Running Head: DISBELIEF IN FREE WILL INCREASES CONFORMITY 1
Determined to Conform: Disbelief in Free Will Increases Conformity
Jessica L. Alquist, Sarah E. Ainsworth, & Roy F. Baumeister
Florida State University
1107 West Call St
Tallahassee FL 32301
Jessica Alquist, corresponding author: firstname.lastname@example.org, (570) 335-2333
Sarah Ainsworth: email@example.com, (407) 222-8805
Roy Baumeister: firstname.lastname@example.org, (850) 644-4200
Word count: 7,000
DISBELIEF IN FREE WILL INCREASES CONFORMITY 2
Does disbelief in free will reduce people’s willingness to exert the effort needed for autonomous
thought and action rather than simply conforming to group norms? Three studies tested the
hypothesis that disbelief in free would be associated with greater conformity than a belief in free
will. In Study 1 (correlational), participants who expressed a greater belief in free will reported
that they were less likely to conform in a variety of situations than participants who expressed
greater disbelief in free will. In Study 2 (experimental), participants who were induced to
disbelieve in free will conformed significantly more to the opinions of ostensible other
participants when judging paintings than participants in free will and control conditions. In Study
3 (experimental), participants who were induced to disbelieve in free will conformed
significantly more to experimenter-provided examples than participants in a meaning-threat
control condition, as well as more than those encouraged to believe in free will. These findings
suggest that belief in free will contributes to autonomous action and resisting temptations and
pressures to conform.
Keywords: Free Will, Conformity
DISBELIEF IN FREE WILL INCREASES CONFORMITY 3
Many situations offer powerful cues to guide behavior, but people can and sometimes do
resist these so as to act based on their own inner thoughts, feelings, and motives. As highly social
creatures, humans find it easy to copy the behavior of others, and (for better or worse)
conformity is a common pattern. People also respond to subtle social cues that tell them how to
act in standard ways. Yet deviance, defiance, and novelty-seeking creativity are also part of the
The questions of whether and in what sense humans have free will have been discussed
for centuries. Although philosophical views about free will have become quite complex and
subtle, laypersons tend to see free will as being able to act based on one’s own inner thoughts,
feelings, and choices, rather than being driven by external pressures (Monroe & Malle, 2010;
Stillman, Baumeister, & Mele, 2011). The present investigation tested the hypothesis that
people’s degree of belief in free will would contribute to whether they acted based on inner
thoughts or simply went along with external cues. Specifically, we predicted that conformity
would increase (while effortful, creative, original thought would decrease) as people’s belief in
free will declined.
Psychology has recently begun to investigate the effect of people’s beliefs about free will
on their behavior. It seems unlikely that psychology experiments will establish whether people
have free will, but such studies can show whether people who believe in free will behave
differently than people who do not. The value of such research does not depend on metaphysical
truths about the reality of free will. Psychology has a long history of investigating how people’s
beliefs affect their behavior, regardless of the truth of those beliefs. As examples, research on
positive illusions, just world beliefs, and self-esteem are almost entirely concerned with the
effects of a belief on behavior, independent of the objective truth of those beliefs.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 4
Previous research has shown that a belief in free will, whether pre-existing or
experimentally manipulated, can affect behavior. Vohs and Schooler (2008) found that
participants who were induced to believe that they did not have free will were more likely than
others to take advantage of an opportunity to cheat. Subsequent research has demonstrated that a
disbelief in free will is associated with more aggression, less helping, failing to change behavior
after transgressions, and relatively poor work quality, even as assessed by a supervisor
(Baumeister, Masicampo, & DeWall, 2009; Stillman & Baumeister, 2010; Stillman, et al., 2010).
The effects of free will on behavior are not due to changes in perceived accountability, feelings
of agency, Protestant work ethic, conscientiousness, or locus of control (Baumeister et al., 2009;
Stillman et al., 2010). Instead, it seems that reducing people's belief in free will makes them less
willing to exert effort as needed for volition and self-control. The apparent deficit in effortful
volition led us to think that disbelieving in free will should promote conformity and weaken
Forming one’s own opinion and asserting it require effort. Nonconformists and political
minorities often bemoan the laziness of a population that seems unwilling to expend the effort to
generate, much less express, a unique opinion. Nonconformity requires effort in a few different
ways. First, nonconformists may have to expend effort to ignore or deliberately resist the
influence of others. Research has shown that making a decision on an Asch-style task in the
presence of others’ conflicting opinions requires more effort than making a decision without
knowledge of others’ choices (Kahan, Polivy, & Herman, 2003). Second, forming and
expressing an opinion requires the person to consider multiple options and find a basis for
choosing one above the others, and this process is effortful and depleting (see Vohs et al., 2008).
Last, individuals who choose not to conform may be asked to justify their behavior (e.g.,
DISBELIEF IN FREE WILL INCREASES CONFORMITY 5
Tetlock, 1983; Tetlock, Skitka, & Boettger, 1989). Agreeing with majority consensus rarely
needs to be explained, and so deviating from the majority carries the risk of further demands for
mental effort. For all these reasons, people can save themselves considerable effort and energy
by going along with the crowd.
We proposed that a disbelief in free will would make individuals disinclined to exert the
effort necessary to form and express their own opinions. The first study tested the hypothesis that
high trait disbelief in free will would correlate with a tendency to conform to others in general.
The second study was an experimental test of the hypothesis that induced disbelief in free will
would cause an increase in conformity to ostensible judgments of peers, specifically in the
context of evaluating abstract art. The third study tested the hypothesis that inducing people to
disbelieve in free will would make them conform to salient examples (provided by the
experimenter) on a creativity task. This study also addressed an alternative explanation that any
threat to people’s beliefs, rather than just specifically messages denying free will, would
decrease their likelihood of forming their own opinions. Across these studies, we defined
conformity broadly so as to encompass both copying the behaviors of others and copying the
forms of response presented in social cues. People can resist such tendencies and think for
themselves, but it is often easier just to conform. We reasoned that disbelief in free will might
reduce the inclination to put forth the effort to think for oneself.
Participants. Thirty-nine participants (22 women; mean age = 37.8) were recruited for
participation through Amazon.com’s Mechanical Turk. Mechanical Turk is an on-line service
DISBELIEF IN FREE WILL INCREASES CONFORMITY 6
where individuals can solicit help with various tasks, including completing surveys, in exchange
for a small payment.
Conformity scale. The Conformity Scale is an 11-item self-report measure assessing
individuals’ tendencies to conform to those around them (Mehrabian & Stefl, 1995). Participants
were asked to respond to questions such as “I tend to rely on others when I have to make an
important decision quickly” and “I don’t give in to others easily” (reverse-coded) on a scale of 1
(not at all true of me) to 5 (extremely true of me).
FAD-Plus. In order to measure belief in free will, participants were given the FAD-Plus
(Paulhus & Carey, 2011). The FAD-Plus consists of 27 items designed to measures four
constructs related to free will: free will, scientific determinism, fatalistic determinism, and
unpredictability. For the free will subscale, participants were asked to respond to statements such
as “People have complete control over the decisions they make” and “People have complete free
will” on a scale of 1 (strongly disagree) to 5 (strongly agree).
Procedure. Participants completed the conformity scale, the FAD-Plus, and some
demographic questions on-line. The order of the conformity and FAD-Plus scales was
Results and Discussion
There was a significant negative correlation between belief in free will (M = 3.54, SD =
.75) and conformity (M = 2.46, SD = .56), r (37) = -.34, p = .03. Participants who expressed a
stronger belief in free will reported conforming less than participants with a weaker belief in free
DISBELIEF IN FREE WILL INCREASES CONFORMITY 7
We also checked the other subscales on the FAD-Plus. There was no significant
correlation between conformity and scientific determinism (M = 3.06, SD = .54), r(37) = .08, ns,
or fatalism (M = 2.52, SD = .71), r(37) = -.07, ns. There was a significant negative (and
unpredicted) correlation between people’s belief that the world is unpredictable (M = 3.17, SD =
.64) and their self-reported conformity, r(37) = -.37, p = .02: The more unpredictable the world
seemed, the less people reported conforming. It is possible that believing the world is
unpredictable is associated with believing in more free will, insofar as other people’s free
choices might make the world seem unpredictable. Supporting this, the unpredictability and free
will subscales of the FAD-Plus were marginally correlated, r(37) = .28, p = .08.
It may be surprising that, although free will belief predicted participants’ self-reported
conformity, belief in scientific determinism did not. Theoretically, some might assume that a
belief in free will should necessitate a lack of belief in determinism. However, previous research
has found no negative correlation between self-reported belief in free will and self-reported
belief in determinism (Paulhus & Carey, 2011). Individuals do not seem to assume that
determinism necessitates a lack of free will. Even in studies where researchers present
participants with hypothetical deterministic scenarios, where individuals’ values are determined
entirely by their genes and environment, the majority of participants still indicate that the
characters in that world have free will (Nahmias, Morris, Nadelhoffer, & Turner, 2005).
Individuals’ belief in their freedom seems to be somewhat unimpeded by their beliefs about the
strength of the causes of their behavior.
Two drawbacks to Study 1 must be noted. First, self-reports of conformity are not
entirely reliable, insofar as people may claim to conform or resist conformity due to social
desirability issues, wishful thinking, or lack of awareness (Nolan, Schultz, Cialdini, Goldstein, &
DISBELIEF IN FREE WILL INCREASES CONFORMITY 8
Griskevicius, 2008; Pronin, Berger, & Molouki, 2007). Second, the finding is correlational and
hence precludes causal inference. Study 2 sought to rectify these concerns.
Study 2 was designed to test the hypothesis that disbelief in free will would cause an
increase in conformity. Adapting procedures developed by Vohs and Schooler (2008), we sought
to create experimental groups that differed as to the degree of belief in free will. Instead of
merely passively exposing participants to statements supporting or denying free will, as in the
original procedure, we had participants read the sentences and then re-state them in their own
words. We reasoned that the active construction of sentences stating such views would constitute
a good way to prime the thoughts, as opposed to merely playing a tape of them and hoping that
participants would heed it.
Conformity was measured with an artwork evaluation task. Participants were shown a
series of abstract pictures and instructed to rate them. Supposedly as an accident of convenience,
they were able to see how previous participants had ostensibly rated the items. The ratings were
bogus and contrived and thus were not based on any actual merit of the works. Participants could
thus form and express their own opinion on each work, which requires some mental exertion —
or they could rather easily just write opinions conforming to what others had already seemingly
expressed. The prediction was therefore that participants who had been induced to disbelieve in
free will would conform to others’ opinions more than participants whose belief in free will had
been either strengthened or left untouched.
Participants. Fifty-six students participated in exchange for credit toward their
introductory psychology course. Participants were run individually and were randomly assigned
DISBELIEF IN FREE WILL INCREASES CONFORMITY 9
to one of three conditions: pro-free will sentence condition, anti-free will sentence condition, or
control sentence!condition. Two participants expressed suspicion about the conformity measure
and were excluded from the analyses, leaving the final sample size at fifty-four students (mean
age = 18.8 years).
Free will belief manipulation. When participants arrived for the experiment, they were
told they were participating in a study of written expression and art preferences. After
completing an informed consent, participants were seated in front of a computer. The study was
administered using MediaLab research software (Jarvis, 2006). Participants were shown a
sentence on the computer screen for thirty seconds and were then asked to re-write the sentence
in their own words. This was repeated ten times for ten different sentences.
The sentences for all three conditions were taken from Vohs and Schooler (2008). In the
pro-free will condition, participants were asked to rewrite sentences such as, “I have free will to
control my actions and, ultimately, to control my destiny in life,” and “There are many things
that science still cannot explain so it does not trouble me that science cannot offer an explanation
for free will.” In the anti-free will condition, participants were asked to rewrite sentences such as,
“Science has demonstrated that free will is an illusion,” and “Everything a person does is a direct
consequence of their environment and genetic makeup.” In the control condition, participants
were asked to rewrite sentences such as, “Oceans cover 71% of the earth’s surface.”
Painting ratings. After completing this manipulation, participants were told that they
would be rating a series of paintings as a measure of their art preferences. Six abstract paintings
by Paul Klee and Wassily Kandinsky were presented to participants one at a time on the
DISBELIEF IN FREE WILL INCREASES CONFORMITY 10
computer screen. Participants were asked to rate the extent to which they liked each painting on a
scale of 1 (not at all) to 9 (extremely).
Participants were given a packet that contained ratings for each painting, supposedly
completed by twenty-three other students. This measure of conformity was adapted from the
research of Arndt, Schimel, Greenberg, and Pyszczynski (2002). The ratings of the ostensible
preceding 23 students averaged around either a seven or a three for each painting. In half the
packets, paintings 1, 2, and 4 were averaged around a rating of seven and paintings 3, 5, and 6
were averaged around a rating of three. In the other packets, these were reversed, with paintings
1, 2, and 4 getting the lower ratings. The counterbalancing should rule out any effects of
coincidence between actual quality and ostensible ratings.
Each participant was told that he or she was participant number 24 and should put his or
her rating next to number 24 for each painting. The experimenter explained that the ratings were
being collected this way in order to conserve paper for environmental reasons.
We computed a value that reflected the extent to which each participant matched his or
her ratings to the ratings of the ostensible previous participants. Specifically, we assessed the
extent to which participants gave high ratings to paintings that were ostensibly rated highly by
previous participants and low ratings to paintings that were ostensibly rated poorly by previous
participants. In order to obtain a number that reflected this, we reverse-coded the ratings each
participant gave the ostensibly low-rated paintings and summed them with the ratings each
participant gave the ostensibly high-rated paintings. As predicted, there was significant variation
in conformity among the three conditions, F(2, 51) = 5.55, p < .01.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 11
The main prediction of this experiment was that participants who rewrote anti-free will
statements would conform more than participants who rewrote pro-free will statements. Planned
comparisons supported this prediction. Participants in the anti-free will condition conformed
significantly more (M = 34.58, SD = 4.27) than participants in the pro-free will condition (M =
30.41, SD = 4.10), F(2, 51) = 9.47, p < .01. Participants in the anti-free will condition also
conformed significantly more than participants in the control condition (M = 29.82, SD = 3.55),
F(2, 51) = 8.10, p < .01. There was no significant difference in conformity between the control
condition and the pro-free will condition, F(2, 51) = .18, p = .68.
Participants in the anti-free will condition conformed significantly more in their painting
ratings than participants in either the pro-free will or control condition. This indicates that
inducing people to disbelieve in free will made them more likely to conform to others’ opinions.
The lack of difference between the control and free will conditions is similar to findings in
previous research (e.g. Baumeister, et al., 2009) and is likely due to the fact that most people
already believe in free will (Nahmias, et al., 2005). This unfortunately raises the possibility that
the effects of the manipulation had less to do with free will beliefs per se and more to do with
threatening people’s already established worldviews. Study 3 was designed in part to address that
One alternative explanation for the effects of Study 2 is that the anti-free will condition
constitutes a worldview threat, and that any worldview threat would cause conformity.
Consistent with that view, prior work has shown that inducing an existential threat causes people
to conform (Renkema, Stapel, & Van Yperen, 2008). Study 3 added a meaning-threat control
DISBELIEF IN FREE WILL INCREASES CONFORMITY 12
condition to test the hypothesis that threatening participants’ belief in free will would increase
conformity more than threatening other kinds of important personal beliefs. Previous research
has shown that believing one’s life to be meaningful contributes to psychological well-being
(Low & Molzahn, 2007; Steger & Frazier, 2005; Jim & Andersen, 2007) and that it is
threatening for this belief to be questioned (Routledge et al., 2011).
Study 3 also extended our previous findings by measuring conformity to stimulus
examples rather than to other people’s behaviors. Creativity often requires thinking for oneself in
an effortful fashion, and there are degrees as to how far one can push oneself to stray from the
familiar paths. We provided participants with supposedly illustrative examples of the kinds of
product names that would be suitable. The examples had some common structural features (i.e.,
similar endings). The measure was whether participants created product names that were quite
similar to the examples we gave them or departed from them in more novel ways. We reasoned
that it was easier to generate names similar to the examples than to depart from them in a more
original, creative manner.
Our main prediction was that undermining belief in free will would increase participants’
conformity to the examples. We also predicted that this would occur specifically for the
manipulation of free will beliefs and not for a different worldview threat.
Participants. Participants were 75 undergraduate students who participated in exchange
for partial credit for their Introduction to Psychology course. Two students were excluded,
including one who correctly stated the purpose of the conformity measure and another who did
not understand the experiment instructions due to a language barrier. This left a total of 73
participants (50 women).
DISBELIEF IN FREE WILL INCREASES CONFORMITY 13
Free will belief manipulation. Groups of participants completed this experiment in a
classroom. Participants first completed the manipulation from Study 2, which was adapted for
classroom use. Participants received a packet that contained 10 sentences printed one per page
with 10 additional pages with blank lines to rewrite those sentences. Participants were instructed
to think about the meaning of each sentence for 30 seconds and then to turn the page in the
packet and rewrite each sentence for 45 seconds. The time spent writing was held constant to
make it possible for the experimenter to administer the manipulation to groups of participants. In
addition to the pro-free will, anti-free will, and control conditions, a fourth condition was
included to manipulate meaning threat. Participants assigned to the meaning threat condition
rewrote 10 sentences that were designed to portray life as meaningless. The sentences were taken
directly from an essay used by Routledge et al. (2011) to manipulate meaning threat (e.g., “After
I am dead, what mark will I have left on the world to show that I have existed?” or “Sometimes
the world seems like a colony of ants, each individual endlessly repeating his behavior until he
Conformity measure. Following the experimental manipulation, participants completed
the conformity measure (Galinsky, Magee, Gruenfeld, Whitson, & Liljenquist, 2008; Rubin,
Stoltzfus, & Wall, 1991). Participants were told to imagine that they were interviewing with a
top marketing firm and that this interview included a business aptitude test in which they would
generate novel product names. Participants received three types of product categories, including
pasta, nuclear elements, and analgesics, and were asked to generate at least one and up to three
novel product names for each category. Each product category included a list of six example
products, but critically, all six items on each list included one of two or three common endings.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 14
The pasta examples ended in either ‘na’, ‘ti’, or ‘ni’ (e.g., lasagna, spaghetti, fettuccini); the
nuclear element examples ended in ‘ium’ or ‘on’ (e.g., plutonium, radon), and the analgesic
examples ended in ‘ol’ or ‘in’ (e.g., Tylenol, aspirin). To discourage conformity, the instructions
stressed that participants need not copy aspects of the examples, but rather generate novel
product names. The total number of responses across product categories that included the
example endings served as the dependent measure of conformity.
Mood and threat measures. Additionally, participants completed the Brief Mood
Introspection Scale (BMIS; Mayer & Gaschke, 1988) to test the hypothesis that differences in
mood or arousal account for any observed effects. Participants then answered two questions to
check for the effectiveness of the experimental manipulation which included: “How much did
the sentences cast doubt on the belief that you have free will?” and “How much did the sentences
cast doubt on the belief that life is full of meaning?” (Routledge et al., 2011). Responses to these
questions were recorded on a 7-point Likert scale (1 = not at all, 7 = very much).
Conformity. We predicted that participants assigned to the anti-free will condition would
display more conformity to the product examples than participants in the pro-free will condition,
meaning threat condition, or control condition. There was a significant variation among
conditions in conformity, F(3, 69) = 2.73, p = .05 (see Figure 2). The most important and novel
finding was that participants in the anti-free will condition (M = 4.5, SD = 2.48) conformed
significantly more than participants in the meaning threat condition (M = 2.79, SD = 2.04), F(1,
69) = 5.05, p = .03. Thus, disbelief in free will is more than just meaning or worldview threat.
Replicating the results from the previous study, participants in the anti-free will condition
generated more conforming responses than participants in the pro-free will condition (M = 2.72,
DISBELIEF IN FREE WILL INCREASES CONFORMITY 15
SD = 2.52), F(1, 69) = 5.31, p = .02. Participants in the anti-free will condition also generated
more conforming examples than participants in the control condition (M = 2.61, SD = 2.2), F(1,
69) = 6.0, p = .02. The meaning threat condition did not differ from the anti-free will and control
One alternative explanation of our results could be that participants in the anti-free will
condition may have just been generating more examples in general, which would explain why
they had a greater number of conforming examples. However, there was no difference in the
number of product labels generated by participants in the various conditions, F(3, 69) = .97, p =
.41. This provides evidence that the level of effort was consistent across conditions and that any
differences in the number of conforming responses were not due to a difference in the overall
number of responses provided. Thus, the main findings indicate differences in conformity, not
differences in total output.
Mood. We also tested the hypothesis that the effect of the manipulation on conformity
would persist even when controlling for mood. There were no differences in mood valence, p =
.96, or mood arousal, p = .49, by experimental condition. Furthermore, the effect of condition on
conformity remained significant even after controlling for mood valence and arousal, F(3, 67) =
2.69, p = .05.
Threat measures. There was a significant difference among conditions on the extent to
which participants indicated that the sentences they read cast doubt on their belief in free will,
F(3, 69) = 3.51, p = .02, ƞ2 = .13. To interpret this main effect, we conducted three simple
contrasts comparing the anti-free will condition (M = 2.89, SD = 1.97) to the pro-free will (M =
2.06, SD = 1.43), meaning threat condition (M = 2.16, SD = 1.21), and control condition (M =
1.39, SD = .61). Participants in the anti-free will condition reported that the sentences cast
DISBELIEF IN FREE WILL INCREASES CONFORMITY 16
significantly more doubt on their belief in free will than participants in the control condition,
t(69) = 10.46, p < .01. Participants in the anti-free will condition reported that the sentences cast
marginally more doubt on their belief in free will than participants in the pro-free will condition
t(69) = 3.23, p = .08, and meaning threat condition t(69) = 2.83, p = .10. This suggests that the
anti-free will condition caused participants to doubt free will more than the pro-free will, control,
or meaning threat conditions.
There was a marginally significant difference among conditions in the extent to which
participants indicated that the sentences they read cast doubt on their belief that their life was
meaningful, F(3, 69) = 2.4, p = .08. To interpret this effect, we conducted three simple contrasts
comparing the meaning threat condition (M = 2.58, SD = 1.77) to the anti-free will condition (M
= 2.83, SD = 1.86), the pro-free will condition (M = 2.44, SD = 1.42), and the control condition
(M = 1.56, SD = .78). Participants in the meaning threat condition reported that the sentences
cast more doubt on their belief that life was meaningful than participants in the control condition,
t(69) = 4.17, p < .05. There were no significant differences between the meaning threat condition
and the anti-free will condition, t(69) = .26, p = .61, or the pro-free will condition, t(69) = .07, p
= .79. The meaning threat condition threatened meaning more than the control condition.
However, the pro and anti-free will conditions also seemed to have cast some doubt on
participants’ belief that their life was meaningful.
We also tested the hypothesis that there was a relationship between free will threat and
meaning threat and the amount of conformity participants demonstrated on the product label
generation task. The relationship between participants’ self-reported meaning threat and
conformity was not significant, p = .71, r = .05. There was also no relationship between
participants’ self-reported free will threat and conformity, p = .83, r = .03.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 17
Participants in the anti-free will condition conformed to the examples provided by the
experimenter significantly more than participants in the pro-free will, control, and meaning threat
conditions. This provides some evidence that the effect of manipulated free will beliefs on
conformity may be specific to the issue of free will, rather than being the effect of threatening
any important worldview. It also suggests that the conformity in these studies is not so much an
attunement to other people as a yielding to external influences and guides.
Although there was a significant difference in reported threat to meaning between
participants in the meaning threat and control conditions, there was not a significant difference in
reported meaning threat between the meaning threat, free will, and anti-free will conditions. This
could be because even bringing up the concept of free will makes people less convinced that
their life is meaningful. Recent experimental studies have confirmed that manipulating beliefs in
free will alter people’s sense of the meaningfulness of life (Crescioni et al., 2012).
Among laypersons, the concept of free will essentially involves making choices based on
inner thoughts and values, rather than being coerced by external forces (Monroe & Malle, 2010).
Indeed, if there is external pressure to act in a certain way, people associate free will with going
against those pressures (Stillman, Baumeister, & Mele, 2012). Meanwhile, pressure to conform
to norms, standards, and other people’s opinions is often felt as an external force against which
people must strive in order to behave in an autonomous fashion based on their inner thoughts and
values (e.g., Asch, 1956; Deutsch & Gerard, 1955). Hence we began this investigation with the
hypothesis that free will beliefs would alter conformity.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 18
Three studies provided evidence linking belief in free will to (non) conformity. A
questionnaire survey (Study 1) found that people who habitually have higher levels of belief in
free will reported lower conformity across a broad variety of everyday situations. Two laboratory
experiments (Studies 2 & 3) showed that manipulated decreases in free will beliefs caused
increases in conformity, broadly defined. In both studies, some participants had their belief in
free will weakened by the exercise of expressing anti-free-will assertions in their own words.
These participants were later significantly more likely than others to conform to other people’s
opinions or to examples given by the experimenter. Conformity was measured in terms of rating
abstract paintings quite similarly to ostensible but bogus ratings made by previous participants
(Study 2) and by generating possible product names that conformed structurally to patterns given
in the examples (Study 3). In both cases, all participants had been encouraged to think for
themselves and depart from norms, but people tended to conform more to those norms if their
belief in free will had been reduced by the manipulation.
One alternative interpretation might be that reducing free will beliefs with an
experimental manipulation would amount to threatening an important value and worldview
element, and that this threat was the operative part of the manipulation – and so the increase in
conformity might have nothing to do with free will per se. Several findings speak against that
interpretation. First, the survey results indicate that chronic beliefs were linked to conformity in
the same manner as experimentally manipulated ones. The trait measure of free will beliefs does
not pose a threat in the manner that one might speculate that the experimental manipulations
aimed at altering beliefs could. Second, Study 3 included a meaning threat manipulation that was
included to provide a direct test of the alternative interpretation, and it did not increase
conformity like the anti-free-will manipulation. Our manipulation check found that the meaning
DISBELIEF IN FREE WILL INCREASES CONFORMITY 19
threat manipulation did make people doubt that their lives were meaningful. But participants in
that condition conformed about as much as neutral controls and even as much as those in the pro-
free will condition. Thus, the anti-free will manipulation did not resemble the meaning threat
condition, making it unlikely that the effects of the anti-free will manipulation can be explained
as a meaning threat. (Indeed, we failed to find that meaning threat by itself had any effect on
conformity.) Third, self-reports of feeling threatened by the manipulations failed to mediate
conformity on the behavioral measure.
Many alternative interpretations for experimental findings invoke possible mood and
emotion effects, based on the widespread assumption that mood and emotion are the direct
causes of behavior (cf. Baumeister, Vohs, DeWall, & Zhang, 2007). It seemed plausible a priori
that our manipulations of belief in free will could evoke emotional reactions that might
contribute to conformity. Against that possibility, however, Study 3 contained a measure of
mood and emotional state, and it yielded no differences among conditions. Moreover, the main
findings remained significant even after controlling for mood and emotion. Hence we conclude
that mood and emotion did not drive our results.
The most plausible and parsimonious explanation is therefore that belief in free will
motivates people to put forth the greater effort required to act on their own volition and think for
themselves rather than taking the easier path of conforming to externally provided guidelines.
Low or reduced belief in free will has been linked to low levels of effort for volition and self-
control, whereas high or increased belief in free will is associated with higher levels of volitional
effort (Baumeister et al., 2009; Stillman et al., 2010). When one is given prior ratings by others
or examples that one can follow, it is relatively easy to conform one’s own thoughts to match
DISBELIEF IN FREE WILL INCREASES CONFORMITY 20
these. It takes more effort to disregard the models and come up with a fresh, novel thought.
Apparently, weakening people’s belief in free will undermines the motivation to do that.
Our results (and in Study 3 our measures) resemble those of Galinsky et al. (2008), who
showed that having or thinking of having power makes people more willing to act on their
internal states and less susceptible to external, situational pressures. We found much the same
with belief in free will. Both power and free will beliefs invoke agency, and so although at first
blush a philosophical opinion about human nature and a type of interpersonal relationship would
seem to be quite different, they may in fact bring about similar effects. Further work may explore
additional parallels between free will beliefs and social power. For present purposes, the parallel
lends further support to the notion with which we began this manuscript, namely that sometimes
people go along with the group and succumb to external pressures, whereas other times they
resist and act on their own — and belief in free will (like power) maybe an important factor in
guiding people to take the latter rather than the former path.
Several limitations must be noted. Our theory and methods were based on the idea that
conformity is easier than nonconformity. Hence we are reluctant to generalize to cases in which
conforming requires more effort than deviating. (In fact, we conducted one further study in
which conformity required more effort than nonconforming, and it failed to replicate the pattern
of increased conformity among participants whose belief in free will had been reduced.) Also, as
with many laboratory procedures, the actual behaviors measured had relatively low personal
significance to participants, and it is quite plausible that conformity would be less influential in
cases involving strong personal opinions and values. In other words, disbelief in free will might
encourage participants to go along with other people’s ratings of unfamiliar, abstract paintings,
DISBELIEF IN FREE WILL INCREASES CONFORMITY 21
but that does not mean they would as easily betray loved ones, renounce their religious faith, or
abandon strongly held political views. In other words, we assume there are boundary conditions.
Furthermore, our findings are irrelevant to the ongoing debates about whether and in
what sense people actually have free will. The present research shows that personal beliefs about
free will have behavioral consequences, and that evidence is entirely independent of the validity
or accuracy of those beliefs.
Last, we do not view our results as depicting conformity (and by extension disbelief in
free will) as inherently bad. Although conforming has a negative reputation, it is often a useful
and advantageous pattern. As cultural animals, humans get much information from others and
can save themselves considerable exertion and energy by conforming. More often than not, the
conclusions of the group are likely to be reasonably correct and beneficial. When Deutsch and
Gerard (1955) introduced the concept of informational social influence, they noted that getting
evidence about reality from other people was a standard and usually helpful practice.
Conforming can be harmful, but it is often beneficial and prosocial.
In the present experimental situation, participants were encouraged to think for
themselves, and so conforming was indeed making the lazy and in some ways less desirable
response. Still, the finding presumably reflects the more general pattern that going along with the
group is appealing as an energy-conserving strategy.
In our results, the manipulation that bolstered belief in free will failed to reduce
conformity from the level of the neutral control group. This pattern, in which the neutral control
furnishes the same results as the pro-free will condition has been observed in many other studies
(e.g., Vohs & Schooler, 2008; Baumeister et al., 2009), and it presumably indicates that most
people already believe in free will to some extent. Making people aver belief in free will thus
DISBELIEF IN FREE WILL INCREASES CONFORMITY 22
changes relatively little, whereas having them deny free will does constitute a departure from the
Hence we conclude that belief in free will encourages people to think for themselves and
depart from social norms and pressures, as they may often be inclined to do anyway. In contrast,
disbelief in free will reduces the motivation to exert mental effort, especially when the situation
presents an easy path of going along with what others have said or done. The cultural importance
of belief in free will may thus reside partly in how it pushes people to think independently and
thereby perhaps make unique contributions to the collective enterprise.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 23
We thank Shannon Harris, Lauren Genduso, and Ali McCully for their help with data collection.
We also thank the Templeton Foundation for grant support for this work.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 24
Arndt, J., Schimel, J., Greenberg, J., & Pyszczynski, T. (2002). The intrinsic self and
defensiveness: Evidence that activating the intrinsic self reduces self-handicapping and
conformity. Personality and Social Psychology Bulletin, 28, 671-683.
Asch, S. E. (1956). Studies of independence and conformity: A minority of one against a
unanimous majority. Psychological Monographics: General and Applied, 70, 1-70.
Baumeister, R. F., Masicampo, E. J., & DeWall, C. N. (2009). Prosocial benefits of feeling free:
Disbelief in free will increases aggression and reduces helpfulness. Personality and
Social Psychology Bulletin, 35, 260-268.
Baumeister, R. F., Vohs, K. D., DeWall, C. N., Zhang, L. (2007). How emotion shapes behavior:
Feedback, anticipation, and reflection, rather than direct causation. Personality and
Social Psychology Review, 11, 167-203.
Crescioni, A.W., Baumeister, R.F., Ainsworth, A.E., Ent, M. R., & Lambert. N.M. (2012).
Subjective correlates and consequences of belief in free will. Manuscript submitted for
publication, Florida State University, Tallahassee, FL.
Deutch, M. & Gerard, H. B. (1955) A study of normative and informative social influence upon
judgment. Journal of abnormal and social psychology, 51, 629-636.
Galinsky, A. D., Magee, J. C., Gruenfeld, D. H., Whitson, J. A., & Liljenquist, K. A. (2008).
Power reduces the press of the situation: Implications for creativity, conformity, and
dissonance. Journal of Personality and Social Psychology, 95, 1450-1466.
Jarvis, B. G. (2006). MediaLab (Version 2006) [Computer Software]. New York, NY:
DISBELIEF IN FREE WILL INCREASES CONFORMITY 25
Jim, H. S. & Andersen, B. L. (2007). Meaning in life mediates the relationship between social
and physical functioning and distress in cancer survivors. British Journal of Health
Psychology, 12, 363-381.
Kahan, D., Polivy, J., & Herman, C. P. (2002). Conformity and dietary disinhibition: A test of
the ego-strength model of self-regulation. The International Journal of Eating Disorders,
Low, G. & Molzahn, A. E. (2007). Predictors of quality of life in old age: A cross-validation
study. Research in Nursing and Health, 30, 141-150.
Mayer, J. D. & Gaschke, Y. N. (1988). The experience and meta-experience of mood. Journal of
Personality and Social Psychology, 55, 102-111.
Mehrabian, A. & Stefl, C. A. (1995). Basic temperament, components, of loneliness, shyness,
and conformity. Social Behavior and Personality, 23, 253-264.
Monroe, A. E. & Malle, B. F. (2010). From uncaused will to conscious choice: The need to
study, not speculate about people’s folk concept of free will. Review of Philosophy and
Psychology, 1, 211-224.
Nahmias, E., Morris, S., Nadelhoffer, T., & Turner, J. (2005). Surveying freedom: Folk
intuitions about free will and moral responsibility. Philosophical Psychology, 18, 561-
Nolan, J. M., Schultz, W., Cialdini, R. B., Golstein, N. J., & Griskevicius, V. (2008). Normative
social influence is underdetected. Personality and Social Psychology Bulletin, 34, 913-
Paulhus, D. L. & Carey, J. M. (2011). The FAD-Plus: Measuring lay beliefs regarding free will
and related constructs. Journal of Personality Assessment, 93, 96-104.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 26
Pronin, E., Berger, J., & Molouki, S. (2007). Alone in a crowd of sheep: Asymmetric perceptions
of conformity and their roots in an introspection illusion. Journal of Personality and
Social Psychology, 92, 585-595.
Renkema, L. J., Stapel, D. A., & Van Yperen, N. W. (2008). Go with the flow: Conforming to
others in the face of existential threat. European Journal of Social Psychology, 38, 747-
Routledge, C., Arndt, J., Wildschut, T., Sedikides, C., Hart, C. M., Juhl, J., Vingerhoets, A. J. J.
M., Schlotz, W. (2011). The past makes the present meaningful: Nostalgia as an
existential resource. Journal of Personality and Social Psychology, 101, 638-652.
Rubin, D. C., Stolzfus, E. R., & Wall, K. L. (1991). The abstraction of form in semantic
categories. Memory and Cognition, 19, 1-7.
Steger, M. F. & Frazier, P. (2005). Meaning in life: One link in the chain from religiousness to
well-being. Journal of Counseling Psychology, 52, 574-582.
Stillman, T. F. & Baumeister, R. F. (2010). Guilty, free, and wise: Determinism and psychopathy
diminish learning from negative emotions. Journal of Experimental Social Psychology,
Stillman, T. F., Baumeister, R. F., & Mele, A. (2011). Free will in everyday life:
Autobiographical accounts of free and unfree actions. Philosophical Psychology, 24, 381-
Stillman, T. F., Baumeister, R. F., Vohs, K. D., Lambert, N. M., Fincham, F. D., & Brewer, L. E.
(2010). Personal philosophy and personnel achievement: Belief in free will predicts better
job performance. Social Psychological and Personality Science, 1, 43-50.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 27
Tetlock, P. E. (1983). Accountability and complexity of thought. Journal of Personality and
Social Psychology, 45, 74-83.
Tetlock, P. E., Skitka, L., & Boettger, R. (1989). Social and cognitive strategies for coping with
accountability: Conformity, complexity, and bolstering. Journal of Personality and
Social Psychology, 57, 632-640.
Vohs, K. D., Baumeister, R. F., Schmeichel, B. J., Twenge, J. M., Nelson, N. M., & Tice, D. M.
(2008). Making choices impairs subsequent self-control: A limited-resource account of
decision making, self-regulation, and active initiative. Journal of Personality and Social
Psychology, 94, 883-898.
Vohs, K. D. & Schooler, J. W. (2008). The value of believing in free will: Encouraging a belief
in determinism increases cheating. Psychological Science, 19, 49-54.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 28
Figure 1. Mean difference in conformity by condition; Study 2.
DISBELIEF IN FREE WILL INCREASES CONFORMITY 29
Figure 2. Mean difference in conformity by condition; Study 3.