ArticlePDF Available

Compensating for Innovation: Extreme Product Incongruity Encourages Consumers to Affirm Unrelated Consumption Schemas

Authors:

Abstract and Figures

New products are often extremely incongruent with expectations. The inability to make sense of these products elevates anxiety and leads to negative evaluations. Although scholars have predominantly focused on combatting the negative response to extreme incongruity, we propose that extreme incongruity may have implications that extend beyond the category. We base our predictions on the concept of fluid compensation, which suggests that when people struggle to make sense of something, they will non‐consciously reinforce highly accessible schemas in unrelated domains. Four studies confirm that extreme incongruity encourages fluid compensation, such that it elevates preference for dominant brands (study 1), green consumption (studies 2 and 4), and ethnocentric products (study 3). We isolate the causal role of anxiety using moderation tasks and biometric feedback. Furthermore, we demonstrate that compensation has an immediate dampening effect on arousal intensity. Thus, if consumers can compensate before explicitly evaluating an extremely incongruent product, their evaluations tend not to be negative. Taken together, we document that extreme innovations encourage compensation, and in compensating, consumers can become more receptive to extreme innovations. This article is protected by copyright. All rights reserved.
Content may be subject to copyright.
Compensating for Innovation: Extreme Product Incongruity Encourages
Consumers to Afrm Unrelated Consumption Schemas
N
ukhet Taylor , and Theodore J. Noseworthy
York University
Accepted by S. Christian Wheeler, Editor; Associate Editor, Chris Janiszewski
New products are often extremely incongruent with expectations. The inability to make sense of these prod-
ucts elevates anxiety and leads to negative evaluations. Although scholars have predominantly focused on
combating the negative response to extreme incongruity, we propose that extreme incongruity may have
implications that extend beyond the category. We base our predictions on the concept of uid compensation,
which suggests that when people struggle to make sense of something, they will nonconsciously reinforce
highly accessible schemas in unrelated domains. Four studies conrm that extreme incongruity encourages
uid compensation, such that it elevates preference for dominant brands (study 1), green consumption (studies 2
and 4), and ethnocentric products (study 3). We isolate the causal role of anxiety using moderation tasks and
biometric feedback. Furthermore, we demonstrate that compensation has an immediate dampening effect on
arousal intensity. Thus, if consumers can compensate before explicitly evaluating an extremely incongruent
product, their evaluations tend not to be negative. Taken together, we document that extreme innovations
encourage compensation, and in compensating, consumers can become more receptive to extreme innova-
tions.
Keywords Schema incongruity; Fluid compensation; Compensatory consumption; Brand domi-
nance; Green consumption; Ethnocentrism; Product innovation; Anxiety
New products often challenge consumersexisting
beliefs about a product concept. For example,
Heinzs purple ketchup violated consumersexpec-
tations for the category given that people have only
ever known ketchup to be red, and the color red
relates to schematic expectations for the primary
ingredient in ketchup. The consequence of violating
product schemas has been well documented. A gen-
eral nding is that new products can stimulate a
favorable response in consumers, but only if con-
sumers do not have to abandon or recongure their
existing beliefs for the product category (Campbell
& Goodstein, 2001; Jhang, Grant, & Campbell, 2012;
Meyers-Levy & Tybout, 1989; Noseworthy, Cotte, &
Lee, 2011; Noseworthy & Trudel, 2011; Peracchio &
Tybout, 1996). If a new product is extremely incon-
gruent with existing set of beliefs, evaluations tend
to suffer. This nonmonotonic relationship between
the severity of incongruity and product evaluations
sets the basis for the well-documented schema con-
gruity effect (Meyers-Levy & Tybout, 1989).
At its core, the schema congruity effect is built
on the general notion that people actively seek
meaning (Bruner, 1990). Meanings are beliefs that
shape expectations and allow people to predict and
make sense of their experiences (Proulx & Inzlicht,
2012). However, sometimesas with Heinz purple
ketchupexperiences conict with these expecta-
tions and thus threaten ones sense of meaning. The
growing body of literature on meaning mainte-
nance suggests that the human motivation to main-
tain a sense of meaning is so compelling that when
a schema is sufciently threatened, people will
often compensate by afrming other beliefs in
memory (Heine, Proulx, & Vohs, 2006; Randles,
Proulx, & Heine, 2011). Such beliefs may reside
within the same domain, or even across domains
Received 17 December 2018; accepted 21 June 2019
Available online 28 June 2019
This research was based on part of the rst author's disserta-
tion, conducted under the guidance of the second author. This
research was funded by the Social Sciences and Humanities
Research Council of Canada (SSHRC insights grant #435-2017-
0835). The authors would like to thank Sean T. Hingston for his
assistance in the previous versions of this manuscript.
Correspondence concerning this article should be addressed to
N
ukhet Taylor, Schulich School of Business, York University, 99
Ian Macdonald Blvd, Toronto, ON M3J 1P3, Canada. Electronic
mail may be sent to nagar17@schulich.yorku.ca.
©2019 The Authors. Journal of Consumer Psychology published by Wiley
Periodicals, Inc. on behalf of Society for Consumer Psychology
This is an open access article under the terms of the Creative Commons
Attribution License, which permits use, distribution and reproduction in
any medium, provided the original work is properly cited.
1057-7408/2020/1532-7663/30(1)/7795
DOI: 10.1002/jcpy.1127
referred to as uid compensation (Heine et al., 2006;
Randles et al., 2011). For example, having a person
do something as benign as read semantically inco-
herent word pairings can lead them to double
down on their belief in morality (Randles et al.,
2011). Thus, if extreme incongruity indeed threatens
existing beliefs, extremely incongruent products
may have unintended consequences that extend
beyond the product category.
In testing this possibility, we make several
important advances. First, we unify schema con-
gruity and meaning maintenance literatures under
a common emotional mechanism: arousal-induced
anxiety. Specically, schema congruity literature
suggests that extreme incongruity lends itself to a
heightened physiological arousal state which mani-
fests as anxiety (Noseworthy, Muro, & Murray,
2014), and meaning maintenance literature posits
that arousal-induced anxiety encourages consumers
to compensate by afrming unrelated belief systems
(Proulx & Heine, 2008). In bridging these disparate
literatures, we demonstrate that consumers may
cope with the anxiety that stems from extreme
incongruity by afrming established beliefs in unre-
lated consumption domains. Furthermore, using
actual biometric feedback, we add to meaning
maintenance literature by providing the rst empir-
ical evidence that the act of uid compensation
results in an immediate dampening of physiological
arousal. Finally, we bring our lens back to the
schema congruity effect and explore what compen-
sation means for consumersproduct evaluations.
Given that the negative reaction to extremely incon-
gruent products is the result of arousal-induced
anxiety (Noseworthy et al., 2014) and that compen-
sation alleviates this tension, we demonstrate that
allowing consumers to compensate in unrelated
domains can mitigate the well-documented nega-
tive response to extreme incongruity. Taken as a
whole, we document that extreme incongruity
encourages compensation in unrelated consumption
domains, and compensation, in return, makes con-
sumers more open to extremely incongruent prod-
ucts.
Conceptual Development
The Schema Congruity Effect
The schema congruity effect was originally pro-
posed as a dynamic relationship between arousal
and affective judgment (Mandler, 1982). Mandler
posited that incidental affect can result from an
autonomic physiological response to seeing an
object that violates existing beliefs. His thesis cen-
tered on the notion that although people generally
like things that match their expectations, moderate
incongruity can be arousing and thus intriguing.
Furthermore, because moderate incongruity can be
resolved with minimal effort, it tends to result in
favorable evaluations (Mandler, 1982; Meyers-Levy
& Tybout, 1989). In instances of extreme incon-
gruity, expectations can be violated to such an
extent that the object cannot be resolved without
reconstructing the existing schema, or forming an
entirely new schema (Mandler, 1982). This tends to
coincide with anxiety, which often results in unfa-
vorable evaluations (Mandler, 1982; Noseworthy
et al., 2014). It is in this sense that extreme product
incongruity can threaten a product concept and
may even render it meaningless. While extant work
has explored several boundary conditions to the
schema congruity effect (Campbell & Goodstein,
2001; Galli & Gorn, 2011; Maoz & Tybout, 2002;
Meyers-Levy & Tybout, 1989; Noseworthy, Murray,
& Di Muro, 2018; Peracchio & Tybout, 1996; Pri-
ester, Godek, Nayakankuppum, & Park, 2004), the
core-dependent variable has invariably been the
evaluations of the target product. Researchers have
yet to consider that there may be more to extreme
incongruity than what it means for the product in
question.
A critical insight from the schema congruity liter-
ature is that the affective response to product incon-
gruity is determined by the consumers ability to
resolve the product within an existing set of beliefs.
The positive affect that results from successful reso-
lution not only increases task satisfaction (Maoz &
Tybout, 2002), but also elevates evaluations for
incongruent products (Jhang et al., 2012). The
inability to resolve a product elevates autonomic
arousal to the point of anxiety, resulting in a nega-
tive affective response (Noseworthy et al., 2014).
These ndings are best characterized as within-do-
main compensation, whereby a consumer resolves
an incongruent product within their beliefs about
the category. This view holds, for example, that
consumers who see a smartphone that challenges
their preconceptions will either update their beliefs
or develop an entirely new set of beliefs by subtyp-
ing the phone as a special case.
The within-domain focus of schema congruity
theory predominantly stems from Mandlers (1982)
adoption of classic notions of assimilation and
accommodation (Piaget, 1960). Assimilation in the
sense that people try to resolve the source of
incongruity within an activated schema (e.g., try-
ing to reconcile the rst-generation tablet within a
78 Taylor and Noseworthy
laptop schema); accommodation in the sense that
when the former fails, people will then modify the
schema or develop a new schema (e.g., looking at a
tablet as a laptop without a keyboard, and then
later galvanizing the tablet concept once marketers
coin a name for the category). It was Piaget (1960)
who originally proposed that things that do not
make sense cause tension and that effort to regain a
sense of meaning constitutes a fundamental devel-
opmental motive. That is, people generally have a
sense-making impulse. This can operate at the
extremes as a need for coherence (Antonovsky,
1979) or a need for closure (Kruglanski & Webster,
1996), but in general, it is a basic human motivation
to retain a sense of meaning when faced with infor-
mation that is inconsistent with a given set of
beliefs. The importance of understanding that the
within-domain focus is more a by-product of Man-
dlers (1982) theoretical lens than a necessary
parameter of his model comes from an entirely dif-
ferent, but related stream of research on meaning
maintenance.
Meaning Maintenance Model
The meaning maintenance model (MMM)
expands on the notion of a sense-making impulse
by proposing that when a persons beliefs are suf-
ciently violated, they will nonconsciously attempt
to regain a sense of meaning. Meaning, in this con-
text, refers to the ability to conrm expected rela-
tionships (Proulx & Inzlicht, 2012). Building on
Piagets (1960) concepts of assimilation and accom-
modation, the MMM posits a third way in which
people can deal with expectancy violations: uid
compensation (McGregor, Prentice, & Nash, 2012;
Proulx & Heine, 2008; Steele, 1988). Fluid compen-
sation is predicated on the assumption that schemas
are functionally interchangeable, such that afrming
one schema (e.g., ones culture or morality) can sat-
isfy the inability to resolve another (e.g., a percep-
tual anomaly; Proulx & Heine, 2008). Researchers
have shown that this can occur without conscious
activation (Proulx, Heine, & Vohs, 2010). For exam-
ple, Proulx and Heine (2008) found that people
tend to afrm cultural norms after they witness a
perceptual anomaly, even without them consciously
noticing the anomaly. In keeping with the smart-
phone example, this view holds that American con-
sumers who see something that challenges their
preconceptions for the smartphone category may
instead choose to afrm their cultural identity and
buy Americanas a way of reinforcing something
that does make sense.
The above example relates to a second premise
of the MMM, which is that uid compensation is
instigated by an autonomic physiological response
to things that are difcult to make sense of, which
manifests broadly as nonconscious anxiety (Proulx
& Heine, 2008; Proulx & Inzlicht, 2012). This is con-
sistent with evidence that anxiety can impact peo-
ples judgments in unrelated domains, but only
when they are not consciously aware of the source
of tension (Raghunathan, Pham, & Corfman, 2006).
Conscious awareness of the source dampens the
reparative nature of afrming in unrelated domains
unless these domains have some strong resem-
blance to the original source of the anxiety-ridden
experience (Tritt, Inzlicht, & Harmon-Jones, 2012).
Thus, it is not that people are resolving the expec-
tancy violation by compensating in unrelated
domains, but rather are engaging in a nonconscious
strategy aimed at reducing the aversive tension that
stems from expectancy violations (Major & Town-
send, 2012). Further to this point, research conrms
that arousal-induced anxiety predicts a specic
need to afrm meaning, which cannot be satised
with positively valenced stimuli decoupled from
the process of regaining meaning (McGregor et al.,
2012). Collectively, these ndings point to the
intriguing possibility that consumers may attempt
to alleviate the tension that stems from extreme
incongruity by afrming belief systems in unrelated
domains.
Hypotheses
To recap, extant work on schema congruity doc-
uments that extreme incongruity elicits an auto-
nomic physiological response that manifests as
anxiety (Noseworthy et al., 2014). Meaning mainte-
nance literature suggests the arousal-induced anxi-
ety that originates from expectancy violations
engenders a nonconscious need to regain meaning,
which can be accomplished by afrming schemas
unrelated to the violationcoined as uid compensa-
tion (Proulx & Heine, 2008). Both theories rely on
the motivational aspects of the aversive tension
(Noseworthy et al., 2014; Proulx & Inzlicht, 2012).
However, it is not that people can only resolve
incongruent products within a product schema, but
that they just tend to do so when the violation is
moderate enough to make sense of (Noseworthy
et al., 2014). In the case of extreme product incon-
gruity, the violation cannot be resolved within pre-
existing beliefs. It is our contention that consumers
will use different strategies to alleviate this tension,
such as uid compensation. Stated formally,
Compensating for Innovation 79
H1: Consumers will compensate by afrming
existing beliefs in unrelated domains after view-
ing an extremely incongruent, as opposed to a
congruent, product.
H2: The effect of extreme product incongruity on
compensation is mediated by state anxiety, such
that viewing an extremely incongruent (vs. con-
gruent) product elevates state anxiety, which, in
turn, leads to compensation in unrelated
domains.
Our nal step was to explore what the act of
compensation may mean for extremely incongruent
products. Indeed, consumers are not forced to
explicitly evaluate products, so there is the question
of whether they would naturally compensate if
given the chance, and if so, whether the act of com-
pensation would have any lasting effects on prefer-
ence. There is reason to believe that it would.
Recent evidence suggests that alleviating the auto-
nomic response to extremely incongruent products
can lower anxiety and thus elevate evaluations
(Noseworthy et al., 2014). Given that uid compen-
sation is believed to be a nonconscious strategy to
alleviate the tension that arises from expectancy
violations (Proulx & Heine, 2008), it follows that
consumers who engage in compensatory behavior
prior to explicitly evaluating the product would
respond more favorably. Stated formally,
H3: The time at which consumers compensate
moderates the impact of schema congruity on
product evaluations, such that consumers will
evaluate extremely incongruent products more
favorably after (vs. before) they can compensate
in unrelated domains. In contrast, time of com-
pensation will have no impact on the evaluation
of congruent products.
In sum, if extreme product incongruity violates
expectations to such an extent that it instigates anx-
iety (Noseworthy et al., 2014), and if the anxiety
caused by expectancy violations creates a noncon-
scious need to regain meaning by afrming preex-
isting beliefs (Proulx & Heine, 2008), then extremely
incongruent products may elevate consumerspref-
erences for products that can afrm unrelated
beliefs. Furthermore, if compensatory behaviors
indeed have an immediate dampening effect on
arousal-induced anxiety, then the act of compensa-
tion should ameliorate the negative response to
extreme product incongruity. Researchers exploring
uid compensation have yet to examine the impact
of compensation on the expectancy violation. The
following studies were designed to test these pre-
dictions.
Overview of the Studies
In focusing on the compensation process, we
crossed distinct-dependent variables which unite
under the notion that they are highly accessible
belief systems. Some of the more notable beliefs
explored as compensatory domains are strong
semantic associations (Randles et al., 2011), ethical-
ity (Proulx & Heine, 2008), and cultural identity
(Proulx et al., 2010). In adapting these paradigms,
we focus on brand dominance (study 1), green con-
sumerism (studies 2 and 4), and ethnocentric prefer-
ence (study 3). Each study shows the underlying
role of anxiety in eliciting compensatory acts. Study
2 elaborates on the mechanism by having con-
sumers attribute their arousal to a salient source.
Study 3 directly moderates anxiety using a relax-
ation task and further shows that compensation has
an immediate dampening effect on psychological
arousal. Study 4 brings our inquiry full circle by
showing that because compensation can immedi-
ately dampen arousal intensity, allowing consumers
to compensate prior to making an explicit evalua-
tive judgment eliminates the negative response to
extreme incongruity.
In all studies, we validate that the extremely
incongruent products are indeed extreme. Speci-
cally, we conduct pilot studies to conrm prior
work that extremely (vs. moderately) incongruent
products cannot be resolved within existing beliefs,
which corresponds to elevated anxiety compared
to congruity (see Appendix S1, pp. 2230). These
tests also conrm that when a schema violation is
not severe enough to induce anxiety (i.e., moderate
incongruity), people tend not to compensate. We
then simplify the design of our main studies to
include the two levels of product congruity that is
the focus of our hypotheses (i.e., congruity and
extreme incongruity). The main studies also con-
rm that extreme incongruity is perceived less typ-
ical, is less likely to be resolved, and receive less
favorable evaluations than congruity. However,
unlike traditional work in this area, the contribu-
tion is not in empirically demonstrating con-
sumersperceptions of extreme incongruity, but
more so to isolate that we are truly tapping
extreme product incongruity and then to test for
downstream compensatory effects. Nevertheless,
80 Taylor and Noseworthy
these measures allowed us to conrm that com-
pensatory acts are the result of arousal-induced
anxiety that originates from extreme incongruity.
Specically, given that variations in state anxiety
do not alter perceived typicality and resolution
(Noseworthy et al., 2014), we expected these mea-
sures to not be impacted by anxiety-relevant fac-
tors throughout. In contrast, the negative
evaluations of extremely incongruent products
have a direct relationship with felt anxiety (Man-
dler, 1982; Noseworthy et al., 2014). Thus, in stud-
ies that expand on the role of anxiety, we further
elaborate on the impact of these manipulations on
product evaluations. Finally, we provide additional
analyses pertaining to state and trait anxiety in the
Appendix S1 (p. 21).
Study 1
Fluid compensation occurs at a nonconscious level,
such that consumers can double down on semantic
associations in unrelated schemas after experiencing
an expectancy violation (Randles et al., 2011). To
this point, one of the strongest semantic associa-
tions to a given product category is the most domi-
nant brand (e.g., Campbells in soup category; Herr,
Farquhar, & Fazio, 1996). Brand dominance is
dened as the strength of the directional associa-
tion between the parent category and the branded
product(Herr et al., 1996, 135). As such, category
dominant brands are often named earlier, recalled
more frequently, classied faster, and recognized
sooner upon the presentation of the product cate-
gory (Bagga, Noseworthy, & Dawar, 2016; Herr
et al., 1996). In what follows, study 1 thus sets the
stage for the current investigation by exploring an
important, yet previously unanticipated, down-
stream consequence of extreme incongruity: enhanc-
ing the choice of category dominant brands in
unrelated domains.
Method
Participants and Design. An a priori sample
was estimated based on the standard type 1 error
rate (a=.05), estimated power (1 b=0.80), and
the ability to identify a medium-to-small effect
should one exist (d=0.35; Cohen, 1988). The end
result was a sample (N) of two hundred and ten
participants (45% females; M
age
=37.5), recruited
through Amazons Mechanical Turk, and paid $1
for their participation. Participants were randomly
assigned to one of two conditions in a 2 (product
congruity: congruent vs. extreme incongruity)
between-subjects factorial design.
PretestBrand Dominance. A pretest (N=198)
was conducted to determine product categories
with a single category dominant brand (see
Appendix S1 for details). The end result was ve
categories with an identiable dominant brand
(dominant brands in parentheses): soup (Camp-
bells), smartphone (Apple), battery (Duracell), vac-
uum cleaner (Dyson), and coffee (Starbucks). These
categories were selected for the compensation task.
StimuliProduct Congruity. The product cate-
gory chosen for the congruity manipulation was
facial tissues. This category is not only familiar to
consumers, but also has a strong perceptual proto-
typicality in that the standard color of the category
(i.e., white) has varied little over time. This gave us
an opportunity to incorporate a classic schema con-
gruity manipulation (i.e., non-normative color) into
our design (Jhang et al., 2012; see Appendix S2),
which, notably, was also used by prior work to
explore the impact of product congruity on state
anxiety (Noseworthy et al., 2014).
Procedures. Participants took part in the study
through their individual computers. All participants
were informed that they were participating in a
consumer survey designed to explore consumers
opinions about a new facial tissue product. The
facial tissue category was explicitly conveyed to
evoke the necessary schema (Meyers-Levy & Tyb-
out, 1989). Participants were then directed to the
next page, where they were presented with the pic-
ture of the facial tissue which was either congruent
(i.e., white) or extremely incongruent (i.e., black)
with the category.
Participants began the questionnaire by respond-
ing to 10 randomized items (anchored: 1 =not at
all,7=very much), seven of which capturing over-
all evaluations (e.g., is likeable,”“is appealing),
and the remaining three capturing the perceived
typicality of the facial tissue (e.g., is common;
Campbell & Goodstein, 2001). Participants then
responded to 20 randomized items, 10 indicating
their trait anxiety (e.g., I am nervous and restless,
I am a steady person;anchored: 1 =almost never;
4=almost always), and 10 indicating their current
state of anxiety (e.g., tense,”“nervous;anchored:
1=not at all;4=very much so; Spielberger &
Reheiser, 2009). Consistent with prior work, the
basic idea was to control for participantstrait anx-
iety levels while testing state changes (Noseworthy
et al., 2014). Resolution was captured with two 7-
point scale items that indicated participantsability
to make sense (makes no sense/makes sense),
Compensating for Innovation 81
and understand the rationale of the facial tissue
(disagree/agree;Jhang et al., 2012). Participants
also typed their open-ended opinions about the
product, which served to reinforce the explicit
measure of resolution (Jhang et al., 2012; see
Appendix S1 page 5 for details of the coding pro-
cedure).
Participants were then thanked for reviewing
the facial tissue and told that we wanted to gener-
ate a general consumption prole on them before
they nalized the survey. In reality, this served as
the main compensation task. Specically, partici-
pants responded to the following question for ve
unrelated product categories: Consider if you were
going to buy a product from [Insert product category
name] today. Which brand would be your top choice
(please select Otherif your preferred brand does not
appear in the list)?In accordance with the brand
dominance pretest, the product categories piped
into the text were soup, battery, vacuum cleaner,
smartphone, and coffee categories. For each cate-
gory, participants were presented with a choice set
consisting of 10 brand names and their logos
(taken from the brand dominance pretest), plus an
Otheroption (see Appendix S2). The order in
which product categories and brand options
appeared was randomized. For each category, par-
ticipants merely clicked their brand of preference.
The questionnaire concluded with basic demo-
graphic questions.
Results
Manipulation ChecksProduct Congruity. Inde-
pendent samples t-tests conrmed that compared to
the white facial tissue, the black facial tissue was
perceived as less typical (M
ext.inc.
=2.54 vs. M
cong.
=
6.20), t(208) =19.61, p<.001, d=2.72, was less
likely to be resolved (ratings: M
ext.inc.
=3.28 vs.
M
cong.
=6.31), t(208) =14.73, p<.001, d=2.04
(coded responses: M
ext.inc.
=.40 vs. M
cong.
=.62), t
(208) =10.01, p<.001, d=1.38, and was evaluated
less favorably (M
ext.inc.
=3.18 vs. M
cong.
=5.10), t
(208) =10.11, p<.001, d=1.40.
State Anxiety. Controlling for trait anxiety, F(1,
207) =12.02, p<.001, g
2
=.05, an ANCOVA on
state anxiety (a=.88) conrmed that participants
were more anxious after evaluating the extremely
incongruent facial tissue (M=1.86) than the con-
gruent facial tissue (M=1.64), F(1, 207) =7.42,
p<.01, g
2
=.03.
Compensation TestDominant Brand
Choice. The typicality, resolution, evaluation, and
state anxiety measures reect previous work on
schema congruity (Noseworthy et al., 2014). Where
we now diverge from existing research is in explor-
ing beyond what this means for the target product.
To reiterate, the current study tested the prediction
that being exposed to extreme incongruity would
enhance the choice of dominant brands in unrelated
categories.
A multilevel binary logistic regression was con-
ducted on participantsbrand choices. This analysis
was deemed appropriate, as it permits the use of
repeatedly measured categorical-dependent vari-
ables (Sommet & Morselli, 2017). The model was
specied such that product category served as a
within-subject variable, and the product congruity
served as the between-subjects variable. The brand
choice-dependent variable was coded as binary
(coded: 1 =dominant), as the only constant across
product categories was the existence of a dominant
brand, and it would be inappropriate to statistically
equate the remaining brands across categories
(however, see Table 1 for multinomial logistic
regression analyses, which included all brands).
The analysis conrmed that participants were more
likely to select dominant brands after being exposed
extremely incongruent facial tissue (50.4%) than
after being exposed to the congruent facial tissue
(34%), b=0.68,SE=0.12, p<.001, OR =1.98. Crit-
ically, the effect of product category was not signi-
cant (p=.77), suggesting that the impact of product
congruity on brand selection did not vary between
different product categories used in the compensa-
tion task.
Mediation. To explore whether state anxiety
accounted for the observed variations in dominant
brand choice, we tested for multilevel mediation
(10,000 simulations; Mediation Package for R; Ting-
ley, Yamamoto, Hirose, Keele, & Imai, 2014). This
analysis was deemed appropriate, as it allows the
clustering of separate binary-dependent variables
for each participant. The analysis conrmed a sig-
nicant indirect effect of product congruity (coded:
1=extremely incongruent) through state anxiety
on dominant brand choice, 95% CI: .01, .10. Thus,
viewing extreme incongruity elevated state anxiety,
which enhanced the choice of dominant brands.
Discussion
The results of study 1 support the prediction that
consumers will compensate for the anxiety caused
by extreme product incongruity in unrelated pro-
duct schemas (H1 and H2). Specically, the experi-
ence of the black facial tissue, which was far less
typical, resolvable, and favorable than the white
82 Taylor and Noseworthy
facial tissue, coincided with an increase in the
choice of category dominant brands in unrelated
product categories.
Although compelling, one potential shortcoming
of this study was that the mediator was only mea-
sured. Thus, other correlated factors may have con-
tributed to the observed effects. One possibility is
that because extreme incongruity is difcult to
make sense of, it results in cognitive depletion
(Muraven, Tice, & Baumeister, 1998). Thus, con-
sumers may be afrming existing schemas because
they feel cognitively depleted (cf., Coleman, Wil-
liams, & Morales, 2018). One way in which deple-
tion can be ruled out is by considering an
important parameter of uid compensation, which
is that the anxiety elicited by expectancy violations
nonconsciously activates a need to seek meaning in
unrelated domains (Proulx et al., 2010). If the physi-
ological tension is consciously attributed to a
source, people tend not to compensate. Some of the
classic examinations of this was done by having
participants attribute their tension to a salient
sourcecoined as a misattribution task (Proulx &
Heine, 2008). Critically, attributing anxiety to a sali-
ent source neither alleviates tension nor assists in
making sense of the original schematic violation
(Tritt et al., 2012). Thus, if this is depletion, people
should still be depleted. However, if this is a non-
conscious compensatory act (Proulx et al., 2010),
then consciously attributing arousal to a salient
source should alleviate the need to compensate.
Study 2 tested this possibility.
Study 2
Study 2 was designed with two primary objectives
in mind: First, we sought to further explore the
underlying mechanism by employing a classic
misattribution paradigm (cf., Proulx & Heine, 2008).
This also afforded the opportunity to test a poten-
tial implication for schema congruity effect. Speci-
cally, we tested the possibility that bringing the
physiological tension to conscious awareness and
attributing it to a different source may dampen the
well-documented negative evaluations for extre-
mely incongruent products (cf., Noseworthy et al.,
2014). Second, we explored another important, and
previously unanticipated, consequence of extreme
incongruity. Specically, extant research suggests
that in addition to afrming semantic associations,
people may attempt to alleviate the tension that
originates from expectancy violations by afrming
ethical beliefs (Heine & Proulx, 2008). Ethical beliefs
often predict the preference for environmentally
friendly (i.e., green) products, which are deemed to
serve the betterment of society (Mazar & Zhong,
2010). Thus, study 2 explored the intriguing
Table 1
Means, Standard Deviations, Choice Percentages, and Cell Sizes for Study 1
Congruent Extremely incongruent Dof dominant brand choice
d
Typicality 6.20 (0.92) 2.54 (1.66)
Resolution 6.31 (0.93) 3.28 (1.87)
Evaluations 5.10 (1.13) 3.18 (1.56)
State anxiety
b
1.64 (0.56) 1.86 (.64)
Coding for Resolution
a
0.62 (0.66) 0.40 (0.81)
9.7% 60.7%
Dominant brand choice
c
34% 50.4% +16.4% (b=0.68, SE = 0.12, p<.001)
Soup (Campbell's) 35.9% 52.3% +16.4% (b=1.71, SE = 0.69, p= .01)
Battery (Duracell) 38.8% 57% +18.2% (b=2.36, SE = 1.08, p= .03)
Vacuum (Dyson) 33% 46.7% +13.7% (b=0.88, SE = 0.44, p= .05)
Smartphone (Apple) 33% 49.5% +16.5% (b=0.79, SE = 0.44, p= .055)
Coffee (Starbuck's) 29.1% 46.7% +17.6% (b=1.16, SE = 0.45, p= .01)
Cell sizes 103 107
Note. Standard deviations are reported in parentheses.
a
Percent of unsuccessful resolution (1).
b
Means with adjusted covariate for trait anxiety (M=1.83).
c
Bold values indicate the overall choice percentage of dominant brands across ve product categories, and the results of a multilevel
binary regression of dominant brand choice as a function of product congruity.
d
Reported statistics for each product category based on separate multinomial regressions (i.e., with all brand options included in the
model) as a function of product congruity (all ps<.06, see above).
Compensating for Innovation 83
possibility that extreme product incongruity may be
enhancing the preference of greenproducts.
Method
Participants and Design. Using the same
parameters as study 1 (with the effect size parame-
ter updated to f=.18; Cohen, 1988), two hundred
and eight students (56% females; M
age
=19.3) were
recruited in return for partial course credit. Partici-
pants were randomly assigned to one of four condi-
tions in a 2 (product congruity: congruent vs.
extreme incongruity) 92 (misattribution: yes vs.
no) between-subjects factorial design.
Stimuli. The target product chosen for this
study was running shoes. Following the procedures
of prior research (Jhang et al., 2012), a pilot test
conrmed the three-level operationalization of pro-
duct congruity (see Appendix S1). We then adapted
the two levels that were the focus of our hypothe-
ses (Appendix S3). The stimuli for the compensa-
tion task consisted of environmentally friendly (i.e.,
green) products (Griskevicius, Tybur, & van den
Bergh, 2010).
Procedures. Participants were brought into a
behavioral laboratory and were informed that they
would be taking part in a study on the effects of
subliminal sounds on memory. They were told
that the study consisted of participants examining
a new running shoe, and recalling specic things
about the shoe in a later task. The shoe was con-
veyed as a prototype designed by a major manu-
facturer, which did not include any brand-related
information on it. Participants were then
instructed to put on a pair of headphones and to
keep them on while viewing the shoe. This set
the foundation for our misattribution manipulation
(cf., Savitsky, Medvec, Charlton, & Gilovich,
1998).
The task began with participants being directed
to a computer screen where they read that they will
be exposed to a high-frequency sound (above
20 kHz). They were informed that while they can-
not hear this sound, their subconscious would pick
up on it. In actuality, participants were not exposed
to any sound. Participants in the misattribution con-
dition read that the subliminal sound may lead to a
temporary increase in their heart rate and that peo-
ple exposed to these sounds report feeling some-
what anxious and nervous. In contrast, participants
in the no misattribution condition read that sublimi-
nal sound had no discernible physiological impact
(manipulation directly taken from Savitsky et al.,
1998, 531). After conveying the misattribution guise,
the electronic instrument then transitioned to a pho-
tograph of a shoe (one of the two congruity condi-
tions) and then to an electronic questionnaire. The
questionnaire consisted of the same product evalua-
tion, typicality, anxiety, and resolution measures
described in study 1.
Participants were then informed that they would
be taking a short break to give their brains time to
process the subliminal sounds before moving on to
the recollection task, and thus, we were interested
in taking this time to get to know their general con-
sumption habits better. In reality, this served as the
core compensation task. Specically, participants
were asked to state their preference for what was
conveyed as random consumer goods. In fact, what
they evaluated were three items from different cate-
gories (a dishwasher, a car, and a household clea-
ner) which incorporated environmentally friendly
attributes (Griskevicius et al., 2010; Appendix S3; to
ensure that we were not inadvertently tapping into
a general increase in preference for unrelated prod-
ucts, a pilot study was conducted. The pilot used
the same stimuli, with the only difference being
that the products in the pilot did not contain any
environmentally friendly attributes. The results con-
rmed that preference of the three items was not
signicantly altered by whether participants were
exposed to an extremely incongruity versus congru-
ent product beforehand; see Appendix S1, pilot B
for study 2). For each, participants were asked to
rate, How attractive is this item to you?(an-
chored: 1 =not attractive at all,9=very attractive;
Griskevicius et al., 2010). The order of presentation
of items was electronically randomized. The
questionnaire concluded with basic demographic
questions.
Results
Manipulation ChecksProduct Congruity. Two-
way ANOVAs conrmed that compared to the con-
gruent running shoe, the extremely incongruent
running shoe was perceived less typical (M
ext.inc.
=
2.04 vs. M
cong.
=5.18), F(1, 203) =262.04, p<.001,
g
2
=.56, and was less likely to be resolved (ratings:
M
ext.inc.
=2.54 vs. M
cong.
=4.25), F(1, 203) =72.37,
p<.001, g
2
=.26 (coded responses: M
ext.inc.
=.52
vs. M
cong.
=.51), F(1, 203) =77.49, p<.001, g
2
=
.27. No other effects were signicant (ps >.15). This
supports the notion that misattribution does not
alter typicality or resolution, but instead alters con-
sumersreaction to the resolution process (Nose-
worthy et al., 2014). We expected that this would
be reected in product evaluations.
84 Taylor and Noseworthy
Product Evaluations. A two-way ANOVA on
product evaluations (a=.93) yielded a signicant
main effect of congruity, such that participants
evaluated the extremely incongruent running shoe
less favorably (M=2.63) than the congruent run-
ning shoe (M=3.23), F(1, 203) =11.87, p<.001,
g
2
=.05. The main effect of misattribution was not
signicant (p=.11). Importantly, there was a signif-
icant misattribution 9product congruity interac-
tion, F(1, 203) =5.09, p<.05, g
2
=.02. As
illustrated in Table 2, in the no misattribution con-
dition, participants evaluated the extremely incon-
gruent running shoe less favorably (M=2.30) than
the congruent running shoe (M=3.29), F(1,
203) =16.50, p<.001, g
2
=.08. This once again
conrmed the operationalization of product con-
gruity. Consistent with the notion that the negative
response to extreme incongruity is the result of
nonconscious arousal attribution (Noseworthy
et al., 2014), having participants consciously attri-
bute their arousal to a different source eliminated
the negative evaluations for extreme incongruity
(M
ext.inc
.=2.97 vs. M
cong.
=3.18, p=.40).
State Anxiety. Controlling for trait anxiety, F(1,
202) =32.42, p<.001, g
2
=.13, a two-way
ANCOVA on state anxiety (a=.88) revealed that
participants were more anxious after viewing the
extremely incongruent shoe (M=1.86) than the
congruent shoe (M=1.71), F(1, 202) =4.37, p<.05,
g
2
=.02. No other effects were signicant (ps >.36).
This supports the notion that misattribution does
not alter ones state anxiety (Proulx & Heine, 2008).
Compensation TestPreference for GreenProd-
ucts. Once again, the typicality, resolution, anxi-
ety, and evaluation measures aligned with the
classic schema congruity literature, and the misattri-
bution task varied evaluations in the predictable
manner. We again shifted our attention to what this
means beyond the target product of interest. To
reiterate, we predicted that extreme product incon-
gruity would elevate preference for green products,
but not when participants attributed their anxiety
to a salience source. To assess green product prefer-
ence, we averaged scores for the three environmen-
tally friendly items (see Appendix S1 page 10 for
details).
A two-way ANOVA yielded a main effect of
product congruity, such that participants preferred
green products more after viewing the extremely
incongruent shoe (M=5.99) than the congruent
shoe (M=5.45), F(1, 203) =5.13, p<.05, g
2
=.02.
The main effect of misattribution was not signi-
cant (p=.35). There was also a signicant product
congruity 9misattribution interaction, F(1,
203) =6.33, p<.05, g
2
=.03. Specically, partici-
pants in the no misattribution condition preferred
green products more after viewing the extremely
incongruent shoe (M=6.39) than the congruent
shoe (M=5.27), F(1, 203) =11.54, p<.005, g
2
=.05.
Critically, this effect was eliminated in the misattri-
bution condition (M
ext.inc
=5.58 vs. M
cong
=5.64,
p=.85).
Mediated Moderation. To explore whether anxi-
ety accounted for the observed variations in green
preference, we tested for mediated moderation
(Hayes, 2012; Model 15; bootstrapped with 10,000
draws). This model permits a back-ended modera-
tor (bpath) to an overall mediated effect. Theo-
retically this was appropriate given the
misattribution task did not alter state anxiety, but it
Table 2
Means, Standard Deviations, and Cell Sizes for Study 2
No misattribution Misattribution
Congruent Extremely incongruent Congruent Extremely incongruent
Typicality 5.90 (1.43) 1.92 (1.42) 5.27 (1.37) 2.18 (1.34)
Resolution 4.18 (1.58) 2.43 (1.33) 4.32 (1.48) 2.66 (1.34)
Evaluations 3.29 (1.35) 2.31 (1.19) 3.18 (1.21) 2.98 (1.29)
State anxiety
c
1.68 (.50) 1.90 (.76) 1.73 (.50) 1.82 (.53)
Coding for Resolution
a
.42 (.90) .60 (.79) .60 (.77) .43 (.90)
26.9% 79.2% 17.3% 70.6%
Green preference
b
5.27 (2.02) 6.39 (1.57) 5.64 (1.53) 5.58 (1.58)
Cell sizes 52 53 52 51
Note. Standard deviations are reported in parentheses.
a
Percent of unsuccessful resolution (1).
b
Preference of three environmentally friendly items combined into green preference composite.
c
Means with adjusted covariate for trait anxiety (M=2.25).
Compensating for Innovation 85
did interact with product congruity to alter green
preference. As illustrated in Figure 1, there was a
conditional indirect effect of product congruity
(1 =extremely incongruent) on green preference
through state anxiety, 95% CI: .42; .01. Speci-
cally, state anxiety explained why viewing the
extremely incongruent (vs. congruent) running shoe
elevated green preference in the no misattribution
condition, 95% CI: .01; .36. As predicted, this effect
was eliminated when participants consciously
attributed their arousal to a salient source, 95% CI:
.10; .18.
Discussion
The results of study 2 conceptually replicated the
core ndings of study 1 and further demonstrated
the process by using a classic misattribution task.
Specically, the results conrmed that having par-
ticipants consciously attribute their arousal to a sali-
ent source reduced compensatory preference to
levels consistent with the control (i.e., congruity
condition). Importantly, the misattribution task
dampened green product preference without alter-
ing resolution or alleviating tension. This is consis-
tent with the literature on uid compensation,
which suggests that although people may be no
better at making sense of an expectancy violation,
compensation in unrelated domains is unlikely if
anxiety can be consciously attributed to a salient
source (Proulx & Heine, 2008). The nding that the
misattribution task elevated evaluations for extreme
incongruity also corroborates with the notion that
the negative response to extreme incongruity is the
result of nonconsciously experienced tension that
manifests as anxiety (Noseworthy et al., 2014).
Given this, study 3 aimed to directly manipulate
the mechanism and explore the underlying role of
physiological arousal.
Study 3
In addition to semantic associations and ethical
beliefs, another well-established means of compen-
sating for an expectancy violation is afrming ones
culture. Specically, extant research shows that peo-
ple identify more with their culture after experienc-
ing an expectancy violation (Proulx et al., 2010). To
this point, people often afrm their cultural identity
through ethnocentricconsumption, which is
dened as consumption activities that show defer-
ential preference for ones cultural in-group (Shimp
& Sharma, 1987). One key example of ethnocentric
consumption is in research on the country-of-origin
effect (COE; Schooler, 1965), where people indicate
a greater willingness to pay for products that origi-
nate from regions that they identify with (e.g., Wall
& Heslop, 1986). Thus, if extreme incongruity
indeed threatens beliefs, then extremely incongruent
products may inadvertently have an impact on
COE effect.
Furthermore, up to this point, we have adopted
paradigms from the uid compensation literature.
However, there is a specic paradigm in schema
congruity literature that may lend important
insights. Specically, in testing Mandlers (1982)
seminal prediction, Noseworthy et al. (2014)
demonstrated that putting consumers in a relaxed
state dampens the physiological response to
extreme incongruity and thus reduces anxiety and
increases preference. This nding is particularly rel-
evant for two reasons: First, if uid compensation
is a means by which people cope with arousal-in-
duced anxiety, then putting someone in a relaxed
state should reduce compensation by dampening
the aroused state. Second, researchers have yet to
conrm what is actually accomplished by compen-
sating through unrelated schemas. Does compensa-
tion allow people to manage the intensity of their
emotional state? Indeed, the results from study 2
Product congruity
(Extreme incongruity = 1)
Compensation
(“Green” preference)
State
anxiety
Arousal
(Misattribution = 1)
b= .16*
b= – 1.18*
b= – 1.05*
( = – .95*)
Figure 1. Mediated moderation results for Study 2. Note. Unstandardized betas are reported with superscripts: * = (p<.05), ** =
(p<.05), and *** = (p<.001).
86 Taylor and Noseworthy
support this possibility. However, only by using
actual physiological measures would it be possible
to isolate the moment-to-moment changes in ones
physiological state and thus test whether compensa-
tion has an immediate dampening effect on arousal
intensity.
Method
Participants and Design. Using the same
parameters as study 2 to determine our sample (N),
two hundred participants (55% females; M
age
=
26.3) were recruited through posters and public
advertisements and paid $20 for taking part in the
study. Each participant was randomly assigned to
one of four conditions in a 2 (product congruity:
congruent vs. extremely incongruent) 92 (aroused
state: low arousal vs. control) between-subjects fac-
torial design.
Stimuli. The target product chosen for this
study was soft drinks. Not only are soft drinks rele-
vant to most consumers, but they are also featured
extensively in the schema congruity literature (e.g.,
Campbell & Goodstein, 2001; Meyers-Levy & Tyb-
out, 1989; Noseworthy et al., 2014). For this study,
we took the congruent and extremely incongruent
manipulations from Noseworthy et al.s (2014) web
appendix (Appendix S4 [top]), which notably, are
the same stimuli used to conrm the autonomic
and emotional underpinnings of the schema con-
gruity effect. This gave us an operationalization
and a methodological template to test for the com-
pensatory acts that follow the autonomic responses
elicited by extreme incongruity.
The stimulus for the compensation task consisted
of a basket of strawberries. We used this stimulus
because strawberries are grown in many locales
and thus are not associated with a particular coun-
try of origin. Furthermore, there is a mandatory
labeling law in the United States, requiring vendors
to disclose the origins of produce items. This setting
allowed us to explore ethnocentric preference
through products from regions consumers identify
with (Wall & Heslop, 1986).
Procedures. Participants were brought into a
behavioral laboratory where a lab tech adminis-
tered a prescreening questionnaire under the guise
that we were screening for physical and psycho-
logical health. The true intent was to capture the
geographical region participants best identied
with. We kept the geographical regions broad
(e.g., South America), as more focused piloting
led to several people being uncategorized or unfa-
miliar with precise regions. Specically, the regions
were broken down into 22 categories, directly
taken from the United Nationsofcial geo-scheme
list (United Nations Statistics Division, 2018). This
variable served as a cultural identication marker
which would later be piped into the compensation
task.
After completing the prescreening, participants
were informed that we would be measuring
their physiological response to different products
using a galvanic skin response (GSR) system (see
Appendix S1 for details of the procedure). They
were further told that before beginning the study,
the lab tech needed to calibrate this equipment,
during which time participants were to put on
noise cancellation headphones. Participants in the
low arousal condition were exposed to Albinonis
Adagio in G Minor through the headphones and fur-
ther instructed to relax during this time. Partici-
pants in the control condition were not exposed to
any sounds. A pretest (N=63) conrmed that the
low arousal and the control condition induced simi-
lar feelings of pleasantness, but different feelings of
arousal (see Appendix S1 for details). This was
important to ensure, given that cognitive exibility
from positive affect can alter how people process
extreme incongruity (Jhang et al., 2012). The cali-
bration task ran for exactly 3 min.
Afterwards, participants were informed that they
would be evaluating a soft drink concept. The
study commenced with an image of the congruent
(extremely incongruent) soft drink, followed by a
questionnaire which consisted of the same items
discussed in study 1 (i.e., typicality, evaluations,
resolution, state, and trait anxiety). Participants
were then immediately administered what was con-
veyed as a brief consumer taste prole. They were
told that several different produce, some from the
same product category, will ash one at a time,
and they would simply list what they would be
willing to pay for each product as if they were
shopping at their local grocery store. Willingness to
pay (WTP) was captured on a single open-ended
item (Wertenbroch & Skiera, 2002). Embedded in
the product array were two identical pictures of a
basket of strawberries, one with the statement
Grown in [piped in cultural-identication mar-
ker]!and the other with Grown in [piped in ran-
dom non-cultural identication marker]!These
stimuli were presented in a counterbalanced order.
Skin conductance responses (SCRs) were sam-
pled for the duration of the task. The key compar-
ison was the 15 s window following the moment
at which participants were exposed to the rst bas-
ket of strawberries, which either did or did not
Compensating for Innovation 87
correspond to participantscultural origin (see
Appendix S1 page 16 for a visual representation of
the procedures). Consistent with prior work (Nose-
worthy et al., 2014), the goal was to measure partic-
ipantsphysiological reactions immediately
following the onset of the stimulus. Counterbalanc-
ing of the compensation task served two purposes.
First, it allowed us to explore whether compensa-
tion indeed dampens the physical tension elicited
by extreme incongruity. If true, then we should see
physiological arousal diminish immediately after
someone is permitted to compensate (i.e., if the bas-
ket of strawberries corresponds to their cultural ori-
gin) than otherwise (i.e., if the basket of
strawberries does not correspond to their cultural
origin). Second, counterbalancing allowed us to rule
out temporal effects by testing whether physiologi-
cal arousal is naturally diminishing, or diminishing
only when someone can cope by compensating.
Results
Manipulation ChecksProduct Congruity. Two-
way ANOVAs conrmed that the extremely incon-
gruent soft drink was perceived less typical
(M
ext.inc.
=2.77 vs. M
cong.
=5.69), F(1, 196) =1482.11,
p<.001, g
2
=.88, and was less likely to be resolved
(ratings: M
ext.inc.
=3.11 vs. M
cong.
=5.40), F(1,
196) =209.84, p<.001, g
2
=.52, (coded responses:
M
ext.inc.
=.11 vs. M
cong.
=.38), F(1, 196) =19.00,
p<.001, g
2
=.09, than the congruent soft drink.
No other effects were signicant (ps>.18). Thus,
consistent with prior work (Noseworthy et al.,
2014), aroused state did not inuence perceived
typicality or resolution.
Product Evaluations. A two-way ANOVA on
product evaluations (a=.92) yielded a signicant
main effect of aroused state, such that participants
liked the product more in the low arousal condition
(M=3.61) than in the control (M=3.26), F(1,
196) =4.84, p<.05, g
2
=.02. The main effect of
product congruity was not signicant (p=.59). The
analysis also revealed a signicant product con-
gruity 9aroused state interaction, F(1, 196) =11.56,
p<.001, g
2
=.05. As illustrated in Table 3, partici-
pants in the control condition liked the congruent
soft drink (M=3.57) more than the extremely
incongruent soft drink (M=2.96), F(1, 196) =7.79,
p<.01, g
2
=.04. However, participants in the low
arousal condition liked the extremely incongruent
soft drink (M=3.83) more than the congruent soft
drink (M=3.39), F(1, 196) =4.08, p<.05, g
2
=.02.
Another way to look at this is that, consistent with
prior work (Noseworthy et al., 2014), relaxation
elevated evaluations for extreme incongruity (M
low
arousal
=3.83 vs. M
control
=2.96), F(1, 196) =15.69,
p<.001, g
2
=.07, but not for congruity (p=.40).
State Anxiety. Controlling for trait anxiety, F(1,
195) =5.65, p<.05, g
2
=.03, a two-way ANCOVA
on state anxiety (a=.90) yielded a main effect of
aroused state, such that participants were less anx-
ious in the low arousal condition (M=2.10) than in
the control (M=2.37), F(1, 195) =10.90, p<.005,
g
2
=.05. There was also a main effect of product con-
gruity, such that participants were more anxious
after seeing the extremely incongruent soft drink
(M=2.39) than the congruent soft drink (M=2.08),
F(1, 195) =12.12, p<.005, g
2
=.05. These effects
were qualied by a signicant product congruity 9
aroused state interaction, F(1, 195) =4.18, p<.05,
g
2
=.02. Specically, participants in the control con-
dition were more anxious after seeing the extremely
incongruent soft drink (M=2.62) than the congruent
soft drink (M=2.13), F(1, 195) =15.26, p<.001,
g
2
=.07. This effect did not emerge in the low arou-
sal condition (M
ext.inc
=2.16 vs. M
cong
=2.04,
p=.31). Thus, relaxation dampened anxiety elicited
by extreme incongruity by bringing it down to levels
consistent with the congruent condition, as expected.
Compensation TestEthnocentric Preference. The
anxiety and evaluation results aligned with the
prior studies, and putting participants in a low
aroused state altered these effects in a predictable
way. We again shifted our focus to what this means
beyond the focal product. We predicted that damp-
ening the severity of the autonomic response to
product incongruity would inhibit compensation.
Compensation was tested through ethnocentric
preference, calculated by taking the average differ-
ence in WTP between the national and international
versions of the same basket of strawberries. Thus, a
positive number would indicate stronger ethnocen-
tric preference, whereas a negative number would
indicate stronger international preference.
A two-way ANOVA revealed a main effect of
arousal, such that the low arousal condition damp-
ened ethnocentric preference (M=.11) relative to
the control condition (M=.73), F(1, 196) =4.86,
p<.05, g
2
=.02. The main effect of product con-
gruity was not signicant (p=.16). The analysis
also yielded a signicant interaction, F(1,
196) =3.92, p<.05, g
2
=.02. Although most partici-
pants were willing to pay more for the culturally
consistent versions (denoted by positive means), the
nature of the interaction came down to a single cell,
which related directly to our core prediction. Specif-
ically, participants in the control condition
expressed greater ethnocentric preference after
88 Taylor and Noseworthy
evaluating the extremely incongruent soft drink
(M=$1.20) than the congruent soft drink (M=$.25),
F(1, 196) =5.72, p<.05, g
2
=.03. As predicted, put-
ting consumers in a relaxed state eliminated this
effect (M
ext.inc
=$0.02 vs. M
cong
=$0.18, p=.68). In
fact, the only condition to positively differ from $0
was the extremely incongruent control (M=$1.20),
t(49) =3.20, p<.005, d=0.57 (all other cells
ps>.26). Thus, putting participants in a relaxed state
reduced compensatory preference in a predictable
manner.
Mediated Moderation. To explore whether anxi-
ety accounted for the observed variations in ethno-
centric preference, we tested for mediated
moderation (Hayes, 2012; Model 8; bootstrapped
with 10,000 draws). Of notable distinction from
study 2 is that this model incorporates a moderat-
ing effect of the mechanism (i.e., an apath mod-
erator). As illustrated in Figure 2, there was a
conditional indirect effect of product congruity
(1 =extremely incongruent) through state anxiety
on ethnocentric preference, 95% CI: .002; .37. Specif-
ically, state anxiety mediated the increase in ethno-
centric preference following the evaluation of the
extremely incongruent soft drink in the control con-
dition, 95% CI: .03; .44, but not in the low arousal
condition, 95% CI: .02; .16. As predicted, low
arousal alleviated anxiety and subsequently damp-
ened ethnocentric preference.
Physiological Arousal. Our nal goal in this
study was to answer the elusive question of whether
compensation truly is a means by which a person
can immediately regulate the intensity of the
emotional response to extreme incongruity. To
explore this, we used the counterbalance of the com-
pensation task as a between-subjects condition and
added it to our model. This allowed us to isolate
whether physiological arousal varied when partici-
pants were able to compensate at t
1
(time point 1) in
the taste prole taskin essence, whether the rst
exposure to the basket of strawberries was the one
that corresponded to the participantscultural origin.
A three-way ANOVA on participantsSCRs
yielded a signicant main effect of aroused state,
such that participants at t
1
were less aroused in the
low arousal condition (M=1.24) than in the control
(M=1.69), F(1, 196) =8.55, p<.005, g
2
=.04. This
conrmed that the manipulation of arousal held for
the duration of the task. There was also a main
effect of product congruity, such that participants at
t
1
were more aroused after seeing the extremely
incongruent soft drink (M=1.69) than after seeing
the congruent soft drink (M=1.25), F(1,
196) =7.58, p<.01, g
2
=.04. This is consistent with
evidence of a direct physiological consequence to
extreme incongruity (Noseworthy et al., 2014).
These two effects were qualied by a marginally
signicant three-way interaction, F(1, 196) =2.97,
p=.086, g
2
=.01. The only signicant simple effect
was in the extremely incongruent control. As pre-
dicted, participants were less aroused immediately
following t
1
when they could afrm ethnocentric
preference (M=1.60) than when they could not
(M=2.26), F(1, 196) =4.59, p<.05, g
2
=.02.
This did not occur in the low arousal condition
(M
no-afrmation
=1.24 vs. M
afrmation
=1.60, p=.24).
Table 3
Means, Standard Deviations and Cell Sizes for Study 3
Low arousal (relaxed) Moderate arousal (control)
Congruent Extremely incongruent Congruent Extremely incongruent
Typicality 5.65 (0.66) 2.20 (0.69) 5.73 (0.52) 2.37 (0.72)
Resolution 5.47 (0.68) 3.22 (1.69) 5.32 (0.61) 2.99 (1.14)
Evaluations 3.39 (1.12) 3.83 (0.93) 3.57 (0.98) 2.96 (1.36)
State anxiety
c
2.04 (0.47) 2.16 (0.60) 2.13 (0.55) 2.62 (0.91)
SCRs pre-compensation 1.02 (1.36) 1.24 (1.01) 1.39 (1.00) 2.26 (1.49)
SCRs postcompensation 1.09 (1.05) 1.60 (0.83) 1.50 (1.05) 1.60 (0.57)
Coding for Resolution
a
0.34 (0.75) 0.08 (0.85) 0.42 (0.67) 0.14 (0.83)
16% 40% 10% 42%
Ethnocentric Preference
b
0.18 (2.19) 0.02 (1.27) 0.25 (1.56) 1.20 (2.66)
Cell Size 50 50 50 50
Note. Standard deviations are reported in parentheses.
a
Percent of unsuccessful resolution (1).
b
Calculated as the difference in willingness to pay for the domestic versus international version of the same basket of strawberries.
c
Means with adjusted covariate for trait anxiety (M=2.31).
Compensating for Innovation 89
No other simple effects were signicant (ps>.73).
These results suggest that the diminishing SCRs
when participants afrmed ethnocentric preference
could not be the result of a natural decrease in
ones aroused state over time given that the condi-
tions were counterbalanced. Thus, the results con-
rmed that compensation had a dampening effect
that was similar to that elicited by relaxation. This
represents some of the rst empirical evidence that
compensation combats a heightened physiological
response, which corresponds with anxiety.
Discussion
In further support of our core prediction, the
instance of extreme product incongruity coincided
with greater willingness to pay for a basket of
strawberries from the consumersself-identied cul-
tural region (H1 and H2). Furthermore, relaxation
inhibited compensatory preference in a predictable
way by reducing state anxiety. Thus, building on
study 2, we tested for our mechanism using direct
moderation. Critically, the results also revealed that
extreme incongruity elevates actual physiological
arousal, and that this is immediately dampened
after participants compensate in an unrelated
domain. This represents some of the rst denitive
evidence that compensation indeed serves as a true
coping mechanism for expectancy violations.
Study 4
One of the most important insights gleaned from
study 3 was that compensation dampens the arou-
sal intensity elicited by extreme incongruity. Impor-
tantly, arousal-induced anxiety is the driving factor
for why evaluations of extremely incongruent prod-
ucts tend to be negative (Noseworthy et al., 2014).
This raises the possibility that if a consumer can
compensate prior to explicitly evaluating an
extremely incongruent product, such a judgment
may not be as negative as it would be otherwise
(H3), as compensating in unrelated domains would
alleviate the tension. Our previous studies do not
reect this possibility, as we have exclusively col-
lected product evaluations and anxiety ratings
before consumers compensated. In study 4, we thus
sought to come full circle and bring the lens back
to what compensation may mean for product evalu-
ations.
Method
One hundred and ninety university students
(49% females; M
age
=21.9) participated in the study
in return for partial course credit. Participants were
randomly assigned to one of four conditions in a 2
(product congruity: congruent vs. extremely incon-
gruent) 92 (compensation: pre- vs. postevaluation)
between-subjects factorial design. The study used
the compensation task from study 2 (green con-
sumption) to explore the implications of the core
nding observed in study 3 (dampened arousal
intensity) on product evaluations.
The product category chosen for this study was
smartphones because smartphones are relevant to
consumers, and there is strong prototypicality
within the category in that the standard design of a
smartphone (a black box) has varied little over
time. This afforded the opportunity to develop an
extremely incongruent product design that violates
this prototype. As with prior studies, a pilot test
conrmed the three-level operationalization of pro-
duct congruity (see Appendix S1). We then adapted
the two levels that were the focus of our hypothe-
ses (Appendix S5).
Participants were informed that they were going
to examine a new smartphone. The questionnaire
consisted of the same evaluation, typicality, anxiety
(state and trait), and resolution measures used in
Product congruity
(Extreme incongruity = 1)
Compensation
(Ethnocentric preference)
State
anxiety
Aroused state
(Low = 0)
b= .18*
b= .56*
b= .77***
= .42)
Figure 2. Mediated moderation results for Study 3. Note. Unstandardized betas are reported with superscripts: * = (p<.05), ** =
(p<.01), and *** = (p<.001).
90 Taylor and Noseworthy
our prior studies. The compensation task followed
the procedures outlined in study 2. Specically,
participants learned that in addition to examining a
new smartphone, we sought to develop a general
consumption prole. In reality, this served as the
compensation task, wherein consumers were asked
their preference for three products with environ-
mentally friendly attributes (Griskevicius et al.,
2010). The compensation task was administered
after participants viewed the phone, but either
before or after participants lled out the question-
naire. Specically, in the postevaluation condition,
participants were administered the compensation
task after viewing and providing their opinions on
the smartphone (in line with prior studies). In con-
trast, and of specic interest to the current study,
participants in the pre-evaluation condition were
administered the compensation task after viewing,
but before explicitly providing their opinions on,
the smartphone (To ensure we were not inadver-
tently tapping temporal effects leading to differ-
ences in evaluations, this design was piloted using
the pilot test products from study 2, which were
not linked to compensation. The results showed
evaluations for extreme incongruity did not alter by
whether they were captured pre/post a standard
preference task). The session concluded with basic
demographic questions.
Results
Manipulation ChecksProduct Congruity. Two-
way ANOVAs conrmed that the extremely incon-
gruent smartphone was indeed perceived as less
typical (M
ext.inc.
=2.66 vs. M
cong.
=5.39), F(1,
186) =174.78, p<.001, g
2
=.56, and was less likely
to be resolved (ratings: M
ext.inc.
=3.49 vs. M
cong.
=
4.85), F(1, 186) =42.81, p<.001, g
2
=.19 (coded
responses: M
ext.inc.
=.43 vs. M
cong.
=.54), F(1,
186) =78.31, p<.001, g
2
=.29, than the congruent
smartphone. No other effects were signicant (all
ps >.15). Thus, as expected, compensation task did
not inuence perceived typicality or the resolution
process.
Compensation TestPreference for GreenProd-
ucts. A two-way ANOVA on green preference
yielded a main effect of product congruity, such
that participants preferred the green product more
after viewing the extremely incongruent phone
(M=6.24) than the congruent phone (M=5.67), F
(1, 186) =4.81, p<.05; g
2
=.03. No other effects
were signicant (ps >.52). Thus, regardless of when
the compensation task took place, extreme incon-
gruity elevated green preference. The critical
question was, unlike study 2, whether the compen-
sation task had a meaningful impact on consumers
product evaluations and state anxiety.
Product Evaluations. A two-way ANOVA on
product evaluations (a=.95) yielded a main effect
of product congruity, such that participants liked
the extremely incongruent phone less (M=3.96)
than the congruent phone (M=4.51), F(1,
186) =13.11, p<.001; g
2
=.03. The main effect of
compensation was not signicant (p=.46). The
analysis also yielded a signicant interaction, F(1,
186) =6.76, p<.05; g
2
=.03. As illustrated in
Table 4, and consistent with previous studies, par-
ticipants who conveyed their evaluations before
they compensated liked the extremely incongruent
phone far less (M=3.59) than the congruent phone
(M=4.71), F(1, 186) =13.11, p<.001; g
2
=.07. This
effect was eliminated when evaluations were cap-
tured after compensation (M
ext.inc.
=4.31 vs. M
cong.
=
4.30, p=.99). Another way to look at this is that
compensating elevated the evaluations for extreme
incongruity (M
pre-eval.
=4.31 vs. M
post-eval.
=3.59),
F(1, 185) =5.65, p<.05, g
2
=.02, but not for con-
gruity (p=.20).
State Anxiety. Controlling for trait anxiety, F(1,
185) =33.51, p<.001; g
2
=.14, an ANCOVA on
state anxiety (a=.87) yielded a signicant main
effect of product congruity, such that participants
were more anxious after viewing the extremely
incongruent phone (M=1.91) than the congruent
phone (M=1.76), F(1, 185) =5.66, p<.05; g
2
=.02.
The main effect of compensation was not signicant
(p=.33). There was also a signicant interaction, F
(1, 185) =7.67, p<.01; g
2
=.03. Specically, before
compensating, participants were more anxious
upon viewing the extremely incongruent phone
(M=2.06) than the congruent phone (M=1.66), F
(1, 185) =13.01, p<.001; g
2
=.07. Consistent with
the results of study 3, this effect disappeared after
compensation (M
ext.inc.
=1.77 vs. M
cong.
=1.80,
p=.78). Another way to look at this is that, com-
pensation reduced anxiety for participants exposed
to extreme incongruity (M
pre-eval.
=1.77 vs.
M
post-eval.
=2.06), F(1, 185) =7.16, p<.01, g
2
=.03,
but not congruity (p=.21).
Mediated Moderation. To determine whether
state anxiety accounted for variations in product
evaluations, we conducted a mediated moderation
analysis (Model 8; Hayes, 2012; bootstrapped 10,000
draws). As illustrated in Figure 3, there was a con-
ditional indirect effect of product congruity (coded:
1=extremely incongruent) on evaluations through
anxiety, 95% CI: .04, .56. Specically, and replicat-
ing prior research (Noseworthy et al., 2014), state
Compensating for Innovation 91
anxiety explained why evaluations of extremely
incongruent phone suffered in the postevaluation
condition, 95% CI: .49, .05. As predicted, this
pattern was eliminated for participants who could
compensate before they explicitly provided their
evaluations, 95% CI: .11, .15.
Discussion
Study 4 conrmed that because the unfavorable
evaluations received by extreme incongruity is the
result of arousal-induced anxiety (Noseworthy
et al., 2014), and because compensation can allevi-
ate this tension (study 3), if consumers can compen-
sate after viewing extremely incongruent products,
their evaluations may not be as negative as they
would be otherwise (H3). Importantly, the compen-
sation task altered neither the perceived typicality
nor resolution, but nevertheless ameliorated evalua-
tions by dampening the anxiety elicited by extreme
incongruity. Thus, in bringing the investigation full
circle, study 4 illustrated that by compensating in
unrelated domains, consumers can become more
open to extreme incongruity.
General Discussion
Evidence from four studies suggests that extremely
incongruent products, which are inherently difcult
to make sense of, can lead consumers to compen-
sate by afrming other aspects of consumption,
such as brand dominance (study 1), green con-
sumption (studies 24), and ethnocentric preference
(study 3). Importantly, we conrm that these effects
are the result of arousal-induced anxiety. Speci-
cally, all studies showed that anxiety mediates com-
pensatory preference. Furthermore, study 2
illustrated that by allowing consumers to con-
sciously attribute their tension to a salient source,
Table 4
Means, Standard Deviations, and Cell Sizes for Study 4
Postevaluation (compensation after evalua-
tions)
Preevaluation (compensation before evalua-
tions)
Congruent Extremely incongruent Congruent Extremely incongruent
Typicality 5.53 (1.30) 2.50 (1.49) 5.24 (1.26) 2.81 (1.60)
Resolution 5.01 (1.51) 3.37 (1.55) 4.69 (1.36) 3.42 (1.68)
Evaluations 4.71 (1.41) 3.59 (1.64) 4.30 (1.31) 4.31 (1.59)
State anxiety
c
1.66 (0.48) 2.06 (0.72) 1.80 (0.55) 1.77 (0.55)
Coding for Resolution
a
0.61 (0.65) 0.36 (0.87) 0.48 (0.77) 0.49 (0.71)
8.7% 61.7% 16.7% 61.2%
Green preference
b
5.80 (2.02) 6.28 (1.66) 5.54 (1.91) 6.21 (1.56)
Cell size 46 47 48 49
Note. Standard deviations are reported in parentheses.
a
Percent of unsuccessful resolution (1).
b
Preference of three environmentally friendly items combined into green preference composite.
c
Means with adjusted covariate for trait anxiety (M=2.16).
Product congruity
(Extreme incongruity = 1)
Product evaluations
State anxiety
b=.43**
b= 1.12**
b=.59***
=.89*)
Time of compensation
(Pre-evaluation = 1)
Figure 3. Mediated moderation results for Study 4. Note. Unstandardized betas are reported with superscripts: * = (p<.05), ** =
(p<.01), and *** = (p<.005).
92 Taylor and Noseworthy
we were able to attenuate compensatory preference.
Further still, study 3 conrmed that compensatory
preference did not occur when consumers were put
in a relaxed state to begin with. This same study
extracted actual physiological readings to show that
the moment of compensation corresponds with an
immediate reduction in arousal intensity. This rep-
resents some of the most conclusive evidence that
compensation operates almost as a form of coping.
Study 4 brought our inquiry full circle by docu-
menting that because compensation dampens the
tension elicited by extreme incongruity, having con-
sumers compensate in unrelated domains can
alleviate the negative response to extremely incon-
gruent products.
From a theoretical perspective, this work con-
tributes to the schema congruity literature by
demonstrating that, in addition to the traditional
predictions of assimilation and accommodation (Man-
dler, 1982), consumers can employ uid compensa-
tion as a means of alleviating the tension that
originates from schematic violations. This nding
supports the notion that consumption schemas are
functionally interchangeable in alleviating anxiety,
such that the inability to resolve an inconsistency in
one belief structure can be satiated by afrming
another belief structure. An additional study
(Appendix S1, pp. 3034) conrmed that compensa-
tion is not predicted by conscious factors such as a
need for consistency, underlining the nonconscious
nature of compensatory acts that follow extreme
incongruity. Furthermore, the current work con-
tributes to the uid compensation literature by pro-
viding the rst empirical evidence that
compensation indeed alleviates the physiological
response elicited by an expectancy violation. Build-
ing on this insight, this work also shows that allow-
ing consumers to compensate in unrelated domains
may ameliorate the negative response to extreme
incongruity. Thus, this work highlights how the lit-
erature on schema congruity may be narrowing
itself by focusing solely on resolving the product
within its respective schema. Rather than change or
update a product schema, consumers can afrm
other meaningful aspects of consumption.
From a managerial perspective, this work lends
important insights for marketers. Past research has
noted that extreme incongruity results in a product
being irreconcilable within its broader category
(Noseworthy & Trudel, 2011), which tends to make
evaluations suffer (Meyers-Levy & Tybout, 1989).
Thus, it is not surprising that marketers often caution
against the introduction of radical innovation in the
marketplace (Cornescu & Adam, 2013). The current
work offers an alternate point of view by suggesting
that consumers may quite adept at resolving the ten-
sion that originates from extreme incongruity
through other consumption practices, and that this
may even result in positive outcomes for companies
that operate in completely unrelated categories.
Notable beneciaries of extreme product innovations
covered in this work ranges from category dominant
brands, to culturally relevant products, and to envi-
ronmentally friendly items. The last point has impor-
tant implications for public policymakers, as it
suggests that the strategic use of extreme incongruity
may indirectly have a positive impact on society.
Future research can expand on our ndings by
investigating other compensatory domains. For
example, consumers may compensate for the anxi-
ety elicited by violations by afrming their self-es-
teem (Mandel, Rucker, Levav, & Galinsky, 2017).
Research may also explore just how unrelated a
belief system needs to be for compensatory acts to
kick in. For instance, could a category dominant
brand benet from an extremely incongruent pro-
duct introduced within its own category? Examin-
ing this possibility would reveal the true boundaries
of uid compensation, while affording unique man-
agerial insights into how companies may directly
benet from extreme innovations within their own
product category. Finally, research can benet from
exploring other consumption experiences that may
prompt compensatory acts. One interesting possibil-
ity is promotional lotteries, which may threaten
meaning by shaking the perception that consumers
have control over desirable outcomes (Taylor,
Noseworthy, & Pancer, 2019). Overall, one can
imagine a variety of experiences that may threaten
meaning, leading to unforeseen consequences. Cer-
tainly, more research is needed in this area.
References
Antonovsky, A. (1979). Sense of Coherence scale. San Fran-
cisco, CA: JosseyBass.
Bagga, C. K., Noseworthy, T. J., & Dawar, N. (2016).
Asymmetric consequences of radical innovations on
category representations of competing brands. Journal of
Consumer Psychology,26,2939. https://doi.org/10.
1016/j.jcps.2015.04.005
Bruner, J. S. (1990). Acts of meaning (Vol. 3). Cambridge,
MA: Harvard University Press.
Campbell, M. C., & Goodstein, R. C. (2001). The moderat-
ing effect of perceived risk on consumersevaluations
of product incongruity: Preference for the norm. Journal
of Consumer Research,28, 439449. https://doi.org/10.
1086/323731
Compensating for Innovation 93
Cohen, J. (1988). Statistical power analysis for the behavioral
sciences. Hillsdale, NJ: Lawrence Erlbaum.
Coleman, N. V., Williams, P., & Morales, A. C. (2018).
Identity threats, compensatory consumption, and work-
ing memory capacity: How feeling threatened leads to
heightened evaluations of identity-relevant products.
Journal of Consumer Research,46,99118.
Cornescu, V., & Adam, C. R. (2013). The consumer resis-
tance behavior towards innovation. Procedia Economics
and Finance,6, 457465. https://doi.org/10.1016/S2212-
5671(13)00163-9
Galli, M., & Gorn, G. (2011). Unconscious transfer of
meaning to brands. Journal of Consumer Psychology,21,
215225. https://doi.org/10.1016/j.jcps.2010.12.004
Griskevicius, V., Tybur, J. M., & van den Bergh, B. (2010).
Going green to be seen: Status, reputation, and conspic-
uous conservation. Journal of Personality and Social Psy-
chology.,98, 392401. https://doi.org/10.1037/a0017346
Hayes, A. F. (2012). PROCESS: A versatile computational
tool for observed variable mediation, moderation, and
conditional process modeling.
Heine, S. J., Proulx, T., & Vohs, K. D. (2006). The meaning
maintenance model: On the Coherence of social motiva-
tions. Personality and Social Psychology Review,10,88
110. https://doi.org/10.1207/s15327957pspr1002_1
Herr, P. M., Farquhar, P. H., & Fazio, R. H. (1996).
Impact of dominance and relatedness on brand exten-
sions. Journal of Consumer Psychology,5, 135159.
https://doi.org/10.1207/s15327663jcp0502_03
Jhang, J. H., Grant, S. J., & Campbell, M. C. (2012). Get
It? Got It. Good! Enhancing new product acceptance by
facilitating resolution of extreme incongruity. Journal of
Marketing Research,49, 247259. https://doi.org/10.
1509/jmr.10.0428
Kruglanski, A. W., & Webster, D. M. (1996). Motivated
closing of the mind: Seizingand freezing.Psycho-
logical Review,103, 263. https://doi.org/10.1037/0033-
295X.103.2.263
Major, B., & Townsend, S. S. (2012). Meaning making in
response to unfairness. Psychological Inquiry,23, 361
366. https://doi.org/10.1080/1047840X.2012.722785
Mandel, N., Rucker, D. D., Levav, J., & Galinsky, A. D.
(2017). The compensatory consumer behavior model:
How self-discrepancies drive consumer behavior. Jour-
nal of Consumer Psychology,27, 133146. https://doi.
org/10.1016/j.jcps.2016.05.003
Mandler, G. (1982). The structure of value: Accounting
for taste. In M. S. Clark, & S. T. Fiske (Eds.), Affect and
cognition: The seventeenth annual carnegie symposium on
cognition (pp. 336). Hillsdale, NJ: Lawrence Erlbaum.
Maoz, E., & Tybout, A. M. (2002). The moderating role of
involvement and differentiation in the evaluation of
brand extensions. Journal of Consumer Psychology,12,
119131. https://doi.org/10.1207/S15327663JCP1202_05
Mazar, N., & Zhong, C. (2010). Do green products make
us better people? Psychological Science,21, 494498.
https://doi.org/10.1177/0956797610363538
McGregor, I., Prentice, M., & Nash, K. (2012). Approach-
ing relief: Compensatory ideals relieve threat-induced
anxiety by promoting approach-motivated states. Social
Cognition,30, 689714. https://doi.org/10.1521/soco.
2012.30.6.689
Meyers-Levy, J., & Tybout, A. M. (1989). Schema con-
gruity as a basis for product evaluation. Journal of Con-
sumer Research,16,3954. https://doi.org/10.1086/
209192
Muraven, M., Tice, D. M., & Baumeister, R. F. (1998).
Self-control as a limited resource: Regulatory depletion
patterns. Journal of Personality and Social Psychology,74,
774. https://doi.org/10.1037/0022-3514.74.3.774
Noseworthy, T. J., Cotte, J., & Lee, S. H. (2011). The
effects of ad context and gender on the identication of
visually incongruent products. Journal of Consumer
Research,38, 358375. https://doi.org/10.1086/658472
Noseworthy, T. J., Muro, F. D., & Murray, K. B. (2014).
The role of arousal in congruity-based product evalua-
tion. Journal of Consumer Research,41, 11081126.
https://doi.org/10.1086/678301
Noseworthy, T. J., Murray, K. B., & Di Muro, F. (2018).
When two wrongs make a right: Using conjunctive
enablers to enhance evaluations for extremely incongru-
ent new products. Journal of Consumer Research,44,
13791396.
Noseworthy, T. J., & Trudel, R. (2011). Looks interesting,
but what does it do? Evaluation of incongruent product
form depends on positioning. Journal of Marketing
Research,48, 10081019. https://doi.org/10.1509/jmr.
10.0384
Peracchio, L. A., & Tybout, A. M. (1996). The moderating
role of prior knowledge in schema-based product eval-
uation. Journal of Consumer Research,23, 177192.
https://doi.org/10.1086/209476
Piaget, J. (1960). General problems of the psychological
development of the child. In J. M. Tanner & B. Inhelder
(Eds), Discussion on child development, proceedings of the
world health organization study group on psychological
development of the child (Vol. 4, pp. 327). New York,
NY: International Universities Press.
Priester, J. R., Godek, J., Nayakankuppum, D. J., & Park,
K. (2004). Brand congruity and comparative advertis-
ing: When and why comparative advertisements lead
to greater elaboration. Journal of Consumer Psychology,
14, 115123. https://doi.org/10.1207/s15327663jc
p1401&amp;2_13
Proulx, T., & Heine, S. J. (2008). The case of the transmo-
grifying experimenter afrmation of a moral schema
following implicit change detection. Psychological
Science,19, 12941300. https://doi.org/10.1111/j.1467-
9280.2008.02238.x
Proulx, T., Heine, S. J., & Vohs, K. D. (2010). When is the
unfamiliar the uncanny? Meaning afrmation after
exposure to absurdist literature, humor, and art. Person-
ality and Social Psychology Bulletin,36, 817829.
https://doi.org/10.1177/0146167210369896
94 Taylor and Noseworthy
Proulx, T., & Inzlicht, M. (2012). The ve As of meaning
maintenance: Finding meaning in the theories of sense-
making. Psychological Inquiry,23, 317335. https://doi.
org/10.1080/1047840X.2012.702372
Raghunathan, R., Pham, M. T., & Corfman, K. P. (2006).
Informational properties of anxiety and sadness, and
displaced coping. Journal of Consumer Research,32, 596
601. https://doi.org/10.1086/500491
Randles, D., Proulx, T., & Heine, S. J. (2011). Turn-frogs
and careful sweaters: Non-conscious perception of
incongruous word pairings provokes uid compensa-
tion. Journal of Experimental Social Psychology,47, 246
249. https://doi.org/10.1016/j.jesp.2010.07.020
Savitsky, K., Medvec, V. H., Charlton, A. E., & Gilovich,
T. (1998). What, Me Worry?: Arousal, misattribution,
and the effect of temporal distance on condence. Per-
sonality and Social Psychology Bulletin,24, 529536.
https://doi.org/10.1177/0146167298245008
Schooler, R. D. (1965). Product bias in the central Ameri-
can common market. Journal of Marketing Research,2,
394397. https://doi.org/10.1177/002224376500200407
Shimp, T. A., & Sharma, S. (1987). Consumer ethnocen-
trism: Construction and validation of the CETSCALE.
Journal of Marketing Research,24, 280289. https://doi.
org/10.1177/002224378702400304
Sommet, N., & Morselli, D. (2017). Keep calm and learn
multilevel logistic modeling: A simplied three-step
procedure using Stata, R, Mplus, and SPSS. International
Review of Social Psychology,30, 229230. https://doi.
org/10.5334/irsp.162
Spielberger, C. D., & Reheiser, E. C. (2009). Assessment of
emotions: Anxiety, anger, depression, and curiosity.
Applied Psychology: Health and Well-Being,1, 271302.
Steele, C. M. (1988). The psychology of self-afrmation:
Sustaining the integrity of the self. In L. Berkowitz
(Ed.), Advances in Experimental Social Psychology (Vol.
21, pp. 181227). San Diego, CA: Academic Press.
Taylor, N., Noseworthy, T. J., & Pancer, E. (2019). Super-
size my chances: Promotional lotteries impact product
size choices. Journal of Consumer Psychology,29,7988.
https://doi.org/10.1002/jcpy.1063
Tingley, D., Yamamoto, T., Hirose, K., Keele, L., & Imai,
K. (2014). Mediation: R package for causal mediation
analysis. Journal of Statistical Software,59(5), 138.
Tritt, S. M., Inzlicht, M., & Harmon-Jones, E. (2012).
Toward a biological understanding of mortality sal-
ience (and other threat compensation processes). Social
Cognition,30, 715733. https://doi.org/10.1521/soco.
2012.30.6.715
United Nations Statistics Division. (2018). Standard coun-
try and area codes classications methodology.
Retrieved May 7, 2018, from https://unstats.un.org/
unsd/methodology/m49/.
Wall, M., & Heslop, L. A. (1986). Consumer attitudes
toward canadian-made versus imported products. Jour-
nal of the Academy of Marketing Science,14,2736.
https://doi.org/10.1007/BF02722154
Wertenbroch, K., & Skiera, B. (2002). Measuring con-
sumerswillingness to pay at the point of purchase.
Journal of Marketing Research,39, 228241. https://doi.
org/10.1509/jmkr.39.2.228.19086
Supporting Information
Additional supporting information may be found in
the online version of this article at the publishers
website:
Appendix S1. Methodological Details Appendix
Appendix S2. Stimuli for Study 1
Appendix S3. Stimuli for Study 2
Appendix S4. Stimuli for Study 3
Appendix S5. Stimuli for Study 4
Compensating for Innovation 95
... Make "Incongruent" to Be "Excellent" of the same category (Noseworthy and Trudel, 2011;Jhang et al., 2012;Taylor and Noseworthy, 2020). Statistics also show that the failure rate of product innovation is very high in both traditional and emerging industries (Gourville, 2006), with failure estimates ranging from 40 to 90%. ...
... Although moderately incongruent products have higher acceptance, extremely incongruent products are also meaningful to enterprises (Jhang et al., 2012;Huang and Wan, 2019;Zhu et al., 2020). From the subjective perspective, due to the serious homogenization of products, enterprises will carry out extremely incongruent product design in order to better distinguish their products from similar products (Noseworthy and Trudel, 2011;Taylor and Noseworthy, 2020), which will lead to better brand awareness (innovative brand; Clemente et al., 2013;Huang and Wan, 2019). From the objective point of view, the invention of new technology will inevitably make product design extremely incongruent with the original product schema. ...
... Fluid compensation refers to compensation made with the assistance of beliefs outside the field when it is unable to be made through the beliefs within the field after the meaning is violated (Tesser, 2000). For example, once the participants encounter obstacles in reading surreal short stories, they tend to identify more with their culture (Proulx and Inzlicht, 2012), and it is a pattern that unconsciously enhances connectivity in unrelated domains (Taylor and Noseworthy, 2020). The extant means of fluid compensation mainly include semantic associations (i.e., dominant brand), ethical beliefs (i.e., green consumption), affirming personal culture (i.e., ethnocentric theory), etc. ...
Article
Full-text available
Under fierce market competition, firms are accelerating the pace of product innovation, which may bring extremely incongruent new products. Generally, consumers are unfamiliar with extreme incongruence, resulting in passive product evaluation and a low success rate. This study draws on the meaning maintenance model of fluid compensation theory to examine how extremely incongruent new products influence product evaluation. Three experiments indicate that consumers’ product evaluation of extremely incongruent new products is significantly lower than that of congruent products, and it can be notably improved by three ways of fluid compensation. The findings revealed that anxiety moderates the relationship between extremely incongruent new products and product evaluations, and further self-esteem moderates the relationship on the main effects. Overall, our research reconciles divergent findings on the effect of extremely incongruent new products on the effects of product evaluation and further provides rich managerial implications for product innovation management.
... Consumer morality is defined as a set of rules, ideals, and guidelines that apply to all stages of consumer decision-making: acquisition, consumption, and disposal. Taylor and Noseworthy (2020) found that when 208 students were exposed to extreme product inconsistency or expectancy violations (e.g. belief versus evidence), they attempted to confirm their ethical convictions by making environmentally friendly consumption decisions. ...
Article
Full-text available
This research aims to investigate the relationship between moral deficiency and moral consumption. Consumers' moral values cannot be separated from their consumption activities. In other words, consumers' spending preferences may be an expression of their beliefs about what is right and wrong. A less explored concept within moral consumption behavior theory is 'moral deficiency'. To the best of our knowledge, this is the first research effort to integrate green purchasing and religious purchasing under the banner of moral consumption behavior. There are two studies: Study 1 aimed to measure the moral deficiency of participants through moral scenarios (implicit) and then test its relationship with the green purchase and religious purchase, two proxies of moral consumption. A total of 121 academia were chosen via the nonprobability sampling method. To improve the results of the prior study, Study 2 aimed to measure the moral deficiency of participants through moral deficiency self-report (explicit) and then test its effects on green purchase and religious purchase. A total of 208 participants from the general public were recruited via the nonprobability sampling method. The findings of the two studies suggest that participants with high moral deficiency showed more intention to engage in moral consumption behavior.
... A related principle in categorization theory that fits the mental accounting lens could be how psychological budgets interact. If consumers see calorie budgets similar to financial budgets, sharing in one may lead people to compensate for a biased shortfall in the other-re: fluid compensation (Taylor & Noseworthy, 2020). Another possibility relates to egocentric categorization theory (Weiss & Johar, 2016), which shows that people assimilate to products they own, and contrast products they do not own. ...
Article
Full-text available
Food sharing has become quite popular over the last decade, with companies offering food options specifically designed to be shared. As the popularity has grown, so too has concerns over the potential negative impact on consumer health. Despite companies’ explicit claims to the contrary, critics maintain that food sharing may be encouraging excessive caloric intake. The current article provides the first systematic exploration of why this may be happening. Three main and two supplementary studies suggest that food sharing reduces perceived ownership, which, in turn, leads people to mentally decouple calories from their consequence. Thus, sharing can reduce the perceived fattening potential of a consumption episode without biasing caloric estimates. This phenomenon persists even when explicit caloric information is provided, and it applies to both healthy and unhealthy foods. Importantly, we establish a relevant downstream consequence by illustrating that people tend to subsequently select calorie‐dense foods after underestimating the fattening potential of a shared consumption episode. A roadmap for future research and practical implications are discussed.
... For the "partially" green product, there are no differences in WPP, PPP and PI for both individuals (Moser, 2015). Interestingly, for the non-green product, the individual with higher protected values is not willing to pay more, has lower purchase intentions, and does not perceive the product as having a better performance (Taylor & Noseworthy, 2020). It is worth mentioning that, as in the moderation analysis of the effects of the GSCM practices on the dependent variables, the effects of the moderator on consumers' PPP are the weakest ones compared with the effects of the moderator on WPP and PI. ...
Thesis
Full-text available
Widespread efforts are being made to mitigate the environmental crisis mainly driven by human activities. From a supply chain management perspective, companies improve their sustainable performance while increasing their organizational performance. Specifically, companies are interested in understanding if consumers care about their green practices because consumers are the sources of the companies’ profitability. It is supposed that consumers would be willing to pay a premium (WPP), or would have higher purchase intentions (PI), for environmentally differentiated products, but that is not always the case. Moreover, there is scant evidence regarding the integrated effect of intra- and inter-organizational Green Supply Chain Management (GSCM) practices on green consumerism. Therefore, this study adopts two psychological approaches (i.e., protected values and halo effect) to describe the mentioned relationship in two models that encompass mediation and moderation. Data were collected from 351 Colombian university students through a behavioral experiment with three product-based conditions, and the hypothetical models were tested using two-instances repeated-measures linear regressions and non-parametric tests. The results indicated that the perceived product performance mediates the effect of the GSCM practices on consumers’ WPP and PI (halo effect). Additionally, consumers’ moral orientation toward the environment (protected values) moderates the effect of GSCM practices on consumers’ WPP, PI and perceived product performance. In other words, the study found that people who held protected values evaluate the product better not just for its green attributes, but because of their increased perception of the product performance. The contributions are centered in the role of psychological approaches in the GSCM studies to understand consumers’ preferences.
... dampen) adoption decisions when hope is also strong. These findings augment prior studies that show that that anxiety can cause individuals to reject new products (e.g., Lee, Kim, and Vohs 2011;Meuter et al. 2003), as well as studies that document methods to reduce anxiety when new products are incongruent with expectations (i.e., Taylor and Noseworthy 2020). Our findings suggest that, in some situations, anxiety can induce an approach (vs. ...
Article
Full-text available
New products can evoke anticipatory emotions such as hope and anxiety. On the one hand, consumers might hope that innovative offerings will produce goal-congruent outcomes; on the other hand, they might also be anxious about possible outcomes that are goal-incongruent. The authors demonstrate the provocative and counterintuitive finding that strong anxiety about potentially goal-incongruent outcomes from a new product actually enhances (vs. weakens) consequential adoption intentions (Study 1) and actual adoption (Studies 2 and 3) when hope is also strong. The authors test action planning (a form of elaboration) and perceived control over outcomes as serial mediators to explain this effect. They find that the proposed mechanism holds even after they consider alternative explanations, including pain/gain inferences, confidence in achieving goal-congruent outcomes, global elaboration, affective forecasts, and motivated reasoning. Managerially, the findings suggest that when bringing a new product to market, new product adoption may be greatest when hope and anxiety are both strong. The findings also point to ways in which marketers might enhance hope and/or anxiety, and they suggest that the use of potentially anxiety-inducing tactics such as disclaimers in ads and on packages might not deter adoption when hope is also strong.
Article
In recent years, research focusing on consumer reactions to threats due to self-discrepancy has received growing academic interest. These threats and consumer reactions are discussed in terms of compensatory consumption in consumer research. A synthesis of the findings in a review of studies on compensatory consumption published between 2017 and 2022 identified three issues related to: (1) determinants of coping strategies, (2) successive compensatory consumption, and (3) ways to avoid specific coping strategies. The future research agenda is discussed based on these issues raised in prior research.
Article
Full-text available
Previous research has shown that perceived incongruity affects product evaluations in an inverted U shape. However, it remains unclear whether this relation also occurs at individual levels with continuous incongruity measures, and for products with repurposed materials. Five within-subjects studies do not show the inverted U relation across all participants. Instead, consumer subgroups show a monotonic relation: higher congruity leads to higher product evaluations. This aligns with processing fluency theory. Additionally, we demonstrate that the degree of processing from raw to end materials and the extent to which materials fulfil product functions mediate the effect of repurposed materials on perceived incongruity.
Article
An increasing number of advertisements imply that market competition is becoming fiercer. This study identifies different types of advertising elements to suggest advertising design strategies that can attract consumers' attention. In this paper, we investigate the effects of schema congruity and attribute relevance on attitude toward advertising. We refer to schema congruity that is created by product color design as “color congruity” and verify the influence of color congruity. To identify the underlying psychological mechanism by which these effects occur, we investigate the mediating effect of perceived novelty and typicality on advertisement attitude. Study 1 identifies the effect of color congruity and product attribute relevance on advertising attitude. When presenting secondary attributes, low color congruity leads to a more favorable advertising attitude than that of high color congruity, whereas when presenting primary attributes, high color congruity is more effective. In Study 2, for secondary attributes, perceived novelty explains the positive impact of low color congruity on advertising attitude. However, for primary attributes, the effect of high congruity on advertising attitude is mediated by perceived typicality. These findings may be useful in proving guidelines for advertising companies and marketers that want to create fresh content by suggesting how to combine advertising elements that will attract consumers' attention.
Article
Full-text available
What motivates people to consume and engage with food media on social networks? We adopt an evolutionary lens to suggest that the valence of people’s affective state varies by the implied caloric density of food media, which has a direct impact on social media engagement. First, we analyze a catalog of Buzzfeed’s Tasty videos based on nutritional content derived from the dish’s ingredients and find that visualizing caloric density (i.e., calories per serving) positively influences likes, comments, and shares on Facebook. We then replicate this phenomenon in an experiment, providing preliminary evidence for the role of affect as an explanatory mechanism. We conclude by isolating the role of affect with a classic misattribution task, which attenuates the elevated engagement resulting from exposure to calorie‐dense food media. These findings contribute to the dialogue on the antecedents of social media engagement and offer implications for content developers, advertisers, consumer health advocates, and policy makers.
Article
Full-text available
While there is an ample amount of consumer behavior research that recruits processing fluency as an explanatory construct, the question how to best measure the fluency experience has received little attention. Therefore, there is a lack of consistency in measuring the construct, particularly with regard to the use of single‐item versus multi‐item measures. The current research, thus, aims to investigate how processing fluency can be consistently measured across different experimental fluency manipulations and which type of measure has the highest validity. Based on classic scale development procedures, we propose a reliable and valid multi‐item measure and compare this measure against a single‐item measure in terms of predictive validity. We show that both measures mediate the effect of five established fluency manipulations and that the single‐item measure is sufficient. In addition to providing a measure for future research that can be adapted to different empirical settings, we provide empirical evidence on the replicability of fluency effects and on the theoretical conjecture that people have a uniform fluency experience across different manipulations of fluency.
Article
Full-text available
Promotional lotteries offer consumers a chance to win one of many prizes along with their purchase. Critically, as is often the case, these campaigns not only include an assortment of prizes, but also an assortment of offerings that one can buy to enter the lottery—such as a small or an extra‐large coffee. While companies regularly advertise that the objective odds of winning do not vary by the size of their product offerings, recent anecdotal evidence suggest that consumers behave as if it does. The net result is that consumers seem to be supersizing during promotional lotteries, and thus purchasing larger‐sized items. Eight studies (four core and four supplementary) and a single‐paper meta‐analysis confirm that the supersizing phenomenon is indeed real and provides evidence that this behavior is the manifestation of consumers elevating their sense of control. Specifically, supersizing serves to gain psychological control over the pursuit of a desirable, but seemingly unobtainable, outcome. This article is protected by copyright. All rights reserved.
Article
Full-text available
The success of new incongruent products hinges largely on whether consumers can efficiently make sense of the product. One of the most efficient ways that people make sense of new objects is through feature-based association. Such associations often incorporate an enabler (e.g., the color green) to help make sense of a semantically related feature (e.g., vitamin enriched). Evidence from three studies suggests that marketers can strategically incorporate enablers in product design to help consumers make sense of an extremely incongruent feature. As a result, consumers tend to reflect more favorably on the product. Furthermore, the authors find that even if the enabler itself is incongruent and leads to lower evaluations on its own, when combined with an atypical feature the effect can still be positive. Thus, a small but semantically meaningful adjustment in design can help marketers successfully introduce extremely incongruent innovations. © The Author 2017. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved.
Article
Full-text available
This article details a correction to the article: Sommet, N. & Morselli, D., (2017). Keep Calm and Learn Multilevel Logistic Modeling: A Simplified Three-Step Procedure Using Stata, R, Mplus, and SPSS. 'International Review of Social Psychology'. 30(1), pp. 203–218. DOI: https://doi.org/10.5334/irsp.90
Article
Full-text available
Consumer goods and services have psychological value that can equal or exceed their functional value. A burgeoning literature demonstrates that one source of value emerges from the capacity for products to serve as a psychological salve that reduces various forms of distress across numerous domains. This review systematically organizes and integrates the literature on the use of consumer behavior as a means to regulate self-discrepancies, or the incongruities between how one currently perceives oneself and how one desires to view oneself (Higgins, 1987). We introduce a Compensatory Consumer Behavior Model to explain the psychological consequences of self-discrepancies on consumer behavior. This model delineates five distinct strategies by which consumers cope with self-discrepancies: direct resolution, symbolic self-completion, dissociation, escapism, and fluid compensation. Finally, the authors raise critical research questions to guide future research endeavors. Overall, the present review provides both a primer on compensatory consumer behavior and sets an agenda for future research.
Article
This article shows that regional fears, jealousies, and animosities constitute invisible barriers to increased trade within CACM, and that the attitude toward the people of a given country is a factor in existing preconceptions regarding the products of that country.
Article
The concept of consumer ethnocentrism is introduced and a corresponding measure, the CETSCALE, is formulated and validated. Four separate studies provide support for the CETSCALE's reliability and convergent and discriminant validity. A series of nomological validity tests show consumer ethnocentrism to be moderately predictive of theoretically related constructs.