Available online at www.sciencedirect.

com

Journal of Consumer Psychology 20 (2010) 508 – 520

Understanding the impact of self-concept on the stylistic properties

of images☆
Xiaojing Yang a,⁎,1 , Jing Zhang b,1 , Laura A. Peracchio a
a
Marketing at the Sheldon B. Lubar School of Business, University of Wisconsin-Milwaukee, Milwaukee, WI 53201, USA
b
Department of Communication Studies at California State University, Sacramento, Sacramento, CA 95819, USA
Received 28 November 2009; revised 25 June 2010; accepted 29 June 2010
Available online 1 August 2010

Abstract

Much research has shown that the stylistic properties of visual images can serve as an important source of persuasion. However, the existing
literature seldom incorporates characteristics of consumers, such as self-concept, as a determining factor in the effectiveness of images' stylistic
properties. In the context of exploring a specific stylistic property, camera angle, we demonstrate across three experiments that when an upward-
looking camera angle is employed to depict a product, participants with an ought-self (motivated by duties and obligations) generate more
favorable product evaluations than participants with an ideal-self (motivated by hopes and aspirations). The reverse is found when a downward-
looking angle is utilized. We also explore the mechanism underlying these effects. That is, individuals with an ideal-self and ought-self use
different control strategies (influencing versus adapting to existing realities) impacting their product evaluations.
© 2010 Society for Consumer Psychology. Published by Elsevier Inc. All rights reserved.

Keywords: Aesthetic experiences; Persuasion; Stylistic properties; Self-concept; Feeling of control

Recently, there has been a steady increase in the study of the
stylistic properties of visual images on advertising persuasive-
ness. Stylistic properties refer to a variety of factors that impact
the manner in which visual material is displayed, such as
camera angles, visual perspectives, the orientation (e.g.,
vertical, diagonal) of objects displayed in a scene, as well as
various other production elements (Peracchio & Meyers-Levy,
2005). This stream of research has demonstrated that the
persuasive power of ads often resides in pictorial depictions,
and visual images in ads convey semantically meaningful
concepts to consumers via their stylistic properties (Messaris,
1997; Phillips & McQuarrie, 2002, 2004; Scott, 1994; Scott &
Batra, 2003). Consider, for example, the stylistic property of
camera angle. When a product is photographed from an
upward-looking angle, viewers tend to associate the product
with potency and powerfulness (Peracchio & Meyers-Levy,
☆
C.W. Park served as Editor for this article.
⁎ Corresponding author.
E-mail addresses: yangxiao@uwm.edu (X. Yang), zhang@csus.edu
(J. Zhang), LPeracch@uwm.edu (L.A. Peracchio).
1
The first two authors contributed equally to the research and are listed
alphabetically.
1057-7408/$ - see front matter © 2010 Society for Consumer Psychology. Published by Elsevier Inc. All rights reserved.
doi:10.1016/j.jcps.2010.06.012
2005). This association may be due to viewers' experiences
with those they literally “look up to.” For example, young
children must often look up to view an adult's face.
Although many insights have emerged from the investiga-
tion of stylistic properties, few researchers have examined
characteristics of the consumers themselves, such as self-
concept, as a determining factor in the persuasiveness of such
elements. It seems possible that, depending on the aspect of self-
concept that is accessible, consumers may interpret the same
stylistic property in very different ways. In this paper, we study
a particular aspect of self-concept, the ideal-self versus the
ought-self, and examine its effect on people's interpretations
and evaluations of the stylistic properties of images in the
context of camera angles. We also explore the mechanism
underlying how consumers with different self-concepts process
stylistic properties.
We propose that an ideal-self versus an ought-self prompts
consumers to recruit different paths to achieve control, primary
vs. secondary, which in turn affects their judgments of the
stylistic properties employed in the ad, and ultimately, impacts
evaluations of the advertised product. Specifically, consumers
with an ought-self achieve control by fitting into existing
 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
realities and aligning with powerful others (products, people
etc; i.e., secondary control). For instance, consumers sometimes
purchase a product to fit into a desirable social group (e.g., a
prestigious social club). Accordingly, an upward-looking
camera angle may foster more favorable product evaluations
among consumers with an ought-self because such an angle
confers the perception of power and potency (Kraft, 1987;
Meyers-Levy & Peracchio, 1992). On the other hand,
consumers with an ideal-self achieve a sense of control by
influencing existing realities in the surrounding environment to
fit self (i.e., primary control). As an example, consumers buy a
product to sustain their autonomy (e.g., buying a car to facilitate
transportation). Thus, a downward-looking camera angle may
trigger more favorable product evaluations among consumers
with an ideal-self because an object presented at this angle is
perceived to be less threatening and more malleable.
The remainder of the paper is organized as follows. First, we
review the literatures related to stylistic properties and the
implications of self-concept for persuasion. We use feeling of
control (primary vs. secondary) as a theoretical framework to
examine how self-concept affects consumers' processing of ads
employing a specific stylistic property, camera angle, and
investigate how this processing ultimately affects product
evaluations. Three experiments were conducted to test our
hypotheses and provide robust and converging evidence for our
theorizing regarding the effect of self-concept on the evalua-
tions of products depicted from different camera angles. Study 2
reveals the mediating effect of a primary- vs. secondary-control
mechanism, and study 3 manipulates control strategies and
provides evidence for the impact of the proposed control
mechanism on product evaluations.
Theoretical background
Stylistic properties and persuasion
Recent research suggests that the stylistic properties of an ad
can serve as central arguments by conveying descriptive concepts
that are diagnostic for product evaluations (Peracchio & Meyers-
Levy, 2005; Wang & Peracchio, 2008). Thus, stylistic properties
subsequently affect consumers' perceptions of the product and the
effectiveness of an ad. For instance, Peracchio and Meyers-Levy
(2005) found that the stylistic property of product orientation in an
ad (e.g., the advertised product is displayed in a diagonal
orientation) affects viewers' perceptions by communicating
descriptive concepts (e.g., dynamism and activity). However,
this effect appears to occur only when viewers engage in ample
processing of the ad and the accessibility of the concept (e.g.,
dynamism) is heightened by ad copy. In addition, this research
revealed that the stylistic elements of an ad can affect how
consumers process other ad elements, such as ad copy. Thus,
those highly motivated individuals, who tend to engage in
detailed and extensive processing, use the stylistic properties of an
ad as a starting point for examining the ad and forming their
product evaluations.
Despite this progress, the existing research on stylistic
properties seldom takes into consideration characteristics of the
509
consumers themselves. For example, individual differences
among consumers are likely to systematically affect the
perceptions of the stylistic properties of an ad. In this paper,
we argue that a particular consumer characteristic, self-concept,
can play a critical role in affecting how individuals perceive
stylistic properties and may ultimately impact their evaluations
of the advertised product.
Self-concept and information processing
Self-concept, or how consumers view and understand
themselves, has been found to play an intriguing role in
explaining consumer behavior, ranging from the persuasiveness
of ad messages, to evaluations of brand extensions, as well as
perceptions of product attributes (e.g., Monga & John, 2007;
Zhu & Meyers-Levy, 2009). This paper focuses on one aspect
of self-concept, the ideal-self and the ought-self (Higgins,
1987). The ideal-self is the representation of the set of attributes
that one (or someone else) believes an individual would ideally
possess. It is closely related to one's hopes, wishes, and
aspirations (e.g., wanting a luxury sedan, dreaming of an exotic
vacation). The ought-self is the representation of the set of
attributes that one (or someone else) believes an individual
should possess and is closely related to one's obligations,
duties, and responsibilities (e.g., taking care of aging parents,
behaving professionally at work).
The ideal- versus the ought-self and consumers'
feeling of control
Consumers with an ideal-self versus an ought-self rely on
different types of information when forming product evalua-
tions (Pham & Avnet, 2004). We theorize that self-concept
should also affect how consumers achieve a feeling of control.
Control is defined as “causing an intended event” (Skinner &
Chapman, 1984; Weisz, Rothbaum, & Blackburn, 1984). The
intended event may involve “influencing objective or external
realities or influencing the personal, psychological impact of
those realities” (Weisz et al., 1984, p. 958). There are two paths
to achieve control: primary control and secondary control
(Chipperfield & Perry, 2006; Weisz et al., 1984). Consumers
with primary control gain a sense of control by influencing
existing realities, e.g., exerting influence over other people,
objects, and circumstances. Primary control is achieved “via
acts involving personal agency, dominance, or even aggression.
These acts are often intended to express, enhance, or sustain
individualism and personal autonomy” (Weisz et al., 1984, p.
956; also demonstrated in Rothbaum, Weisz, & Snyder, 1982).
By contrast, consumers with secondary control gain a sense of
control by accommodating to existing realities and maximizing
satisfaction or goodness of fit with things as they are. This
secondary control is achieved “via acts that limit individualism
and personal autonomy, but enhance perceived alignment or
goodness of fit with people, objects, or circumstances in their
world” (Weisz et al., 1984, p. 956; also demonstrated in
Rothbaum et al., 1982). Thus, the fundamental difference
between primary and secondary control is altering the
510 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
environment (primary control) or altering self (secondary
control).
We propose that the ideal-self versus the ought-self might
prompt consumers to use either primary or secondary control.
To explain, an ideal-self promotes an exploration stance,
leading to more risk-seeking behaviors (Crowe & Higgins,
1997) and an emphasis on ways to challenge the status quo
(Liberman et al., 1999). Consumers with an ideal-self are thus
more likely to focus on seeking personal agency, identifying
ways to exert influence on existing realties and implement
changes to challenge the status-quo (i.e., primary control). In an
effort to influence existing realities and challenge the status-
quo, consumers with an ideal-self are likely to employ primary
control, and consider how a product can be used for their own
purposes.
On the other hand, an ought-self activates vigilance, resulting
in more risk-averse behaviors (Crowe & Higgins, 1997) and a
focus on how to maintain the status quo (Liberman et al., 1999).
Consumers with an ought-self are more likely to emphasize ways
to fit into existing realities and adhere to the status-quo (i.e.,
secondary control). As suggested by Weisz et al. (1984) and
Rothbaum et al. (1982), people who exercise secondary control
tend to align themselves with powerful entities in the social
environment, e.g., leaders as well as cultural symbols that convey
the notion of power and potency. In the context of consumer
behavior, these consumers with an ought-self are inclined to align
with entities they deem prestigious or important when they
evaluate or purchase a product.
Hypotheses
Research focused on feeling of control has been used to
explain many consumption phenomena, such as individual
differences in consumer tolerance (Chan, Wan, & Sin, 2009)
and the likelihood of using consumption behavior to change or
regulate mood (Luomala et al., 2004). The feeling of control
when dealing with powerful entities is particularly helpful in
understanding what consumers experience psychologically after
being exposed to a stylistic property either conveying power
(e.g., the upward-looking camera angle) or a lack of power (e.g.,
the downward-looking camera angle).
As of now, little research has looked into how self-concept,
e.g., the ideal- vs. ought-self, may interact with stylistic
properties of visual images in ads and impact product
evaluations. In this research, we study one specific stylistic
property, camera angle (upward-looking vs. downward-look-
ing). Past studies found that when a product is photographed
from an upward-looking angle, it is generally evaluated more
favorably compared to when it is depicted from a downward-
looking angle (Kraft, 1987; Meyers-Levy & Peracchio, 1992).
This effect occurs because consumers tend to attribute greater
power and performance to objects that are viewed from an
upward-looking angle. Presumably, consumers have accumu-
lated such attributions through their past experience, e.g.,
looking up to view an adult's face when they were young or
associating skyscrapers with the notion of strength or potency.
We argue, however, this view is somewhat static and a more
comprehensive view of stylistic properties needs to incorporate
the dynamic role of viewers themselves. To explain, when an
upward-looking camera angle is employed to portray a product
in an ad, such a camera angle conveys powerfulness (of the
product) that is external to audiences, i.e., the powerful entity is
present in the viewer's surrounding environment. By contrast,
when a downward-looking camera angle is used, it portrays the
product in a less powerful position, which may elicit feelings of
powerfulness from audiences, i.e., viewers experience the
feeling of powerfulness and personal agency.
Upward-looking angle
Existing literature suggests that a product photographed from
an upward-looking camera angle conveys that the product is
powerful and strong. Among viewers with an ought-self (as
opposed to those with an ideal-self), who are seeking fit with the
environment and alignment with powerful figures, this
depiction is likely to elicit secondary control. What this implies
is that when processing the power or potency message of the
product, consumers with an ought-self are likely to consider
what accommodations (either behavioral or psychological) they
need to make in order to acquire or better use the product. As a
consequence of secondary control, consumers with an ought-
self may prefer a product portrayed from an upward-looking
camera angle, which confers the perception of power. As such,
products photographed from an upward-looking camera angle
are likely to elicit more favorable evaluations among consumers
with an ought-self.
Conversely, an upward-looking angle may not be perceived
as favorable by consumers with an ideal-self, who are seeking
primary control and personal agency. To enhance their own
sense of strength and achieve control, consumers with an ideal-
self focus on influencing existing realities. In an effort to sustain
personal agency, consumers with an ideal-self may see powerful
figures as threatening and obstructive to their effort, and they
are likely to react less favorably towards a product photo-
graphed from an upward-looking camera angle. Thus, after
viewing a product from an upward-looking angle, we expect
that consumers with an ought-self are likely to form more
favorable product evaluations than consumers with an ideal-
self.
Downward-looking angle
By contrast, a downward-looking angle conveys a different
set of perceptions. Existing literature suggests that a downward-
looking angle conveys an absence of power, which leads to less
favorable evaluations (Kraft, 1987; Meyers-Levy & Peracchio,
1992). We argue that this is particularly true for consumers with
an ought-self. Because these individuals attempt to achieve a
feeling of control by relating to powerful others, the lack of
power in a product depiction may make them feel less powerful
or unimportant. Hence, products viewed from a downward-
looking angle would trigger less favorable evaluations among
consumers with an ought-self.
On the other hand, consumers with an ideal-self tend to think
about how a product fits into their lives and whether a product is
helpful in sustaining personal agency. As discussed, consumers
 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
with an ideal-self achieve a sense of power and a feeling of
control by challenging the status-quo and influencing existing
realities. Thus, they are more likely to generate favorable
evaluations of objects that they can easily exert influence on,
including products and their own possessions. When consumers
with an ideal-self see a product portrayed using a downward-
looking angle, they are more likely to feel that they themselves
are in a more powerful and stronger position in relation to the
product. Indeed, viewing an object from a downward-looking
angle can enhance their feeling of control over the environment.
After viewing a product from a downward-looking camera
angle, we predict that consumers with an ideal-self will have
more favorable product evaluations than consumers with an
ought-self. Thus, we expect a two-way interaction between the
type of self (ought-self vs. ideal-self) and the camera angle used
to portray a product on product evaluations. Specifically,
H1. When an upward-looking angle is employed, consumers
with an ought-self will have more favorable product evaluations
than consumers with an ideal-self.
H2. When a downward-looking angle is employed, consumers
with an ideal-self will have more favorable product evaluations
than consumers with an ought-self.
Study 1
Method
Study design
This study used a 2 (self-concept: ideal-self vs. ought-
self) × 2 (camera angle: downward-looking vs. upward-looking)
between-subjects ANOVA design. One hundred fifteen under-
graduate students from a mid-western university participated in
the experiment.
Stimulus
A target ad for a fictitious cereal product called Vicore was
used in the study (see Peracchio & Meyers-Levy, 2005). The ad
featured a headline as well as a large box of the cereal
photographed from either an upward-looking or a downward-
looking camera angle. Similar to previous stylistic property
studies (Meyers-Levy & Peracchio, 1992; Peracchio & Meyers-
Levy, 2005), camera angle was manipulated by varying the
manner in which the advertised cereal was photographed. An
upward-looking photo of the product was taken from below
eye-level. A downward-looking photo of the product was taken
from above eye-level. The graphics on the box displayed the
brand name, a bowl of the cereal, and two ad claims, which
stated that the cereal was fortified with calcium and folic acid
and was low in fat. Below the cereal box was brief ad copy
encouraging product trial.
Procedure
Participants were told to complete a series of unrelated
studies at their own pace. On the first page of the experimental
booklet, participants read the instructions for the manipulation
511
of the ideal-self versus the ought-self. Specifically, as in Pham
and Avnet's (2004) study, in the ideal-self priming condition,
participants were asked to think about their past hopes,
aspirations, and dreams, and list two of them. They were then
asked to think about their current hopes, aspirations, and
dreams, and again list two of them. In the ought-self priming
condition, participants were asked to think about their past
duties, obligations, and responsibilities, and list two of them.
They were then asked to think about their current duties,
obligations, and responsibilities, and again list two of them. A
separate pretest (n = 40) conducted showed our manipulation of
the ideal-self versus the ought-self was successful.2
Participants were then presented with the target ad for a new
cereal in which camera angle was manipulated. They were
asked to view the ad as they normally would and indicate their
evaluations toward the advertised cereal on five, seven-point
semantic differential scales, anchored at “bad/good,” “unfavor-
able/favorable,” “negative/positive,” “worthless/valuable,” “not
interesting/interesting.” The responses were averaged to form a
single index for product evaluations (alpha = .94).
Results
A 2 (self-concept: ideal-self vs. ought-self) × 2 (camera
angle: downward-looking vs. upward-looking) ANOVA was
performed. The means are portrayed in Fig. 1. Unless otherwise
noted, the degrees of freedom are 1 and 111.
Product evaluations
As predicted, there was a significant self-concept by camera
angle interaction on product evaluations (F = 8.56, p b .01).
Further analysis of participants' product evaluation ratings
confirmed our expectations. Specifically, when a product was
photographed using an upward-looking angle, participants
whose ought-self (vs. ideal-self) was activated reported more
favorable product evaluations (Mideal = 2.98 vs. Mought = 3.48;
F = 4.27, p b .04). Conversely, when a product was depicted
using a downward-looking angle, participants whose ideal-self
(vs. ought-self) was activated generated more favorable product
evaluations (Mideal = 3.54 vs. Mought = 3.05; F = 4.29, p b .04).
Hence, both hypotheses 1 and 2 were supported.
2
Pretest participants from the same respondent pool as the main studies were
exposed to the manipulation of the ideal-self versus the ought-self and then
completed the manipulation check measures (Pham & Avnet, 2004).
Specifically, they were presented with three different “personal choices” meant
to capture conflicts between the ideal-self and ought-self. These seven-point
scales were anchored by: (1) I prefer to “take a trip around the world” (ideal)
versus “pay back my loans” (ought); (2) I prefer to “go wherever my heart takes
me” (ideal) versus “do whatever it takes to keep my promises” (ought); and
(3) I prefer to “do what is right” (ought) versus “whatever I want” (ideal)
(reverse coded). Responses were averaged into a single index (alpha = .73),
which was subject to a one-way ANOVA with the manipulation as the
independent variable. Results showed that, as expected, participants who
received the ought-self (ideal-self) instructions reported higher (lower) scores
on the index (Mideal = 3.60 vs. Mought = 4.55; F(1,38) = 9.03, p b .04), indicating
their reliance on the ought-self (ideal-self) when making choices. Thus, our
manipulation of self-concept was successful.
512 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
Fig. 1. Study 1, product evaluations as a function of camera angle and self-concept.
Discussion
As expected, we find that when an upward-looking camera
angle is employed, participants report more favorable product
evaluations if their ought-self is accessible. Conversely, when a
downward-looking camera angle is utilized, participants report
more favorable product evaluations if their ideal-self is
accessible. This interaction provides support for our central
hypothesis that consumers' self-concept should be incorporated
to predict the effectiveness of portraying the product from a
particular camera angle.
Though we have found evidence for the hypothesized
interaction effect between self-concept and camera angle, we
still need to provide evidence for the underlying psychological
processes: type of control. We theorize that consumers with
activated ideal-self versus ought-self differ in how to achieve a
sense of control. In addition, we provide evidence for the
robustness of our results using an alternative product category.
Study 2
There are two objectives in study 2. First, by analyzing
participants' thoughts, we aim to provide evidence for the sense
of control that we propose underpins the results of study 1.
Second, by employing a different product category, we replicate
the results from study 1, showing the robustness of the
hypothesized camera angle and self-concept interaction.
Our main contention is that when exposed to a product
photographed from either an upward-looking or downward-
looking camera angle, the two types of self-concept, the ideal-
self versus ought-self, prompt consumers to recruit different
paths to achieve control and elicit different views of products.
This in turn affects their product evaluations and, ultimately, the
effectiveness of the stylistic properties of a product image. We
expect the thoughts generated by participants to reveal evidence
for the proposed processing mechanism.
Specifically, consumers with an ideal-self pursue primary
control, emphasizing how they can influence existing realities
(change environment) in an effort to achieve a sense of control
(e.g., feel more powerful or stronger). Consumers with an
ought-self are more likely to use secondary control, focusing on
what they can do to adapt to their surroundings (change self) to
achieve a sense of control. This fundamental difference between
primary and secondary control, influencing vs. adapting to
existing realities, changing environment or changing self,
implies that different types of thoughts may be elicited when
the two types of self-concept are activated.
To explain, pursuing primary control and sustaining personal
agency, individuals with an ideal-self tend to generate primary-
control thoughts that emphasize how to influence existing
realities, situations and problems to fit themselves. Thus, their
thoughts focus on trying to understand how they can employ a
product for their own use. By contrast, individuals with an
ought-self pursing secondary control generate more secondary-
control thoughts that focus on how to change themselves
psychologically or behaviorally to fit the environment, and how
to align with powerful others.
In the current study, the pattern of primary-control versus
secondary-control thoughts is expected to depend on the
activated self-concept and the particular camera angle that is
employed to display the advertised product. To explain, a
product photographed from a downward-looking camera angle
makes viewers feel powerful and strong in relation to the
product. And, this may be particularly true among individuals
with an ideal-self because they are inclined to influence their
surroundings to achieve a sense of control (primary control), as
well as personal agency and autonomy. Individuals with an
ideal-self are more likely to attempt to understand what changes
in their immediate environment they can exert, and how
personal agency and autonomy can be achieved by using the
advertised product. By contrast, individuals with an ought-self
are less likely to generate primary-control thoughts. Instead,
individuals with an ought-self focus on aligning with powerful
others to achieve a sense of control, and they consider how to
change themselves to fit the environment or situation. Thus,
when a product is portrayed with a downward-looking angle,
individuals with an ideal-self (vs. ought-self) should generate
more primary-control thoughts.
On the other hand, an upward-looking camera angle
activates different types of thoughts among individuals with
an ideal-self versus an ought-self. The literature suggests an
upward-looking angle conveys power and potency. Individuals
with an ought-self are expected to interpret such an angle more
favorably because they are oriented towards adapting to existing
realities, and powerful products help them achieve a sense of
control (secondary control). These individuals may want to
understand the status of the brand or who the primary users are.
These types of thoughts may help consumers with an ought-self
evaluate a product. By contrast, individuals with an ideal-self,
who focus on influencing existing realities to achieve control, as
well as personal agency and autonomy, are less likely to
generate secondary-control thoughts. Thus, in the upward-
looking camera angle condition, we predict that individuals
 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520 513

with an ought-self will have more secondary-control thoughts Table 1

than those with an ideal-self. The previous discussion leads to Study 2. Treatment means and standard deviations.
the following predictions: Upward-looking Downward-
looking
H3. When a downward-looking angle is employed, participants Ideal Ought Ideal Ought
with an ideal-self generate a greater proportion of primary- Product Evaluations 4.09 4.78 4.95 4.23
control thoughts than those with an ought-self. When an (1.05) (.81) (1.01) (.95)
upward-looking angle is employed, such a difference in Proportion of Positive-Negative −.39 .03 .15 −.32
Thoughts (.64) (.59) (.66) (.62)
primary-control thoughts will diminish.
Proportion of Primary Control .13 .17 .39 .12
Thoughts (.20) (.27) (.34) (.18)
H4. When an upward-looking angle is employed, participants Proportion of Secondary Control .12 .34 .08 .13
with an ought-self generate a greater proportion of secondary- Thoughts (.14) (.26) (.13) (.18)
control thoughts than those with an ideal-self. When a Total Thoughts 4.00 3.88 4.27 4.20
(2.45) (1.78) (1.71) (2.51)
downward-looking angle is employed, such a difference in
Cell Size 17 16 15 15
secondary-control thoughts will diminish.
Numbers in parentheses are standard deviations.
Method
to portray the product, participants reported more favorable
The experiment employed a 2 (self-concept: ideal-self vs. product evaluations when their ought-self (vs. ideal-self) was
ought-self) × 2 (camera angle: downward-looking vs. upward- activated (Mideal = 4.09 vs. Mought = 4.78; F(1, 59) = 4.13,
looking) between-subjects ANOVA design. Sixty-three under- p b .05). Conversely, when a downward-looking angle was
graduate students from a mid-western university participated in employed, participants reported more favorable product evalua-
the experiment. Participants were told to complete a series of tions when their ideal-self (vs. ought-self) was activated
seemingly unrelated studies at their own pace. The study (Mideal = 4.95 vs. Mought = 4.23; F(1, 59) = 4.20, p b .04). We
procedure was the same as study 1 except that: (1) participants replicated the study 1 findings and provided support for
viewed a fictitious new motorbike called KMT; and (2) hypotheses 1 and 2.
participants were instructed to list the thoughts that occurred
to them when viewing the target ad. The order of the thought-
listing and product evaluation (alpha = .92) tasks were Valence of thoughts
counterbalanced. Thoughts were categorized by two independent judges based
A new ad stimulus was developed for study 2. The new ad on valence (r = .90): positive, negative and neutral thoughts. A
featured a headline and ad copy, as well as a motorbike valence index was obtained by subtracting the number of
photographed from either an upward-looking or a downward- negative thoughts from the number of positive thoughts and then
looking camera angle. As in study 1, camera angle was dividing by the total number of thoughts. This index was subject
manipulated by varying the way in which the advertised to arcsine transformation prior to analysis. An analysis of the
motorbike was photographed: an upward-looking angle photo valence index revealed a pattern similar to the evaluation results.
was taken from below eye-level; a downward-looking angle That is, a two-way interaction between self-concept and camera
photo of the motorbike was taken from above eye-level. The ad angle emerged (F(1, 59) = 9.28, p b .01). Contrast analysis
featured the photo of the advertised motorbike and the brand further confirmed our expectations. Specifically, when an
name. The ad claims stated that the motorbike responds quickly upward-looking angle was employed, participants had more
to any terrain and is equipped with quality parts. net favorable thoughts when their ought-self was activated

Results

Preliminary analysis showed that the order of the thought

listing and product evaluations tasks was not significant (F b 1).
Thus, the data were collapsed across different orders for
subsequent analysis. A 2 (self-concept: ideal-self vs. ought-
self) × 2 (camera angle: downward-looking vs. upward-looking)
ANOVA was performed. The means are presented in Table 1
and Fig. 2.

Product evaluations
As expected, the ANOVA analysis revealed a significant
two-way interaction between self-concept and camera angle (F
(1, 59) = 8.32, p b .01). Contrast analysis confirmed our
expectations. When an upward-looking angle was employed Fig. 2. Study 2, product evaluations as a function of camera angle and self-concept.
514 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
compared to when their ideal-self was activated (Mideal = −.39 vs.
Mought = .03; F(1, 59) = 5.12, p b .03). Conversely, when a
downward-looking angle was employed, participants reported
significantly more net favorable thoughts when their ideal-self
was activated compared to when their ought-self was activated
(Mideal = .15 vs. Mought = −.32; F(1, 59) = 4.17, p b .05). Hence,
hypotheses 1 and 2 were also supported by the valence of
participants' thoughts.
Primary- versus secondary-control thoughts
Thoughts were categorized by two independent judges into
the following groups (r = .89): primary-control (influencing
existing realities; altering the environment; personal agency and
autonomy) and secondary-control (accommodating to existing
realities; altering self; alignment with powerful or aspirational
entities) thoughts. Consistent with earlier discussion, examples
of primary-control thoughts include: “This bike enables me to
ride in tough places that people can't go otherwise.” “With the
bike, I can do many more outdoor sports and off-road
adventures.” Examples of secondary-control thoughts include:
“I want to look as cool as those people riding the bike on the
street.” “Some of my best friends have similar bikes and owning
this bike definitely helps me to fit in.”
The proportions of primary-control and secondary-control
thoughts were subject to arcsine transformation and submitted
to the same ANOVA as in prior analyses. The analysis of
participants' primary- versus secondary-control thoughts
revealed patterns consistent with the hypotheses. As predicted
in hypothesis 3, the two-way interaction effect between camera
angle and self-concept on primary-control thoughts was
significant (F(1, 59) = 5.63, p b .02). Specifically, when a
downward-looking angle was employed, participants had a
greater proportion of primary-control thoughts when their ideal-
self was activated compared to when their ought-self was
activated (Mideal = .39 vs. Mought = .12; F(1, 59) = 7.16, p b .01).
Conversely, when an upward-looking angle was employed,
there was no significant difference among participants' primary-
control thoughts regardless of which self-concept was activated
(Mideal = .13 vs. Mought = .17; F b 1) supporting hypothesis 3.
To provide further support for the underlying process that
primary-control mechanism explains the effect of camera angle
and self-concept on product evaluations under the downward-
looking angle condition, a mediation analysis was conducted
following Baron and Kenny (1986). In the downward-looking
angle condition, we first ran a regression with product
evaluations as the dependent variable, and self-concept as the
independent variable. We found that self-concept was a
significant predictor of product evaluations (β = .36, t = 2.00,
p b .05). Then, we regressed primary-control thoughts on self-
concept and found that the self-concept was a significant
predictor of primary-control thoughts (β = .44, t = 2.57, p b .02).
As a final step, we ran a regression with product evaluations as
the dependent variable, and both self-concept and primary-
control thoughts as independent variables. The results showed
that primary-control thoughts were a significant predictor of
product evaluations (β = .51, t = 2.87, p b .01). However, the
original significant effect of self-concept on product evaluations
(as shown in the first regression) became insignificant (β = .14,
t = .76, p N .46). These findings showed that under the
downward-looking camera angle condition, primary-control
thoughts mediated the effect of self-concept on product
evaluations. We conducted the same analysis under the
upward-looking angle condition and found that the primary-
control thoughts did not mediate the effect of self-concept on
product evaluations.
In addition, as predicted in hypothesis 4, there was a marginal
two-way interaction between self-concept and camera angle on
secondary-control thoughts (F(1, 59) = 3.85, p b .06). Contrast
analysis further confirmed our expectations. Specifically, when
an upward-looking angle was employed, participants had a
greater proportion of secondary-control thoughts when their
ought-self was activated compared to when their ideal-self was
activated (Mideal = .12 vs. Mought = .34; F(1, 59) = 13.16, p b .01).
Conversely, when a downward-looking angle was employed,
there was no significant difference among participants regard-
less of the self-concept that was activated (Mideal = .08 vs.
Mought = .13; F b 1). Hence, hypothesis 4 was supported.
To investigate the mediation effect of the secondary-control
mechanism in explaining how individuals with an ought-self
evaluate products under the upward-looking angle condition, a
similar mediation analysis was performed. We first ran a
regression with product evaluations as the dependent variable,
and self-concept as the independent variable. We found that
self-concept was a significant predictor of product evaluations
(β = −.35, t = −2.07, p b .05). Then, we regressed secondary-
control thoughts on self-concept and found that the self-concept
was a significant predictor of secondary-control thoughts (β = −.49,
t=−3.16, p b .01). As a final step, we ran a regression with product
evaluations as the dependent variable, and both self-concept and
secondary-control thoughts as independent variables. The results
showed that secondary-control thoughts were a significant
predictor of product evaluations (β = .41, t =2.27, p b .03). Howev-
er, the effect of self-concept on product evaluations (as found in the
first regression) became insignificant (β =−.15, t =−.80, p N .43).
Again these results showed that under the condition of an upward-
looking camera angle, secondary-control thoughts mediated the
effect of self-concept on product evaluations. We did the same
analysis under the downward-looking angle condition and found
that the secondary-focused thoughts did not mediate the effect of
self-concept on product evaluations.
Discussion
Study 2 replicated the findings of study 1. The results
support the hypothesized two-way interaction between self-
concept and camera angle on product evaluations. Specifically,
products depicted using an upward-looking angle generated
more favorable evaluations among participants with an ought-
self, while products portrayed using a downward-looking angle
elicited more positive evaluations among participants with an
ideal-self. More importantly, we also found evidence for our
proposed theorizing by analyzing participants' thoughts.
Specifically, participants whose ideal-self was activated
generated more primary-control thoughts when the product
 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
was depicted using a downward-looking angle, while partici-
pants whose ought-self was activated had more secondary-
control thoughts when the product was portrayed using an
upward-looking angle. Thus, our claim that the ideal-self versus
the ought-self activates different types of product control among
participants is supported.
Study 3
Our central theorizing proposes that an ideal-self leads
people to generate primary-control thinking, which leads these
individuals to form more favorable evaluations of products
portrayed with a downward-looking angle. Conversely, an
ought-self induces secondary-control thinking, which results in
more positive evaluations of products displayed from an
upward-looking angle. To provide stronger evidence for the
process we proposed, we manipulate the extent to which
participants engage in primary- versus secondary-control in
study 3. Extant research has adopted analogous procedures to
offer support for similar theorizing (Monga & John, 2007;
Zhao, Hoeffler, & Zauberman, 2007).
Study 3 employs a 3 (prime: primary-control vs. secondary-
control vs. no control prime) × 2 (self-concept: ideal-self vs.
ought-self) × 2 (camera angle: downward-looking vs. upward-
looking) between-subjects factorial design. To be clear, in the
no control prime condition (“no prime condition” henceforth) in
which participants receive neither primary- nor secondary-
control primes, we expect the results from studies 1 and 2 to be
replicated. Specifically, when an upward-looking angle is used
to display a product, participants with an ought-self (vs. ideal-
self) will report more positive product evaluations. When a
downward-looking angle is employed, participants with an
ideal-self (vs. ought-self) will generate more favorable product
evaluations.
When participants receive either primary- or secondary-
control primes, we expect the interaction pattern identified in
earlier studies (the same as in the no prime condition) to be
replicated or reversed, dependent on whether the primed control
mechanism is consistent with participants' prevailing control
strategies. To explain, offering participants a control mecha-
nism that is redundant and consistent with their prevailing
control strategies (priming primary-control [secondary-control]
mechanism among participants with an ideal-self [ought-self])
should replicate the interaction patterns identified in studies 1
and 2 (as shown in the no prime condition). By contrast,
inducing participants with a control mechanism that is different
from their prevailing control strategies (priming primary-
control [secondary-control] mechanism among participants
with an ought-self [ideal-self]) should reverse the interaction
we identified earlier (also shown in the no prime condition). We
next present our rationale.
Upward-looking angle
According to our theory, an effort to achieve secondary
control explains why individuals with an ought-self favorably
evaluate a product displayed from an upward-looking angle. On
515
the other hand, individuals with an ideal-self are expected to
evaluate a product portrayed with an upward-looking angle less
favorably because they adopt a primary-control mechanism.
Unless otherwise mentioned, the discussion below applies to an
upward-looking camera angle.
Ideal-self
Asking participants with an ideal-self to pursue primary
control should reinforce the control mechanism typically
employed by these individuals (i.e., primary control). Our
theorizing suggests that an upward-looking angle elicits less
favorable product evaluations when individuals with an ideal-
self pursue primary control. However, asking participants with
an ideal-self to pursue secondary control will orient them
towards a control strategy not spontaneously associated with an
ideal-self. In line with our theorizing that a secondary-control
mechanism results in self-alignment with powerful entities and
an upward-looking angle conveys such meaning, individuals
with an ideal-self (when pursuing secondary control) will form
more favorable evaluations towards products depicted with an
upward-looking angle. In sum, among participants with an
ideal-self who view an upward-looking camera angle, those
with a secondary-control prime are expected to form more
favorable product evaluations than those in the primary-control
and no prime conditions.
Ought-self
Asking participants with an ought-self to pursue secondary
control should reinforce the control mechanism typically
employed by these individuals (i.e., secondary control). Based
on our earlier theorizing, an upward-looking angle results in
more favorable evaluations when individuals with an ought-self
are guided by secondary control. However, asking participants
with an ought-self to pursue primary control, a control
mechanism not spontaneously associated with an ought-self,
orients these individuals towards influencing the environment
and sustaining personal agency. Our earlier theorizing suggests
that this effort to pursue primary control results in less favorable
evaluations of products depicted with an upward-looking angle.
Therefore, among participants with an ought-self, those in the
secondary-control prime and those in the no prime conditions
are expected to form more favorable product evaluations than
those exposed to the primary-control prime. Therefore, when
evaluating a product portrayed with an upward-looking angle,
we expect:
H5a. Among participants with an ideal-self, those in the
secondary-control prime condition should report more favor-
able product evaluations than those in the primary-control and
no prime conditions after viewing an upward-looking camera
angle.
H5b. Among participants with an ought-self, those in the
primary-control prime condition should report less favorable
product evaluations than those in the secondary-control and no
prime conditions after viewing an upward-looking camera
angle.
516 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
Downward-looking angle
To begin, replicating the results of studies 1 and 2, we expect
that in the no prime condition (in which participants do not
receive either the primary- or secondary-control primes),
participants with an ideal-self (vs. an ought-self) will develop
more favorable attitudes towards products with a downward-
looking angle. Unless otherwise mentioned, the discussion
below applies to a downward-looking camera angle.
Ideal-self
Similar to the rationale we presented in the earlier section,
asking participants with an ideal-self to pursue primary control
reinforces the control mechanism typically employed by these
individuals (i.e., primary control). And, based on our earlier
findings, individuals with an ideal-self who adhere to a primary
control strategy form favorable evaluations when viewing a
product from a downward angle. However, asking participants
with an ideal-self to pursue secondary control, a control
mechanism not spontaneously associated with an ideal-self,
invites these individuals to focus on self-alignment with
powerful entities. Our earlier theorizing suggests that this effort
to pursue secondary control leads to less favorable evaluations
of products depicted from a downward-looking angle. There-
fore, among participants with an ideal-self, those in the primary-
control prime and those in the no prime conditions are expected
to form more favorable product evaluations than those in the
secondary-control prime condition after viewing a product from
a downward angle.
Ought-self
Asking participants with an ought-self to pursue secondary
control should reinforce the control mechanism typically
employed by these individuals (i.e., secondary control). Based
on our theorizing, an ought-self relies on a secondary control
mechanism leading to less favorable evaluations of products
portrayed from a downward-looking camera angle. However,
asking participants with an ought-self to pursue primary control
results in participants relying on a control mechanism not
spontaneously associated with an ought-self. In line with our
theory, this effort to pursue primary control leads to more
favorable evaluations of products portrayed from a downward-
looking angle. Therefore, when viewing a product from a
downward angle, participants with an ought-self in the
secondary-control prime and in the no prime conditions are
expected to form less favorable product evaluations than those
with an ought-self exposed to the primary-control prime. Thus,
when evaluating a product portrayed from a downward-looking
angle, we expect:
H6a. Among participants with an ideal-self, those in the primary-
control and no prime conditions should report more favorable
product evaluations than those in the secondary-control condition
after viewing a downward-looking camera angle.
H6b. Among participants with an ought-self, those in the
secondary-control and no prime conditions should report less
favorable product evaluations than those in the primary-control
condition after viewing a downward-looking camera angle.
Method
The study was 3 (prime: primary-control vs. secondary-
control vs. no prime) × 2 (self-concept: ideal-self vs. ought-
self) × 2 (camera angle: downward-looking vs. upward-looking)
between-subjects factorial design. One hundred seventy-one
undergraduate students took part in the experiment.
The procedure and the task used in this study were the same
as those employed in study 1 (i.e., cereal advertising) with the
addition of a control mechanism priming task. The control
mechanism priming task was introduced prior to exposing
participants to the target ad. Specifically, participants assigned
to the primary control condition were asked to list three things
that they have done in the past in which they achieved control of
things by actively influencing existing realities (e.g. “I buy
decorations to change the appearance of my room and to make it
more enjoyable to be in.”). Participants in the secondary-control
condition were asked to write down three things they have done
in the past in which they achieved control of things by adapting
themselves to existing realities (e.g. “I buy the latest fashion and
high heels in order to fit in with the “club scene” and my girl
friends when we go out.”). Participants in the no prime
condition did not receive either priming task.
To provide evidence for the control priming manipulation,
two independent coders (r = .91), who had received training on
different types of control mechanisms, rated the situations
recalled by participants using the following five-point scale
anchored by: 1 = “in general these experiences involved
primary-control strategies” to 5 = “in general these experiences
involved secondary-control strategies.” Results of a one-way
ANOVA with type of control manipulation as the independent
variable (only performed on participants who received control
manipulation) revealed a main effect of type of control such that
participants in the primary-control (secondary-control) condi-
tion recalled more primary-control (secondary-control) experi-
ences (Mprimary = 2.46 vs. Msecondary = 3.49, F(1,108) = 52.05,
p b .01), confirming the success of our manipulation.
Results
The data were analyzed as a 3 (prime: primary-control vs.
secondary-control vs. no prime) × 2 (self-concept: ideal-self vs.
ought-self) × 2 (camera angle: downward-looking vs. upward-
looking) between-subjects factorial design. See Fig. 3 for a
depiction of the interaction.
Product evaluations
An analysis of product evaluations (alpha = .91) revealed a
three-way interaction effect of control prime, self-concept and
camera angle (F(2, 159) = 3.08, p b .05). Planned contrasts
performed on product evaluations confirmed our predictions.
In the no prime condition, our earlier findings from studies 1
and 2 were replicated, and we found the same two-way interaction
between camera angle and self-concept (F(1, 159) = 9.36, p b .01).
 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
Fig. 3. Study 3, product evaluations as a function of camera angle, self-concept, and control strategy.
Specifically, when an upward-looking angle was employed,
participants with an ought-self (vs. ideal-self) generated more
positive evaluations (Mideal = 3.38 vs. Mought = 4.37, F(1, 159)=
5.87, p b .02). When a downward-looking angle was viewed,
participants with an ideal-self (vs. ought-self) reported
marginally more favorable product evaluations (Mideal = 4.29
vs. Mought = 3.52, F(1, 159) = 3.61, p b .06). These results
provide further support for hypotheses 1 and 2.
Consistent with hypothesis 5a, when an upward-looking
angle was employed, participants with an ideal-self generated
marginally more favorable product evaluations in the second-
ary-control condition, than the primary-control and no prime
conditions (Mprimary = 3.64 vs. Msecondary = 4.50, F(1, 159) =
3.64, p b .06; Msecondary = 4.50 vs. Mno prime = 3.38, F(1, 159) =
7.48, p b .01). Meanwhile, there was no significant difference in
product evaluations between the latter two conditions (Mprimary =
3.64 vs. Mno prime = 3.38, F b 1). Thus hypothesis 5a was
supported.
As stated in hypothesis 5b, when an upward-looking angle
was viewed, participants with an ought-self reported more
positive product evaluations in the secondary-control and no
prime conditions, as compared with primary-control condition
(Mprimary = 3.51 vs. Msecondary = 4.51, F(1, 159) = 5.53, p b .02;
Mprimary = 3.51 vs. Mno prime = 4.37, F(1, 159) = 4.29, p b .04). At
517
the same time, there was no significant difference in product
evaluations between the secondary-control and no prime
conditions (Msecondary = 4.51 vs. Mno prime = 4.37, F b 1). Thus
hypothesis 5b was supported.
As predicted by hypothesis 6a, when a downward-looking
angle was employed, participants with an ideal-self reported
more favorable product evaluations in the primary-control and
no prime conditions, compared to their counterparts in the
secondary-control condition (Mprimary = 4.48 vs. Msecondary = 3.46,
F(1, 159) = 5.55, p b .02; Msecondary = 3.46 vs. Mno prime = 4.29, F(1,
159)= 4.08, p b .05). At the same time, there was no difference
between the primary-control and no prime conditions (Mprimary =
4.48 vs. Mno prime = 4.29, F b 1). Thus, hypothesis 6a was
supported.
As suggested by hypothesis 6b, when a downward-looking
angle was viewed, participants with an ought-self had more
favorable product evaluations in the primary-control condition, as
compared to their counterparts in the secondary-control and no
prime conditions (Mprimary = 4.37 vs. Msecondary = 3.48, F(1,159) =
4.74, p b .03; Mprimary = 4.37 vs. Mno prime = 3.52, F(1, 159) =
4.32, p b .04). Meanwhile, there was no significant difference
in product evaluations between the latter two conditions
(Msecondary =3.48 vs. Mno prime = 3.52, F b 1). Thus hypothesis 6b
was also supported.
518 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
Discussion
The results of study 3 provided support for our claims that
individuals with an ideal-self and ought-self employ different
control mechanisms, primary versus secondary, underlying the
results obtained in studies 1 and 2. Specifically, we found that
when we induced participants to engage in a control mechanism
that was consistent with their own (priming ideal-self [ought-
self] with primary [secondary] control mechanism), the results
from studies 1 and 2 were replicated. By contrast, when
participants were induced to pursue a control mechanism that
was inconsistent with their own (priming ideal-self [ought-self]
with secondary [primary] control mechanism), the results from
1 and 2 (as well as in the no prime condition) were reversed.
General discussion
Across three studies, using different product categories, we
find converging evidence for the hypothesized interaction
between self-concept and camera angle. Specifically, when an
upward-looking camera angle is utilized, participants generated
more favorable product evaluations if their ought-self (the
representation of the set of attributes that one, or someone else,
believes one should possess) is accessible. Conversely, when a
downward-looking camera angle is employed, participants had
more favorable product evaluations if their ideal-self (the
representation of the set of attributes that one, or someone else,
believes one would ideally possess) is accessible. We further
reveal how a feeling of control (primary versus secondary)
explains these findings. Specifically, a product photographed
from an upward-looking camera angle conveys the impression
of power or potency (Kraft, 1987; Meyers-Levy & Peracchio,
1992). This perception resonates well among participants with
an ought-self (vs. an ideal-self) as they tend to employ
secondary control of the product and generate thoughts related
to adapting to powerful others (e.g., how to align with a social
group associated with the product, or adapt themselves to using
the product). Thus, they evaluate the product photographed
from an upward-looking angle more favorably. By contrast, a
product photographed from a downward-looking camera angle
is less threatening and more malleable (Kraft, 1987; Meyers-
Levy & Peracchio, 1992). Such a perception may resonate well
among participants with an ideal-self (vs. an ought-self) as they
engage in primary control of the product and generate thoughts
relaying the perception of personal agency and autonomy (e.g.,
the employment of the product to solve existing problems).
Accordingly, they evaluate a product depicted from a downward-
looking angle more favorably. In an unreported study, we
examined the effect of involvement and found that under high
processing motivation, the two-way interaction between camera
angle and self-concept was replicated. However, the results
dissipated under low involvement situations. This finding
underscores that the control mechanism we are proposing is an
effortful process and requires a higher level of involvement and
motivation.
Our findings contribute to the literature on consumer
aesthetic experiences. Aesthetic experiences entail a wide
range of visual properties that are either integral to or external
to the product. One line of research in this area has shown that
aesthetic elements internal to the product, e.g., product design,
play a critical role in determining a product's sales success
(Hoegg & Alba, 2008; Simonson & Schmitt, 1997). Along
these lines, stylistic properties (the manner in which visual and
production elements are displayed in ads) have been consis-
tently found to affect consumers' product evaluations, inde-
pendent of the actual attributes or benefits offered by the
product. For example, in this special issue on aesthetics, Hoegg
et al. (2010) examined how product design and product
attributes interact to affect consumers' product evaluations.
Our studies offer new insights to this stream of aesthetic
research in several ways.
First, our findings demonstrate the critical role that self-
concept (in our case the ideal-self vs. ought-self) plays in
determining the effectiveness of the stylistic properties of a
visual image. Current research on stylistic properties mainly
focuses on the aesthetic elements per se and little research has
incorporated consumers' individual characteristics such as self-
concept in pinpointing the effect of stylistic properties. For
instance, much research has shown that an upward-looking
angle (vs. a downward-looking angle) is generally more
effective as it is associated with power and potency. Our
findings suggest that consumers' self-concept should be
considered to understand the effect of camera angle. At a
broader level, the incorporation of consumers' characteristics in
understanding the effect of stylistic properties is consistent with
a general persuasion framework which claims that persuasion
should be understood across message factors (i.e., stylistic
properties in our case), situational factors (e.g., processing
motivation) and message processor factors (e.g., self-concept)
(Petty & Cacioppo, 1986).
Second, our studies also reveal the mechanism by which the
stylistic property of camera angle affects consumers. The results
of this research show that an ideal-self versus an ought-self
motivates consumers to recruit particular paths to achieve a
sense of control, which in turn impacts their perceptions of and
relationship with the product and, ultimately, their product
evaluations. This particular finding enhances our understanding
of aesthetic experiences by suggesting that such experiences can
be engaging ones, as these aesthetic elements can symbolize
different aspects of the consumer–product relationship, either
consumer-driven or product-driven, which might be consistent
with consumers' inherent control preferences.
Our results also have implications for practitioners. When
promoting a product, marketers should not choose stylistic
properties in an ad randomly. When deciding which camera angle
to utilize, for example, they should consider the characteristics of
consumers—are they motivated by ideals or oughts? If for some
reason a particular camera angle must be chosen (e.g., for a better
portray of the products' features, or simply for artistic reasons),
practitioners must be careful to ensure the appropriate self-
concept is elicited among consumers. The elicitation of the self-
concept can be achieved by putting the ad in a context that
promotes either an ideal- or an ought-self, or crafting ad claims
that emphasize these different selves.
 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520
There are several potential paths for future research. First, one
possible direction of future research would be to examine the
effect of the “strength” of self-concept on stylistic properties. As
an example, future research can look into how the size of self-
discrepancy (Higgins, 1987) might affect our findings. It could be
that our theorizing regarding camera angle and self-concept holds
true only if individuals' self-discrepancy (e.g., ideal vs. actual
self) passes a certain threshold. Second, we examine the effect of
self-concept in the context of one particular stylistic property,
camera angle. One path for future research may involve
examining how other stylistic properties (e.g., orientation of
objects, visual perspective, etc.) are influenced by self-concept.
For instance, a consumer with an ideal-self may prefer an actor-
like visual perspective (e.g., Meyers-Levy & Peracchio, 1996)
because this perspective offers them a stronger feeling of control
over the environment. On the other hand, a consumer with an
ought-self may prefer an observer-like visual perspective because
this perspective allows them to align themselves with the
environment. Third, our focus in the paper is product evaluations.
One path to extend our findings is by studying dependent
variables other than product evaluations. For example, in this JCP
special issue on aesthetics, Madzharov and Block (2010)
examined the effect of product unit image and found that the
number of units displayed on the product package (i.e., snack
items) influences consumers' perception of snack quantity
contained in the package, as well as their subsequent snack
consumption. Likewise, to extend our work, future research could
be conducted to investigate whether the findings of our research
will carry over and affect consumers' behaviors while shopping
and making purchase decisions. Alternatively, future research
could present participants with products depicted from a variety of
camera angles and could let them choose the angle most
appropriate for enhancing the appeal of the product. Finally, our
paper examines the role of self-concept in processing stylistic
properties. Another possible route to enhance the understanding
of stylistic properties is to examine other individual difference
variables. For instance, in this special issue on athletics, Meyers-
Levy and Zhu (2010) examined the role of gender and need for
cognition on consumers' inferences of aesthetic features. In future
research, these and other related individual difference variables
should be investigated. For example, individuals with alternative
cognitive styles (e.g., Wyer et al., 2008; Childers & Jiang, 2008)
might react differently to stylistic properties.
Acknowledgments
The authors wish to thank the Editor C.W. Park, Associate
Editor, and three anonymous reviewers for their suggestions
and guidance. The authors also thank Melissa Bublitz and Elise
Boutchard for their comments on earlier versions of the paper.
References
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable
distinction in social psychological research: Conceptual, strategic and
statistical considerations. Journal of Personality and Social Psychology, 51,
1173−1182.
519
Chan, H., Wan, L. C., & Sin, L. Y. (2009). The contrasting effects of culture on
consumer tolerance: Interpersonal face and impersonal fate. Journal of
Consumer Research, 36, 292−304.
Childers, T. L., & Jiang, Y. (2008). Neurobiological perspectives on the nature
of visual and verbal processes. Journal of Consumer Psychology, 18,
264−269.
Chipperfield, J. G., & Perry, R. (2006). Primary- and secondary-control
strategies in late life: Predicting hospital outcomes in men and women.
Health Psychology, 25, 226−236.
Crowe, E., & Higgins, E. T. (1997). Regulatory focus and strategic inclinations:
Promotion and prevention in decision making. Organizational Behavior and
Human Decision Processes, 69, 117−132.
Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect.
Psychological Review, 94, 319−340.
Hoegg, J., & Alba, J. (2008). A role for aesthetics in consumer psychology. In C. P.
Haugtvedt, P. M. Herr, & F. Kardes (Eds.), Handbook of consumer psychology
(pp. 733−754). New York, NY: Lawrence Erlbaum Associates.
Hoegg, J., Alba, J. W., & Dahl, D. W. (2010). The good, the bad, and the ugly:
Influence of aesthetics on product feature judgments. Journal of Consumer
Psychology, 20(4), 419−430.
Kraft, R. N. (1987). The influence of camera angle on comprehension and
retention of pictorial events. Memory & Cognition, 15, 291−307.
Liberman, N., Idson, L. C., Camacho, C. J., & Higgins, E. T. (1999). Promotion
and prevention choices between stability and change. Journal of Personality
and Social Psychology, 77, 1135−1145.
Luomala, H. T., Kumar, R., Worm, V., & Singh, J. D. (2004). Cross-cultural
differences in mood-regulation: An empirical comparison of individualistic
and collectivistic cultures. Journal of International Consumer Marketing,
16, 39−62.
Madzharov, A. V., & Block, L. G. (2010). Effects of product unit image on
consumption of snack foods. Journal of Consumer Psychology, 20(4),
398−409.
Messaris, P. (1997). Visual persuasion: The role of images in advertising.
Thousand Oaks, CA: Sage.
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, 454−461.
Meyers-Levy, J., & Peracchio, L. A. (1996). Moderators of the impact of self-
reference on persuasion. Journal of Consumer Research, 22, 408−423.
Meyers-Levy, J., & Zhu, R. (2010). Gender differences in the meanings
consumers infer from music and other aesthetic stimuli. Journal of
Consumer Psychology, 20(4), 495−507.
Monga, A. B., & John, D. R. (2007). Cultural differences in brand extension
evaluation: The influence of analytic versus holistic thinking. Journal of
Consumer Research, 33, 529−536.
Peracchio, L. A., & Meyers-Levy, J. (2005). Using stylistic properties of ad
pictures to communicate with consumers. Journal of Consumer Research,
32, 29−40.
Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central
and peripheral routes to persuasion. New York, NY: Springer-Verlag.
Pham, M. T., & Avnet, T. (2004). Ideals and oughts and the reliance on affect
versus substance in persuasion. Journal of Consumer Research, 30,
503−518.
Phillips, B. J., & McQuarrie, E. F. (2002). The development, change, and
transformation of rhetorical style in magazine advertisements 1954–1999.
Journal of Advertising, 31, 1−13.
Phillips, B. J., & McQuarrie, E. F. (2004). Beyond visual metaphor: A new
typology of visual rhetoric in advertising. Marketing Theory, 4, 113−136.
Rothbaum, F., Weisz, J. R., & Snyder, S. S. (1982). Changing the world and
changing the self: A two-process model of perceived control. Journal of
Personality and Social Psychology, 42, 5−37.
Scott, L. M. (1994). Images in advertising: The need for a theory of visual
rhetoric. Journal of Consumer Research, 21, 252−273.
Scott, L. M., & Batra, R. (2003). Persuasive imagery. Mahwah, NJ: Erlbaum.
Simonson, A., & Schmitt, B. H. (1997). Marketing aesthetics: The strategic
management of brands, identity and image. New York, NY: the Free Press.
Skinner, E. A., & Chapman, M. (1984). Control beliefs in an action-perspective.
Human Development, 27, 129−133.
520 X. Yang et al. / Journal of Consumer Psychology 20 (2010) 508–520

Wang, K., & Peracchio, L. A. (2008). Reading pictures: Understanding the
stylistic properties of advertising images. In E. F. McQuarrie, & B. J.
Phillips (Eds.), Go figure: New directions in advertising rhetoric
(pp. 205−266). New York, NY: M.E. Sharpe Inc.
Weisz, J. R., Rothbaum, F. M., & Blackburn, T. C. (1984). Standing out and
standing in: The psychology of control in America and Japan. American
Psychologist, 39, 955−969.
Wyer, R. S., Jr., Hung, I. W., & Jiang, Y. (2008). Visual and verbal processing
strategies in comprehension and judgment. Journal of Consumer Psychology,
18(4), 244−257.
Zhao, M., Hoeffler, S., & Zauberman, G. (2007). Mental simulation and
preference consistency over time: The role of process- versus outcome-
focused thoughts. Journal of Marketing Research, 44(3), 379−388.
Zhu, R., & Meyers-Levy, J. (2009). The influence of self-view on context
effects: How display fixtures can affect product evaluations. Journal of
Marketing Research, 46, 37−45.