Effects of the perspective of visual stimuli on consumer
evaluations
Master’s Thesis
Maaike Reijlink
30 JANUARI 2015
GRADUATE SCHOOL OF COMMUNICATION – UNIVERSITY OF AMSTERDAM
Master’s programme Communication Science
Track: Persuasive Communication
Supervisor: dr. A.M. Wennekers
1
Abstract
Despite the research developments on mental simulation in the fields of psychology and neurosciences, the application of mental simulation in advertising is still in its infancy. Prior research already showed effects of small visual variations in advertisements on consumer evaluations. The current research extends these findings for reasons of generalizability and uncovering underlying processes. Therefore a between-subjects experiment was conducted whereby the visual orientation of a straw in a smoothie was manipulated, and effects on consumer evaluations were measured. Self-referencing, positive affect and embodied simulation were considered as underlying processes that mediate effects on consumer evaluations. Imagination ability and style of processing were considered as individual differences that could strengthen or weaken effects of the manipulation.
The main finding of this research is that the advertisement with the straw in the smoothie directed toward the consumer, positively affects the consumers’ attitude toward the product. Underlying effects have not been extracted by this research. However, self-referencing, positive affect and embodied simulation often have positive effects on consumer evaluations. A higher ability to imagine and a more visual style of processing do not strengthen or weaken effects of the manipulation. Therefore, conclusions in accordance to this research are:
evoking self-referencing, positive affect and embodied simulation in advertising can lead to positive consumer evaluations; and orientating a product toward the consumer would be advisable, but evidence for generalizations is still too scarce.
2
An Unconscious Influence in Advertising: Effects of the Perspective of Visual Stimuli on Consumer Evaluations
Most human communication is non-verbal. Marketers already use forms of non-verbal communication to reach the consumer in a more implicit way, such as the curves of a Coca
Cola bottle or the “plop”-sound of opening a Grolsch beer bottle. Also cues like color (green
for sustainability), sensory cues such as smell (freshly baked bread in the supermarket), or touch (puppies to refer to the softness of toilet paper) are cues referring to existing schemas in the brain, so that the consumer creates stronger associations with the product.
Implicit cues in advertising could be an answer on how to reach the consumer effectively, since explicit cues are not always effective for a cognitive overloaded mind (Petty &
Cacioppo, 1986). Implicit cues possibly result in consumers who unconsciously generate positive brand attitudes themselves, rather than positive brand attitudes verbally provided by advertisers (Krishna, 2012). Self-generated engagement like this, may be more persuasive than statements about the product provided by the advertiser (Sengupta & Gorn, 2002).
Implicit triggers can evoke mental simulations of the product or event (for example the image of you drinking a beer when hearing the “plop”-sound of a Grolsch beer bottle). To underline the strength of mental simulation, Barsalou (2008) argues that mental simulation may be enough to activate the body. There is evidence provided by neuro-imaging studies at mental simulations, whereby the processing of sensory perceptions activates corresponding neural areas in the brain. For example the taste cortices are activated by only seeing pictures of chocolate chip cookies (Rolls, 2005; Simmons, Martin & Barsalou, 2005). Also, imagining the playing of Beethoven leads to activation of the auditory cortex (Zatorre & Halpern, 2005). Furthermore, reading (not aloud) words associated with strong cinnamon or garlic smells activates the primary olfactory cortex (González et al., 2006).
Despite the research developments on mental simulation in the fields of psychology and neurosciences, the application of mental simulation in advertising is still in its infancy. With implicit cues in advertising, consumers could be able to encode the message unconsciously, rather than be provided by a message and actively need to encode the message. More research on mental simulation within the marketing field would therefore possibly lead to major developments in effectively reaching the consumer.
Especially interesting and relevant to the present research are small visual variations in advertising, that lead to different evaluations of the product or advertisement. Prior research showed that small variations appeal on different existing schemas in the brain, which lead to different associations. For example, a product is evaluated more positively when it is viewed from a camera angle looking up, than from a camera angle looking down (Meyers-Levy & Peracchio, 1992). Meyers-Levy and Peracchio (1992) suggest these associations are stored in our memory since early childhood. Products that we visually look up to (e.g. our parents
3 when we are children) are evaluated positively and on the contrary, products that we visually look down on (e.g. younger siblings) are evaluated negatively or subordinate.
Another example is that the orientation of a product toward the dominant hand of the respondent (a cup with the handle to the right or left) increases purchase intentions (Elder & Krishna, 2012). Furthermore, the absence of an attribute that facilitates mental simulation (a spoon) reduces the impact on purchase intentions (Elder & Krishna, 2012). These effects occur because products oriented toward the dominant hand make consumers easier process a message, even if they do not actually grab the product (Eelen et al., 2013).
These examples show that the first steps have been taken in research on implicit
processes and mental simulation. However, since the research topic is relatively new, only a few effects have been shown. The underlying processes of mental simulation remain for the biggest part unclear and need to be explored to truly understand how mental simulation works in the field of advertising. Exploring mental simulation in marketing will possibly lead to major contributions to this relatively new research field. Besides, when we know which implicit processes underlay mental simulation, grounded implications for marketers and advertisers could be made.
The current research will continue on the topic of visual orientation of an image,
subsequent to handedness and orientation of the attributes in an advertisement. However, the current research tries to seek effects that occur without handedness attributes, since handedness has practical limitations in excluding left-handed consumers and excluding products that do not have a handedness attribute. Either interesting is whether effects occur independently of handedness attributes, for example with a specific visual perspective toward the product. The visual attribute which will be examined, will be the straw in a smoothie that is positioned in two different directions: toward and opposed from the consumer. In addition, an advertisement with no straw present will be added to control for mere effects of the position of the straw. The current research will also look at underlying processes that could explain effects, or possible factors that could strengthen or weaken effects.
RQ: To what extent affects the orientation (toward or opposed or neutral) of visual stimuli in an advertisement, consumer evaluations?
Theoretical Framework Visual Depictions
The assumption that implicit stimuli could be more effective in reaching the consumer than explicit stimuli, can be explained by the Elaboration Likelihood Model of Petty and Cacioppo (1986): when the consumer has low elaboration likelihood to process a message, due to little
4 motivation, opportunity and/or ability, the processing of the message occurs via the heuristic route rather than the central route. The Elaboration Likelihood Model shows that attitudes are formed in a heuristic way, rather than a cognitive way, when a consumer does not have the motivation, opportunity or ability to process a message. Therefore, triggers which appeal on the unconscious mind could be critical influencers in a world where people are bombarded with advertisements and presumably be cognitively overloaded to process all this
information.
Visual depictions could be used to trigger the unconscious and thereby affect consumer evaluations of an advertisement or product. For example, Elder and Krishna (2012) showed that purchase intentions can be increased by facilitating mental simulation through specific visual depictions of an object. They showed this by orienting a product toward the dominant hand of the respondent (a cup with the handle to the right). They also showed that the absence of an attribute that facilitates mental simulation (a spoon) reduces the impact of a visual depiction on purchase intentions. Elder and Krishna (2012) mention that other
instruments which appeal on embodied cognition could be all products with handles (bottles, mugs, containers) or even hands that interact with the product. Eelen et al. (2013) showed that products oriented toward the dominant hand affect process fluency positively. In other words, respondents find it easier to process information when mental simulation is facilitated by the right visual depictions of an object.
Next to better process fluency as an result of the right visual depictions, image perspective also plays an important role in evaluations of the product. Meyers-Levy and Peracchio (1992) found that the perspective of the mental image based on a photograph, is often determined by the camera angle from which the photograph is taken. People tend to have positive experiences with looking up to things in the natural visual world (e.g. looking up to our parents when we are children) and this relates to looking up to things in visual
depictions such as advertisements. Therefore, evoking different perspectives in which the consumer imagines the product, could be used to implicitly influence consumer evaluations of a product or advertisement.
Next to perspective defined by camera angle, perspective by first or third person could also play a role in effects on consumer evaluations of an advertisement. However, evoking different perspectives in which the consumer imagines the product has not been researched yet by the solely use of visual depictions, but by descriptions combined with (mental) visual depictions. For example, Libby, Shaeffer and Eibach (2009) gave participants concrete or abstract descriptions of actions depicted in photographs. For each description, participants had to choose which of the photographs depicted the action better. The participants were more likely to choose first-person images when they received concrete as opposed to abstract descriptions.
5 Likewise, McIsaac and Eich (2002) manipulated the dimension of perspective in referring to the first or third person. They asked participants to remember the task they just performed from their own point of view or from an observer point of view. More information was recalled by participants about bodily sensations, psychological states and affective reactions when participants imagined events from the first-person perspective rather than the third-person perspective.
An explanation for the different evaluations when imagining an event from the first-person perspective or the third-person perspective, is a bottom-up and top-down style of making meaning (Libby & Eibach, 2011). Imagining an event from the first-person perspective involves a style of making meaning that is bottom-up. One connects the information to his or her own experiences stored in memory. People make meaning by incorporating information about the experience by concrete features of the picture. Imagining an event from the third-person perspective involves a style of making meaning that is top-down. People integrate the illustrated event with a broader context and define the event with abstract meanings. Libby and Eibach (2011) suggest that a distinction can be made between the “I” and “me”, wherein the “I’’ can be understood as experiential and emerging in a bottom-up fashion. The ‘’I’’ is evoked by concrete features and one’s actions on it. The ‘’me’’ is a representation of the self in a more abstract fashion.
In accordance of the previous theory and research, an appeal on the self by referring to the experiential “I” would lead to a mental simulation of experiences. Whereas prior research refers to the “I” by explicitly asking the participants to imagine the event from the perspective of the self, the current research seeks a way of evoking mental images of the self with visual depictions. Small variations in visual depictions have shown effects in prior research (a cup oriented toward the dominant hand, looking up or down to a product). Therefore, the current research seeks to discover whether an advertisement that is visually directed toward the self, will possibly lead to better evaluations of the ad, since the consumer is better capable to mentally experience the product. This leads to the following hypothesis:
H1: A visual stimulus directed toward the consumer affects consumer evaluations more positive than a visual stimulus opposed from the consumer.
Underlying Processes
We have seen that mental simulation often plays a role in effects of visual depictions on the consumer. When one sees an image, mental simulations of experiences with the product or event could be evoked. However, mental simulation is a very broad concept. Prior
research already showed examples of mental simulation in the form of embodiment (cup oriented to the dominant hand), self-referencing (imagining oneself using the product or
6 imagining the event from the experiential “I”), or positive affect (looking up to a product simulates prior experiences from earlier childhood: e.g. looking up to your parents). Though, a one-way definition of mental simulation is not available. Therefore, the current research tries to discover which underlying processes play a role in effects on consumer evaluations, under the umbrella of mental simulation. The concepts of embodied simulation,
self-referencing and positive affect are further being explained in the following sections. Findings from prior research lead to the following hypothesis:
H2: Effects on consumer evaluations are mediated by mental simulation.
Embodied simulation. Elder and Krishna (2012) suggest that we automatically activate
previous bodily experiences when we observe a product. This indicates that bodily experiences are in the basis of thinking, and also that thinking is contextual and occurs in interaction with the environment (Barsalou, 2008). An increasing number of studies have shown effects of embodied simulation on consumer behavior. It appears that consumers prefer products that are easy to grasp. For example, when consumers have the choice between two products with the handle oriented rightwards or leftwards, right-handers were more likely to pick up the product with the handle oriented rightwards (Ping, Dhillon &
Beilock, 2009). Furthermore, also in the absence of a physical environment the effects of
visual orientation appear. Advertisements that appeal on usage of the product with one’s dominant hand (e.g. a spoon on the right side of a bowl of yoghurt) affect purchase intentions positively compared to an appeal on the non-dominant hand (Elder & Krishna, 2012). The side that is dominant may be positive in the mind of a person, since simulations of actions on the dominant side tend to be more fluent than actions on the non-dominant side (Casasanto & Chrysikou, 2011; Eelen et al., 2013).
These findings are in agreement with Beilock and Holt (2007), who also found that fluent perceptual motor processes often lead to more positive evaluations and feelings. They tested evaluation on letter dyads for experienced typists and unexperienced typists. Beilock and Holt (2007) found that letter dyads resulted in sensorimotor simulations of typing the letters by experienced typists. Experienced typists preferred letter dyads that would not create motoric interference. Casasanto and Chrysikou (2011) explain these effects in terms of motor affordances: when people are presented to an object that would afford to act on it (typing letters or picking up a spoon), they mentally simulate the performance and their evaluations vary according to how fluent this action would be.
More generally, we represent our surroundings partly through motor simulations of how we might act in association of the products we encounter (Ellis & Tucker, 2000). Therefore, the present research tries to discover whether a straw toward the consumer is evaluated
7 more positive, than a straw opposed from the consumer or no straw. The assumption is that embodied simulation will occur more easily when the straw is positioned toward the
consumer, since we might act upon this visual attribute.
Self-referencing. Self-referencing is “a process inducing relative significance where a
consumer relates message information to his or her self-experience or expectation” (Burnkrant & Unnava, 1989, p.632). The distinction between self-referencing and other-referencing was first identified by Bone and Ellen (1992). They examined the use of self-related and other-self-related imagery in the processing of advertisements. Their results were that self-related images of advertisements were easier to produce, more vivid, and they had a positive effect on ad attitude. Burnkrant and Unnava (1995) also explored the positive effects of self-referencing in advertising. They found better memory of the advertisement and more positive attitudes toward the ad and the product in the ad.
Next to studies with self-referencing as an independent variable, Yoon and Park (2012) studied the effects of sensory appeals on consumers’ attitudes toward a brand, with self-referencing as a mediator. Sensory appeals are stimuli that appeal on the senses and can unconsciously evoke mental images of (associations of) the product (Yoon & Park, 2012). They found that preferred sensory appeals (smell for a coffee-brand) enhanced the
effectiveness of ads, mediated by self-referencing. This means that people spontaneously refer the product to themselves by imagining the smell of coffee.
However, not only appeals on the senses result in mental images. Dahl and Hoeffler (2004) suggest that consumers who encounter ads for a service or product might form image-based representations of the different aspects of the ad. In their study Dahl and Hoeffler (2004) showed that for familiar products, self-imagery led to higher evaluations of the product than other-imagery. Consumers were better able to mentally see themselves instead of the other using the product, when they already knew the product. This
self-imagery subsequently led to more positive evaluations than other-self-imagery. Dahl and Hoeffler (2004) speculate that visualization based on others should be less likely to include thoughts about how the product fits within existing personal using patterns.
As we can conclude from these studies, self-referencing plays an important positive role in the effects of mental simulation. Self-referencing mostly results in more positive effects than for example other-referencing. This could be explained by an appeal on existing schemas in the brain, when one refers mental images to him or herself. Mental images from a person that referred a product in an advertisement to him or herself, are easier produced and more vivid, as Bone and Ellen (1992) found. A person already has experiences with certain products, and therefore an image could be directly associated with the existing associations one has.
8
Positive affect. Positive affect refers to the extent to which an individual subjectively
experiences positive mood such as joy, interest, and alertness (Miller, 2011). In ads, both strength and preference of creative appeals influence affective reactions toward ads. If there is a match between the affective feature in the stimulus and existing schemata in the brain, the affective feature could cause a relevant affective reaction (Hoffman, 1986). Positive affect can occur without cognitive efforts if the stimulus appeals on existing schemata. This is a consistent view in perspective of classical condition and affective reaction (Yoon & Park, 2012). For example, if consumers watch a commercial of toothpaste with a fresh taste and they are highly involved in a fresh taste, they can easily place themselves into the ad and get emotionally assimilated with it. In other words, a positive relationship between the self and the brand is formed by mental simulation of themselves with the product. According to this theory, Yoon and Park (2012) found enhanced effectiveness of ads by sensory preferences, with positive affect as a mediator.
A second theory of the workings of positive affect is that it causes a broadening of cognitive processes in the brain. Hypothetically it means that positive affect makes it more likely for the mind to see the forest, whereas negative affect is making the mind more narrowly focused and it will only see the trees (Fredrickson, 2001). Mild positive affect, the sort that people can experience every day, facilitates recall of positive and neutral material and improves creativity in problem solving (Ashby, Isen & Turken, 1999).
However, the two theories of positive affect (1: affective features in stimuli cause relevant affective reactions and 2: positive affect as a mood state before processing the stimulus causes broader cognitive processing of the stimulus) are slightly different, in having different moments of working when a consumer processes an ad. For the current study, the first theory is most relevant, since the ad itself evokes positive affect. Consumers will be able to compare physical characteristics of the stimulus (direction) and the smoothie itself with existing schemata in the brain (previous experiences with similar products). This match can lead to a relevant affective reaction: mental simulation of an positive experience with the advertised product and could therefore have an effect on the dependent variables (consumer evaluations).
Individual Differences in Visualizing Skills
The current research seeks to discover whether the orientation of the straw affects consumer evaluations of the product and advertisement. The prior section illustrated what underlying processes could mediate effects of the stimuli on consumer evaluations (mental simulation). However, there are possible factors that could strengthen or weaken effects of the manipulated stimuli on consumer evaluations. The manipulated stimuli are visual aspects that appeal on imagination (straw directed toward or opposed from consumer). Therefore,
9 the question is whether individual differences in how one processes a message could
moderate effects of the manipulated stimuli on consumer evaluations. Prior research already made some steps in discovering which individual factors could play a role in this process. In accordance to prior research, the current research further examines imagination ability and style of processing as possible moderators of effects on consumer evaluations.
Imagination ability. Libby and Eibach (2011) found that the differences in referring to the
first or third person according to the visual point of view, were more pronounced among respondents with better visualization skills. Earlier, Marks (1973) discovered the existence of individual differences in how someone experiences visual imagery. He developed a self-report scale which he called the “Vividness of Visual Imagery Questionnaire” (Marks, 1973). The scale distinguishes persons with high and low ability to imagine. Individuals who have high imagination ability, experience clear and extensive visual imagery, whereas individuals with low imagination ability are less likely to experience vivid visual imagery. Contributing to Marks (1973), Fennis, Das and Fransen (2012) argue that individuals have a habitual, possibly innate, tendency to experience vivid visual mental simulations, whereby some individuals experience more vivid visual mental images than others. The theory of individual differences in imagination ability is adopted by multiple researchers (McKelvie, 1994; Richardson, 1995; McKelvie, 1998; Johnson & Puddifoot, 1998).
Those able to evoke vivid images are able to imagine events as if they were actually occurring. Schlosser (2003) found positive effects of high imagination ability on purchase intention. She researched the effects of object interactivity on purchase intentions: respondents got assistance in simulating events by virtually interacting (clicking, moving, commanding) with the object. Schlosser (2003) found that the occurring effects were mediated by mental imagery, whereas the ability to evoke mental images moderated the effects. Schlosser (2003) compared interactive objects with static objects and found moderating effects of imagery ability on purchase intentions in a study on different website designs. Respondents with higher capability of evoking vivid mental images showed higher purchase intentions.
H3a: Effects on consumer evaluations are moderated by imagination ability, whereby
respondents with higher imagination ability evaluate the advertisement and product more positive than respondents with lower imagination ability.
Style of processing. Next to the ability to imagine, individual differences in style of
processing can also play a role in how consumers process the information from an
advertisement. Visualizers may evaluate the advertisement more positive than verbalizers, since the advertisement appeals on mental images of visual stimuli.
10 Childers, Houston and Heckler (1985) note that previous studies have found significant differences in how individuals process information. In a critical review, Ernest (1977) extracts three facets of processing: imagery vividness (the clearness of the mental image), imagery control (self-generation of a mental image or performing of mental simulations such as mental rotation), and imagery style (willingness to engage in verbally or imaginal oriented processing). The facets he extracted can have a significant impact on memory, perception, learning, and problem solving. Especially the third facet, imagery style, is important to the current research. The manipulated stimuli appeal on visual mental images, therefore we can assume that persons with a more visual style of processing will evaluate the advertisement more positive.
There is little research available on style of processing in the field of advertising. However, individual differences in style of processing are widely researched in the field of
neurosciences. Miller, Donovan, Bennett, Aminoff, and Mayer (2012) found that individuals have a particular style of thinking or a preferred set of cognitive operations that affects their pattern of brain activity. The visualizer-verbalizer dimension of cognitive style is based on the idea that some people are better at processing visual stimuli, whereas other people are better at processing verbal stimuli. In their research, Miller et al. (2012) showed that
individual differences in encoding strategy and cognitive style during a memory retrieval task were significant factors in explaining the variability in different patterns of brain activity.
H3b: Effects on consumer evaluations are moderated by style of processing, whereby
respondents with a visual style of processing evaluate the advertisement and product more positive than respondents with a verbal style of processing.
Theoretical Model and Implicit Measures
Figure 1 shows the theoretical model of the current research. This research tries to explore effects of the manipulated advertisements on evaluations of the advertisement and product. Mental simulation is hypothesized as a mediator of the effects, whereas the individual differences imagination ability and style of processing are hypothesized as moderators of the effects.
11 Figure 1
Theoretical model
The current research is concerned with automatic, implicit and unconscious processes. Therefore, implicit measures could be of great value for research on processes which
consumers themselves are not consciously aware of (Gawronski & Payne, 2010). Consumer evaluations are expected to be different when one is exposed to a stimulus toward or
opposed from oneself, or when no stimulus is present, due to these unconscious processes. Therefore, the current research uses several implicit measures to test for effects. However, implicit measures are very often explorative, since the research field is relatively new. The next section will elaborate on which implicit tests were used to test for consumer evaluations.
Method
Participants. A diverse group of Dutch participants was recruited on Facebook, by e-mail,
and by personally contacting the immediate social circle of the researcher. A total of 116 participants (31 men and 85 women) completed the survey. The participants were between 18 and 80 years old (M = 30.76, SD = 13.75). Educational level varied from primary school to university. However, a larger part of the group was highly educated (WO = 60%, HBO = 29%, lower education = 11%). By participating in this research, a voucher from a Dutch supermarket, worth 20 euros, could be won by one participant.
Experimental design. An experimental between-subjects design was used to test the
hypotheses. In order of this between-subjects experiment, the survey software system automatically randomized the participants between the three advertisements. The three
12 manipulated advertisements (straw toward respondent, straw opposed from respondent, no straw present) were the independent variables. The mediating variable mental simulation was measured with items on self-referencing, embodied simulation and positive affect. Imagination ability and style of processing were measured as moderating variables. Finally, the dependent variable consumer evaluations was measured with items on ad attractiveness, perceived percentage of strawberries, perceived distance from the smoothie, perceived volume of the smoothie and purchase intention. These items will be further explained in the procedure section.
Stimuli. For this study, an image of a strawberry smoothie was used and three
advertisements were created (Appendix A). The image of the strawberry smoothie was centered in the middle and the brand logo was shown in the upper right corner. The advertisements were similar, except from the key manipulation. The manipulation was a straw that was positioned toward the respondent (advertisement 1) or opposed from the respondent (advertisement 2). The third advertisement showed the same smoothie without a straw (advertisement 3). This third advertisement was to control for effects that were due to the mere presence of the straw, instead of the direction of the straw (toward vs. opposed). An unfamiliar brand name was chosen: Pom Wonderful. This brand is only being sold in the USA and Canada, therefore it was expected to be unfamiliar for the Dutch participants. Furthermore, it was assumed that all the respondents would be familiar with the product category smoothies. The familiarity of the product is important for tests on mental simulation, whereby associations with existing schemata in the brain are evoked.
Procedure. The experiment was set in an online survey. After agreeing with the terms
and voluntary participation, the survey started. First, the respondents were explicitly told that they had to remember a 9-digits number before seeing the advertisement for 8 seconds. Participants were told not to write down the number. This way, the participants were cognitively distracted while seeing the advertisement (Eelen et al., 2013). This cognitive distraction method was used in order for the respondents to implicitly process the
advertisement. Implicitly processing of advertisements approaches the daily environment of the participants, where they see most of the advertisements implicitly, simply due to the lack of time to process all the advertisements they pass by during the day. In Eelen et al. (2013), the stimuli were exposed for 5 seconds. In this research, exposure time was 8 seconds, since the loading time of the picture in the browser was taken into account.
After being exposed to the advertisement, the participants had to write down the 9-digit number. A control question was asked, to check whether the advertisement had been fully visible to the respondents, or perhaps had not been fully loaded in the screen. Then, to measure the dependent variable consumer evaluations, an item on ad attractiveness followed. The participants had to indicate how attractive they considered the advertisement
13 on a visual analogue scale. This scale ranged from not attractive at all to very attractive (200 points) (Eelen et al., 2013). Only the slider, and not the scale numbers, was visible to the respondents.
The second implicit item to test for consumer evaluations was based on the perceived percentage of strawberries in the smoothie (Hammond’s error test: Hammond, 1948). The test involves response options that are indeterminate, but valued as positive or negative (further explained by Gawronski and Payne (2010), p. 146). An example shown by
Gawronski and Payne (2010), is: “President George W. Bush is known around the world for his (a) affable (b) bumbling demeanor.” Hammond’s error test in the current research involved a visual analogue scale on which the participants had to indicate what the
percentage of strawberries in the smoothie was. It was considered that a higher estimated percentage indicated a more positive evaluation of the advertisement than a lower estimated percentage of strawberries. Hence, this test implicitly extracts the attitude toward the product.
The next part of the survey contained items on the moderating variable imagination ability. Based on the Vividness of Visual Imagery Questionnaire of Marks (1973), 4 items on a 5-point scale were created. The reliability of the items was acceptable (Cronbach’s alpha = .66). Participants had to think of the front of a shop they often go to. Then they had to consider the picture that came before there mind’s eye and tick the box that was true for them: 1 (perfectly clear and vivid as normal vision) to 5 (no image at all, you only “know” that you are thinking of the object).
After the items on imagination ability, three items on a 4-point scale were created to measure the moderating variable style of processing. These items were based on previous research of Childers et al. (1985), in order to examine the degree of visual processing of the participants (Cronbach’s alpha = .77): (1) “I find it helps to think in terms of mental pictures when doing many things”, (2) “My thinking often consists mental pictures or images”, (3) “When I have forgotten something I frequently try to form a mental picture to remember it.”
Again, the advertisements were shown to clear up the mental images of the participants, so that they were able to fill in the items. This time, participants were able to click to the next page in their own pace. To further test for the dependent variable consumer evaluations, three items followed. The first implicit item was based on research of Balcetis and Dunning (2010). In their research, they found that participants estimated a desirable object as closer by, than a undesirable object. Therefore the participants in the current research were asked to note on a visual analogue scale from 0 to 100 cm, how far away they were from the smoothie if the situation was real. Similar to this item, Van Koningsbruggen, Stroebe and Aarts (2011) found that participants estimate an object as bigger, when they desire this object. Therefore the participants in this research were also asked to indicate on a visual analogue scale from 200 to 500 ml what the volume of the smoothie was, to implicitly test
14 how much the respondents desired the smoothie. The third item was an item on purchase intention (7-point Likert scale; not likely at all – very likely): “Imagine you are going to buy a strawberry smoothie in a shop, how likely would it be that you would buy a Pom Wonderful smoothie?”. Since implicit tests are still very explorative, it seemed wisely to include one widely used explicit measure.
Then, items that measured the mediating variables self-referencing, positive affect and embodied simulation followed. Three items on a 7-point Likert scale (strongly disagree – strongly agree) were created to examine self-referencing (Cronbach’s alpha = .82). The items were based on previous research of Debevec and Iyer (1988), Burnkrant and Unnava (1989), and Chang (2011): (1) “I related the advertisement to earlier personal experiences with smoothies”, (2) “I thought about drinking the smoothie while seeing the ad”, (3) “The advertisement seemed to relate to me personally”. Three items on a 5-point Likert scale (not at all – very strong) were created to examine positive affect (Cronbach’s alpha = .92). The items examined feelings of happiness, joy and cheerfulness. Two items based on previous research of Elder and Krishna (2012) were created to examine embodied simulation (Cronbach’s alpha = .85). The items were created on a 9-point scale (1 not at all – 9 to a great extent): (1) “As you viewed the ad, to what extent did images of drinking the smoothie come to mind (for example, picking it up, taking a sip, etc)?”, (2) “To what extent while viewing the ad could you imagine drinking the smoothie?”.
Finally, to test for the control variables, respondents had to fill in three items on age, education and sex. Then, respondents were thanked and they had completed the survey.
Results Design Checks
For some participants, the advertisement did not fully load in the screen (e.g. only half of the picture was shown; it took at least 5 seconds for the picture to load). Data from these participants was erased. Data was also erased from participants that did not fully complete the survey. This resulted in the deletion of data from 19 participants. Data from 116
participants was left for analyses. For one item on perceived volume of the smoothie, a participant noted a volume that was more than three standard deviations away from the mean (M = 287.73, SD = 47.35). Following the rule of thumb, this number was erased and replaced by ‘missing value’.
The three conditions (straw toward respondent, straw opposed from respondent, and no straw present) were equally divided with respect to gender (x2 (1, N = 116) = 2.71, p = .26), education (x2 (12, N = 116) = 17.04, p = .15), and age (F(2, 113) = .29, p = .75). Bivariate correlation tests were conducted to check for alternative explanations of effects. There were no significant correlations between the control variable education and the dependent
15 variables. However, significant correlations were found between gender and ad
attractiveness (rs = .19, n = 116, p = .046), and gender and perceived distance from smoothie
(rs = -.28, n = 115, p = .003). These correlations mean that girls perceive the advertisements
as more attractive than boys, and girls perceive the distance from the smoothie closer by than boys. No significant correlations were found between gender and perceived percentage of strawberries or purchase intention. The third control variable, age, showed significant correlations with ad attractiveness (r = -.23, p = .013), perceived volume of the smoothie (r = -.31, p = .001), and perceived percentage of strawberries (r = -.28, p = .003). Therefore, older people perceive the advertisements as less attractive than younger people, they perceive the smoothie smaller, and they perceive a lower percentage of strawberries in the smoothie. No significant correlations were found between age and perceived distance from the smoothie or purchase intention. In further analyses, there will be controlled for the variables that showed significant relations with the dependent variables.
Main Effects of the Manipulations on the Dependent Variables
Univariate analyses were conducted to test whether the manipulation in the advertisements affected consumer evaluations. The independent variable was the advertisement with the manipulation: one advertisement with the straw in the smoothie directed toward the respondent, one advertisement with the straw in the smoothie directed opposed from the respondent, and one advertisement without a straw in the smoothie. The dependent variables were: ad attractiveness, distance from smoothie, volume of smoothie, strawberry percentage and purchase intention. Means and standard deviations are shown in table 1.
No effect of the manipulation was found on ad attractiveness, with age and gender as control variables, F<1, n.s. Respondents found the advertisements in the three groups equally attractive. Furthermore, no effect of the manipulation was found on perceived distance from the smoothie, with gender as control variable, F<1, n.s. Therefore,
respondents perceived the distance from the smoothie similar in the three groups. No effect of the manipulation was found on perceived percentage of strawberries, with age as a control variable, F<1, n.s. Therefore, respondents estimated the percentage of strawberries in the smoothie similar in each group. Finally, no effect of the manipulation was found on purchase intentions, F<1, n.s. Therefore, respondents reported similar purchase intentions in each group.
A significant effect of the manipulation on the perceived volume of the smoothie was found, F(2, 109) = 4.33, p = .016. The group with the straw toward oneself estimated the volume of the smoothie as bigger (M = 302.84, SD = 54.47), than the group with the straw opposed from oneself (M = 283.25, SD = 43.11), or the group with no straw in the smoothie
16 (M = 277.19, SD = 40.91). A post hoc test (LSD) showed that the differences between the group with the straw toward oneself and the straw opposed from oneself were significant (p = .027). The differences between the group with the straw toward oneself and the group
without a straw were also significant (p = .006). The differences between the group with the straw opposed from oneself and without a straw were not significant (p = .55). Therefore, when the straw was positioned towards oneself, one estimated the smoothie as bigger than when the straw was positioned opposed from oneself or then when there was no straw present. The mere presence of the straw was not related to perceived volume of the smoothie, since there were no significant differences between the group with the straw opposed from oneself and the group without a straw.
Specifically, H1 was approved for effects on perceived volume of the smoothie. On the contrary, H1 was rejected for insignificant effects on the other dependent variables.
Therefore, the position of the straw does not cause any differences in perception of attractiveness of the advertisement, the estimated percentage of strawberries in the smoothie, the estimated distance between the respondent and the smoothie, or purchase intention of Pom Wonderful smoothies.
Table 1
M and SD of the groups on the dependent variables
Group Dependent variable M SD
Straw towards respondent Straw opposed from respondent No straw Ad attractiveness 108.62 108.88 104.31 54.09 41.61 50.67 Straw towards respondent
Straw opposed from respondent No straw Perceived percentage of strawberries 51.26 45.55 48.36 25.82 24.39 21.48 Straw towards respondent
Straw opposed from respondent No straw Perceived distance from smoothie (cm) 46.66 42.07 45.69 22.45 19.99 21.56 Straw towards respondent
Straw opposed from respondent No straw Perceived volume of smoothie (ml) 302.84 283.25 277.19 54.47 43.11 40.91 Straw towards respondent
Straw opposed from respondent No straw Purchase intention 3.36 3.78 3.50 1.61 1.64 1.54
17
Effects Explained by the Mediating Processes of Mental Simulation
Effects of the manipulations on the mediating variables. Multivariate analyses showed
that there were no effects of the manipulations (straw towards oneself, opposed from oneself, or no straw present) on the variables self-referencing F<1, n.s., positive affect F<1,
n.s., or embodied simulation F<1, n.s. Hence, mental simulation, defined by self-referencing,
positive affect and embodied simulation, did not mediate effects on consumer evaluations. Therefore H2 was rejected. The mean and standard deviations of the group scores on the mediating variables are shown in table 2.
Table 2
M and SD of the three groups on the mediating variables
Group Dependent variable M SD
Straw towards respondent Straw opposed from respondent No straw Self-referencing 3.62 3.69 3.53 1.74 1.43 1.47 Straw towards respondent
Straw opposed from respondent No straw Positive affect 1.91 2.20 2.10 .83 .93 .86 Straw towards respondent
Straw opposed from respondent No straw Embodied simulation 4.06 4.73 4.43 2.73 1.99 2.18
Effects of the mediating variables on the dependent variables.
Ad attractiveness. A multiple regression analysis was conducted with self-referencing, positive affect and embodied simulation as independent variables and ad attractiveness as dependent variable, with control for gender and age. Self-referencing, positive affect and embodied simulation explained a significant part of the variance in scores on ad
attractiveness, F(5, 110) = 9.55, p = .000. The strength of the prediction is strong: 55% of the variation in ad attractiveness can be predicted on the basis of self-referencing, positive affect and embodied simulation (R2 = .55). Self-referencing did not show to be a significant
predictor of ad attractiveness, b* = .16; t = 1.41, p = .16. Thus, there are no indications of a relation between self-referencing and ad attractiveness. Positive affect showed to be a significant predictor of ad attractiveness, b* = .26; t = 2.85, p = .005. Therefore, a higher score on positive affect is moderately related to perceiving the ad as more attractive.
18
t = 1.78, p = .079. Therefore, there are no indications of a relation between embodied
simulation and ad attractiveness.
Perceived percentage of strawberries. A multiple regression analysis was conducted to test for effects of self-referencing, positive affect and embodied simulation on perceived percentage of strawberries, with control for age. Self-referencing, positive affect and embodied simulation explained a significant part of the variance in scores on perceived percentage of strawberries, F(4, 109) = 5.63, p = .000. The strength of the prediction is moderate: 41% of the variation in perceived percentage of strawberries can be predicted on the basis of self-referencing, positive affect and embodied simulation (R2 = .41).
Self-referencing did not show to be a significant predictor of perceived percentage of
strawberries, b* = .09; t = .73, p = .46. Thus, there are no indications of a relation between self-referencing and perceived percentage of strawberries. Positive affect showed to be a significant predictor of perceived percentage of strawberries, b* = .28; t = 2.87, p = .005. Therefore, when respondents were positive affected after being exposed to the
advertisement, they perceived the percentage of strawberries as higher. Embodied
simulation did not show to be a significant predictor of perceived percentage of strawberries,
b* = -.04; t = -.34, p = .74. Thus, there are no indications of a relation between embodied
simulation and perceived percentage of strawberries.
Perceived distance from the smoothie. A multiple regression analysis was conducted to test for effects of self-referencing, positive affect and embodied simulation on perceived distance from the smoothie, with control for gender. Self-referencing, positive affect and embodied simulation explained a significant part of the variance in scores on perceived distance from the smoothie, F(4, 110) = 4.45, p = .002. The strength of the prediction is moderate: 37% of the variation in perceived distance from the smoothie can be predicted on the basis of self-referencing, positive affect and embodied simulation (R2 = .37).
Self-referencing showed to be a significant predictor of perceived distance from the smoothie, b* = .27; t = 2.15, p = .034. Therefore, when respondents referred the advertisement to
themselves, they perceived the smoothie as further away. Positive affect did not show to be a significant predictor or perceived distance from the smoothie, b* = -.07; t = -.68, p = .50. Thus, there are no indications of a relation between positive affect and perceived distance from the smoothie. Embodied simulation showed to be a significant predictor of perceived distance from the smoothie, b* = -.38; t = -3.05, p = .003. Therefore, when respondents reported higher scores on embodied simulation, they perceived the smoothie as closer by.
Perceived volume of the smoothie. A multiple regression analysis was conducted to test for effects of self-referencing, positive affect and embodied simulation on perceived volume of the smoothie, with control for age. Self-referencing, positive affect and embodied simulation explained a significant, but small part (13%) of the variance in scores on
19 perceived volume of the smoothie, R2 = .13, F(4, 108) = 4.13, p = .004. Self-referencing showed to be a moderately significant predictor of perceived volume of the smoothie, b* = .24; t = 1.95, p = .054. Therefore, when respondents referred the smoothie more to themselves, they perceived the volume of the smoothie as bigger. Positive affect did not show to be a significant predictor of perceived volume of the smoothie, b* = -.07; t = -.65, p = .52. Thus, there are no indications that positive affect and perceived volume of the smoothie are related. Embodied simulation also did not show to be a significant predictor of perceived volume of the smoothie, b* = -.06; t = -.51, p = .61. Therefore, there were no indications of a relation between embodied simulation and perceived volume of the smoothie.
Purchase intention. A multiple regression analysis was conducted to test for effects of self-referencing, positive affect and embodied simulation on purchase intention. Self-referencing, positive affect and embodied simulation explained a significant part of the variance in scores on purchase intention, F(3, 112) = 15.53, p = .000. The strength of the prediction is strong: 54% of the variation in purchase intention can be predicted on the basis of self-referencing, positive affect and embodied simulation (R2 = .54). Self-referencing showed to be a significant predictor of purchase intention, b* = .33; t = 2.98, p = .004. Thus, when respondents referred the advertisement more to themselves, they showed a higher intention to purchase the Pom Wonderful smoothie. Positive affect also showed to be a significant predictor of purchase intention, b* = .28; t = 3.07, p = .003. Therefore, when respondents were more positive affected after being exposed to the advertisement, they showed more intention to purchase the Pom Wonderful smoothie. Embodied simulation did not show to be a significant predictor of purchase intention, b* = .04; t = .36, p = .72.
Therefore, there were no indications of a relation between embodied simulation and purchase intention.
Correlations between the three mediating variables. To test whether self-referencing,
positive affect and embodied simulation could be placed under the umbrella of mental simulation, a multiple correlation test was conducted. Embodied simulation was significantly correlated with self-referencing, r = .68, p < .001. Positive affect also was significantly correlated with self-referencing, r = .43, p < .001. Similarly, embodied simulation was
significantly correlated with positive affect, r = .44, p < .001. Hence, the tests showed that the three variables were related to each other. This indicates that self-referencing, positive affect and embodied simulation could be placed under the umbrella of mental simulation.
Effects Explained by the Moderating Processes of Individual Differences
Imagination ability and style of processing were tested for being possible moderators of the effects of the manipulations in the advertisements on the dependent variables. In order to test for effects, Anova tests were conducted. To conduct the tests, the variables were divided
20 by a median split (imagination ability: M = 2.21, SD = .67, Mdn = 2.25; processing style: M = 2.77, SD = .66, Mdn = 2.67). For the purpose of clarity, the total amount of results of the Anova tests are shown in Appendix B, and significant results are explained in this section. The tests were conducted with control for age and/or gender, as explained in earlier sections.
Respondents with higher imagination ability scored higher on perceived ad attractiveness (M = 117.96, SD = 39.40), than respondents with lower imagination ability (M = 98.15, SD = 53.90). This effect was significant, F(2, 108) = 3.83, p = .053. Therefore, respondents with high ability to imagine perceive the POM Wonderful advertisements as more attractive. No other significant main effects of imagination ability on consumer evaluations were found. Neither were effects of style of processing on consumer evaluations found.
A significant interaction effect between imagination ability and the three groups on perceived percentage of strawberries was found, F(2, 107) = 3.36, p = .038. Respondents with low imagination ability who were exposed to the advertisement with the straw in the smoothie toward themselves, perceived the percentage of strawberries higher than the respondents with high imagination ability. This effect was reversed for respondents who were exposed to the smoothie with the straw opposed from themselves, or without a straw
(Appendix B). This finding was opposite from the expected effect. No other significant
interaction effects were found between imagination ability and the three groups on consumer evaluations, or between style of processing and the three groups on consumer evaluations. Therefore, H3a and H3b were rejected.
Discussion
This research continues on the topic of mental simulation in advertising, focused on the visual orientation of an image. Prior research already showed effects of visual orientation on consumer evaluations of products. Therefore, a start has been made on investigating the underlying processes of the effects of these implicit visual stimuli. However, the research topic is rather new, and therefore theory is still scarce or not grounded enough. The aim of this research was to contribute to theories on how to reach the consumer effectively, by seeking effects of the visual orientation of an object in an advertisement on consumer
evaluations. Whereas prior research focused on handedness attributes, the current research focuses on the perspective from how the object is seen: toward the consumer or opposed from the consumer. Specifically, the manipulation in the current research was a straw in a smoothie that points in a specific direction.
The main effect shown by the experiment, was that when a straw is positioned toward oneself, one perceives the advertisement as more positive then when the straw was
21 with an explorative implicit test wherein the estimated size of the product plays a role,
developed by Van Koningsbruggen et al. (2011).
This finding indicates that products oriented toward the consumer, affect the evaluation of advertisements positively. Besides, the mere presence of an attribute that indicates what the orientation of a product is (e.g. a straw positioned forwards or backwards), does not lead to more positive evaluations. Namely, there were no significant differences between the evaluations of the group with the straw opposed from themselves or without a straw. Thus, specifically the direction of the straw affects consumer evaluations, and not the mere presence of the straw. This finding leads to the recommendation for marketers and advertisers, that adding attributes that are directed toward the consumer could lead to positive effects in advertising.
However, other explorative implicit measures used in the current research, did not show main effects. For example Hammond’s error test (1948) and evaluations by distance from the product (Balcetis & Dunning, 2010). Future research should further investigate the workings of these measures, to contribute to the research field of implicit measures on consumer evaluations and behavior. Furthermore, the test on the visual analogue scale (Eelen et al., 2013) and the explicit test for purchase intention neither showed effects. Therefore, it could be possible that the test of Van Koningsbruggen et al. (2011) lacks internal validity and the perceived size of the smoothie is not related to the attitude toward the advertisement, but that it is related to other aspects. For example, the smoothie could be optical bigger, when the straw is positioned towards the consumer, and therefore not relate to the consumers’ attitude, but to mere optical aspects. On the other side, the test of Van Koningsbruggen et al. (2011) showed positive effects in prior research, which indicates that there is a relationship between attitude and perceived size of an object. Since this test is still very explorative, more research is needed to validate the relationship between perceived size of an object and attitude toward that object.
Underlying processes in the consumers’ mind, when being exposed to the
advertisements, have not been revealed by this research. This research did not show that self-referencing, positive affect and/or embodied simulation mediate the effect of the
manipulations on consumer evaluations. Therefore, there are no indications for explanations of the positive effect of the straw positioned toward the consumer.
Although self-referencing, positive affect and embodied simulation could not be explained as mediating processes in the current research, interesting is that they sometimes do play a role in evaluation of the product or advertisement. We have seen for example that self-referencing leads to positive evaluations: consumers think the advertisement is more attractive, the attitude toward the product is higher and their purchase intentions are higher. We also have seen that positive affect leads to positive evaluations: consumers think the
22 advertisement is more attractive, the attitude towards the product is higher and consumers show higher purchase intentions. Finally, embodied simulation also leads to positive
evaluations: the advertisement is perceived as more attractive, the desirability of the product is higher and consumers show higher purchase intentions.
The research showed one peculiar finding: higher self-referencing was related to perceiving the smoothie as further away. Therefore, consumers desire the product less, when they refer the advertisement to themselves. This finding is in contrast with the other positive findings. A reason could be that self-referencing is related to perceiving the distance from the smoothie in another way than by the desirability of the product. However, an
explanation remains unclear.
In accordance to the other findings, the current research indicates that advertisements that appeal on self-referencing, positive affect and embodied simulation, evoke more positive consumer evaluations than advertisements that do not appeal on these factors. This is an important finding for practitioners. However, future research should indicate what visual details in advertisements could appeal on these factors, since the current research did not show effects of the different visual stimuli on self-referencing, positive affect or embodied simulation. Knowing what aspects in an advertisement appeal on these implicit processes, would lead to deeper insights in the consumers’ mind and better recommendations for practitioners.
Next to effects of these implicit processes, self-referencing, positive affect and embodied simulation were related to each other in the current research. This finding is an addition to prior research, wherein these different concepts were examined independently (Elder & Krishna, 2012; Yoon & Park, 2012; Eelen et al., 2013;). Mental simulation is a very diverse concept, but the mutual relation of these processes is a good addition to the research field, to better define the concept of mental simulation.
To check for moderating effects in the experiment, the current research investigated what individual differences could influence the consumer evaluations. A main finding of
imagination ability on ad attractiveness was found: respondents with a higher ability to imagine perceived the ad as more attractive. This finding is interesting to the research field, since persons with a higher imagination ability see more vivid mental images, and apparently therefore find the ad more attractive. It would be interesting to investigate whether a high imagination ability always evokes a higher attraction toward the picture, or whether a high imagination ability also could evoke feelings of disfavor, e.g. when a picture is negative. Are people with low ability to imagine more difficult to reach with advertisements that appeal on mental images?
A moderating effect of imagination ability was shown with the implicit test of Hammond (1948). However, this effect was opposite from what was expected: Individuals with better
23 ability to imagine who were exposed to the advertisement with the straw in the smoothie toward themselves, showed a lower attitude toward the smoothie than the respondents with less imagination ability. This effect was reversed in the groups who were exposed to the smoothie with the straw opposed from themselves, or without a straw. An explanation remains unclear.
Prior research has shown that imagination ability (Marks, 1973; Schlosser, 2003; Libby & Eibach, 2011; Fennis et al., 2012) and style of processing (Childers et al., 1985; Miller et al., 2012) could moderate effects of factors that appeal on mental simulation. The current research showed that the direction of the moderating effects should be further examined. Why is a higher ability to imagine related to a lower attitude toward the product, when an attribute is directed toward the consumer? Future research should continue on explaining why and when these moderating effects occur, since the current research does not show indications for explanations.
In the current research, the moderating effect of imagination ability was only shown on attitude (perceived percentage of strawberries). Imagination ability did not moderate effects on attractiveness of the advertisement, perceived distance from the smoothie, perceived volume of the smoothie or purchase intention. Style of processing did not moderate effects on any dependent variable. A clear explanation for why effects only occur on perceived percentage of strawberries, and not on the other variables, remains absent. Also here, the explorative characteristics of the implicit measurements could definitely play a role.
Thus, this research contributes to the marketing research field by showing that an attribute positioned toward the consumer can affect consumer evaluations positively. This effect can be moderated by the imagination ability of an individual. Often, self-referencing, positive affect and embodied simulation, lead to positive evaluations of advertisements. It highly depended on the style of measuring, whether effects were shown. This research was very explorative in its styles of measuring, and thereby adds value to the research field of implicit measurements. We have seen that all the explorative measures did show effects on
consumer evaluations. However, the internal validation of these measurements remains a matter of debate. Research is still too scarce to use these explorative implicit measures with certainty. Nevertheless, implicit measures are getting more important to tab into a
consumers’ mind, since consumers nowadays are bombarded with advertisements and often do not consciously know why they prefer a certain product.
Finally, this research shows indications for higher consumer evaluations of objects in an advertisement positioned toward the consumer. The mere use of a straw in a smoothie, positioned differently, does not enable to generalize. However, first steps have been taken in discovering indicators for an unconscious influence in advertising, to reach the consumer in an implicit way. Other ways in how reaching the consumer implicitly could be encouraged,
24 could be a pair of warm fuzzy slippers with the opening towards the consumer, having the bottle top off of a soda, an open door toward the consumer in a car advertisement or the sheets folded down on the side of the bed toward the consumer (Elder & Krishna, 2012). Possibilities are endlessly, and therefore recommendations for marketers and advertisers would be to position a product in an advertisement toward the consumer, or to appeal on mental simulation, since particularly positive effects have been shown.
References
Ashby, F. G., Isen, A. M. & Turken, U. (1999). A neuropsychological theory of positive affect and its influence on cognition. Psychological review, 106(3), 529.
Balcetis, E., & Dunning, D. (2010). Wishful seeing: More desired objects are seen as closer.
Psychological Science, 21(1), 147-152.
Barsalou, L. (2008). Grounded cognition. Annual Review of Psychology, 59(1), 617–645. Beilock, S. L., & Holt, L. E. (2007). Embodied preference judgments can likeability be
driven by the motor system?. Psychological Science, 18(1), 51-57.
Bone, P. F., & Ellen, P. S. (1992). The generation and consequences of communication-evoked imagery. Journal of Consumer Research, 19(1), 93-104.
Burnkrant, R. E., & Unnava H. R. (1989). Self-referencing: A strategy for increasing
processing of message content. Personality and Social Psychology Bulletin, 5(4), 628–38. Burnkrant, R. E., & Unnava, H. R. (1995). Effects of self-referencing on persuasion. Journal
of Consumer Research, 22(1): 17–26.
Casasanto, D., & Chrysikou, E. G. (2011). When left is “right”: Motor fluency shapes abstract concepts. Psychological Science, 22(4), 419-422.
Childers, T. L., Houston, M. J., & Heckler, S. E. (1985). Measurement of individual differences in visual versus verbal information processing. Journal of Consumer
Research, 12(2), 125-134.
Dahl, D. W., & Hoeffler, S. (2004). Visualizing the self: Exploring the potential benefits and drawbacks for new product evaluation. Journal of Product Innovation Management, 21(4), 259-267.
Eelen, J., Dewitte, S., & Warlop, L. (2013). Situated embodied cognition: Monitoring orientation cues affects product evaluation and choice. Journal of Consumer
Psychology, 23(4), 424-433.
Elder, R. S., & Krishna, A. (2012). The “visual depiction effect” in advertising: Facilitating embodied mental simulation through product orientation. Journal of Consumer
Research, 38(6), 988-1003.
Ellis, R., & Tucker, M. (2000). Micro-affordance: The potentiation of components of action by seen objects. British Journal of Psychology, 91, 451–471.
25 Ernest, C. H. (1977). Imagery ability and cognition: A critical review. Journal of Mental
Imagery, 1(2), 181-215.
Fennis, B. M., Das, E., & Fransen, M. L. (2012). Print advertising: Vivid content. Journal of
Business Research, 65(6), 861-864.
Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The broaden and build theory of positive emotions. American Psychologist, 56(3), 218–226.
doi:10.1037/0003-066X.56.3.218
Gawronski, B., & Payne, B. K. (Eds.). (2010). Handbook of implicit social cognition:
Measurement, theory, and applications. New York: Guilford Press.
González, J., Barros-Loscertales, A., Pulvermüller, F., Meseguer, V., Sanjuán, A., Belloch, V., et al. (2006). Reading cinnamon activates olfactory brain regions. NeuroImage, 32(2), 906–912.
Hammond, K. R. (1948). Measuring attitudes by error choice: An indirect method. Journal of
Abnormal and Social Psychology, 43(1), 38-48.
Hoffman, M. L. (1986). Affect, cognition, and motivation. In R. M. Sorrentino & E.T. Higgins (Eds.), Handbook of motivation and cognition: Foundations of social behavior (pp. 244-280). New York: Guilford Press.
Johnson, M. P., & Puddifoot, J. E. (1998). Miscarriage: Is vividness of visual imagery a factor in the grief reaction of the partner?. British Journal of Health Psychology, 3(2), 137-146. Krishna, A. (2012). An integrative review of sensory marketing: Engaging the senses to affect
perception, judgment and behavior. Journal of Consumer Psychology, 22(3), 332–351. Libby, L. K., & Eibach, R. P. (2011). Visual perspective in mental imagery: A
representational tool that functions in judgment, emotion, and self-insight. Advances in
experimental social psychology, 44, 185.
Libby, L. K., Shaeffer, E. M., & Eibach, R. P. (2009). Seeing meaning in action: A bidirectional link between visual perspective and action identification level. Journal of
Experimental Psychology, 138(4), 503–516.
McIsaac, H. K., & Eich, E. (2002). Vantage point in episodic memory. Psychonomic Bulletin
& Review, 9(1), 144–150.
McKelvie, S. J. (1994). The Vividness of Visual Imagery Questionnaire as a predictor of facial recognition memory performance. British Journal of Psychology, 85(1), 93-104.
McKelvie, S. J. (1998). Effects of gender on reported vividness of visual imagery for parents. Journal of Mental Imagery, 22(3-4), 99-112.
Marks, D. F. (1973). Visual imagery differences in the recall of pictures. British journal of
Psychology, 64(1), 17-24.
Meyers-Levy, J., & Peracchio, L. A. (1992). Getting an angle in advertising: The effect of camera angle on product evaluations. Journal of Marketing Research, 29(4), 454-461.
