Received: 19 June 2019 | Revised: 4 October 2019 | Accepted: 11 October 2019

DOI: 10.1111/jasp.12639

ORIGINAL ARTICLE

Helping the “couch potato”: A cognitive dissonance approach to

increasing exercise in the elderly

Joel Cooper | Lauren A. Feldman

Department of Psychology, Princeton

University, Princeton, NJ, USA Abstract
In this manuscript, we present the results of the research that employs a technique
Correspondence
Lauren A. Feldman, Department of inspired by the cognitive dissonance theory to increase older adults’ intentions to
Psychology, Princeton University, Princeton, exercise. The failure to engage in physical activity is at the root of many psycho‐
NJ 08540, USA.
Email: [email protected] logical and physical ailments, from depression to diabetes. This is especially true of
older adults for whom sedentary behavior is increasingly common. The current work
reports the results of an experiment and a preliminary study that adopted the hypoc‐
risy paradigm of dissonance to motivate increased behavioral intentions to exercise.
Elderly participants volunteered to generate strong arguments in favor of engaging in
physical exercise. As predicted by the dissonance theory, when urged to recall times
in the past in which they had chosen not to exercise, participants increased their in‐
tention to exercise in the future and became more positive about the importance of
exercise in their lives. Participants who only advocated for greater exercise but were
not made aware of their past transgressions did not show increases in behavioral
intentions. The results were interpreted as evidence for the effectiveness of a social
psychological theory on a significant social problem.

1 | I NTRO D U C TI O N as well as psychological disorders such as depression (de Wit, van

Straten, Lamers, Cuijpers, & Penninx, 2011). The risk created by
From its inception, social psychological research has entertained the physical inactivity occurs at every adult age group, but is especially
dual mission of advancing theory about human behavior and using troublesome as people age (Evenson, Buchner, & Morland, 2012).
theory as a lens through which social problems could be understood Research shows that sedentary behavior increases with age and
and influenced. The current work adopts and extends the theory accelerates chronic aspects of age‐related declines (Booth, Laye,
of cognitive dissonance (Festinger, 1957) to a problem of social and & Roberts, 2011; Evenson et al., 2012). Fewer than 35% of adults
personal significance: The increased prevalence of physical and aged 65–75 meet the age‐appropriate physical activity guidelines of
mental illness caused by the lack of physical activity and exercise. In the U.S. Department of Health and Human Services and fewer than
this research, we take the position that cognitive dissonance can be 18% of adults 75 years of age and older do so (White, Wójcicki, &
brought to bear to create a theoretically based intervention at the McAuley, 2012).
service of increasing people's motivation and intention to get off the The issue addressed in the current study is to derive and test a
couch and increase their level of physical activity. theoretically based intervention to increase physical activity in an
Sedentary behavior is increasingly acknowledged as a precursor elderly population. Despite the positive physical and mental health
to a wide array of physical and mental health concerns. Research benefits of physical activity, elderly adults are remarkably inactive.
has been accumulating that insufficient physical activity is associ‐ For those without physical impediments to exercise, psychological
ated with increased risk of chronic physical diseases like diabetes barriers exist that reduce their motivation. Some of these are based
and cancer (Fried, 2000; Thorp, Owen, Neuhaus, & Dunstan, 2011) on individual differences such as low self‐esteem or diminished

J Appl Soc Psychol. 2019;00:1–8. wileyonlinelibrary.com/journal/jasp

© 2019 Wiley Periodicals, Inc. | 1
2 | COOPER and FELDMAN

levels of self‐efficacy (Bandura, 1997, 2004; McAuley, Lox, & 1.2 | Dissonance and physical exercise
Duncan, 1993; White et al., 2012). A relatively small literature has
In a typical induced compliance experiment, people are induced
explored psychological approaches to increase exercise in adult and
to change their attitudes by advocating for a position with which
elderly populations. For example, researchers have found that adults
they disagree. The dilemma for applying the well‐established in‐
over the age of 65 who exercised with a group and perceived high
duced compliance procedure to health research is that advocating
task and social cohesion showed improved attitudes and intentions
for healthy life choices is generally proattitudinal for most people
to exercise (Estabrooks & Carron, 1999); other research on action
(Azdia, Girandola, & Andraud, 2002). The challenge is increasing
planning and exercise has shown that telephone‐delivered “booster
people's motivation to take action consistent with their beliefs. For
sessions” focusing on positive exercise experiences and action plan
example, teens are in favor of taking precautions to prevent HIV/
adherence led to improved exercise maintenance in individuals in
AIDs but often act in ways that put themselves at risk (Becker &
physical rehabilitation (Fleig, Pomp, Schwarzer, & Lippke, 2013; see
Joseph, 1988; Beckman, Harvey, & Tiersky, 2010). People who seek
also Bélanger‐Gravel, Godin, & Amireault, 2013).
to change their diet believe that doing so is important, but many
While promising, many of these interventions require a high time
have difficulty motivating themselves to take the steps needed to
commitment or some such barrier to entry (i.e., several follow‐up
change their eating habits. Similarly, people of all ages believe that
sessions, or the consistent presence of a social group). Given the low
physical exercise is valuable but nonetheless remain sedentary. The
cost and high efficiency of cognitive dissonance procedures, our goal
term “couch potato” vividly invokes the cultural truism that physical
in this research is to adopt a situationally based dissonance interven‐
activity is preferred to indolence. This is especially true of the elderly
tion that bolsters older adults’ motivation and intention to engage in
who become increasingly sedentary relative to their capacity to be
the physical activity.
active (Evenson et al., 2012).
The central question in this research is whether we can use dis‐
1.1 | Cognitive dissonance as intervention
sonance to motivate the elderly to increase their intention to en‐
Cognitive dissonance (Festinger, 1957) is a general motivational ap‐ gage in the physical activity. Stone, Aronson and their colleagues
proach that has been studied for several decades (Cooper, 2007; (Aronson, 1999; Focella, Stone, Fernandez, Cooper, & Hogg, 2016;
Harmon‐Jones, Amodio, & Harmon‐Jones, 2009). In the labora‐ Fried & Aronson, 1995; Stone, Aronson, Crain, Winslow, & Fried,
tory, it has typically been invoked as a theory of attitude change, 1994) have shown that people experience cognitive dissonance
but its general motivational properties make it well suited to de‐ when they are made aware of situations in their past in which they
rive techniques that can lead to change in important health‐related have acted inconsistently with their current attitudes. The apparent
behaviors. hypocrisy between people's stated attitudes and their past behav‐
When people realize that they have cognitions that are incom‐ ior leads to motivation to reduce the inconsistency (see Priolo et al.,
patible with each other, cognitive dissonance ensues. The motiva‐ 2019, for a meta‐analytic review). We propose that the elderly can
tion to reduce dissonance leads to a change in the cognition that is become more motivated to engage in the physical activity if they are
most readily changeable. Beginning with Festinger and Carlsmith's confronted with the inconsistency between their positive attitudes
(1959) classic experiment, myriad research studies have shown that toward the physical activity and their lack of activity in the recent
people who make public statements in contrast to their attitudes past.
reduce their dissonance by changing their attitudes to make them The study by Stone et al. (1994) is particularly relevant to the
more compatible with their behavior. Similarly, people who expend current goal. Their project was designed to show the impact of cog‐
unpleasant effort to reach a goal psychologically elevate the attrac‐ nitive dissonance in promoting greater safe‐sex behaviors among
tiveness of the goal (Aronson & Mills, 1959); people who make a university students. Male university students were recruited to
difficult choice between two alternatives subsequently distort the create what they believed were public service announcements that
value of the alternatives to justify their choice (Brehm, 1956) and would be used in local high schools to promote the use of condoms.
people who refrain from a pleasant activity for barely sufficient jus‐ The students had positive attitudes toward the use of condoms and
tification resort to derogating the activity as a way of dealing with believed that they should always be used during sex. After making
their dissonance (Aronson & Carlsmith, 1963). their attitude‐consistent speech about always using condoms, par‐
Prior research has shown that dissonance can motivate peo‐ ticipants in the key dissonance arousing condition were asked to re‐
ple to take action that benefits their physical and mental health. member times that they had engaged in sex without a condom. The
Axsom (1989) demonstrated that dissonance motivated university juxtaposition of recalling instances discrepant with the attitude‐con‐
students to overcome their anxiety about public speaking. Cooper sistent speech aroused dissonance. Later, when they were given an
(1980) showed that dissonance helped people overcome phobias opportunity to purchase condoms at a reduced cost, students in the
toward snakes. In a meta‐analytic review, Stice and Shaw (2004) dissonance conditions purchased significantly more condoms than
showed that dissonance interventions were effective at help‐ participants in any of the various control conditions.
ing people with a goal of weight loss in a variety of weight loss In a replication and extension of this study, Stone, Wiegand,
programs. Cooper, and Aronson (1997) provided research participants with
COOPER and FELDMAN | 3

other means of reducing their dissonance rather than committing a no‐treatment control condition to the research design, allowing us to
to using condoms. They found that when dissonance was aroused test whether there is a positive impact of having people advocate their
by the perception of inconsistency between past behavior and cur‐ prohealth position in the absence of the hypocrisy manipulation. We
rent attitudes, participants preferred to reduce their dissonance in predict that being made aware of the times that they failed to exercise
as direct a manner as possible, that is, committing to using condoms. following their advocacy of exercise (hypocrisy condition) will increase
Even when there were easier alternatives that might have alleviated participants’ attitudes toward exercise, as well as their behavioral in‐
their dissonance, the students preferred to reduce their dissonance tentions to exercise in the future, more so than those who only advo‐
by changing their safe sex behaviors. cate in favor of exercise (advocacy‐only condition) or a no treatment
We chose to apply the theoretical concept of hypocrisy to mo‐ group (control condition).
tivate the exercise activity in the elderly because older adults are at
increased risk of physical and mental decline associated with inac‐
2 | M E TH O D
tivity. Although elderly populations have been generally overlooked
in dissonance research, recent work has shown that the elderly
2.1 | Participants
are as susceptible to the effects of induced compliance as younger
adults (Cooper & Feldman, 2019). We believe that the induction of We conducted our research with a special population, recruiting
hypocrisy can be similarly successful with older adults and will lead 63 healthy residents of an “over‐55” community in central New
to greater motivation to increase their magnitude of the physical Jersey who volunteered for the study.1 They were contacted by a
activity. female experimenter who was a resident of the community.
Inclusion criteria included being older than 65 years of age and
with the ability to understand the instructions. The experimenter
1.3 | Preliminary research
explained that she was conducting research on physical exercise
A preliminary study gave us confidence in the approach we devel‐ for a research group at Princeton University and asked if they
oped. We asked volunteers at a Jewish Community Center to help would be willing to volunteer. The mean age of the participants
us with creating a public service pamphlet on the benefits of exer‐ was 71.4 (SD = 6.7) with a range of 65–85. Two‐thirds were female.
cise. Forty‐four participants (28 women) between the ages of 60 All of the participants who volunteered for the study completed
and 92 were recruited (M = 75.61, SD = 9.29). We utilized a stand‐ the questionnaires and none needed to be excluded for the final
ard hypocrisy paradigm that induced hypocrisy through commit‐ analyses. Neither the age nor gender distribution varied by experi‐
ment to their prohealth attitude toward physical exercise, as well mental condition.
as the mindful awareness that their past behavior had not always
been consistent with this commitment. To establish commitment,
2.2 | Procedure
participants were asked to write a statement advocating for the
health benefits of physical exercise, ostensibly to be compiled and The participants were tested individually in public areas within their
distributed as motivation to another group of older adults trying community. The experimenter told participants that the researchers
to improve their health through exercise. The participants in the at Princeton were making a booklet to convince seniors about the im‐
hypocrisy condition were then presented with a list of reasons portance of the physical activity. Rather than merely using facts and
that one could potentially give for not having exercised enough statistics related to exercise, the researchers wanted a booklet that
and were asked to indicate the reasons that applied to them, thus contained arguments written by seniors for seniors extolling the ben‐
establishing awareness about times in which they hadn't exer‐ efits of exercise to seniors. She asked each person if he or she would
cised enough. In the advocacy‐only control condition, participants be willing to generate approximately four proexercise arguments that
skipped this part of the procedure and immediately continued to would be included in the book. All agreed to do so. The participants
the dependent measures; that is, they were only asked to write generated their arguments orally and the experimenter converted
the statement, but did not recall past instances in which they had them into written bullet points.
the opportunity to exercise but failed to do so. At the conclusion Following the bullet point arguments, participants who had been
of the study, participants were asked about their attitudes related randomly assigned to the Hypocrisy condition (n = 21) were told, “We
to exercise. Although the small number of participants limited the are also trying to find out what prevents some people from exercis‐
generalizability of the results, the data showed that participants ing even when they have the opportunity. The researchers at the
who went through the hypocrisy induction rated the physical activ‐ university thought it would be helpful if you could think of one or
ity as more important than participants who did not, t(42) = 2.56, two times that you had the opportunity to exercise, but you chose
p = .014, Cohen's d = 0.77.
Encouraged by the preliminary data, we conducted a new study to
1
examine the impact of hypocrisy‐induced dissonance on older adults' A priori power analysis indicates that, assuming a small effect size, in order to achieve
80% power we would have needed to recruit 159 participants in total (53 per condition).
attitudes and behavioral intentions toward exercise. We conducted We were not able to recruit this number of participants, due to the size and restrictions
the main study in the community in which participants lived. We added of the senior citizen community where our special population was recruited.
4 | COOPER and FELDMAN

not to. Just think of some instances that you really remember.” All 1994; McAuley, Elavksy, Motl, Konopack, Hu, & Marquez, 2005), the
participants were able to generate occasions that they had not exer‐ participants completed a single‐item self‐esteem measure created
cised and they were recorded by the experimenter. and validated by Robins, Hendin, and Trzesniewski (2001). They also
In the Advocacy‐Only condition (n = 21), participants generated filled out single‐item measures asking about their current level of
their proexercise arguments but were not asked to generate memo‐ comfort and their feelings of self‐efficacy.
ries of having failed to exercise. Instead, they completed the depen‐
dent measures immediately following their bullet‐point arguments in
3 | R E S U LT S
favor of exercising.
Participants who had been randomly assigned to the Control con‐
3.1 | Behavioral intentions
dition (n = 21) agreed to participate in research on physical exercise
and proceeded directly to the dependent measure questionnaire. The data provide strong support for the effectiveness of the dis‐
sonance procedure to produce changes in participants’ behavior
intentions. We asked participants to estimate the number of hours
2.2.1 | Dependent measures
that they intended to exercise during the following week. Hypocrisy
The dependent measures consisted of a questionnaire that in‐ participants expected to exercise for 5.2 hr, advocacy‐only partici‐
quired about several issues, including participants’ intentions to pants intended to exercise for 4.0 hr and the control group planned
increase their exercise behavior, their attitudes regarding the im‐ to exercise for 3.7 hr. An ANOVA performed on the behavioral inten‐
portance of exercise, and the number of hours that they planned tion data showed that the difference was significant, F(2, 60) = 3.28,
to exercise per week in the future. To measure specific changes in p = .044, η2 = .098. Post hoc comparisons using Tukey's HSD test
participants' intentions to alter the time they intended to devote showed, as predicted, that the hypocrisy condition differed signifi‐
to exercise, they were asked to indicate specifically how many cantly from the no‐treatment control condition (p = .018). The dif‐
hours they had exercised per week in the recent past and then ference between the hypocrisy and advocacy‐only conditions was
to “indicate how many times per week you plan to exercise in the marginal (p = .064), and the difference between the control and ad‐
future,” thus providing us with an estimated difference between vocacy‐only conditions was not significant (p = .584).
participants' past exercise levels and their expected exercise lev‐ We also collected data on participants’ recollection of the num‐
els in the future. ber of hours that they had exercised during the week prior to the
We also included four items that asked about participants' atti‐ intervention. The pre and postintervention data are presented in
tudes toward the importance of exercise and their general intentions Figure 1. The results support our predictions. A 2 × 3 mixed model
to exercise in the future. As a measure of their behavioral intentions ANOVA of the data shows no difference in prior exercise behavior,
to exercise in the future, participants were asked, “In the future I plan a main effect for time, F(1, 61) = 4.24, p < .05, η 2 = .112, and, as
to make an increased effort to do regular physical exercise”, “Going predicted, a significant time × condition interaction, F(2, 61) = 8.81,
forward, I plan to exercise more often than I did before” and “When p < .001, η2 = .213. Post hoc tests comparing the effects of the hy‐
I exercise in the near future, I plan to exercise for longer periods of pocrisy, advocacy‐only, and control conditions on participants’ pre
time than I had been before.” As a measure of their attitudes toward and postchange scores are reported below.
exercise, participants were asked, “In my opinion, regular physical An analysis of simple change scores for each participant is con‐
exercise is important for maintaining good health.” Each item was sistent with the mixed model ANOVA. We subtracted the number of
asked on 7‐point Likert type scales ranging from “strongly agree” to hours each participant recalled exercising prior to the intervention
“strongly disagree.” from their intention to exercise in the following week. Participants
Summarizing our hypotheses, we predicted that participants in in the hypocrisy condition expected to exercise 1.7 hr per week
the hypocrisy condition would show the highest degree of inten‐ more than they had done in the past. Advocacy‐only participants
tion to increase their exercise behavior compared to participants expected to increase their exercise by 32 min and control partici‐
in the control or advocacy‐only conditions. Second, we predicted pants expected to increase their exercise by 23 min. The ANOVA
that participants would express more favorable attitudes about on the change scores shows that the difference was significant, F(2,
the importance of exercising under hypocrisy condition relative to 62) = 8.59, p < .001, η2 = .223. In addition, post hoc comparisons
those in the other two conditions. We had no dissonance‐based using Tukey's HSD procedure showed that the increase in intended
prediction as to whether advocacy‐only participants would express hours for the hypocrisy condition was significantly greater than
greater intention to exercise than participants in the control con‐ the control condition (p < .001) and the advocacy‐only condition
dition but tentatively expected that advocating for the importance (p < 002). There was no difference between the control and advo‐
of exercise might result in somewhat improved intentions and cacy‐only conditions (p = .49).
attitudes. Other questions that inquired about behavioral intentions
Finally, in line with previous research on the role of discomfort in showed similar patterns of results but were not statistically signif‐
the dissonance process, as well as the role of self‐esteem and self‐ icant. Participants in the hypocrisy condition estimated that, in gen‐
efficacy in the maintenance of prohealth behaviors (Elliot & Devine, eral, they planned to exercise for longer periods of time, planned to
COOPER and FELDMAN | 5

F I G U R E 1 Reported hours of
exercise. Dark bars = number of hours
reported prior to the intervention; light
bars = number of hours participants
intended to exercise following week

TA B L E 1 Ratings of the importance of exercise There were no differences in any of the dependent measures as
a function of participants' gender or age.
Condition M SD

Hypocrisy 6.76 0.54

Advocacy Only 6.43 0.75 3.3 | Discomfort
Control 6.05 0.80 The response to the question inquiring about discomfort showed al‐
Note: Higher numbers indicate greater importance, scale = 1–7. most no variance. About 61 of the 63 participants responded with
a 1 on the 1–5 scale (M = 1.03) where 1 was labeled as “very com‐
fortable”. Similarly, participants asserted a uniformly high sense of
make more of an effort to exercise and planned to exercise more self‐esteem and self‐efficacy.
often than participants in the other conditions, but those differences
were not statistically significant.
4 | D I S CU S S I O N

3.2 | Attitudes
Overall, the pattern of results supports the prediction that having
We asked our participants how important they thought exercise is people confront the discrepancy between their positive attitudes to‐
for maintaining good health (see Table 1). ward exercise and their past reluctance to engage creates a positive
As indicated by the control condition, our elderly participants force for change. Members of an elderly community advocated for
expressed extreme support for the importance of regular physical the importance of exercise, but when made aware of past behaviors
exercise, rating its importance a 6.05 (M = 6.05, SD = 0.80) on a scale discrepant with their attitude, they increased their specific commit‐
of 1–7, where 7 was extremely important. Nonetheless, the manip‐ ment toward exercise and became even more positive about how
ulation of hypocrisy increased the perceived importance of exercise much they believed in its importance. To our knowledge, this is the
(M = 6.76, SD = 0.54). The advocacy‐only group fell in between the first study to show that hypocrisy occurs in an elderly population
control and hypocrisy group (M = 6.43, SD = 0.75). Although the and that the hypocrisy paradigm of dissonance can create prohealth
ratings of importance were near the ceiling of the scale, the one‐ changes in the behavioral intentions and attitudes of the elderly.
way ANOVA was nonetheless significant, F(2, 60) = 5.38, p = .007, We were agnostic about predictions for proattitudinal advocacy
η2 = .152. Tukey's HSD procedure further showed that the hypocrisy in the advocacy‐only condition. Surely, advocacy focuses partici‐
condition led to a greater increase in importance compared to the pants' attention on the issue of exercise and, consistent with self‐
control condition (p < .002). In addition, the advocacy‐only condi‐ perception theory (Bem, 1972), may well lead to positive changes.
tion produced a marginally greater increase in importance than the Our prediction was that by connecting proattitudinal advocacy
control condition (p < .09), although this difference was not statis‐ with reminders of past transgressions, dissonance would be evoked
tically significant. There was no difference between the importance and lead to greater changes than occurred for self‐perception. The
ratings in the hypocrisy condition and the advocacy‐only conditions results supported our predictions but also gave some credence to
(p < .13). the mere advocacy of prohealth goals. On most of the dependent
6 | COOPER and FELDMAN

measures, advocacy‐only showed greater prohealth attitudes and to obtain behavioral data tracking real‐time changes in aerobic or
intentions than the control condition, but always less than the hy‐ other physical activities. Priolo et al.'s (2019) meta‐analysis found
pocrisy condition. that hypocrisy interventions have been effective at changing both
Participants in the hypocrisy conditions had little difficulty re‐ intentions and behavior. Overall, researchers have found a robust
calling instances in which they felt they could have exercised but relationship between behavioral intentions and behavior (Ajzen,
chose not to do so. Reasons ranged from those that might affect Albarracin, & Hornik, 2007; Armitage & Conner, 2001; Webb &
anyone, such as, “the phone rang just as I was about to go out. By the Sheeran, 2006). For example, in research on safe sex behaviors,
time the call was over, I didn't feel like walking anymore.” Other rea‐ Albarracín, Johnson, Fishbein, and Muellerleile (2001) found a
sons included physical problems related to age such as being afraid highly reliable correlation between the intentions to use condoms
of falling on wet pavement or worrying about a backache. Several and the actual use of condoms. At the aggregate level, Webb and
just said they felt lazy at the time they could have gone swimming Sheeran’s (2006) meta‐analysis of 47 studies found that a medium‐
or gone for a walk. What is important for establishing a dissonance‐ to‐large change in intention (d = 0.66) led to a small‐to‐medium
based intervention is that people were easily able to recall the oc‐ change in behavior (d = 0.36). In subsequent practical applications
casions in which physical activity was a distinct possibility but they of hypocrisy to exercise in the population of elderly adults, imple‐
had declined to follow through. The combination of recalling past mentation should pay attention to factors that have been shown
decisions that were inconsistent with their advocacy of exercise es‐ to strengthen the link between behavioral intentions and behav‐
tablishes the important conditions to motivate positive changes in ior. For example, adopting implementation plans and strengthening
their behavioral intentions. habits can serve to augment further the practical link between in‐
In principle, hypocrisy is driven by the need to reduce the un‐ tentions and actual behaviors.
comfortable tension state of dissonance (Aronson, 1999). We did Future work applying hypocrisy to elderly participants’ commit‐
not find differences in discomfort across conditions, with almost all ment to exercise should include potential individual differences in
of our elderly participants reporting a high degree of comfort. It is our participants that might augment or diminish the effect. For ex‐
possible that the context of the present study—that is, conducted ample, there may be a range of individual differences in the mere
at a community recreation area rather than a lab and conducted enjoyment of physical exercise, as well as participants' specific his‐
by a member of the residential community—influenced people to tories and experiences with exercise that play moderating roles.
report high levels of comfort and satisfaction. Although there has Additionally, it would be interesting to determine the cognitions
been some evidence that hypocrisy is accompanied by discomfort that are activated when participants recall the instances that pre‐
(Focella et al., 2016; Fried & Aronson, 1995), Priolo and colleagues' vented them from exercising and assess whether different types of
(2019) meta‐analysis of hypocrisy research concluded that evidence cognitions (e.g., body shaming) affect the success of the interven‐
for discomfort is equivocal. Establishing whether the discomfort tion. Future studies could examine the motivational impacts of such
is associated with hypocrisy‐driven effects is worthy of further concern types.
studies. We also note that two‐thirds of our participants were women.
Subsequent research seeking ways to leverage cognitive disso‐ Future research may wish to study gender more systematically. We
nance to improve elderly people's decision making about exercise found no differences as a function of gender and also note that, in
should be considered in the context of psychological factors that the age group of our participants, women in the U.S. population out‐
have been shown to affect healthy decision making. Bandura (2004) number men (100 women to every 79 men).
has suggested that the core determinants of healthy decision making Cognitive dissonance has been a powerful theoretical concept
are people's knowledge of the health risks and benefits of a partic‐ in social psychology for more than 60 years. It has led to provoca‐
ular action, their perceived self‐efficacy, and their expectations for tive hypotheses and findings of how people resolve inconsistencies
successful outcomes. White et al. (2012) showed that among a sam‐ and has clashed with other theoretical approaches (e.g., learning and
ple of older men and women, self‐efficacy and outcome expecta‐ self‐perception theories) for acceptance as a fundamental process in
tions were the factors that were directly related to changes in levels human social behavior (Cooper, 2007). At the heart of social psychol‐
of physical activity. A route for further research would be to inte‐ ogy has been the twin goals of expanding theory and using theory
grate the psychological factors that promote physical exercise with for the betterment of personal and social issues. Psychologists have
the dissonance‐based technique we have presented. For example, paid attention to Lewin's (1951) dictum that there is nothing more
interventions that motivate exercise through cognitive dissonance practical than a good theory. In the current instance, we have been
and that simultaneously facilitate greater self‐efficacy may be espe‐ able to expand the reach of dissonance by showing that people's
cially effective at eliminating the sedentary behaviors that put the fundamental reaction to the inconsistency of hypocrisy motivates
elderly at risk for physical and mental illness. them to increase their intentions to act in ways that can preserve
There are limitations to the current research that warrant and strengthen their physical and psychological health. “Couch
comment. First, we were not able to continue to contact our par‐ potatoes” are resistant to getting off the couch. The current work
ticipants to measure their actual behavior in the days and weeks fol‐ suggests that dissonance can be a powerful motivator to accomplish
lowing our intervention. Subsequent research should be designed that goal.
COOPER and FELDMAN | 7

ORCID de Wit, L., van Straten, A., Lamers, F., Cuijpers, P., & Penninx, B. (2011).
Are sedentary television watching and computer use behaviors as‐
Lauren A. Feldman https://orcid.org/0000-0002-0893-1665 sociated with anxiety and depressive disorders? Psychiatry Research,
186(2), 239–243. https​://doi.org/10.1016/j.psych​res.2010.07.003
Elliot, A. J., & Devine, P. G. (1994). On the motivational nature of cog‐
REFERENCES nitive dissonance: Dissonance as psychological discomfort. Journal
of Personality and Social Psychology, 67(3), 382–394. https​://doi.
Ajzen, I., Albarracin, D., & Hornik, R. (2007). Prediction and change of org/10.1037/0022-3514.67.3.382
health behavior: Applying the reasoned action approach. Mahwah, NJ: Estabrooks, P., & Carron, A. V. (1999). The influence of the group with
Lawrence Erlbaum Associates Publishers. elderly exercisers. Small Group Research, 30(4), 438–452. https​://doi.
Albarracín, D., Johnson, B. T., Fishbein, M., & Muellerleile, P. A. (2001). org/10.1177/10464​96499​03000403
Theories of reasoned action and planned behavior as models of con‐ Evenson, K. R., Buchner, D. M., & Morland, K. B. (2012). Objective mea‐
dom use: A meta‐analysis. Psychological Bulletin, 127(1), 142–161. surement of physical activity and sedentary behavior among US
https​://doi.org/10.1037/0033-2909.127.1.142 adults aged 60 years or older. Preventing Chronic Disease, 9, E26.
Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned Festinger, L. (1957). A theory of cognitive dissonance (Vol. 2). Stanford, CA:
behaviour: A meta‐analytic review. The British Journal of Social Stanford University Press.
Psychology, 40, 471–499. https​://doi.org/10.1348/01446​66011​ Festinger, L., & Carlsmith, J. M. (1959). Cognitive consequences of forced
64939​ compliance. The Journal of Abnormal and Social Psychology, 58(2), 203.
Aronson, E. (1999). Dissonance, hypocrisy and the self‐concept. In E. Fleig, L., Pomp, S., Schwarzer, R., & Lippke, S. (2013). Promoting exer‐
Harmon‐Jones & J. Mills (Eds.), Cognitive dissonance: Progress on a piv‐ cise maintenance: How interventions with booster sessions improve
otal theory in social psychology (pp. 103–126). Washington, DC: APA. long‐term rehabilitation outcomes. Rehabilitation Psychology, 58(4),
Aronson, E., & Carlsmith, J. M. (1963). The effect of severity of threat on 323–333. https​://doi.org/10.1037/a0033885
the devaluation of forbidden behavior. Journal of Abnormal and Social Focella, E. S., Stone, J., Fernandez, N. C., Cooper, J., & Hogg, M. A. (2016).
Psychology, 66, 584–588. Vicarious hypocrisy: Bolstering attitudes and taking action after ex‐
Aronson, E., & Mills, J. (1959). The effect of severity of initiation on liking posure to a hypocritical ingroup member. Journal of Experimental Social
for a group. Journal of Abnormal and Social Psychology, 59, 177–181. Psychology, 62, 89–102. https​://doi.org/10.1016/j.jesp.2015.09.014
Axsom, D. (1989). Cognitive dissonance and behavior change in psycho‐ Fried, C. B., & Aronson, E. (1995). Hypocrisy, misattribution and disso‐
therapy. Journal of Experimental Social Psychology, 25(3), 234–252. nance reduction. Personality & Social Psychology Bulletin, 21, 925–
https​://doi.org/10.1016/0022-1031(89)90021-8 933. https​://doi.org/10.1177/01461​67295​219007
Azdia, T., Girandola, F., & Andraud, O. (2002). Engagement, dissonance Fried, L. P. (2000). Epidemiology of aging. Epidemiologic Reviews, 22(1),
cognitive et identification de l’action: Le cas de l'escalade sport‐ 95–106. https​://doi.org/10.1093/oxfor​djour​nals.epirev.a018031
ive. European Review of Applied Psychology [Revue Europeenne de Harmon‐Jones, E., Amodio, D. M., & Harmon‐Jones, C. (2009). Action
Psychologie Appliquee], 52(2), 81–90. based model of dissonance: A review, integration, and expansion
Bandura, A. (1997). Self‐efficacy: The exercise of control. New York, NY: of conceptions of cognitive conflict. Advances in Experimental Social
Macmillan. Psychology, 41, 119–166.
Bandura, A. (2004). Health promotion by social cognitive means. Health Lewin, K. (1951). Field theory in social science: Selected theoretical papers.
Education & Behavior, 31(2), 143–164. https​://doi.org/10.1177/10901​ New York: Harper & Brothers.
98104​263660 McAuley, E., Elavsky, S., Motl, R. W., Konopack, J. F., Hu, L., & Marquez,
Becker, M. H., & Joseph, J. G. (1988). AIDS and behavior change to re‐ D. X. (2005). Physical activity, self‐efficacy, and self‐esteem:
duce risk: A review. American Journal of Public Health, 78, 394–410. Longitudinal relationships in older adults. The Journals of Gerontology.
Beckman, L. J., Harvey, S. M., & Tiersky, L. A. (2010). Attitudes about con‐ Series B, Psychological Sciences and Social Sciences, 60(5), 268–275.
doms and condom use among college students. Journal of American https​://doi.org/10.1093/geron​b/60.5.P268
College Health, 44(6), 243–250. McAuley, E., Lox, C., & Duncan, T. E. (1993). Long‐term maintenance
Bélanger‐Gravel, A., Godin, G., & Amireault, S. (2013). A meta‐ana‐ of exercise, self‐efficacy, and physiological change in older adults.
lytic review of the effect of implementation intentions on phys‐ Journal of Gerontology, 48(4), 218–224. https​://doi.org/10.1093/
ical activity. Health Psychology Review, 7(1), 23–54. https​://doi. geron​j/48.4.P218
org/10.1080/17437​199.2011.560095 Priolo, D., Pelt, A., Bauzel, R. S., Rubens, L., Voisin, D., & Fointiat, V.
Bem, D. J. (1972). Self‐perception theory. In L. Berkowitz (Ed), Advances in (2019). Three decades of research on induced hypocrisy: A Meta‐
experimental social psychology (Vol. 6, pp. 1–62). New York: Academic analysis. Personality and Social Psychology Bulletin, 1–21. https​://doi.
Press. org/10.1177/01461​67219​8 41621
Booth, F. W., Laye, M. J., & Roberts, M. D. (2011). Lifetime sedentary Robins, R. W., Hendin, H. M., & Trzesniewski, K. H. (2001). Measuring
living accelerates some aspects of secondary aging. Journal of Applied global self‐esteem: Construct validation of a single‐item measure
Physiology, 111(5), 1497–1504. https​://doi.org/10.1152/jappl​physi​ and the Rosenberg self‐esteem scale. Personality & Social Psychology
ol.00420.2011 Bulletin, 27(2), 151–161. https​://doi.org/10.1177/01461​67201​272002
Brehm, J. W. (1956). Postdecision changes in the desirability of alterna‐ Stice, E., & Shaw, H. (2004). Eating disorder prevention programs: A
tives. Journal of Abnormal and Social Psychology, 52, 384–389. https​:// meta‐analytic review. Psychological Bulletin, 130(2), 206. https​://doi.
doi.org/10.1037/h0041006 org/10.1037/0033-2909.130.2.206
Cooper, J. (1980). Reducing fears and increasing assertiveness: The role Stone, J., Aronson, E., Crain, A. L., Winslow, M. P., & Fried, C. B. (1994).
of dissonance reduction. Journal of Experimental Social Psychology, Inducing hypocrisy as a means of encouraging young adults to use
16(3), 199–213. https​://doi.org/10.1016/0022-1031(80)90064-5 condoms. Personality and Social Psychology Bulletin, 20(1), 116–128.
Cooper, J. (2007). Cognitive dissonance: 50 years of a classic theory. https​://doi.org/10.1177/01461​67294​201012
London, UK: Sage. Stone, J., Wiegand, A. W., Cooper, J., & Aronson, E. (1997). When ex‐
Cooper, J., & Feldman, L. A. (2019). Does cognitive dissonance occur in emplification fails: Hypocrisy and the motive for self‐integrity.
older age? A study of induced compliance in a healthy elderly popu‐ Journal of Personality and Social Psychology, 72(1), 54. https​://doi.
lation. Psychology and Aging, 34(5), 709–713. org/10.1037/0022-3514.72.1.54
8 | COOPER and FELDMAN

Thorp, A. A., Owen, N., Neuhaus, M., & Dunstan, D. W. (2011). Sedentary The Journals of Gerontology: Series B: Psychological and Social Sciences,
behaviors and subsequent health outcomes in adults: A system‐ 67(1), 18–26.
atic review of longitudinal studies, 1996–2011. American Journal
of Preventive Medicine, 41(2), 207–215. https​ ://doi.org/10.1016/j.
amepre.2011.05.004
How to cite this article: Cooper J, Feldman LA. Helping the
Webb, T. L., & Sheeran, P. (2006). Does changing behavioral intentions
“couch potato”: A cognitive dissonance approach to
engender behavior change? A meta‐analysis of the experimen‐
tal evidence. Psychological Bulletin, 132(2), 249–268. https​://doi. increasing exercise in the elderly. J Appl Soc Psychol.
org/10.1037/0033-2909.132.2.249 2019;00:1–8. https​://doi.org/10.1111/jasp.12639​
White, S. M., Wójcicki, T. R., & McAuley, E. (2012). Social cognitive influ‐
ences on physical activity behavior in middle‐age and older adults.