most elaboration (the difference between similarity and opposition is under the .10 significance level). Although not significant, these results are still in line with the studies of Chang and Yen (2013) and Lagerwerf & Meijer (2008). Chang and Yen (2013) found that implicit VM ads generated more elaboration than explicit advertisements. Lagerwerf and Meijer (2008) found that rich visual metaphorical advertising evoked more thinking than straightforward advertising. Alternatively Madapu et al. (2013) studied the effect visual structure has on cognitive elaboration and they found different results. More complex structures did not cause more cognitive elaboration there, but surprisingly the proposed least complex structure elicited the most thinking. Although visual structure seems to have a different effect on elaboration, there could be a quite logical explanation for this. Visual
27 structure differs quite a bit from meaning operation in that it is characterized by complexity and not by richness. These are two different things, by which complexity could be influenced by different factors than richness. Complexity is thought to be influenced by need for cognition (Chang & Yen, 2013; McQuarrie & Phillips, 2004) for example whereby richness is more influenced by for example visual processing style (McQuarrie & Phillips, 2004). Because
complexity is something that challenges people’s mental capacity, and richness does this not so much the differences in effects can be understood.
Hypothesis 2 predicted an inverted U-curve. Ad attitude would be higher for similarity than for opposition and connection. Although not completely significant, this study observed a similar U-curve which was also proposed in earlier literature. Ad attitude is affected by meaning operation which causes a weak inverted U-curve, whereby similarity is significantly more
positively influenced than opposition but not more than connection. However, the difference between similarity and connection had a tendency towards significance (significance level was lower than 0.10). Thereby the hypothesis was partly accepted. Although partly accepted, these findings could be potential evidence for the hypothesized inverted U-curve. Which is suggested by some former literature as well (Lagerwerf & Meijer, 2008; Van Hooft, et al, 2011, 2013). Although these articles often talk about an inverted u-curve for complexity levels it is interesting to see that this (as predicted) works the same way for richness. This contradicts McQuarrie & Phillip’s (2004) assumption that the more rich a VM is the more positive the ad attitude is. Which is an interesting finding, because by finding this result it contributes to the typology of McQuarrie & Phillips (2004) by shedding more light on how meaning operation affects outcomes like ad attitude.
The results found in this study were also in line with the research of Lagerwerf and Meijer (2012) where it was found that dissimilar objects that represent a visual metaphor negatively influence ad attitude. Opposition is characterized as making a dissimilar comparison (the two
28 things are not like each other and with this communicate a meaning) and the ad attitude was the lowest of all three the meaning operations. Although hypothesis 2 is only partly accepted, the results of this study point in the same direction as other literature. An inverted u-curve was found with a significant difference in evaluation of the print ad between opposition and similarity
whereby similarity was most positively evaluated.
Meaning operation seemed not to have an significant impact on product perception. The last hypothesis stated that opposition would have caused the most positive product perceptions (cool, exciting). This was however not the case and the hypothesis is therefore not accepted. Similarity however was the meaning operation with the highest product perception scores. Respondents found this product most exciting and cool. Similarity also differed marginal
significant from connection which could mean that there exists a real difference between the two in positive product perceptions.
An inverted u-curve can be observed in graph 3. This is interesting because it contradicts what was predicted in hypothesis 3. In the theory section it was predicted that product
perception would be affected in the same way as cognitive elaboration: an increasing line for increasing richness. Although not predicted, this does not come as a total surprise. In this study product perception was not measured in quantity but with the emphasis on valence. This could somehow have similarities with ad attitude which emphasizes evaluation and which also contains a valence dimension. Another explanation that was partly given in the theory section was that of perception dilution. This means that there were too many interpretations generated by seeing the most rich meaning operation, therefore causing too many perceptions of the product and as a consequence consumers would be confused of how to perceive the product. This could cause people to perceive the product as confusing, causing more negative feelings. Gender seemed to influence the effect of meaning operation on product perception as well. Proctor et al. (2005) already stated in their study that women have different perceptions of visual
29 metaphorical ads and the products advertised than men. This study supports this finding. Also, it is found that women had more negative product perceptions than men.
An increasing line in cognitive elaboration is observed, although no significant differences were found. Brand familiarity contributed to this non-significant result, it seems that familiarity caused people to think less about the ad even when it was advertised in a metaphorical manner. Product perception was also not significantly affected by meaning operation. Contrary to the prediction done in the theory section there was no increasing line observed, but an inverted u-curve instead. Connection and opposition did not seem to cause much positive product perceptions and did not differ much from each other. Similarity caused the highest positive product perceptions of the three.
To answer the question to what extent meaning operation has an impact on the three outcomes, the following can be said. Meaning operation indeed has some effect on the three outcomes albeit not significant (but often has a tendency towards significance). Ad attitude is significantly more positively influenced by similarity than by opposition. Product perception seems to be not affected at all and cognitive elaboration seems to increase by increase in richness but also not significantly. There are however common patterns that occurred in other studies as well which can give an indication of that what we have hypothesized is in the right direction. Only product perception did not go in the direction that was predicted.
Implications
An academic implication could be that similarity not just contributes to a positive ad attitude, but that this positive attitude could spill over to a positive attitude towards the brand (Kim et al., 2012). Using Heider’s balance theory (Heider, 1946), it can be argued that a positive attitude towards the ad can cause a positive brand attitude and positive advertiser attitude. To balance the positive attitude towards the ad, consumers feel urged to also develop a positive attitude
30 towards the brand and advertiser. As an practical implication this could mean people who
evaluate moderate rich VM ads positive, could also evaluate the brand and the advertiser more positively. This proposition is supported by Machleit and Wilson (1988) who state that a positive attitude towards the advertisement can transfer to positive attitudes towards the brand. So it can be recommended to practitioners in the field to use moderate rich visual metaphorical ads very rich ads. It can be recommended to not use meaning operation opposition when you want the ad to be evaluated positively or you want people to have positive perceptions of your product. Academically this study found similar patterns that occurred in former studies as well. This could be an indication for future researchers to go look in the same direction as this study did.
Gender seems to have an impact on product perception. Practitioners should know their target group before implementing visual metaphorical advertisements. It should recommended that women’s products should not be advertised with visual metaphors when the advertiser wants to emphasize certain positive product perceptions. Also, it is recommended to advertise new products/brands instead of established brands/products for more persuasive power of the advertisement.
Limitations & future research
Generalizing the research results should be done with caution. As earlier mentioned the online survey made use of a convenience sample with mostly young (average age was 23) students. It might be so that this particular demographic enjoys generating thoughts about ambiguous messages (like advertisements) and thereby influencing the outcome variables more positively than when the sample was taken from a broader audience. Students are often highly educated (more cognitive resources) which could make them more eager and able to process ambiguous messages and also to take more joy in doing this than the broader population (Jeong, 2008). Which could have an effect on cognitive elaboration, ad attitude and product perception.
31 The significant negative effect of opposition on ad attitude could also be one of an
aesthetic dimension. Some respondents made comments about the opposition ad, in which they thought there was blood illustrated in the print ad while in reality this was red paint. This could have influenced the results more negatively than was suspected and could contribute to the significant difference between opposition and similarity. Also in general, it is not sure if there were differences in aesthetic enjoyment between the ads. A suggestion could be (there is no evidence for this suggestion though), that the same colors should be used in the ads because colors can be perceived as an enjoyable characteristic of an advertisement. The product and the other things where it is compared to can also be placed the same way in all the three ads. With this study one time the two were placed under and above each other, or next to each other. Next time it is wise to put them all for example next to each other.
Choosing the ads that were used in the survey were chosen on the criteria proposed by McQuarrie & Phillips (2004), but the choice for these ads was still arbitrary. Although some ads were used in former research concerning visual metaphors, it is not completely clear if the ads differ on other characteristics besides the meaning operation richness.
There is another limitation concerning the product perception scale. This scale has been a modified scale, based upon Keller’s (1997) brand personality scale. Although it was tested reliable, there are still some doubts about this scale. Can brand personality be seen the same way as perception of a product, can brand personality be product personality as well?
Combining different visual structures with different meaning operations could be a future avenue. We now know some effects of all visual structures with similarity and all meaning operations with juxtaposition. It could be interesting for example to hold visual structure constant (fusion or replacement) and look how different meaning operations (connection, similarity,
opposition) interact with this visual structure. These effects can be compared to the effects found in this study and the study of Madapu et al. (2013) and by doing this more understanding can be
32 gained. Different moderators can also be added to the research model used in this study.
Moderators like visual processing style (Childers, Houston & Heckler, 1985; McQuarrie & Mick, 1999; McQuarrie & Phillips, 2004) and tolerance for ambiguity (McQuarrie & Phillips, 2004; van Hooft et al., 2013). People who often process things visual have a visual processing style which means they are easier in inferring about a visual (Childers et al., 1985). This could moderate the effect meaning operation has on outcome variables of this study.
Also, tolerance for ambiguity could be an interesting moderator to add to the model. When exposed to visual metaphors in advertisements, consumers are faced with an ambiguous message (van Hooft et al., 2013). Ambiguous messages can be seen as messages that are open to multiple interpretations. These weak implicatures are experienced by some people as more desirable than by others (McQuarrie & Phillips, 2004).This psychological personality trait is called tolerance for ambiguity in former literature (Budner, 1962; McLain, 2009; McQuarrie & Mick, 1992; McQuarrie & Phillips, 2004). Tolerance for ambiguity describes how people perceive ambiguous situations (Budner, 1962). There is some evidence that consumers intolerant for ambiguity perceive ambiguous ads as less positive and consumers that have a high tolerance for ambiguity perceive the same ads as more positive (van Hooft et al., 2013). Though not a lot of research has been done at this point yet. Some people like to avoid ambiguous messages (low tolerance for ambiguity) and others will like to resolve the ambiguity and take pleasure in doing so (high tolerance for ambiguity) (McQuarrie & Phillips, 1992).
The results of this study show that, although not significant, there is an observable pattern that is also found in other studies. It gives scholars and practitioners alike more insight in how visual metaphorical richness influences certain consumer responses. It gives some
interesting practical recommendations concerning which group of consumers to target and what kind of brand/product to use (new or established) in order to get the preferred consumer
33 for these patterns are given and suggestions for moderators are included. Although used
frequently by advertisers, there is still a lot to explore regarding this topic. By the end of this study, it is hopefully thought that with this contribution scholars can use the findings in this study to gain more understanding of visual metaphorical advertising and go further were this study stopped.
References
Ajzen, I. (1991). The theory of planned behavior. Organizational behavior and human decision processes, 50(2), 179-211.
Barker, D. C. (2005). Values, frames, and persuasion in presidential nomination campaigns. Political behavior, 27(4), 375-394.
Boozer, R. W., Wyld, D. C., & Grant, J. (1990). Using metaphor to create more effective sales messages. Journal of Services Marketing, 4(3), 63-71.
Budner, S. (1962). Intolerance of ambiguity as a personality variable. Journal of personality.
Chang, C. T., & Yen, C. T. (2013). Missing Ingredients in Metaphor Advertising: The Right Formula of Metaphor Type, Product Type, and Need for Cognition.Journal of Advertising, 42(1), 80-94.
Childers, T. L., Houston, M. J., & Heckler, S. E. (1985). Measurement of individual differences in visual versus verbal information processing. Journal of Consumer Research, 125-134.
34 Elliot, A. J., Maier, M. A., Binser, M. J., Friedman, R., & Pekrun, R. (2009). The effect of red on avoidance behavior in achievement contexts. Personality and Social Psychology Bulletin, 35(3), 365-375.
Enschot, R. V., Beckers, C., & Mulken, M. V. (2010). Rhetorical figures in TV commercials The occurrence of schemes and tropes and their effects on commercial likeability 1. Information Design Journal, 18(2), 138-147.
Forceville, C. (2002). Pictorial metaphor in advertising. Routledge.
Forceville, C. (2006). Non-verbal and multimodal metaphor in a cognitivist framework: Agendas for research. Applications of Cognitive Linguistics, 1, 379.
Gisbergen, M. S., Ketelaar, P. E., & Beentjes,H. (2004). Changes in advertising language: A content analysis of magazine advertisements in 1980 and 2000. Retrieved January 3, 2013, from: http://www.gisbergen.eu/Content.pdf
Gkiouzepas, L., & Hogg, M. K. (2011). Articulating a new framework for visual metaphors in advertising. Journal of Advertising, 40(1), 103-120.
Jeong, S. H. (2008). Visual metaphor in advertising: is the persuasive effect attributable to visual argumentation or metaphorical rhetoric?. Journal of Marketing Communications, 14(1), 59-73.
David, H.A., Hartley, H.O., & Pearson, E.S. (1954). The distribution of the ratio, in a singly
normal sample, of range to standard deviation. Biometrika, 482-493.
Haugtvedt, C.P., Petty, R.E., & Cacioppo, J.T. (1992). Needfor cognition and advertising: Understanding the role of personality variables in consumer behavior. Journal of Consumer
35 Psychology, 1(3), 239-260.
Heider, F. (1946). Attitudes and cognitive organization. The Journal of psychology, 21(1), 107-112.
Joglekar, A. M. (2003). Statistical methods for six sigma: in R&D and manufacturing. John Wiley
& Sons.
Kassin, S. M., Reddy, M. E., & Tulloch, W. F. (1990). Juror interpretations of ambiguous evidence. Law and Human Behavior, 14(1), 43-55.
Kim, J., Baek, Y., & Choi, Y. H. (2012). The Structural Effects of Metaphor-Elicited Cognitive and Affective Elaboration Levels on Attitude Toward the Ad.Journal of Advertising, 41(2), 77-96.
Lagerwerf, L. (2002). Deliberate ambiguity in slogans: recognition and appreciation. Document Design, 3(3), 244-260.
Lagerwerf, L., & Meijers, A. (2008). Openness in metaphorical and straightforward advertisements: Appreciation effects. Journal of Advertising,37(2), 19-30.
Lagerwerf, L., van Hooijdonk, C. M., & Korenberg, A. (2012). Processing visual rhetoric in advertisements: Interpretations determined by verbal anchoring and visual structure. Journal of Pragmatics, 44(13), 1836-1852.
Lee, J. G., & Thorson, E. (2008). The impact of celebrity-product incongruence on the effectiveness of product endorsement. Journal of Advertising Research, 48(3), 433-449.
Leigh, J. H. (1994). The use of figures of speech in print ad headlines. Journal of Advertising, 23(2), 17-33.
36 Leiss, W., Kline, S. and Jhally, S. (1986) Social Communication in Advertising: Persons,
Products, and Images of Well-Being. New York: Methuen.
MacInnis, D. J., Moorman, C., & Jaworski, B. J. (1991). Enhancing and measuring consumers' motivation, opportunity, and ability to process brand information from ads. The Journal of Marketing, 32-53.
MacInnis, D. J., & Price, L. L. (1987). The role of imagery in information processing: Review and extensions. Journal of consumer research.
MacKenzie, S. B., & Lutz, R. J. (1989). An empirical examination of the structural antecedents of attitude toward the ad in an advertising pretesting context. The Journal of M
Machleit, K. A., & Wilson, R. D. (1988). Emotional feelings and attitude toward the
advertisement: The roles of brand familarity and repetition. Journal of Advertising, 17(3), 27-35.
Madupu, V., Sen, S., & Than, S. R. (2013). The impact of visual structure complexity on ad liking, elaboration and comprehension.Marketing Management Journal, 23(2).
McGrath, R., Gunther, I., & MacMillan, I. (2005). Marketbusters: 40 strategic moves that drive exceptional business growth.
McQuarrie, E.F., & Mick, D.G. (1992). On resonance: a critical pluralistic inquiry into advertising rhetoric. Journal of Consumer Research, 180-197.
McQuarrie, E.F., & Mick, D.G. (1996). Figures of rhetoric advertising language. Journal of Consumer Research, 424-438.
37 McQuarrie, E.F., & Mick, D.G. (1999). Visual rhetoric in advertising: Text-interpretive,
experimental, and reader-response analyses. Journal of Consumer Research, 26(1), 37-54.
McQuarrie, E. F., & Mick, D. G. (2003). Visual and verbal rhetorical figures under directed processing versus incidental exposure to advertising. Journal of consumer research, 29(4), 579-587.
McQuarrie, E. F., & Mick, D. G. (2003). The contribution of semiotic and rhetorical perspectives to the explanation of visual persuasion in advertising.Persuasive imagery: A consumer response perspective, 191-221.
McQuarrie, E. F., & Phillips, B. J. (2005). Indirect persuasion in advertising: How consumers process metaphors presented in pictures and words. Journal of Advertising, 34(2), 7-20.
McQuarrie, E. F., & Phillips, B. J. (2008). It's not Your Magazine AD: Magnitude and Direction of Recent Changes in Advertising Style. Journal of Advertising, 37(3), 95-106.
Mitchell, A. A., & Olson, J. C. (1981). Are product attribute beliefs the only mediator of advertising effects on brand attitude?. Journal of Marketing Research, 318-332.
Mothersbaugh, D. L., Huhmann, B. A., & Franke, G. R. (2002). Combinatory and separative effects of rhetorical figures on consumers’ effort and focus in ad processing. Journal of Consumer Research, 28(4), 589-602.
O’Donohoe, S. (2001). Living with Ambivalence Attitudes to advertising in postmodern times. Marketing Theory, 1(1), 91-108.
Petty, R. E., Cacioppo, J. T., & Schumann, D. (1983). Central and peripheral routes to advertising effectiveness: The moderating role of involvement. Journal of consumer research, 10(2), 135.
38 Phillips, B. J., & McQuarrie, E. F. (2004). Beyond visual metaphor: A new typology of visual rhetoric in advertising. Marketing theory, 4(1-2), 113-136.
Phillips, B. J., & McQuarrie, E. F. (2009). Impact of advertising metaphor on consumer belief: delineating the contribution of comparison versus deviation factors. Journal of Advertising, 38(1), 49-62.
Phillips, B. J., & McQuarrie, E. F. (2014).Visual Rhetoric and International Advertising. The Handbook of International Advertising Research, 238.
Pollay, R. (1985) ‘The Subsidizing Sizzle: A Descriptive History of Print Advertising, 1900–1980’, Journal of Marketing 48 (Summer): 24–37.
Poss, S. K. (2005). Theory of visual rhetoric.
Proctor, T., Proctor, S., & Papasolomou, I. (2005). Visualizing the metaphor.Journal of Marketing Communications, 11(1), 55-72.
Putrevu, S., Tan, J., & Lord, K. R. (2004). Consumer responses to complex advertisements: The moderating role of need for cognition, knowledge, and gender. Journal of Current Issues & Research in Advertising, 26(1), 9-24.
Tom, G. (1999). The Use of Rhetorical Devices in Advertising. Journal of Advertising Research, 39(4), 39-41.
Tom, G., & Eves, A. (2012). The use of rhetorical devices in advertising. Cross Currents: Cultures, Communities, Technologies.
39 Törn, F. (2012). Revisiting the Match-Up Hypothesis: Effects of Brand-Incongruent Celebrity Endorsements. Journal of Current Issues & Research in Advertising, 33(1), 20-36.
Van Hooft, A.P.J.V., van Mulken, M.J.P., & Nederstigt, U. (2011). Visual metaphor in advertising: Do complexity and culture matter?
Van Hooft, A., van Mulken, M., & Nedertigt, U. (2013). Cultural Differences? Visual Metaphor in Advertising: Comprehension and Tolerance of Ambiguity in Four European Countries.
In Advances in Advertising Research (Vol. IV) (pp. 351-364). Springer Fachmedien Wiesbaden.
Van Mulken, M., Le Pair, R., & Forceville, C. (2010). The impact of perceived complexity, deviation and comprehension on the appreciation of visual metaphor in advertising across three European countries. Journal of Pragmatics, 42(12), 3418-3430.
Ward, J., & Gaidis, W. (1990). Metaphor In Promotional Communication: A Review of Research on Metaphor Comprehension and Quality. Advances in consumer research, 17(1).
Appendices
Appendix A
40 Appendix B
Juxtaposition - Similarity
41 Juxtaposition - Opposition
Appendix D
TABLE 1: Print ad stimuli
Advertised product Visual structure Meaning operation Description
Persil Juxtaposition Connection The washing detergent is shown to wash
clothing as white as the moon by showing a wolf howling to a almost blinding white shirt on a washing rack. The whitening power of Persil is associated with that of the moon.
