Content uploaded by Matthew Baldwin
Author content
All content in this area was uploaded by Matthew Baldwin on Feb 10, 2018
Content may be subject to copyright.
Journal of Personality and Social
Psychology
Past-Focused Temporal Communication Overcomes
Conservatives’ Resistance to Liberal Political Ideas
Joris Lammers and Matt Baldwin
Online First Publication, January 11, 2018. http://dx.doi.org/10.1037/pspi0000121
CITATION
Lammers, J., & Baldwin, M. (2018, January 11). Past-Focused Temporal Communication Overcomes
Conservatives ’ Resistance to Liberal Political Ideas. Journal of Personality and Social Psychology.
Advance online publication. http://dx.doi.org/10.1037/pspi0000121
Past-Focused Temporal Communication Overcomes Conservatives’
Resistance to Liberal Political Ideas
Joris Lammers and Matt Baldwin
University of Cologne
Nine studies and a meta-analysis test the role of past-focused temporal communication in reducing
conservatives’ disagreement with liberal political ideas. We propose that conservatives are more prone
to warm, affectionate, and nostalgic feelings for past society. Therefore, they are more likely to support
political ideas—including those expressing liberal values—that can be linked to a desirable past state
(past focus), rather than a desirable future state (future focus) of society. Study 1 supports our prediction
that political conservatives are more nostalgic for the past than liberals. Building on this association, we
demonstrate that communicating liberal ideas with a past focus increases conservatives’ support for
leniency in criminal justice (Studies 2a and 2b), gun control (Study 3), immigration (Study 4), social
diversity (Study 5), and social justice (Study 6). Communicating messages with a past focus reduced
political disagreement (compared with a future focus) between liberals and conservatives by between 30
and 100% across studies. Studies 5 and 6 identify the mediating role of state and trait nostalgia,
respectively. Study 7 shows that the temporal communication effect only occurs under peripheral (and
not central) information processing. Study 8 shows that the effect is asymmetric; a future focus did not
increase liberals’ support for conservative ideas. A mixed-effects meta-analysis across all studies
confirms that appealing to conservatives’ nostalgia with a past-focused temporal focus increases support
for liberal political messages (Study 9). A large portion of the political disagreement between conser-
vatives and liberals appears to be disagreement over style, and not content of political issues.
Keywords: conservatism, nostalgia, social influence, ideology, communication
Supplemental materials: http://dx.doi.org/10.1037/pspi0000121.supp
The 2016 United States presidential election has shown the
political power of nostalgia. Promising to bring back jobs, to
rebuild former glory, and make America great again, Donald
Trump has won, apparently by appealing to a widely felt desire
among those on the Right of the political spectrum to restore a lost
era of American greatness. In the current manuscript, we aim to
demonstrate that this nostalgic focus on the past expressed by
those on the political Right is not intrinsically linked to a specific
set of political beliefs or messages. Instead, we view nostalgia as
a general psychological force that can be flexibly harnessed to
increase support for all political ideas—including those considered
to be liberal. Specifically, we propose that political conservatives
experience stronger feelings of nostalgia and therefore favor po-
litical ideas that appeal to this nostalgia by pointing back in
time— even if those ideas are traditionally considered to be polit-
ically liberal.
Nostalgia
Nostalgia is a predominately positive emotion that is associated
with recalling memories of important or momentous events, usu-
ally experienced with close others (Sedikides, Wildschut, Arndt, &
Routledge, 2008; Wildschut, Sedikides, Arndt, & Routledge,
2006). Nostalgia can offer a window to the “true” self (Baldwin,
Biernat, & Landau, 2015) and allow people to see continuity
between their current and past selves. By connecting people with
their authentic past selves, nostalgia can offer people a sense of
meaning, social connectedness, and self-worth that can be re-
cruited in times of experienced threat to the self (Routledge et al.,
2011; Vess, Arndt, Routledge, Sedikides, & Wildschut, 2012;
Zhou, Sedikides, Wildschut, & Gao, 2008).
Nostalgia is not only a personal emotion; people can also be
nostalgic for events in their group’s past. Collective nostalgia is an
emotion focusing on events directly experienced with ingroup mem-
bers (Wildschut, Bruder, Robertson, van Tilburg, & Sedikides, 2014),
or those remembered through collective history (Havlena & Holak,
1996). Whereas personal nostalgia is focused on “the way I was,”
collective nostalgia is focused on “the way we were” (Baldwin et al.,
2015; Wildschut et al., 2014), or even more abstractly on “the way it
was” (Baker & Kennedy, 1994; Havlena & Holak, 1996; Marchegiani
& Phau, 2010). Qualitative research suggests that collective nostal-
gia— even for very distant times—is common. For example, Wilson
(2005) found that several among a sample of college students ex-
pressed a preference for living in past generations, such as the 1940s,
if given the choice of any time in history. Experimental evidence
Joris Lammers and Matt Baldwin, Social Cognition Center Cologne,
Department of Psychology, University of Cologne.
Materials are included in the manuscript Appendix. Raw data are avail-
able at https://osf.io/ptcdg/. Please contact Joris Lammers for details if
needed.
Correspondence concerning this article should be addressed to Joris Lam-
mers, Social Cognition Center Cologne, Department of Psychology, Univer-
sity of Cologne, Richard-Strauss-Str. 2, 50931, Köln, Germany. E-mail:
joris.lammers@uni-koeln.de
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Journal of Personality and Social Psychology © 2018 American Psychological Association
2018, Vol. 1, No. 2, 000 0022-3514/18/$12.00 http://dx.doi.org/10.1037/pspi0000121
1
suggests that collective nostalgia can be a positive resource for the
group by strengthening positive ingroup evaluations and motivating
actions that serve to protect other ingroup members (Wildschut et al.,
2014).
Although constrained by the past, feelings of collective nostal-
gia also provide some room for abstraction because they some-
times focus on indirectly experienced history. This means that
collective nostalgia is based in part on actual past events and in
part based on fantasy. Some people in former East Germany, for
example, are said to experience Ostalgia—a nostalgic longing for
the simpler times before the end of communism that ignores some
of the less redeeming— but historically accurate—aspects of life
behind the Iron Curtain (Bach, 2002). People waxing nostalgic
about their groups past might long for the simplicity, innocence,
liveliness, and morality of those times (Wilson, 2005). In doing so,
collective nostalgia can recreate the past as a golden age by giving
society a “redeemingly benign aura” (Davis, 1979, p. 14).
Nostalgia and Conservatism
In the current article, we aim to show that collective nostalgia
for past society is particularly strong among political conserva-
tives, which has consequences for conservatives’ social and polit-
ical attitudes. This general hypothesis is based on three observa-
tions: that conservative political ideology is associated with a
stronger nostalgic focus on the past, that conservative politicians
tend to wax nostalgic for past society, and that conservative voters
are strongly influenced by nostalgic messages. We elaborate on
these observations in the following sections.
Conservative Political Ideology
In contrast to many other political ideologies, which emerged
from a desire to change society toward a particular ideal or future
state, conservatism is a political ideology that is the product of a
reaction against such attempts at change. Central to conservatism
is a positive evaluation of the past—a valuing of the past status
quo—and a corresponding desire to keep things how they were.
This core element of conservatism can be traced to Edmund Burke,
widely considered the founder of political conservatism, and his
critique of the French Revolution (Burke, 1790/1993). The central
argument in Burke’s treatise drew on an organic view of society,
which holds that all aspects of society serve a function— even if
that function is hidden or difficult to discern. Even though he could
sympathize with some need to change France’s absolute monar-
chy, Burke argued that it was unwise to make radical changes.
Instead, he favored making smaller, piecemeal changes that re-
spect long-standing traditions (Ball & Bellamy, 2003; McClelland,
2005).
Burke’s respect for the past has remained one of the most
defining elements of conservative political philosophy (Hunting-
ton, 1957; Kirk, 1953; Muller, 1997). Conservative philosophers
believe that the past has intrinsic value and should be respected
(Scruton, 1980). In fact, some have even characterized conserva-
tive political philosophy as the “politics of nostalgia” (Schlesinger,
1955). Conservative philosophers criticize liberals for their ideal-
istic attempts at redesigning society from the ground up, because
in their mind this underestimates the complexity of society as an
organism that has developed over time (Hayek, 1944). Rather than
tearing down long established traditions to make way for new
ones, conservatives believe it to be more prudent to respect the past
and only make small changes when they are absolutely necessary
(Ball & Dagger, 1999). Nostalgia is woven into the fabric of
conservative political ideology.
Conservative Politicians
A similar nostalgic focus on the past can be found among
conservative politicians. Drawing on conservative political philos-
ophy, conservative politicians tend to favor returning to an ideal-
ized and cherished past (Von Beyme, 1985). In fact, conservative
politicians across the globe have defended different, and some-
times opposite, political programs, depending on differences in
their own nation’s past. For example, early French conservatives
favored returning to an absolute monarchy (Price, 2005), but
English conservatives at the same time defended parliamentary
monarchy (Eccleshall, 1990), while American conservatives si-
multaneously defended republicanism (Schneider, 2009). In other
words, the most striking similarity between conservative politi-
cians across the globe is their focus on maintaining the past—
irrespective of what that past was like.
This nostalgic focus on the past can also be witnessed in
conservative politicians’ political messages. For example, a con-
tent analysis of 145 State of the Union addresses showed that
Republican presidents referenced the past to a greater extent than
the future, while the opposite was the case for Democratic presi-
dents (Robinson, Cassidy, Boyd, & Fetterman, 2015). Whereas
Democrats predominantly use slogans that point to the future (e.g.,
Obama’s 2012 slogan Forward), Republicans often use slogans
that nostalgically point back to the past, such as Reagan’s 1980
slogan Let’s Make America Great Again (or Trump’s plagiarism of
it in 2016). Perhaps the most dramatic example of this difference
in temporal focus is in Bob Dole’s and Bill Clinton’s acceptance
speeches for nomination as the 1997 presidential candidates. Al-
though Clinton talked about his intentions to build a bridge to the
future—a typically liberal message—Bob Dole expressed his con-
servative desire to build a bridge to past times of tranquility
(Benoit, 2001). Although both focused on change, and used similar
“bridge” metaphors to communicate their plans, one bridge spans
backward and the other forward.
Conservative Voters
Third and finally, this same nostalgic focus on the past can be
observed in conservative voters. In their seminal overview of
political ideology as motivated cognition, Jost and colleagues
proposed that a preference to maintain the past constitutes one of
two core beliefs or tendencies guiding conservative thought (Jost,
Glaser, Kruglanski, & Sulloway, 2003). Conservatives place a
greater emphasis on order and tradition. They have an intuitive
preference for political ideas that contribute to maintaining society
how it was and has been (Graham, Haidt, & Nosek, 2009; Hirsh,
Walberg, & Peterson, 2013; Jost, Federico, & Napier, 2009).
Conservatives’ nostalgic focus on the past is also manifested in
what they read; a content analysis of news websites uncovered that
conservative websites (e.g., Drudge Report) use more past-focused
than future-focused verbs, while liberal sites (e.g., the Huffington
Post) do the opposite (Robinson et al., 2015). In fact, the link
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
2LAMMERS AND BALDWIN
between conservatism and the past is so strong that merely think-
ing about one’s past and how past choices have produced one’s
present position increases participants’ support for a conservative
view on society (Lammers & Proulx, 2013).
This nostalgic focus on the past among political conservatives
also extends to other, nonpolitical aspects of life. For example,
conservatives experience stronger intuitive resistance to new ideas
that break with tradition, such as in modern art or avant-garde
music, compared with liberals (Glasgow & Cartier, 1985; Wilson
& Patterson, 1968). Conservatives also more strongly prefer to
stick with conventional traditions, such as in religion or in family
relations (Wilson, 1973). Conservatives resist innovation and the
introduction of new technologies at work more strongly than do
liberals (Fay & Frese, 2000). Even in basic cognitive functioning,
conservatives tend to show more rigidity (Rock & Janoff-Bulman,
2010; but see Lammers, Koch, Conway, & Brandt, 2016). The
psychology of conservative individuals tends to favor the known,
established, and past (Chirumbolo, 2002; Jost et al., 2007; Kem-
melmeier, 1997; Webster & Kruglanski, 1994).
Temporal Focus
Considering these observations together, we propose that con-
servatives’ nostalgic focus on the past motivates, maintains, and
modulates their evaluation of political ideas. Specifically, we
propose that conservatives, compared with liberals, tend to process
political information and evaluate political ideas based on their
nostalgic preference for the past. This can explain why conserva-
tives intuitively dislike many liberal political ideas: In describing
and defending their political ideas, liberals typically ignore the past
and point to how their plans affect the future (Robinson et al.,
2015). Irrespective of whether it is Barack Obama calling in his
2008 DNC acceptance speech to “once more march into the
future” or Hillary Clinton painting a picture of “the future we’re
working toward” in her 2016 DNC acceptance speech, liberals
tend to focus on describing how their plans will change and
improve the future, without making connections to the past or
tradition. Crucially, however, we aim to show in the current article
that changing the temporal focus of liberal ideas by using past-
focused communication can reduce conservatives’ opposition to
such ideas, even if the content is kept the same.
This hypothesis is in line with research on framing: it is well
demonstrated that communication is more persuasive if its style
matches with the audience’s underlying motivations (Maio &
Haddock, 2015; Maio & Olson, 1995; Schwarz & Clore, 1983,
2007; Watt, Maio, Haddock, & Johnson, 2009). For example, if a
pro-environmental political message is communicated using a moral
framework that connects to conservatives’ preference for a more
binding morality, then conservatives support the pro-environmental
policy that they normally oppose (Wolsko, Ariceaga, & Seiden,
2016). We argue that changing the temporal focus of communication
operates similarly: a political message that uses a past-focused tem-
poral frame connects better with conservatives’ nostalgic focus on the
past and is more easily accepted by them.
1
There is some existing evidence that supports this notion. For
instance, research has shown that although conservatives often
oppose social change, they support such change if such calls for
change appear to be embedded in or arising from the existing
social system (Feygina, Jost, & Goldsmith, 2010). Specifically,
although conservatives may be opposed to pro-environmental cen-
tral government policies, because they break with the status-quo of
free-market capitalism, they may be less opposed to such policies
if they are presented as preserving the American Way of Life
(ibid.) or as being consistent with free-market principles (Camp-
bell & Kay, 2014).
Even more direct support for this idea comes from recent work
showing that past-focused temporal communication can increase
support among conservatives for pro-environment policies aimed
at reducing climate change. Calls for action to reduce global
warming often point to future catastrophic effects of global warm-
ing and easily entice rejection among conservatives. However, a
past-focused temporal focus, for example, by invoking images of
environmental change from a previous (and arguably more pris-
tine) version of the environment, strongly increases conservatives’
support for climate change initiatives (Baldwin & Lammers,
2016).
In the current article, we aim to demonstrate that temporal focus
effects are not limited to this single political issue, but instead can
be applied to a wide range of issues. That is, we aim to show that
a past-focused temporal focus reduces conservatives’ opposition to
many liberal ideas, even while keeping the programmatic content
of these ideas constant. We do so across studies in which we
present participants (both liberals and conservatives) with a variety
of liberal messages, such as calls for greater leniency in criminal
justice or for more social diversity, that are communicated to either
focus on the future or the past. We expect conservatives to reject
these liberal messages (compared with liberals), but to a lesser
degree if the messages make use of a past-focused temporal focus.
Furthermore, across these studies we aim to uncover the process
that drives this effect, by testing a variety of mediators and
moderators, which we describe next.
Mediation by Nostalgia
First, we focus on testing the hypothesized mediating role of
feelings of nostalgia—the sentimental longing or affection for the
past (Baldwin et al., 2015; Wildschut et al., 2006). Although we
are unaware of any empirical research testing this association, our
prediction follows from sound theory. If indeed conservatives are
more inclined to experience nostalgia than liberals, as we expect,
then their stronger nostalgia should explain the effects of past-
focused temporal focus on greater agreement for political mes-
sages. In testing this relation, we follow the common distinction
between nostalgia as a state and nostalgia as a trait (Hamaker,
Nesselroade, & Molenaar, 2007; Mischel, 1968).
Specifically, in Study 5 we present participants with a message
about the value of social diversity that is based on an iconic
Superman comic from the late 1940s. We predict that conserva-
tives are likely to experience a particularly strong (situational)
feeling of nostalgia when they are exposed to a vintage comic, but
not when exposed to a modern comic that expresses the same
1
We use the word communication and focus, instead of the word
framing. Although the word framing is also often used more loosely,
classic framing effects in psychology change only the context of the
stimulus, but not the stimulus itself. It can be argued that our manipulation
also changes the stimulus. We return to this issue in the General Discus-
sion.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
3
POLITICAL PREFERENCE AND TEMPORAL FOCUS
social diversity message. We expect that conservatives in partic-
ular will react positively to the vintage comic expressing diversity,
because they will experience strong (state) feelings of nostalgia
when exposed to that comic.
In Study 6 we instead focus on trait-differences in nostalgia in
the context of a study on social fairness. We test whether conser-
vatives’ chronic nostalgia for the past (independent of condition)
leads to more positive evaluations of a past-focused message about
social fairness, compared with a future-focused message.
Moderation by Central Versus Peripheral
Processing Style
Additionally, we aim to show that the hypothesized effect of
temporal focus only occurs under peripheral processing. This
prediction follows from now-classic research showing that the
effects of more peripheral aspects of information (such as its
temporal focus) emerge especially when participants engage in
superficial processing of information based on affective cues (such
as nostalgia), but not when they engage in central processing based
on careful consideration of the merits of the arguments. When
participants instead engage in central processing, they should
judge a message by its content, analyzing information more deeply
and distrusting any feelings arising from the way a message is
communicated (Chaiken, 1980; Petty & Cacioppo, 1986; Thomp-
son, Roman, Moskowitz, Chaiken, & Bargh, 1994).
An additional reason why peripheral processing may be partic-
ularly important in driving conservatives’ preference for past-
focused messages is that conservatives are thought to differ from
liberals in having (by default) a more peripheral, superficial, and
schematic style of processing political information than liberals
(Jost et al., 2003; Jost & Krochik, 2014; Kemmelmeier, 1997;
Sargent, 2004; Webster & Kruglanski, 1994). Conservatives’ hy-
pothesized tendency to be influenced by the temporal focus of a
message may be reduced by taking them out of their default
peripheral processing mode and instead bringing them in a more
central processing mode, which should be less sensitive to tempo-
ral focus effects.
Consequently, we test in Study 7 whether prompting partici-
pants (conservatives and liberals) to process information more
deeply overcomes these effects and leads participants to evaluate a
message more centrally—in terms of its content and according to
ideological lines. As a result, we predict that we only find strong
effects of the temporal focus manipulation when participants pro-
cess information peripherally, while deep information processing
leads partisans to support or reject a message based on their
dominant ideological differences.
Symmetry Versus Asymmetry Across the
Political Spectrum
Finally, we aim to test whether future-focused temporal manipula-
tion can also reduce liberals’ opposition to conservative messages.
Our theorizing so far has focused on conservatives’ nostalgic focus on
the past, and paid less attention to liberals’ potential focus on the
future. We expect, however, that a future-focused manipulation does
not affect liberals in the same way as a past-focused manipulation
influences conservatives. The reason for this is that we presuppose an
ideological asymmetry in the evaluation of the past and future. Spe-
cifically, although conservatives intuitively value the past and are
intrinsically motivated to defend the status quo, liberals do not intu-
itively value the future and do not value change for the sake of
change. Rather, their interest lies in making material improvements
for the position of disadvantaged groups (Jost, 2006; Jost, Nosek, &
Gosling, 2008). In other words, liberalism does not reflect a desire to
change for the sake of change in the same way that conservatism arose
out of a desire to maintain the past for the sake of the past (Ball &
Bellamy, 2003; Ball & Dagger, 1999; McClelland, 2005). Therefore,
we expect that appealing to the future does not increase liberals’
support for a policy in the same way as appealing to the past does for
conservatives. We test this prediction in Study 8.
Summary and Overview of Studies
We present nine studies to test our predictions. Study 1 tests the
correlation between political conservatism and dispositional nostalgia.
Next, four studies test whether changing the temporal focus of liberal
political messages from future-focused to past-focused increases sup-
port among conservatives. We do so first by testing a classical aspect
of liberal political ideology: a call for greater leniency in criminal
justice (Studies 2a and b). We expect that conservatives oppose
leniency, but not (or less) if the call for greater leniency uses a
temporal focus on the past. To be able to generalize our findings
across countries, we test this prediction in both the United States
(Study 2a) and the United Kingdom (Study 2b). Next, Studies 3 and
4 focus on arguably more contentious political topics in two different
countries. Specifically, Study 3 tests whether past-focused communi-
cation increases American conservatives’ support for restricting the
Second Amendment and gun ownership rights. Study 4 tests whether
past-focused communication increases German conservatives’ sup-
port for immigration. The remaining studies test the mediating and
moderating processes discussed previously: Studies 5 and 6 test the
mediating effect of state and trait nostalgia; Study 7 aims to demon-
strate the moderating role of information processing; and Study 8
focuses on the hypothesized asymmetry between liberals and conser-
vatives and tests whether inducing a future focus also increases
liberals’ support for conservative ideas. A final study aims to quantify
the overall effect of temporal focus manipulations on conservatives’
attitudes toward liberal political messages (Study 9).
Methodological Notes
Across these studies, we report how we determined our sample
size, all data exclusions (if any), all manipulations, and all measures
in the study (Simmons, Nelson, & Simonsohn, 2012). In our Study 9
meta-analysis, we include all effect sizes from the nine studies re-
ported here, as well as those from four studies that are reported in a
supplementary file (that together constitute all studies conducted in
this research project). Doing so helps to estimate the observed overall
power to detect the predicted interactions of interest and estimate the
overall size of the effect, even with so-called “file-drawer” studies
included. Raw data and analyses are available at osf.io.
2
2
https://osf.io/ptcdg/.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
4LAMMERS AND BALDWIN
Study 1
As a first test of our predictions, we examined whether conser-
vatives express higher dispositional nostalgia than liberals, using
an established, nonpolitical measure of nostalgia.
Method
Participants and design. There were 263 American Amazon
Mechanical Turk users (46.4% women, 53.6% men, mean age 36.0
years, 87.8% White, 6.1% Black, 3.4% Asian, 2.7% other/mixed)
who participated for $0.25. Of all participants, 57.8% were self-
identified liberals (combined ideology score lower than the mid-
point; see below), 14.1% were centrist (at the midpoint), and
28.1% were conservative (above midpoint). Sample-size was set a
priori to N⫽262, which would provide sufficient power (1 – ⫽
.95) to observe a small to medium correlation, ⫽0.20. All
participants completed the same items.
Measures. Participants first indicated their political ideology
on two items, between Strongly favor Democrats (1) and Strongly
favor Republicans (7), and between Very Liberal (1) and Very
Conservative (7). The two items correlated strongly, r⫽.82, p⬍
.0001, and were combined in one measure of participant ideology.
Next, participants completed Holbrook’s (1993) eight-item
measure of nostalgia. This scale consists of items that focus on
consumer-related expressions of nostalgia, such as “Products are
getting shoddier and shoddier,” “They don’t make ‘em like they
used to,” and “Modern business constantly builds a better tomor-
row” (reversed). More important, none of the items are explicitly
political, meaning that they do not match onto a liberal or conser-
vative ideological tendency. Instead, the measures assess a general
preference for elements of the past. After recoding, the eight items
showed acceptable reliability (␣⫽.74), and were combined into
one nostalgia index.
Results
Political ideology predicted nostalgia, meaning that the more
conservative participants were, the more nostalgic they were, b⫽
.10, SE ⫽.03, t(261) ⫽3.12, p⫽.002, R
2
⫽.04, 95% confidence
interval (CI
b
) [.03; .17]. For exploratory reasons, we also checked
the effects of age and gender. Older participants were more nos-
talgic than younger ones, r⫽.12, p⫽.05, 95% CI
r
[.01; .23], and
women (M⫽4.12, SD ⫽0.87) were more nostalgic than men
(M⫽3.72, SD ⫽0.87), t(261) ⫽3.73, p⬍.001, d⫽0.46.
Controlling for these effects of age, b
age
⫽.07, SE ⫽.05, p⫽
.226, and gender, b
gender
⫽.44, SE ⫽.11, p⬍.0001, we found
that the effect of political ideology became even more reliable, b⫽
.12, SE ⫽.03, R
2
⫽.11, t(259) ⫽3.62, p⬍.001, 95% CI
b
[.05;
.18].
Studies 2a and 2b: Criminal Justice and the Police
Study 1 showed that conservatives tend to show a stronger
nostalgic preference for aspects of the past. Departing from that
finding, Study 2 and subsequent studies aim to demonstrate that
communicating a liberal political message in terms of the past (vs.
the future) increases support for that message among conserva-
tives. We start by testing this prediction with a message about
criminal justice and police treatment. Progressives tend to support
more permissive police and criminal justice systems. They are less
likely to believe in the value of capital punishment, habitual
offender laws, and deterrence, and instead focus more on rehabil-
itation compared with conservatives (Beckett & Sasson, 2003).
Furthermore, they are more likely to believe that the police should
treat offenders leniently (Flanagan & Longmire, 1996). In Study
2a, we presented conservatives and liberals with the same message
that called for greater leniency in dealing with criminals, but
manipulated the temporal focus of that text. We expected that
conservatives would be less likely to endorse the proleniency
appeal than liberals, but that their disagreement would be reduced
if the appeal drew a comparison with the past, rather than with the
future (e.g., “I would like to go back to how things were then,
when police were more lenient and not so strict toward offend-
ers”). In Study 2b we sought to replicate the findings of Study 2a
in a sample from the United Kingdom, to generalize our findings
beyond the political situation in the United States. Given the
similarity in methodology, we discuss both studies together.
Method
Participants and design Study 2a. There were 200 American
Amazon Mechanical Turk users (43.5% women, 56.5% men, mean
age 33.8 years, 81.5% White, 5.0% Black, 9.0% Asian, 4.5%
other/mixed) who participated for $0.25. Following the same clas-
sification as in Study 1, 63.5% of participants self-identified as
politically liberal, 13.5% were centrist, and 23% were conserva-
tive. Given a lack of prior research, we set sample size to N⫽200
a priori. This ensured that we had more than 95% power to detect
a small to medium sized interaction effect of f
2
⫽.10 Participants
were randomly assigned to either the past-focus or future-focus
condition, with participant ideology (liberal to conservative) mea-
sured as a continuous variable.
Participants and design Study 2b. There were 423 Prolific
Academic users from the United Kingdom (56.7% women, 43.3%
men, mean age 37.3 years, 90.3% White, 2.9% Black, 4.6% Asian,
2.2% other/mixed) who participated for £0.25. Following the same
classification as in Study 1 and 2a, 65.1% of participants self-
identified as politically left, 8.0% were centrist, and 26.9% were
right. We set sample size to N⫽413 a priori, which ensured 90%
power to detect the effect size observed in Study 2a (f⫽.16).
Participants were randomly assigned to either the past-focus or
future-focus condition, with participant ideology (left to right-
wing) measured as a continuous variable.
Manipulation. Participants first indicated their political ide-
ology. In Study 2a, we used the same two items as in Study 1, r⫽
.90, p⬍.0001, that were combined in one measure of participant
ideology. In Study 2b, we adapted the items to the British political
context. Specifically, we first asked participants to place them-
selves on a scale between Very Left Wing (1) and Very Right Wing
(7), and next asked them to indicate their preference for the two
major parties, ignoring all others, between Definitely Prefer Labor
(1) and Definitely Prefer Conservative (7). Both items correlated
well, r⫽.79, p⬍.0001, and were combined in one measure of
participant ideology.
Next, participants read a short, four-sentence political message
on the topic of “the criminal justice system.” In all cases, the
message stressed the undesirable strictness of the current justice
system and called for a more lenient alternative. Depending on
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
5
POLITICAL PREFERENCE AND TEMPORAL FOCUS
condition, this alternative was either inspired by a comparison with
a desirable future or a comparison with a desirable past. In the
future condition, the text read:
Today’s justice system appears to be quite strict, but fortunately we
can change this. Sometime in the near future I can see a shift
happening so that police will be more lenient. If you commit an
offense in traffic or you have a drink in public, the police would not
immediately book you but instead just give you a warning. I’d like this
kind of change, so that in the near future, police will be more lenient
and not so strict toward offenders.
In contrast, in the past condition, the text read:
Today’s justice system appears to be quite strict, but fortunately this
was not always the case. In the old days, police were more lenient. If
you committed an offense in traffic or you had a drink in public, the
police would not immediately book you but instead just give you a
warning. I would like to go back to how things were then, when police
were more lenient and not so strict toward offenders.
Endorsement. After reading the message, participants pro-
vided their endorsement on four 5-point Likert items. The items
asked participants to indicate their agreement with the message
(1 ⫽strongly disagree,5⫽strongly agree), their liking for its
author (1 ⫽strongly dislike,5⫽strongly like), whether they
thought the author would perform well in public office (1 ⫽very
poorly,5⫽very well), and whether they would vote for the author
(1 ⫽very unlikely,5⫽very likely). These four items were reliable
(␣
2a
⫽.90 and ␣
2b
⫽.89), and were combined into one Endorse-
ment index.
Results Study 2a
We used the Process macro for SPSS (Model 1; Hayes, 2013) to
test our prediction that conservatives would be less likely to
endorse the proleniency appeal than liberals, but that their dis-
agreement would be reduced if the appeal drew a comparison with
the past rather than with the future. Consistent with predictions, we
found a significant interaction between participant ideology and
condition on endorsement of the proleniency message, b⫽.19,
SE ⫽.08, t(196) ⫽⫺2.22, p⫽.028, 95% CI
b
⫽[.021; .355] (see
Figure 1, left-side).
To interpret this interaction, we calculated simple slopes of
ideology predicting endorsement within the future- and past-focus
conditions. Conservatism predicted less endorsement in the future
condition, b⫽⫺.203, SE ⫽.06, t(196) ⫽⫺3.23, p⫽.001, 95%
CI
b
⫽[⫺.327; ⫺.079], but this effect was attenuated and no longer
significant in the past condition, b⫽⫺.015, SE ⫽.06,
t(198) ⫽⫺.26, p⫽.793, 95% CI
b
⫽[⫺.127; .098]. In other
words, we found significant disagreement between liberals and
conservatives regarding the proleniency message in the future-
focus condition (a statistically significant effect of ideology on
endorsement), but a temporal focus on the past virtually eliminated
this disagreement, which was no longer significant.
We also utilized the Johnson-Neyman significance region tech-
nique (JN; Johnson & Neyman, 1936; Preacher, Rucker, & Hayes,
2007) to determine the value of political ideology where the effect of
condition on endorsement becomes significant at pⱕ.05 (two-sided).
The past-focus manipulation increased endorsement for participants
scoring 3.83 (z⫽0.30) or above, suggesting that a temporal focus on
Figure 1. Studies 2a and 2b: Conservatives in the United States (left; Study 2a) and the United Kingdom (right;
Study 2b) endorse a proleniency message less than liberals do, but only if it focuses on the future (gray), and
not if it focuses on the past (black). Regression lines indicate simple slopes within condition. The dashed vertical
line represents the Johnson-Neyman (JN) value; differences between conditions to the right of the dashed line
are significant at pⱕ.05. Data-points are jittered to avoid overplotting; regression lines only extend the range
of the data.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
6LAMMERS AND BALDWIN
the past improved conservatives’ opinion about the proleniency mes-
sage, but did not significantly influence liberals’ opinions.
Results Study 2b
As in Study 2a, we found a significant interaction between
ideology and condition, b⫽.10, SE ⫽.05, t(409) ⫽1.98, p⫽
.049, 95% CI
b
⫽[.001; .202] (see Figure 1, right-side). To inter-
pret this interaction, we calculated simple slopes of ideology
predicting endorsement within the future- and past-focus condi-
tions. Conservatism predicted less endorsement in the future con-
dition, b⫽⫺.11, SE ⫽.04, t(409) ⫽⫺3.16, p⫽.002, 95% CI
b
⫽
[⫺.184; ⫺.043], but this effect was attenuated and no longer significant
in the past condition, b⫽⫺.01, SE ⫽.04, t(409) ⫽⫺.32, p⫽.747, 95%
CI
b
⫽[⫺.084; .060]. Again, we found significant disagreement
between liberals and conservatives regarding the proleniency mes-
sage in the future-focus condition, but a temporal focus on the past
virtually eliminated this disagreement to nonsignificance.
Utilizing the Johnson-Neyman significance region technique,
we found that the past-focus manipulation increased endorsement
for participants scoring 3.73 or above (z⫽0.32), at pⱕ.05
(two-sided), suggesting that changing the temporal focus improved
conservatives’ opinion about the proleniency message, while leav-
ing liberals’ opinions largely unaffected.
Discussion
The disagreement between liberals and conservatives about a
message calling for greater leniency in criminal justice was
strongly reduced—to the point of statistical indiscernibility—if the
message was focused on the past, rather than on the future. This is
remarkable, given the strength of opposition that conservatives
typically display toward leniency in criminal justice and their
typical preference for greater severity (Beckett & Sasson, 2003).
Furthermore, we found this effect in both a U.S. sample and in a
higher-powered replication in the United Kingdom, suggesting the
effect is not limited to the American political domain.
Study 3: Gun Rights
Study 2 showed that a past-focused manipulation reduces polit-
ical disagreement over criminal leniency. In the next two studies,
we test the same prediction with even more contentious political
issues. One particularly contentious issue in American politics is
gun rights. American progressives often call for restricting the
Second Amendment of the American Constitution, which protects
the right of the people to keep and bear arms, but this message is
often met with fierce resistance and opposition among conserva-
tives (Shalhope, 1982). Nevertheless, we tested in Study 3 whether
conservatives’ opposition would be reduced if a message in favor
of gun restrictions was fueled by a past-focused comparison (e.g.,
“I would like to go back to the old days, where the ownership of
weapons was more limited”).
Method
Participants and design. There were 200 American Amazon
Mechanical Turk users (56.5% women, 43.5% men, mean age 35.5
years, 77.5% White, 12.5% Black, 5.5% Asian, 4.5% other) who
participated for $0.25. Following the same classification as in prior
studies, 58.7% of participants self-identified as politically liberal,
16.4% were centrist, and 24.9% were conservative. We set sample
size to N⫽200 a priori. Participants were randomly assigned to
either the past-focus or future-focus condition, with participant
ideology (liberal to conservative) measured as a continuous vari-
able.
Manipulation. The procedure was identical to that of Studies
2a and 2b, except for the content of the text. Participants first
indicated their political ideology on the same two items as in
previous studies with American participants, r⫽.78, p⬍.0001,
that we again combined. Next, they read a short call for increased
gun control. The content of the message was identical across
conditions and asked for a ban on assault rifles and recommended
restricting gun ownership to hunting rifles and pistols. The only
difference between conditions was its temporal focus; the text was
either inspired by a comparison with a desirable future or with a
desirable past. Specifically, in the future condition, the text read:
Recent events show the problem of unrestricted access to guns. I
would prefer to make a change, so that in the future people may own
hunting rifles and pistols, but no one will have assault rifles. I would
like to make this change, so that in the near future the ownership of
weapons will be more limited.
In contrast, in the past condition, the text read:
Recent events show the problem of unrestricted access to guns. I
would prefer to go back to the old days, where people may have
owned hunting rifles and pistols, but no one had assault rifles. I would
like to go back to the old days, where the ownership of weapons was
more limited.
Endorsement. After reading the text, participants provided
their endorsement of the message and its author, using the same
four 5-point Likert items as in Studies 2a and 2b (␣⫽.88), that
were combined into one Endorsement index.
Results
We expected that conservatives would reject the call for gun
control, unless it was past-focused. In support of this prediction,
we found a significant interaction between participant ideology
and condition, b⫽.16, SE ⫽.08, t(196) ⫽⫺2.01, p⫽.046, 95%
CI
b
⫽[.003; .326] (see Figure 2). As in Studies 2a and 2b, we
interpreted this interaction by calculating simple slopes of ideol-
ogy on endorsement within condition. Conservatism predicted less
endorsement in the future condition, b⫽⫺.28, SE ⫽.06,
t(196) ⫽⫺4.86, p⬍.0001, 95% CI
b
⫽[⫺.400; ⫺.168], but this effect
was attenuated in the past condition, b⫽⫺.12, SE ⫽.06,
t(196) ⫽⫺2.05, p⫽.042, 95% CI
b
⫽[⫺.232; ⫺.005]. In other
words, we found significant disagreement between liberals and
conservatives regarding the gun control message in the future-
focus condition, but a temporal focus on the past reduced this
disagreement.
Using the JN technique, we found that a temporal focus on the
past significantly increased endorsement for participants scoring
6.78 (z⫽2.18) or above (pⱕ.05, two-sided). However, the
condition effect begins to emerge (at pⱕ.05, one-sided) for
participants scoring around 4.90 (z⫽0.98, p⫽.098), suggesting
that the temporal focus on the past improved conservatives’ opin-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
7
POLITICAL PREFERENCE AND TEMPORAL FOCUS
ion about gun control, while leaving that of centrists and liberals
largely unaffected.
Discussion
Study 3 replicates Study 2 using a different political issue; the
degree to which liberals and conservatives disagreed about a call
for increased gun control was greatly reduced if the message drew
comparisons with the past, instead of pointing toward the future.
Given the passion and emotion that partisans on both sides typi-
cally display in this discussion, the ease by which temporal focus
can reduce such disagreement is remarkable.
Study 4: Migration
In Study 4 we aimed to test a particularly contentious issue in
current German politics: migration. Specifically, political opinion
on how Germany should respond to the Syrian migrant crisis, with
many on the left calling for easing restrictions, while some on the
right call for greater restrictions or even closing of the border
(Geddes & Scholten, 2016). In Study 4 we test whether a call to
ease restrictions on migration is accepted more easily by those on
the right of the German political spectrum if it focused on the past
(compared with the future).
Method
Participants and design. There were 459 German partici-
pants (72.1% women, 27.5% men, 0.4% other, mean age 30.1
years) who took part for partial course credit or participation in a
€25 raffle. We recruited as many participants as possible during a
fixed 2-week period, with a minimum sample size of N⫽360.
3
Participants were randomly assigned to either the past-focus or
future-focus condition, with participant ideology measured as a
continuous variable.
Participant ideology. Participants indicated their opinion on
10 political statements mapping onto the current German political
debate, each on a 5-point scale between 1 ⫽very negative and 5 ⫽
very positive. Five were left-wing political statements (gay-
marriage, gays’ adoption rights, industrial genetic engineering
restrictions, and lignite mining cessation) and five were right-wing
political statements (stricter policing, reduced social security, im-
migration limits, abortion restrictions, and family values). We
combined all items, after recoding the left-wing items, into one
measure of political ideology (␣⫽.74), with higher scores indi-
cating stronger conservatism.
Manipulation. Next, participants read an article calling for
Germany to open its borders to refugees from Syria. The individual
arguments presented were the same in both conditions, except for
their temporal focus. Depending on condition, the text either
presented immigration from Syria as a new phenomenon that
presents new opportunities to forge connections between Germans
and Syrians, that will add to the German economy, and that will
present no major problems (future focus); or as an old phenome-
non and long-existing connection that goes back to the earliest
days in Germany’s recorded history, that have always added to the
German economy, and that have never presented major problems
(past focus; see Appendix 1).
Endorsement. After reading the text, participants provided
their endorsement of the message and its author, using the same
four 5-point Likert items as in Studies 2a, 2b, and 3 (␣⫽.92),
which were combined into one Endorsement index.
Results
We expected that conservatives would reject the call for in-
creased migration, unless it was communicated using a past focus.
In support of this prediction, we found a significant interaction
between participant ideology and condition, b⫽.39, SE ⫽17,
t(455) ⫽2.24, p⫽.025, 95% CI
b
⫽[.048; .733]; see Figure 3.Asin
prior studies, we interpreted this interaction by calculating simple slopes
of ideology on endorsement within condition. Conservatism predicted
less endorsement in the future condition, b⫽⫺1.31, SE ⫽.13,
t(455) ⫽⫺10.13, p⬍.0001, 95% CI
b
⫽[⫺1.561; ⫺1.054], but this
effect was attenuated, although still significant, in the past condition,
b⫽⫺0.92, SE ⫽.12, t(455) ⫽⫺7.83, p⬍.0001, 95% CI
b
⫽
[⫺1.146; ⫺.686]. In other words, although we found statistically
significant disagreement about immigration in both conditions, the
3
We cannot provide a breakdown of the population according to ideol-
ogy, because we do not measure ideology directly, but with Likert valence
items. We thank Aleyna Asan, Lena Blaesen, Rebecca Hector, Serap
Burcak, Seyma Evran, Maria Jansen, Nikola Kargus, Olivia Parczyk,
Isabelle Rockstroh, Frederik Seiter, and Deyan Stoyanov who collected
these data as part of a course requirement. For exploratory reasons, this
study also included an unvalidated German Need for Structure scale,
inspired by Thompson, Naccarato, Parker, and Moskowitz (2001) and an
unvalidated translation of Holbrook’s (1993) measure. These are not dis-
cussed further.
Figure 2. Study 3: Conservatives endorse a message advocating gun
control less than liberals if it focuses on the future (gray), but this effect is
attenuated if the message focuses on the past (black). Regression lines
indicate simple slopes within condition. The dashed vertical line represents
the Johnson-Neyman (JN) value; differences between conditions to the
right of the dashed line are significant at pⱕ.05. Data-points are jittered
to avoid overplotting; regression lines only extend the range of the data.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
8LAMMERS AND BALDWIN
degree of disagreement was significantly stronger in the future
focus than in the past focus condition.
Using the JN technique, we found that past-focused communi-
cation significantly increased endorsement for participants scoring
2.23 (z⫽0.02) or above (pⱕ.05, two-sided), suggesting that the
temporal focus manipulation improved conservatives’ opinion
about immigration, while leaving that of liberals largely unaf-
fected.
Study 5: Diversity and the Mediating Role of
State Nostalgia
In the prior four studies, we presented participants with a polit-
ical message that was fueled by a past or future comparison.
However, in all studies, the message was communicated by an
ostensible person in the present. In Study 5 we modified our
procedure slightly by presenting participants with the same stim-
ulus across conditions, but manipulated whether the stimulus was
perceived to originate in the past or the present. Specifically, all
participants viewed a genuinely vintage comic strip featuring
Superman defending social diversity to a group of schoolchildren
(Institute for American Democracy, 1949). Participants were either
told that the comic was in its original form, and communicated
“old-fashioned values” (past condition) or was being modified to
reflect “modern-day values” (control condition). We expected that
conservatives would reject that comic given that they do not share
liberals’ strong support for social diversity (Parekh, 2002), but not
if they believed the comic to have originated in the past (even
though the content of the comic was held constant).
In addition, we sought to demonstrate the mediating role of state
nostalgia. Specifically, we hypothesized that conservatives would
be likely to experience particularly strong immediate feelings of
nostalgia, but only when shown the vintage comic. In other words,
we expected an interaction of political ideology and condition on
feelings of nostalgia. Furthermore, we expected that these feelings
of nostalgia would explain their greater support for the prodiver-
sity message (in the past-focus condition; moderated mediation)
compared with when shown the allegedly modern-day adaptation
of the comic.
Method
Participants and design. There were 564 American Amazon
Mechanical Turk users (43.1% women, 56.9% men, mean age 32.9
years, 79.1% White, 6.6% Black, 10.1% Asian, 4.3% other) who
participated for $0.25. Following the same classification as in prior
studies, 59.2% of participants self-identified as politically liberal,
15.2% were centrist, and 25.5% were conservative. Participants
were randomly assigned to either the past-focus or future-focus
Condition We set sample size to N⫽564 a priori, which ensured
90% power to detect the effect size observed in a pilot study
(⌬R
2
⫽.025, f⫽.16; N⫽200, see online supplementary file
Study 5b), with participant ideology measured as a continuous
variable.
Manipulation. Participants first listed their political ideology,
using the same two items as in previous studies, r⫽.84, p⬍
.0001. Next, participants were told that an action comic publisher
was considering reprinting old comics and that we were interested
in testing participants’ reaction to it. To illustrate, participants
were shown the same, original Superman comic (Institute for
American Democracy, 1949). In that picture, the superhero tells a
group of schoolchildren to remember that
Your school—like our country—is made up of Americans of many
different races, religions, and national origins. So if you hear anybody
talk against a schoolmate or anyone else because of his religion, race,
or national origin— do not wait: tell him “THAT KIND OF TALK IS
UN-AMERICAN.”
Depending on condition, we told in the control condition that:
What is striking is that although these comics were originally printed
in the 40s and 50s, they can nonetheless be easily adapted to discuss
issues that are socially relevant today. The publisher wants to reprint
these comics, but changing them to an updated, modern-day form.
“We’re updating a popular American icon to communicate good
modern-day values,” said a spokesperson from the company.
In contrast, in the past-focus condition participants read that:
What is striking is that although these comics were originally printed
in the 40s and 50s, they nonetheless discuss issues that are socially
relevant today. The publisher wants to reprint these comics, while
keeping them in their original, authentic form. “We’re using a classic
American icon to communicate good old-fashioned values,” said a
spokesperson from the company.
More important, all participants saw exactly the same picture;
we only changed the temporal focus. Participants were either told
(truthfully) that the picture was authentic and identical to how it
appeared in 1949, or they were led to believe that the picture was
changed and adapted for today.
Figure 3. Study 4: German conservatives endorse a message advocating
easing restrictions on immigration from Syria less than liberals, but this
effect is attenuated if it focuses on the past (black), compared with when
it focuses on the future (gray). Regression lines indicate simple slopes
within condition. The dashed vertical line represents the Johnson-Neyman
(JN) value; differences between condition to the right of the dashed line are
significant at pⱕ.05. Data-points are jittered to avoid overplotting;
regression lines only extend the range of the data. Note the absence of
political orientation scores ⬎4; for historical reasons Germans are less
likely to openly express far-right beliefs.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
9
POLITICAL PREFERENCE AND TEMPORAL FOCUS
Nostalgia and endorsement. Participants first completed a
three-item measure of state nostalgia take from Baldwin, White,
and Sullivan (2017). Specifically, they indicated whether they felt
“nostalgic,” “sentimental,” and “longing” (1 ⫽none at all,5⫽a
great deal). These three items were reliable (␣⫽.90), and were
combined into one State Nostalgia index.
Next, all participants indicated their endorsement of the prodi-
versity message, using four 5-point Likert items that measured
their feelings for the publisher (1 ⫽strongly dislike,5⫽strongly
like), their agreement with the publisher (1 ⫽strongly disagree,
5⫽strongly agree), whether they would like owning a copy of the
reprint (1 ⫽strongly dislike,5⫽strongly like), and whether they
would be likely to donate a copy as a present (1 ⫽very unlikely,
5⫽very likely). These four items were reliable (␣⫽.86), and
were combined into one Endorsement index.
Results
Direct effects. Replicating the results of prior studies, we
found a significant interaction between participant ideology and
condition, b⫽.13, SE ⫽.05, t(560) ⫽3.01, p⫽.003, 95% CI
b
⫽
[.047; .222]. Calculating simple slopes, we found that conservatism
predicted less endorsement in the present condition, b⫽⫺.12, SE ⫽.03,
t(560) ⫽⫺3.77, p⫽.0002, 95% CI
b
⫽[⫺.179; ⫺.056], but this effect
was largely eliminated in the past condition, b⫽.02, SE ⫽.03,
t(560) ⫽.52, p⫽.606, 95% CI
b
⫽[⫺.046; .079]. In other words,
we found significant disagreement between liberals and conserva-
tives regarding the prodiversity message in the future-focus con-
dition, but a temporal focus on the past eliminated this disagree-
ment.
The JN analysis revealed that the vintage message increased
endorsement for participants scoring 1.40 (z⫽⫺1.39) or above on
the political ideology scale (at pⱕ.05, two-sided), suggesting that
the temporal focus manipulation affected conservatives’ and mod-
erates’ opinion about diversity, but left that of liberals unaffected
(see Figure 4).
Mediation analysis. Next, we tested whether conservatives’
positive attitude toward the past-focused message (compared with
the future-focused message) was driven by feelings of nostalgia,
experienced in response to the past-focused message. Note that
because we focus on state feelings of nostalgia, we hypothesize an
interaction between ideology and condition, so that conservatives
experience particularly strong immediate feelings of nostalgia com-
pared with liberals, but only in response to the past-focused message
and not in response to the future-focused message. Indeed, we found
the predicted interaction between ideology and condition, b⫽.12,
SE ⫽.06, t(560) ⫽2.08, p⫽.038, 95% CI
b
⫽[.007; .238].
Calculating simple slopes, we found that conservatism significantly
predicted stronger feelings of nostalgia in the past condition, b
past
⫽
.10, SE ⫽.04, t(560) ⫽2.27, p⫽.023, 95% CI
b
⫽[.013; .178], but
not in the future condition, b
future
⫽⫺.03, SE ⫽.04, t(560) ⫽⫺.65,
p⫽.516. Analyzed differently, the JN analysis revealed that the
vintage message increased feelings of nostalgia for participants scor-
ing 2.05 or above (z⫽⫺0.72), suggesting that only strong liberals
were unaffected by it.
Next, we used Hayes’ (2013) Process macro for SPSS (Model 8;
5,000 bootstrap samples) to test whether the increased nostalgia
experienced by conservatives in the past-focused condition ex-
plained endorsement of the prodiversity message in that condition
(i.e., moderated mediation). We expected that the increased nos-
talgia felt by conservatives in the past-focused condition would
mediate the association between conservatism and increased en-
dorsement in the past-focused condition. To test this model, nos-
talgia was entered as the mediator, and condition was included as
a moderator of both the effect of political ideology on endorsement
(c path, reported above) and the effect of political ideology on
nostalgia (a path, reported above; see Figure 5).
Indeed, we found a significant index of moderated mediation,
index ⫽.06, SE ⫽.03, 95% CI [.001; .124], indicating that the
indirect effects differed significantly between the conditions. Spe-
cifically, we found a significant positive indirect effect of ideology
on endorsement through nostalgia, but only in the past condition,
b
past
⫽.05, SE ⫽.02, 95% CI [.007; .092], and not in the future
condition, b
future
⫽⫺.01, SE ⫽.02, 95% CI [⫺.060; .031]. In
other words, conservatives particularly endorsed the message if
they believed it to be authentic, because in that case they experi-
enced particularly strong feelings of nostalgia.
Discussion
Study 5 showed that conservatives’ opposition to a prodiversity
message strongly depended on the temporal context of the mes-
sage. If participants believed a “vintage” comic was changed to
resonate with “modern day” values, they strongly rejected the
message. However, if they believed the message to be authentic
and rooted in the past— communicating “good old-fashioned”
Figure 4. Study 5: Conservatives endorse a prodiversity message less
than liberals do if it is presented as “adapted for modern-day” (present-
focus; gray) but this effect is attenuated if it is presented as genuinely
vintage (past-focus; black). Regression lines indicate simple slopes within
condition. The dashed vertical line represents the Johnson-Neyman (JN)
value; differences between condition to the right of the dashed line are
significant at pⱕ.05. Data-points are jittered to avoid overplotting;
regression lines only extend the range of the data.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
10 LAMMERS AND BALDWIN
values—they equally supported the prodiversity message as liber-
als. What is striking about this study is that the message itself was
identical in both conditions; only the temporal context changed.
Furthermore, we demonstrated the mediating role of state nostal-
gia. Conservatives (and moderates) endorsed the past-focused
message particular strongly, because that message evoked partic-
ularly strong feelings of nostalgia with them.
It is also noteworthy that the effect of the temporal focus
manipulation in this study appeared to affect not only conserva-
tives, but moderates as well. This is likely because of the nature of
the manipulation in this study compared with the others: in pre-
vious studies, the message always originated in the present, but
either called for returning to the past or looking to the future.
Returning to the past is a particularly conservative ideal, which is
why these temporal focus manipulations only affected conserva-
tives. In this study, however, the message itself either originated in
the past or present. Given that there is a strong general bias to
prefer things and ideas that have existed for a long time (i.e., the
existence bias; Eidelman, Crandall, & Pattershall, 2009), it is
likely that this bias was a factor in moderates’ endorsement of the
message. Nevertheless, the pattern was still strongest for conser-
vatives, suggesting that the preference for the past is stronger for
these participants.
Study 6: Social Justice and the Mediating Role of
Trait Nostalgia
Study 5 demonstrated the mediating role of state nostalgia. We
found that conservatives were particularly likely to endorse the
past-focused message, because they experienced particularly
strong feelings of nostalgia directly after reading that message
(compared with liberals and/or those reading a future-focused
message). The aim of Study 6 was to uncover the underlying
process of dispositional (trait) differences in nostalgia. Building on
the results of Study 1, we expected that conservatives would score
higher on dispositional nostalgia than liberals (independent of
condition) and that this difference would explain their more pos-
itive reaction to a past-focused (vs. future-focused) political mes-
sage. In testing this relation, we focus on a different political issue:
Liberals tend to believe that society disadvantages certain social
groups while privileging others, and as a result demand more
social justice. In contrast, conservatives tend to trust and believe
that current society is just and thus reject the need for progressive
social justice policies (Bobbio, 1996; Hayek, 1944, 1982; Rawls,
1971). They consider inequality to be a fair and justified conse-
quence of differences in merit, willingness to work, and luck.
Therefore, conservatives are more likely to defend those inequal-
ities (Jost & Banaji, 1994; Jost, Banaji, & Nosek, 2004; Jost &
Kay, 2010; Rothmund, Becker, & Jost, 2016). We test whether
disagreement about the issue of social justice is also affected by
temporal focus. Before conducting Study 6, we first conducted two
near-identical studies (N⫽201 and N⫽202), but without mea-
suring nostalgia as a mediator. These studies are reported as
Studies 6b and 6c in an online supplementary file.
Method
Participants and design. There were 200 American Amazon
Mechanical Turk users (54.0% women, 46.0% men, mean age 33.9
years, 81.5% White, 8.0% Black, 6.5% Asian, 4.0% other/mixed)
who participated for $0.40. Following the same classification as in
prior studies, 55.5% of participants self-identified as politically
liberal, 20.5% were centrist, and 24.0% were conservative. Partic-
ipants were randomly assigned to one of two conditions (message
temporal focus: past vs. future) with participant ideology (liberal to
conservative) measured as a second, continuous variable. We set
sample size to N⫽200 a priori.
Nostalgia. Participants first indicated their political ideology
on the same two items as in previous studies, r⫽.86, p⬍.0001,
and then completed Holbrook’s (1993) eight-item measure of
nostalgia, used in Study 1 (␣⫽.73).
Manipulation and endorsement. Next, participants read a
short, four-sentence message stressing the need for more social
justice. In all cases, the message stressed the unfairness of current
society and pointed to an alternative, but the message was either
focused on the past or on the future. Specifically, in the past-focus
condition, the text read:
Today’s society appears to be unfair, but fortunately this was not
always the case. Back when our country was founded like 200
something years ago, society was much more just and fair. In the old
days, everyone had a fair shot at gaining wealth and people usually got
what they deserved. I would like to go back to how things were then,
when everything was more just and fair.
In contrast, in the future-focus condition, the text read:
Today’s society appears to be unfair, but fortunately we can change
this. Sometime in the future we can change society so that it will be
much more just and fair. I can see a shift happening in the near future
so that everyone will have a fair shot at gaining wealth and people will
get what they deserve. I’d like this kind of change, so that everything
will be more just and fair.
Participants provided their endorsement using the same four
items as in previous studies (␣⫽.92).
Results
Direct effects. As in previous studies, we found a significant
Political Ideology ⫻Condition interaction, b⫽.50, SE ⫽.08,
t(196) ⫽6.12, p⬍.0001, 95% CI
b
⫽[.339; .661]. To interpret this
interaction, we calculated simple slopes of ideology predicting
Figure 5. Study 5: Conservatives respond with particularly strong feel-
ings of state nostalgia to a past-focused prodiversity message, and this
explains why they respond particularly well to such a message.
ⴱ
p⬍.05,
ⴱⴱ
p⬍.01,
ⴱⴱⴱ
p⬍.001.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
11
POLITICAL PREFERENCE AND TEMPORAL FOCUS
endorsement within the past- and future-focus conditions. Conser-
vatism predicted less endorsement in the future condition,
b⫽⫺.25, SE ⫽.06, t(196) ⫽⫺4.22, p⬍.0001, 95% CI
b
⫽
[⫺.363; ⫺.132], but this effect was reversed in the past condition,
b⫽.25, SE ⫽.06, t(198) ⫽4.43, p⬍.0001, 95% CI
b
⫽[.140;
.365]. In other words, we found statistically significant disagree-
ment between liberals and conservatives about justice in the future
focus condition, but this disagreement was blocked and even
reversed in the past focus condition, with conservatives calling for
greater social justice.
The JN analysis revealed that the past-focused message in-
creased endorsement for participants scoring 6.27 (z⫽1.77) or
above. In contrast to the previous studies, we found that the
past-focused message also decreased endorsement for participants
scoring 4.56 (z⫽0.72) or below (both at pⱕ.05, two-sided).
Conservatives support a prosocial-justice message if it is focused
on the past, whereas moderates and liberals show the opposite
pattern.
Mediation analysis. Next, we tested whether conservatives’
attitudes toward the messages were driven by nostalgia. To do so,
we used Hayes’ (2013) Process macro for SPSS (Model 15, 5,000
bootstrap samples) to test for the indirect effect of political pref-
erence on endorsement, through nostalgia, whereby the effect of
nostalgia on endorsement was moderated by the temporal focus of
that message (i.e., moderated mediation). Note that because we
now focus on trait-differences in nostalgia, we expected that con-
servatives would overall be more nostalgic than liberals (indepen-
dent of condition) and that this difference would mediate the
association between conservatism and increased endorsement in
the past-focused condition, and between conservatism and de-
creased endorsement in the future-focused condition. To test this
model, nostalgia was entered as the mediator, and condition was
included as a moderator of both the effect of political ideology on
endorsement (c path, reported above) and also the effect of nos-
talgia on endorsement (b path, reported below), as we expected
that nostalgia would predict more support for the past focused
message, and less support for the future-focused message (see
Figure 6).
As in Study 1, political ideology predicted nostalgia (a path),
such that conservatives expressed stronger nostalgia, b⫽.13,
SE ⫽.04, p⬍.001, 95% CI
b
⫽[.058; .207]. We also found a
significant interaction between nostalgia and temporal focus con-
dition on endorsement (b path), b⫽.61, SE ⫽.15, p⫽.0001, 95%
CI
b
⫽[.320; .904], such that nostalgia predicted stronger endorse-
ment of the message in the past condition, b
past
⫽.54, SE ⫽.11,
t(195) ⫽5.05, p⬍.0001, and less support for the message in the
future condition, b
future
⫽⫺.25, SE ⫽.11, t(195) ⫽⫺2.25, p⫽
.026. More important, the index of moderated mediation was
significant, index ⫽.08, SE ⫽.03, 95% CI [.030; .162], indicating
that the indirect effects differed significantly between the condi-
tions. Specifically, we found a significant indirect effect of ideol-
ogy through nostalgia in the past condition, such that conservatism
predicted higher nostalgia, which in turn, predicted stronger en-
dorsement of the past-focused message b
past
⫽.06, SE ⫽.02, 95%
CI [.021; .108]. In the future condition, this indirect effect was
reversed, albeit only trending and nonsignificant, b
future
⫽⫺.03,
SE ⫽.02, 95% CI [⫺.072; .0005].
Discussion
The results of Study 6 uncovered the mediating role of dispo-
sitional differences in trait nostalgia. Conservatives experience
stronger dispositional nostalgia and, therefore, endorse a past-
focused message more than liberals do. Together with the results
of Study 5, which focused on differences in state nostalgia, these
two studies together demonstrate the role of nostalgia in shaping
the current effects. It is also noteworthy that the past-focus induc-
tion reversed the association between political ideology and en-
dorsement in this study, while in previous studies we merely found
attenuation. One reason for this may be that compared with the
issues used in prior studies (e.g., diversity, gun rights), social
justice is more open to differences in construal or definition (Jost
& Kay, 2010; Rothmund et al., 2016). Although a call for greater
social justice is central to liberal political thinking, conservatives
may also call for more, but different forms of justice. For example,
they may perceive a move toward greater equality between social
groups or a change of long-existing institutions as unjust (Ball &
Bellamy, 2003; Ball & Dagger, 1999; McClelland, 2005). This
flexibility in the construal of justice may explain why a temporal
focus on the past even reversed the effect of ideology on support
for the call for more justice, compared with when it was focused
on the future, rather than merely attenuate it—as in previous
studies.
Study 7: Central Versus Peripheral Processing
The degree to which peripheral aspects of information affect
people’s opinion about that information have often been found to
depend on the way in which people process such information.
These peripheral aspects of information have a strong effect if
people rely on affective, superficial, or peripheral information
processing, but are attenuated or erased when people engage in
analytical, deep, or central information processing (Chaiken, 1980;
Petty & Cacioppo, 1986; Thompson et al., 1994). In Study 7, we
sought to demonstrate a similar boundary condition for the ob-
served effect of temporal focus, by using the same materials as in
Study 2 (leniency in criminal justice), but orthogonally manipu-
lating participants’ processing style.
Method
Participants and design. There were 400 American Amazon
Mechanical Turk users (53.3% women, 46.8% men, mean age 37.0
Figure 6. Study 6: Conservatives are more trait nostalgic than liberals
and this explains why they endorse a future-focused prosocial justice
message less, but a past-focused prosocial justice message more than
liberals do.
ⴱ
p⬍.05,
ⴱⴱ
p⬍.01,
ⴱⴱⴱ
p⬍.001.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
12 LAMMERS AND BALDWIN
years, 82.3% White, 6.8% Black, 5.5% Asian, 5.8% other/mixed)
who participated for $0.25. Following the same classification as in
prior studies, 61.5% of participants self-identified as politically
liberal, 14.8% were centrist, and 23.7% were conservative. Partic-
ipants were randomly assigned to one cell of a 2 (message tem-
poral focus: past vs. future) ⫻2 (processing style: peripheral vs.
central) quasi-experimental design, with participant ideology (lib-
eral to conservative) measured as a third, continuous variable. We
used the same procedure as in Study 2, except for the addition of
the processing style manipulation. Given the addition of that
additional factor, we a priori doubled the sample size from prior
studies to N⫽400.
Manipulation. Participants first listed their political ideology,
using the same two items as in previous studies, r⫽.81, p⬍
.0001. Next, participants read a liberal message. In the peripheral
processing style condition we used exactly the same procedure as
in Study 2. In the central processing condition, participants were
first asked to read the message carefully with the goal of under-
standing completely the author’s argument: “Please read the an-
swer well and try to understand what this person wants.” To help
them rely on central information processing, participants answered
two items about the content of the author’s message asking
whether the author proposed that the police should be generally
more lenient and that the criminal justice system should be more
lenient toward minor offenses, both between 1 ⫽definitely false
and 5 ⫽definitely true,r⫽.21, p⫽.003. Participants confirmed
that the author considered these statements to be true (M⫽4.39,
SD ⫽0.62), but as expected, scores on this item did not depend on
the temporal focus of the message, t(198) ⫽0.62, p⫽.534. In the
peripheral processing condition, participants did not receive these
additional instructions and items.
Measures. We used the same four-item measure of endorse-
ment (␣⫽.89), as in prior studies.
Results
We expected that the past-focused manipulation would reduce
liberals’ and conservatives’ disagreement with the call for leniency
in criminal justice, but only in the peripheral and not in the central
processing condition. To test this prediction, we conducted a
regression analysis with temporal focus condition, processing con-
dition, and political ideology as predictors of message endorse-
ment. All possible two- and three-way interactions were included
as predictors as well, but in particular we were interested in the
three-way interaction, and in examining the Focus Condition ⫻
Political Ideology two-way interactions in the peripheral and cen-
tral processing conditions separately. The three-way interaction
(Political Ideology ⫻Temporal Focus ⫻Information Processing)
was significant, b⫽.23, SE ⫽.11, t(392) ⫽2.04, p⫽.042, 95%
CI
b
⫽[.009; .449], and thus we explored each of the two-way
Focus Condition ⫻Political Ideology interactions of interest.
Within the peripheral processing condition (Figure 7, right
pane), the Political Ideology ⫻Temporal Focus interaction was
marginally significant, b⫽.14, SE ⫽.08, t(392) ⫽1.83, p⫽.068,
95% CI
b
⫽[.010; .298], and the pattern replicated those of the
previous studies. Conservatism was associated with lower endorse-
ment of the message in the future condition, b⫽⫺.21, SE ⫽.06,
t(392) ⫽⫺3.77, p⬍.001, 95% CI
b
⫽[⫺.319; ⫺.100], but this
effect was attenuated in the past condition, b⫽⫺.07, SE ⫽.06,
t(392) ⫽⫺1.20, p⫽.233, 95% CI
b
⫽[⫺.175; .042]. The JN
analysis revealed that within this condition, the past-focused mes-
sage increased endorsement for participants scoring 4.38 (z⫽
0.67) or above (at pⱕ.05, two-sided).
In contrast, in the central processing condition (Figure 7, left
pane) the two-way interaction was not significant, b⫽⫺.09, SE ⫽
.08, t(392) ⫽⫺1.06, p⫽.288, 95% CI
b
⫽[⫺.072; .242], and
evidence was in favor of the null hypothesis that b
interaction
⫽0,
scaled JZS Bayes Factor ⫽14.36 (Bayes Factor was scaled on
r⫽1).
Discussion
Study 7 uncovered the moderating role of information process-
ing style. We replicated the results of Study 2 when participants
were allowed to process information peripherally, but this effect
was blocked when participants were instructed and guided to
process the information in a more central manner.
Study 8: Temporal Focus Effects for Liberals
The past seven studies showed that a temporal focus on the past
can reduce conservatives’ opposition to liberal ideas. This leaves
open the question of whether adopting a temporal focus on the
future can also reduce liberals’ opposition to conservative ideas.
As explained previously, we did not expect liberals to respond in
a similarly positive manner to a future-focused communication of
conservative messages as conservatives respond to past-focused
communication of liberal messages, because although conserva-
tives intrinsically value the past over the future, we do not believe
liberals to intrinsically value the future over the past. We test this
in Study 8, by focusing again on leniency in criminal justice (as in
Studies 2a, 2b, and 7), but using either a liberal proleniency
message (as before) or a conservative proseverity message. We
expected to find a three-way interaction between message content
(proleniency vs. proseverity), temporal focus (past vs. future), and
participants’ political preference, so that we replicate the Focus ⫻
Figure 7. Study 7: Under peripheral processing (right pane), conserva-
tives endorse a projustice message less than liberals do if it focuses on the
future (gray), but this effect is attenuated if focuses on the past (black).
This interaction is not significant under central processing (left pane).
Regression lines indicate simple slopes within condition. Data-points are
jittered to avoid overplotting; regression lines only extend the range of the
data.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
13
POLITICAL PREFERENCE AND TEMPORAL FOCUS
Political Preference two-way interaction in the proleniency condi-
tion, but not in the proseverity condition (where we expect only a
main effect, so that conservative participants support the message
more than liberals).
Method
Participants and design. There were 802 American Amazon
Mechanical Turk users (36.0% women, 64.0% men, mean age 34.1
years, 82.8% White, 6.2% Black, 7.2% Asian, 3.7% other/mixed)
who participated for $0.25. Following the same classification as in
prior studies, 55.5% of participants self-identified as politically
liberal, 15.3% were centrist, and 29.0% were conservative. Partic-
ipants were randomly assigned to one cell of a 2 (message content:
liberal vs. conservative) ⫻2 (message temporal focus: past vs.
future) quasi-experimental design, with participant ideology (lib-
eral to conservative) measured as a third, continuous variable. We
used the same procedure as in Study 2, except that participants
now received either a liberal or a conservative message. We
doubled sample size compared with Study 7 to N⫽800 a priori
(and unintentionally received two more observations), given the
predicted null-effect in the conservative proseverity condition.
Manipulation. Participants first listed their political ideology,
using the same two items as in previous studies, r⫽.85, p⬍
.0001. Next, participants read a political message about criminal
leniency that was either proleniency (liberal position) or prosever-
ity (conservative position) and that either focused on the past or
focused on the future. The proleniency message was adopted from
Study 2, with minor changes to make it more comparable with the
proseverity message (see Appendix 2). Next, we measured partic-
ipants’ endorsement of the message, with the same four-item
measure as in prior studies (␣⫽.91), again between 1 ⫽strongly
disagree and 5 ⫽strongly agree.
Results
We expected that a temporal focus on the past would increase
conservatives’ support for the pro-equality (liberal content) mes-
sage, but that temporal focus would not affect liberals’ support for
the proseverity (conservative content) message. All possible inter-
actions were included as predictors as well, but in particular we
were interested in examining the Message Content ⫻Temporal
Focus ⫻Political Ideology interaction, which was significant, b⫽
.24, SE ⫽.08, t(794) ⫽2.99, p⫽.003, 95% CI
b
⫽[.081; .391].
To interpret this interaction, we explored each of the two-way
interactions of interest within the proleniency and proseverity
content conditions, separately (see Figure 8).
Within the pro-equality (liberal content) condition (Figure 8, left
pane), we replicated the Political Ideology ⫻Temporal Focus
interaction of previous studies, b⫽.16, SE ⫽.06, t(794) ⫽2.91,
p⫽.004, 95% CI
b
⫽[.053; .274], and the pattern was consistent
with the previous studies. Conservatism was associated with lower
endorsement of the message in the future condition, b⫽⫺.21,
SE ⫽.04, t(794) ⫽5.17, p⬍.0001, 95% CI
b
⫽[⫺.293; ⫺.132],
but this effect was attenuated in the past condition, b⫽⫺.05,
SE ⫽.04, t(794) ⫽⫺1.27, p⫽.205, 95% CI
b
⫽[⫺.124; .027].
The JN analysis revealed that within this condition, the past-focused
condition increased endorsement for participants scoring 5.19 (z⫽
1.06) or above. Unexpectedly, the past-focused message also reduced
endorsement for participants scoring 2.22 (z⫽⫺0.71) or below (both
at pⱕ.05, two-sided).
In contrast, within the proseverity (conservative content) con-
dition (Figure 8, right pane), the Political Ideology ⫻Temporal
Focus interaction was not significant, b⫽⫺.07, SE ⫽.06,
t(794) ⫽⫺1.30, p⫽.194, 95% CI
b
⫽[⫺.180; .037], and evidence
was in favor of the null hypothesis that b
interaction
⫽0, scaled JZS
Bayes Factor ⫽15.29 (Bayes Factor was scaled on r⫽1).
Discussion
Experiment 8 demonstrated that although a temporal focus on
the past can increase conservatives’ support of liberal ideas (or
rather reduce a large portion of their opposition to liberal ideas), a
temporal focus on the future does not similarly increase liberals’
support of conservative ideas.
Study 9: Omitted Studies and Meta-Analysis
In addition to conducting these nine reported studies, we also
conducted four more studies. For transparency, we discuss them
here shortly and in more detail in an online supplementary file. We
omitted these four studies from the main manuscript because the
reported results of Studies 5, 6, and 7 make them superfluous.
Specifically, Study 5b (N⫽200), Study 6b (N⫽201), and Study
6c (N⫽202) used the same experimental materials as Studies 5
and 6, but these three studies differed from the reported studies in
that they did not include the mediator. Also, Study 6c used a single
item dependent measure, instead of the four-item measure used
throughout the other studies. All three studies showed the same
significant interaction between participant ideology and temporal
focus condition, p
5b
⫽.02, p
6b
⫽.0002, and p
6c
⬍.0001. Finally,
Study 8b (N⫽322) used the same design as Study 8, but different
materials, a single-item DV, and less than half the sample size of
that study. Although the critical three-way interaction was signif-
icant, p⫽.045, and in the expected direction, we felt that the
Figure 8. Study 8: A temporal focus on the past (black) increases
conservatives’ endorsement of a proleniency message (left pane) compared
with a future focus (gray). In contrast, the temporal focus of a message
does not affect liberals’ (or conservatives’) endorsement of a conservative,
proseverity message (right pane). Regression lines indicate simple slopes
within condition. Data-points are jittered to avoid overplotting; regression
lines only extend the range of the data.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
14 LAMMERS AND BALDWIN
sample-size was insufficient (␣⫽.05, ⫽1–0.52) to reliably
demonstrate the small effect we observed.
To test the Political Ideology ⫻Temporal Focus interaction
meta-analytically, including both included and omitted studies, we
used the Metafor package for R (Viechtbauer, 2010) and followed
conventional methods (Borenstein, Hedges, Higgins, & Rothstein,
2009; Lipsey & Wilson, 2001). For this analysis, we included the
results of Studies 2a, 2b, 3, 4, 5, 5b, 6, 6b, 6c, 7 (peripheral
processing only), 8, and 8b (liberal message only). We computed
effect sizes (political ideology on endorsement) by converting
t-values within condition to Cohen’s r, within each temporal focus
condition. Effect sizes and variances were submitted to a mixed-
effects metaregression testing whether temporal focus modified
the effect of political ideology on endorsement.
First, we conducted a random-effects model examining the
overall effect of political orientation on endorsement, without
considering the temporal focus condition variable. The overall
effect was negative, r⫽⫺.13, SE ⫽.03, z⫽⫺3.87, p⬍.0001,
95% CI [⫺.197, ⫺.064], indicating that the more conservative
respondents were, the less they tended to support the (liberal)
messages across studies. However, there was significant variability
in individual effect sizes, Q(23) ⫽186.85, p⬍.0001. Therefore,
we proceeded to test whether the temporal focus condition mod-
erator could account for this variability.
A mixed-effects model indicated that this main effect was
modified by condition, b⫽.23, SE ⫽.05, z⫽4.64, p⬍.0001,
95% CI [.131, .323]. Consistent with our predictions, conservatism
predicted lower endorsement in the future conditions, r⫽⫺.24,
SE ⫽.02, z⫽⫺9.73, p⬍.0001, 95% CI [⫺.291, ⫺.193], but
conservatism had no significant relation to endorsement in the past
conditions, r⫽⫺.02, SE ⫽.04, z⫽⫺0.36, p⫽.72, 95% CI
[⫺.100, .069]. In summary, a temporal focus on the past largely
bridged the political divide in endorsement of liberal messages that
was prevalent in the future focus conditions.
General Discussion
Nine studies and an internal meta-analysis (including four omit-
ted studies) show the importance of past-focused temporal com-
parisons in guiding conservatives’ opinion about political ideas
and programs. A first study demonstrated that conservatives score
higher on a reliable measure of dispositional nostalgia (Holbrook,
1993) than liberals; a measure that is particularly focused on
nostalgia for aspects of society such as consumer goods, business,
and the general welfare of people. Next, six studies showed that
conservatives’ opposition to a host of liberal political messages
and ideas is strongly reduced if those messages adopt a temporal
focus on the past. As would be expected from public political
discourse, conservatives strongly opposed a liberal message that
appealed for greater leniency in criminal justice (Studies 2a and
2b), a liberal message calling for gun rights restrictions (Study 3),
a liberal message to ease restrictions on immigration (Study 4), a
liberal message calling for more social diversity (Study 5), and a
liberal message calling for more social justice (Study 6). However,
conservative opposition to these liberal messages was greatly
reduced, and in some cases eliminated, if these liberal appeals
were focused on the past. We found this effect across three
countries, including the United Kingdom (Study 2b), Germany
(Study 4), and the United States (all other studies). Study 5 and 6
also showed the mediating role of state- and trait-differences
nostalgia. Specifically, Study 5 showed that conservatives are
particularly likely to experience strong state feelings of nostalgia
when reading a vintage comic expressing the value of diversity,
which in turn predicted their endorsement of that diversity mes-
sage. Study 6 focused on trait nostalgia and found that feelings of
nostalgia explain why conservatives respond particularly well to a
past-focused political message calling for greater social justice. On
the basis of these mediation studies, we predicted and found in
Study 7 that conservatives’ change in attitudes in the past-focused
condition is based on superficial (peripheral) processing of infor-
mation, as the effect of temporal focus did not emerge when
participants were encouraged to process information deeply (cen-
trally). Finally, Study 8 showed that the effect of temporal focus is
asymmetrical; as predicted, a past focus increases conservatives’
support for liberal political messages, but a future focus does not
increase liberals’ support for conservative messages. Cumulative
evidence in a meta-analysis supported the notion that the political
divide in endorsement of liberal messages in our studies was
completely eliminated by presenting such messages with a tempo-
ral focus on the past.
Implications
Our findings strongly support the theory of ideology as moti-
vated social cognition, meaning that differences between liberals
and conservatives on policy (i.e., discussions on the direction the
country should head toward) build on underlying differences in
psychological functioning—such as resistance to change and an
intuitive preference to defend the status quo (Jost et al., 2003,
2004; Jost & Hunyady, 2005). Given the ease by which conser-
vatives’ opposition to liberal messages on issues as diverse as gun
ownership, immigration, and leniency in criminal justice could be
reduced by repackaging those messages with a temporal focus on
the past suggests the primacy of differences in psychological
functioning over pragmatic differences. In other words, a substan-
tial part of the disagreement between liberals and conservatives on
policy differences may not reflect differences in pragmatic pref-
erences (e.g., Is it better to have stricter or more lenient gun laws?)
but differences in rhetorical or stylistic preferences (e.g., Does
change in gun laws constitute a return to the past or a move into a
new direction?). Conservatives prefer policies that appear to be
grounded in the past and this strongly influences their support for
political ideas. In contrast, manipulations of temporal focus, even
when focused on a positive future, left liberals’ attitudes largely
unaffected.
This finding may also offer important applied implications for
liberal partisans, by presenting an opportunity to more effectively
communicate liberal political issues across party lines. A wealth of
research in political communication research has demonstrated
that framing can be particularly effective in improving support for
political messages (Kahneman & Tversky, 1984; McDermott,
2004; Quattrone & Tversky, 1988; Tversky & Kahneman, 1981;
for overviews, see Druckman, 2001; Nelson, Oxley, & Clawson,
1997; Pan & Kosicki, 1993; Scheufele, 1999). Our findings on the
effect of temporal focus manipulations fit in this literature and
offer an important opportunity to facilitate pragmatic cooperation
across party lines and overcome political gridlock (McCarty,
Poole, & Rosenthal, 2006; Nivola & Brady, 2008). Of course, we
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
15
POLITICAL PREFERENCE AND TEMPORAL FOCUS
found these effects on a four-item measure focusing on immediate
feelings toward the message; and, thus, future research is needed to
test whether these effects also produce meaningful differences in
the long run. Furthermore, across studies we manipulated the
temporal focus of the message without manipulating the ideolog-
ical background of the messenger (i.e., the politician used as
stimulus). Future research should test whether the effects of tem-
poral focus still hold if that background is known. However, in
other research the political identity of the messenger did not
significantly influence the effect of temporal focus manipulations
on conservatives’ attitudes toward climate change (Baldwin &
Lammers, 2016).
Relationship to Nostalgia Research
On the one hand, our findings fit with literature on nostalgia and
its effect on individual psychological functioning. Specifically,
various findings attest to the positive effects of nostalgia for
individuals’ happiness (Hepper, Ritchie, Sedikides, & Wildschut,
2012; Routledge et al., 2011; Sedikides et al., 2008; Wildschut et
al., 2006; Zhou et al., 2008), while other findings similarly suggest
that a conservative ideology can reduce unhappiness by serving as
an ideological buffer against negative feelings because of witness-
ing inequality (Napier & Jost, 2008) or by being associated with
greater mental health for other reasons (Schlenker, Chambers, &
Le, 2012). On the other hand, our findings appear inconsistent with
earlier work on the effects of nostalgia on group functioning.
Specifically, various findings show that collective nostalgia leads
to a stronger tendency to support the ingroup and a weaker ten-
dency to be open to the perspective of outgroups (Baldwin et al.,
2017; Wildschut et al., 2014). Yet our findings demonstrate that
nostalgia can also lead people to support opinions that run against
those of their ingroup—specifically, by leading conservatives to
support liberal positions. This suggests that feelings of nostalgia
may be a prime mover that can lead people to adopt any idea or
behavior— even those that go against their own ideology, the
interests of their group, or their personal interests, such as their
health or well-being. Future research may test this in more detail.
One factor that we did not investigate systematically is the
degree of temporal distance involved in our past-focused temporal
manipulations. In other words, what is sufficiently “past” to appeal
to conservatives? It has been argued that one defining aspect of
feelings of collective nostalgia is that they are unrestrained in time
and place. Specifically, where feelings of individual nostalgia
typically refer to specific periods of one’s own personal past (e.g.,
place of birth or childhood), feelings of collective nostalgia can
apply to any period that connects to a shared social identity, even
including times that people did not experience themselves, such as
a glorious period in the nation’s past (Baldwin et al., 2017;
Havlena & Holak, 1996; Wildschut et al., 2014). Our findings
reflect this; in some cases we invoked images of a relatively recent
past of a few decades ago (e.g., Study 5), while in others we
pointed to a past situation of several centuries ago (Study 6).
Furthermore, in some studies we invoked vague impressions of
past situations without specifying the year of those situations (e.g.,
Study 2), while in other studies we specified this to the exact year
(Study 6). Irrespective of such differences in temporal distance or
the degree of specificity in formulating that distance, conservatives
showed a relatively consistent preference for ideas that focused on
the past, over those that focused on the future. This suggests that
any situation in the past appeals to conservatives, as long as it
connects to a shared social identity.
Strengths and Limitations
A primary strength of our findings is that we replicated our
effect in three different countries—the United States, Germany,
and Great Britain. An interesting open question is whether the
effect also replicates outside a Western cultural setting. One reason
why the effect may not replicate outside the West is that conser-
vatives in other countries may not share the same preference for
the past, because of differences in those nations’ historical trajec-
tories. For example, it has been noted that in China feelings of
nostalgia for the (Maoist) past are more common among those on
the left, while those on the right want to move away from the past,
toward openness and democracy (Pan & Xu, 2017). In other
countries that in the past decades have seen rapid industrialization,
such as in South-East Asia and South-America, feelings of nos-
talgia may be more strongly tied to an ideology to return or
preserve rural or preindustrial life (Bajpai & Bonura, 2013; Chris-
tie, 2001; Parekh, 2003).
Another strength of our findings is the fact that we provided
conceptual replications of the same effects across many different
political issues, ranging from immigration and diversity to gun
control and criminal justice. By sampling a wide variety of polit-
ical issues that span across the width of the political debate, we
offer stronger evidence of a general effect of temporal focus
communication. This point is related to our choice to use the word
communication and not use the word framing. Classic framing
effects in psychology are found even though the objective content
of a stimulus does not change, but only the context or the way that
stimulus is presented. We did follow this in Study 5, in which
participants saw exactly the same Superman comic and we only
changed whether people believed it to be an original comic of the
50s or a modern adaptation. However, it could be argued that the
effects in our other studies are not framing effects—strictly speak-
ing— because by communicating to participants that a policy was
either part of a desirable past or not (but part of a desirable future),
the objective content of the policy may also change. This is most
clearly the case in Study 3. A past situation without assault rifles
and a future situation that is similarly without assault rifles differ
in more than their temporal location. Specifically, in the former
case assault rifles were simply not yet invented, while in the latter
case they are—and, thus, need to be actively banned by govern-
ment. Our other studies, however, do not suffer from this prob-
lem— or at least much less—and ensured to specify the policy in
great detail, to reduce such effects.
One limitation worth considering is that we relied in these
studies on relatively small samples, mostly drawn from Amazon
Mechanical Turk. We welcome research that can replicate these
findings in a large-scale nationally representative sample. None-
theless, we note that Mechanical Turk samples have been shown to
offer high-quality data that has been shown to mimic results of
more expensive representative samples (Berinsky, Huber, & Lenz,
2012; Buhrmester, Kwang, & Gosling, 2011; Clifford, Jewell, &
Waggoner, 2015). Also, many characteristics of Mechanical Turk
samples are more likely to work against our hypothesis, rather than
in favor of it. In particular, younger, more highly educated, and
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
16 LAMMERS AND BALDWIN
more Internet-savvy citizens are probably less likely to experience
strong nostalgia than older, less well-educated, or less Internet-
savvy citizens and, thus, the positive response to our manipulation
may underestimate the true effect in a national sample.
Another limitation to our findings is that the samples across our
studies were somewhat skewed, with liberals being overrepre-
sented. Across studies, the proportion of conservatives ranged
between 23.0% (Study 2a) and 29.2% (Study 8), while the pro-
portion of liberals ranged between 55.5% (Study 6) and 65.1%
(Study 2b). As a result, the estimation of the effect-size is less-
precise for conservatives than for liberals. It is important to note,
however, that this does not undermine our effect. Across studies
we followed up our overall test of the significance of the interac-
tion with the Johnson and Neyman (1936) method of identifying
regions of significance, which is sensitive to number of observa-
tions. In other words, because we sampled fewer conservatives
than liberals, conservative responses suffered from a larger stan-
dard error of each group and, thus, reduced statistical power to
detect an effect of condition, compared with liberal responses.
Given that we predicted significant differences between conditions
for conservatives (that were undersampled) but weaker differences
for liberals (that were oversampled), any lack of effect among
liberals is unlikely to be because of a lack of power. In other
words, we found our effects despite undersampling conservatives.
Suggestions for Future Research
Although we found strong and consistent support for the idea
that conservatives’ support for political ideas depends on the
temporal focus of these ideas, we found little to no support of a
similar effect among liberals. Specifically, liberals responded
equally negatively to past-focused and future-focused conservative
messages (Study 8) and responded equally positively to past-
focused and future-focused liberal messages (Studies 2–5). The
only exception to this was Study 6, where liberals also responded
with reduced endorsement of a past-focused message about social
justice, compared with their response to a future-focused message.
However, across the remaining studies we found no support for a
significant effect of temporal focus on liberals.
This general finding supports the asymmetry discussed in the
beginning of the article, that although the past has an intrinsic
value for conservatives, the future does not provide intrinsic value
for liberals, and neither is the past intrinsically aversive to them.
Rather, liberals are focused on the pragmatic implications of
policy—for example, whether the policy can help disadvantaged
groups. This finding also supports the hypothesis that liberals
process information more centrally and less heuristically than
conservatives (for overviews, see Jost, 2017; Jost et al., 2003; Jost,
Sterling, & Stern, 2017), which is also supported by the findings of
Study 7.
At the same time, we note that the lack of an effect among
liberals in Study 8 may also be because of the fact that a temporal
focus on the future is less likely to be noticed than a focus on the
past. Whereas calling attention to the past when describing polit-
ical plans and ideas is a rhetorical device that captures attention,
calling attention to the future follows the default conversational
norm. That is, politicians by default defend their political ideas by
pointing to their positive effects for tomorrow. It could be that
future-focused temporal manipulations that are more rhetorically
salient do produce an effect. For example, it may be that liberals
respond better to political plans that are not only future-focused,
but also appeal to hope and idealism. Future research may test this.
Finally, it is important to note that the lack of effects among
liberals is unlikely to be driven by range of instrument effects;
average responses centered around the mean on our scale across all
studies and we did not observe any ceiling or floor effects.
One final question for future research is the degree to which
emotions—in particular feelings of nostalgia—are critical in driv-
ing the current effects. Blanchar and Eidelman (2013) found that
participants (irrespective of their political ideology) judge social
arrangements to be more legitimate if they believe them to be
older. They explained this effect by suggesting that participants
use a rule of thumb that social arrangements that have stood the
test of time must have at least some legitimacy. Our findings
connect to theirs, in the sense that we also find that arrangements
that go back in time are considered more positive (at least by
conservatives), but we explain our findings in more affective
terms, by drawing on differences in nostalgic emotions. In support
of that, we found that state (Study 5) and trait (Study 6) differences
in nostalgia explained why conservatives responded better to past-
focused messages. Conversely, Blanchar and Eidelman (2013)
found effects of longevity even if there was no opportunity to feel
nostalgic. Specifically, in one study American (as well as Indian)
participants indicated that they considered the Indian caste system
to be more legitimate if they thought it to be older. Given that the
Indian caste system is irrelevant to Americans’ shared social
identity, it is hard to explain this finding in terms of collective
nostalgia (Wildschut et al., 2014). In summary, together these
findings suggest that the past can serve as a heuristic that confers
legitimization on political ideas and social arrangements, because
of both cognitive and affective reasons. In any case, it would be
interesting to test whether the effects observed by Blanchar and
Eidelman (2013) are moderated by participants’ political ideology.
Conclusion
Even great nations can feel the occasional pang of nostalgia for
the past. Rather than boldly moving forward, sometimes citizens
desire to return to the old days and restore what was. In a poignant
example of such a phenomenon, Donald Trump recently rode this
wave of nostalgia to promote a political program that seeks to
Make America Great Again. The nostalgic desire to move back-
ward and restore a past is, however, not necessarily tied to isola-
tionist, xenophobic, sexist, racist, or even regressive political
ideas, but can be flexibly used to defend many political ideas, even
those considered to be liberal. Appealing to nostalgia can lead
people to support social diversity, be open to refugees, advocate
leniency in criminal justice, or back calls for increased gun control.
This insight not only provides a better understanding of the psy-
chological processes underlying political preference, but also
opens the possibility for political change.
References
Bach, J. P. (2002). The taste remains: Consumption, (n)ostalgia, and the
production of East Germany. Public Culture, 14, 545–556. http://dx.doi
.org/10.1215/08992363-14-3-545
Bajpai, R., & Bonura, C. (2013). South Asian and Southeast Asian ideol-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
17
POLITICAL PREFERENCE AND TEMPORAL FOCUS
ogies. In M. Freeden & M. Stears (Eds.), The Oxford handbook of
political ideologies (pp. 1–24). New York, NY: Oxford University Press.
http://dx.doi.org/10.1093/oxfordhb/9780199585977.013.0032
Baker, S. M., & Kennedy, P. F. (1994). Death by nostalgia: A diagnosis of
context-specific cases. Advances in Consumer Research Association for
Consumer Research, 21, 169 –174.
Baldwin, M., Biernat, M., & Landau, M. J. (2015). Remembering the real
me: Nostalgia offers a window to the intrinsic self. Journal of Person-
ality and Social Psychology, 108, 128 –147. http://dx.doi.org/10.1037/
a0038033
Baldwin, M., & Lammers, J. (2016). Past-focused environmental comparisons
promote proenvironmental outcomes for conservatives. Proceedings of the
National Academy of Sciences of the United States of America, 113, 14953–
14957. http://dx.doi.org/10.1073/pnas.1610834113
Baldwin, M., White, M. H., & Sullivan, D. (2017). Nostalgia for America’s
past can buffer collective guilt (Working paper). University of Cologne,
Cologne, Germany. http://dx.doi.org/10.1002/ejsp.2348
Ball, T., & Bellamy, R. (2003). The Cambridge history of twentieth-
century political thought (Vol. 4). New York, NY: Cambridge Univer-
sity Press. http://dx.doi.org/10.1017/CHOL9780521563543
Ball, T., & Dagger, R. (1999). Political ideology and the democratic ideal.
New York, NY: Cambridge University Press.
Beckett, K., & Sasson, T. (2003). The politics of injustice: Crime and
punishment in America. Thousand Oaks, CA: Sage.
Benoit, W. L. (2001). Framing through temporal metaphor: The “bridges”
of Bob Dole and Bill Clinton in their 1996 acceptance addresses.
Communication Studies, 52, 70 – 84. http://dx.doi.org/10.1080/
10510970109388541
Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012). Evaluating online
labor markets for experimental research: Amazon.com’s Mechanical
Turk. Political Analysis, 20, 351–368. http://dx.doi.org/10.1093/pan/
mpr057
Blanchar, J. C., & Eidelman, S. (2013). Perceived system longevity in-
creases system justification and the legitimacy of inequality. European
Journal of Social Psychology, 43, 238 –245. http://dx.doi.org/10.1002/
ejsp.1960
Bobbio, N. (1996). Left and right. Cambridge, England: Polity Press.
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009).
Introduction to meta-analysis. Chichester, United Kingdom: Wiley, Ltd.
http://dx.doi.org/10.1002/9780470743386
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s Mechan-
ical Turk: A new source of inexpensive, yet high-quality data?
Perspectives on Psychological Science, 6, 3–5. http://dx.doi.org/10
.1177/1745691610393980
Burke, E. (1993). Reflections on the revolution in France. Oxford, Eng-
land: Oxford Press. (Original published in 1790)
Campbell, T. H., & Kay, A. C. (2014). Solution aversion: On the relation
between ideology and motivated disbelief. Journal of Personality and
Social Psychology, 107, 809 – 824. http://dx.doi.org/10.1037/a0037963
Chaiken, S. (1980). Heuristic versus systematic information processing and
the use of source versus message cues in persuasion. Journal of Per-
sonality and Social Psychology, 39, 752–766. http://dx.doi.org/10.1037/
0022-3514.39.5.752
Chirumbolo, A. (2002). The relationship between need for cognitive clo-
sure and political orientation: The mediating role of authoritarianism.
Personality and Individual Differences, 32, 603– 610. http://dx.doi.org/
10.1016/S0191-8869(01)00062-9
Christie, C. J. (2001). Ideology and revolution in Southeast Asia, 1900 –
1980: Political ideas of the anti-colonial era. London, England: Rout-
ledge.
Clifford, S., Jewell, R. M., & Waggoner, P. D. (2015). Are samples drawn
from Mechanical Turk valid for research on political ideology? Research
& Politics. Advance online publication. http://dx.doi.org/10.1177/
2053168015622072
Davis, F. (1979). Yearning for yesterday: A sociology of nostalgia. New
York, NY: Free Press.
Druckman, J. N. (2001). The implications of framing effects for citizen
competence. Political Behavior, 23, 225–256. http://dx.doi.org/10.1023/
A:1015006907312
Eccleshall, R. (1990). English conservatism since the restoration: An
introduction and anthology. Abingdon-on-Thames, United Kingdom:
Routledge. http://dx.doi.org/10.4324/9780203205181
Eidelman, S., Crandall, C. S., & Pattershall, J. (2009). The existence bias.
Journal of Personality and Social Psychology, 97, 765–775. http://dx
.doi.org/10.1037/a0017058
Fay, D., & Frese, M. (2000). Conservatives’ approach to work: Less
prepared for future work demands? Journal of Applied Social Psychol-
ogy, 30, 171–195. http://dx.doi.org/10.1111/j.1559-1816.2000
.tb02310.x
Feygina, I., Jost, J. T., & Goldsmith, R. E. (2010). System justification, the
denial of global warming, and the possibility of “system-sanctioned
change”. Personality and Social Psychology Bulletin, 36, 326 –338.
http://dx.doi.org/10.1177/0146167209351435
Flanagan, T. J., & Longmire, D. R. (1996). Americans view crime and
justice: A national public opinion survey. London, England: Sage Pub-
lishing. http://dx.doi.org/10.4135/9781483326900
Geddes, A., & Scholten, P. (2016). The politics of migration and immi-
gration in Europe. London, England: Sage.
Glasgow, M. R., & Cartier, A. M. (1985). Conservatism, sensation-
seeking, and music preferences. Personality and Individual Differences,
6, 393–395.
Graham, J., Haidt, J., & Nosek, B. A. (2009). Liberals and conservatives
rely on different sets of moral foundations. Journal of Personality and
Social Psychology, 96, 1029 –1046. http://dx.doi.org/10.1037/a0015141
Hamaker, E. L., Nesselroade, J. R., & Molenaar, P. C. (2007). The
integrated trait–state model. Journal of Research in Personality, 41,
295–315. http://dx.doi.org/10.1016/j.jrp.2006.04.003
Havlena, W. J., & Holak, S. L. (1996). Exploring nostalgia imagery
through the use of consumer collages. Advances in Consumer Research
Association for Consumer Research, 23, 35– 42.
Hayek, F. A. (1944). The road to serfdom. London, England: Routledge.
Hayek, F. A. (1982). Law, legislation and liberty (Vol. 2). London,
England: Routledge.
Hayes, A. F. (2013). Introduction to mediation, moderation, and condi-
tional process analysis: A regression-based approach. New York, NY:
Guilford Press.
Hepper, E. G., Ritchie, T. D., Sedikides, C., & Wildschut, T. (2012).
Odyssey’s end: Lay conceptions of nostalgia reflect its original Homeric
meaning. Emotion, 12, 102–119. http://dx.doi.org/10.1037/a0025167
Hirsh, J. B., Walberg, M. D., & Peterson, J. B. (2013). Spiritual liberals and
religious conservatives. Social Psychological & Personality Science, 4,
14 –20. http://dx.doi.org/10.1177/1948550612444138
Holbrook, M. B. (1993). Nostalgia and consumption preferences: Some
emerging patterns of consumer tastes. Journal of Consumer Research,
20, 245–256. http://dx.doi.org/10.1086/209346
Huntington, S. P. (1957). Conservatism as an ideology. The American
Political Science Review, 51, 454 – 473. http://dx.doi.org/10.2307/
1952202
Institute For American Democracy. (1949). Untitled book cover featuring
an image of Superman. Retrieved from https://www.hakes.com/Auction/
ItemDetail/45031/SUPERMAN-BOOK-COVER
Johnson, P. O., & Neyman, J. (1936). Tests of certain linear hypotheses and
their applications to some educational problems. Statistical Research
Memoirs, 1, 57–93.
Jost, J. T. (2006). The end of the end of ideology. American Psychologist,
61, 651– 670. http://dx.doi.org/10.1037/0003-066X.61.7.651
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
18 LAMMERS AND BALDWIN
Jost, J. T. (2017). Ideological asymmetries and the essence of political
psychology. Political Psychology, 38, 167–208. http://dx.doi.org/10
.1111/pops.12407
Jost, J. T., & Banaji, M. R. (1994). The role of stereotyping in system-
justification and the production of false consciousness. British Journal of
Social Psychology, 33, 1–27. http://dx.doi.org/10.1111/j.2044-8309
.1994.tb01008.x
Jost, J. T., Banaji, M. R., & Nosek, B. A. (2004). A decade of system
justification theory: Accumulated evidence of conscious and uncon-
scious bolstering of the status quo. Political Psychology, 25, 881–919.
http://dx.doi.org/10.1111/j.1467-9221.2004.00402.x
Jost, J. T., Federico, C. M., & Napier, J. L. (2009). Political ideology: Its
structure, functions, and elective affinities. Annual Review of Psychol-
ogy, 60, 307–337. http://dx.doi.org/10.1146/annurev.psych.60.110707
.163600
Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, F. J. (2003). Political
conservatism as motivated social cognition. Psychological Bulletin, 129,
339 –375. http://dx.doi.org/10.1037/0033-2909.129.3.339
Jost, J. T., & Hunyady, O. (2005). Antecedents and consequences of
system-justifying ideologies. Current Directions in Psychological Sci-
ence, 14, 260 –265. http://dx.doi.org/10.1111/j.0963-7214.2005.00377.x
Jost, J. T., & Kay, A. C. (2010). Social justice: History, theory, and
research. In S. T. Fiske, D. Gilbert, & G. Lindzey (Eds.), Handbook of
social psychology (5th ed., Vol. 2, pp. 1122–1165). Hoboken, NJ: Wiley.
http://dx.doi.org/10.1002/9780470561119.socpsy002030
Jost, J. T., & Krochik, M. (2014). Ideological differences in epistemic
motivation: Implications for attitude structure, depth of information
processing, susceptibility to persuasion, and stereotyping. Advances in
Motivation Science, 1, 181–231. http://dx.doi.org/10.1016/bs.adms.2014
.08.005
Jost, J. T., Napier, J. L., Thorisdottir, H., Gosling, S. D., Palfai, T. P., &
Ostafin, B. (2007). Are needs to manage uncertainty and threat associ-
ated with political conservatism or ideological extremity? Personality
and Social Psychology Bulletin, 33, 989 –1007. http://dx.doi.org/10
.1177/0146167207301028
Jost, J. T., Nosek, B. A., & Gosling, S. D. (2008). Ideology: Its resurgence
in social, personality, and political psychology. Perspectives on Psycho-
logical Science, 3, 126 –136.
Jost, J. T., Sterling, J., & Stern, C. (2017). Getting closure on conservatism,
or the politics of epistemic and existential motivation. In C. Kopetz & A.
Fishbach (Eds.), The motivation-cognition interface; From the lab to the
real world: A Festschrift in honor of Arie W. Kruglanski. New York,
NY: Psychology Press.
Kahneman, D., & Tversky, A. (1984). Choices, values, and frames. Amer-
ican Psychologist, 39, 341–350. http://dx.doi.org/10.1037/0003-066X
.39.4.341
Kemmelmeier, M. (1997). Need for closure and political orientation among
German university students. The Journal of Social Psychology, 137,
787–789. http://dx.doi.org/10.1080/00224549709595501
Kirk, R. (1953). The conservative mind: From Burke to Eliot. New York,
NY: Avon Books.
Lammers, J., Koch, A., Conway, P., & Brandt, M. J. (2016). The political
domain appears simpler to the politically extreme than to political
moderates. Social Psychological & Personality Science, 8, 612– 622.
http://dx.doi.org/10.1177/1948550616678456
Lammers, J., & Proulx, T. (2013). Writing autobiographical narratives
increases political conservatism. Journal of Experimental Social Psy-
chology, 49, 684 – 691. http://dx.doi.org/10.1016/j.jesp.2013.03.008
Lipsey, M. W., & Wilson, D. B. (2001). Applied social research methods
series: Volume 49.Practical meta-analysis. Thousand Oaks, CA: Sage.
Maio, G. R., & Haddock, G. (2015). The psychology of attitudes and
attitude change (2nd ed.). London, England: Sage.
Maio, G. R., & Olson, J. M. (1995). Relations between values, attitudes,
and behavioral intentions: The moderating role of attitude function.
Journal of Experimental Social Psychology, 31, 266 –285. http://dx.doi
.org/10.1006/jesp.1995.1013
Marchegiani, C., & Phau, I. (2010). Away from “unified nostalgia”:
Conceptual differences of personal and historical nostalgia appeals in
advertising. Journal of Promotion Management, 16, 80 –95. http://dx.doi
.org/10.1080/10496490903572991
McCarty, N., Poole, K. T., & Rosenthal, H. (2006). Polarized America:
The dance of political ideology and unequal riches. Cambridge, MA:
MIT Press.
McClelland, J. S. (2005). A history of western political thought. London,
UK: Routledge.
McDermott, R. (2004). Prospect theory in political science: Gains and
losses from the first decade. Political Psychology, 25, 289 –312. http://
dx.doi.org/10.1111/j.1467-9221.2004.00372.x
Mischel, W. (1968). Personality and assessment. New York, NY: Wiley.
Muller, J. Z. (1997). Conservatism: An anthology of social and political
thought from David Hume to the present. Princeton, NJ: Princeton
University Press.
Napier, J. L., & Jost, J. T. (2008). Why are conservatives happier than
liberals? Psychological Science, 19, 565–572. http://dx.doi.org/10.1111/
j.1467-9280.2008.02124.x
Nelson, T. E., Oxley, Z. M., & Clawson, R. A. (1997). Toward a psychol-
ogy of framing effects. Political Behavior, 19, 221–246. http://dx.doi
.org/10.1023/A:1024834831093
Nivola, P. S., & Brady, D. W. (Eds.). (2008). Red and blue nation?
Consequences and correction of America’s polarized politics (Vol. 2).
Washington, DC: Brookings Institution Press.
Pan, J., & Xu, Y. (in press). China’s ideological spectrum. The Journal of
Politics. Retrieved from http://ssrn/abstract⫽2593377; http://dx.doi.org/
10.2139/ssrn.2593377
Pan, Z., & Kosicki, G. M. (1993). Framing analysis: An approach to news
discourse. Political Communication, 10, 55–75. http://dx.doi.org/10
.1080/10584609.1993.9962963
Parekh, B. C. (2002). Rethinking multiculturalism: Cultural diversity and
political theory. Cambridge, MA: Harvard University Press.
Parekh, B. C. (2003). Non-Western political thought. In T. Ball & R.
Bellamy (Eds.), The Cambridge history of twentieth century political
thought (pp. 551–578). Cambridge, England: Cambridge University
Press. http://dx.doi.org/10.1017/CHOL9780521563543.028
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of
persuasion. In R. E. Petty & J. T. Cacioppo (Eds.), Communication and
persuasion (pp. 1–24). New York, NY: Springer.
Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing mod-
erated mediation hypotheses: Theory, methods, and prescriptions. Mul-
tivariate Behavioral Research, 42, 185–227. http://dx.doi.org/10.1080/
00273170701341316
Price, R. (2005). A concise history of France. New York, NY: Cambridge
University Press.
Quattrone, G. A., & Tversky, A. (1988). Contrasting rational and psycho-
logical analyses of political choice. The American Political Science
Review, 82, 719 –736. http://dx.doi.org/10.2307/1962487
Rawls, J. (1971). A theory of justice. Cambridge, MA: Harvard University
Press.
Robinson, M. D., Cassidy, D. M., Boyd, R. L., & Fetterman, A. K. (2015).
The politics of time: Conservatives differentially reference the past and
liberals differentially reference the future. Journal of Applied Social
Psychology, 45, 391–399. http://dx.doi.org/10.1111/jasp.12306
Rock, M., & Janoff-Bulman, R. (2010). Where do we draw our lines?
Politics, rigidity, and the role of self-regulation. Social Psychological
& Personality Science, 1, 26 –33. http://dx.doi.org/10.1177/19485
50609347386
Rothmund, T., Becker, J. C., & Jost, J. T. (2016). The psychology of social
justice in political thought and action. In M. Schmitt & C. Sabbagh
(Eds.), Handbook of social justice theory and research (pp. 275–291).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
19
POLITICAL PREFERENCE AND TEMPORAL FOCUS
New York, NY: Springer. http://dx.doi.org/10.1007/978-1-4939-3216-
0_15
Routledge, C., Arndt, J., Wildschut, T., Sedikides, C., Hart, C. M., Juhl,
J.,...Schlotz, W. (2011). The past makes the present meaningful:
Nostalgia as an existential resource. Journal of Personality and Social
Psychology, 101, 638 – 652. http://dx.doi.org/10.1037/a0024292
Sargent, M. J. (2004). Less thought, more punishment: Need for cognition
predicts support for punitive responses to crime. Personality and Social
Psychology Bulletin, 30, 1485–1493. http://dx.doi.org/10.1177/
0146167204264481
Scheufele, D. A. (1999). Framing as a theory of media effects. Journal of
Communication, 49, 103–122. http://dx.doi.org/10.1111/j.1460-2466
.1999.tb02784.x
Schlenker, B. R., Chambers, J. R., & Le, B. M. (2012). Conservatives are
happier than liberals, but why? Political ideology, personality, and life
satisfaction. Journal of Research in Personality, 46, 127–146. http://dx
.doi.org/10.1016/j.jrp.2011.12.009
Schlesinger, A. (1955). The new conservatism: Politics of nostalgia. The
Reporter, 16, 9 –11.
Schneider, G. (2009). The conservative century: From reaction to revolu-
tion. Washington, DC: Rowman and Littlefield.
Schwarz, N., & Clore, G. L. (1983). How do I feel about it? The infor-
mative function of affective states. In K. Fielder & J. Forgas (Eds.),
Affect, cognition and social behavior (pp. 44 – 62). Toronto, Ontario,
Canada: Hogrefe International.
Schwarz, N., & Clore, G. L. (2007). Feelings and phenomenal experiences.
In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Hand-
book of basic principles (2nd ed., pp. 385– 407). New York, NY:
Guilford Press.
Scruton, R. (1980). The meaning of conservatism. London, England: Mac-
millan.
Sedikides, C., Wildschut, T., Arndt, J., & Routledge, C. (2008). Nostalgia:
Past, present, and future. Current Directions in Psychological Science,
17, 304 –307. http://dx.doi.org/10.1111/j.1467-8721.2008.00595.x
Shalhope, R. E. (1982). The ideological origins of the Second Amendment.
The Journal of American History, 69, 599 – 614. http://dx.doi.org/10
.2307/1903139
Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2012). A 21 Word
Solution. http://dx.doi.org/10.2139/ssrn.2160588
Thompson, E. P., Roman, R. J., Moskowitz, G. B., Chaiken, S., & Bargh,
J. A. (1994). Accuracy motivation attenuates covert priming: The sys-
tematic reprocessing of social information. Journal of Personality and
Social Psychology, 66, 474 – 489. http://dx.doi.org/10.1037/0022-3514
.66.3.474
Thompson, M. M., Naccarato, M. E., Parker, K. C. H., & Moskowitz, G.
(2001). The Personal Need for Structure (PNS) and Personal Fear of
Invalidity (PFI) scales: Historical perspectives, present applications and
future directions. In G. Moskowitz (Ed.), Cognitive social psychology:
The Princeton symposium on the legacy and future of social cognition
(pp. 19 –39). Mahwah, NJ: Erlbaum.
Tversky, A., & Kahneman, D. (1981). The framing of decisions and the
psychology of choice. Science, 211, 453– 458. http://dx.doi.org/10.1126/
science.7455683
Vess, M., Arndt, J., Routledge, C., Sedikides, C., & Wildschut, T. (2012).
Nostalgia as a resource for the self. Self and Identity, 11, 273–284.
Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor
package. Journal of Statistical Software, 36, 1– 48. http://dx.doi.org/10
.18637/jss.v036.i03
Von Beyme, K. (1985). Political parties in Western democracies. Alder-
shot, United Kingdom: Gower Publishing Company, Limited.
Watt, S. E., Maio, G. R., Haddock, G., & Johnson, B. T. (2009). Attitude
functions in persuasion: Matching, involvement, self-affirmation, and
hierarchy. In W. Crano & R. Prislin (Eds.), Attitudes and persuasion (pp.
189 –213). New York, NY: Psychology Press.
Webster, D. M., & Kruglanski, A. W. (1994). Individual differences in
need for cognitive closure. Journal of Personality and Social Psychol-
ogy, 67, 1049 –1062. http://dx.doi.org/10.1037/0022-3514.67.6.1049
Wildschut, T., Bruder, M., Robertson, S., van Tilburg, W. A. P., &
Sedikides, C. (2014). Collective nostalgia: A group-level emotion that
confers unique benefits on the group. Journal of Personality and Social
Psychology, 107, 844 – 863. http://dx.doi.org/10.1037/a0037760
Wildschut, T., Sedikides, C., Arndt, J., & Routledge, C. (2006). Nostalgia:
Content, triggers, functions. Journal of Personality and Social Psychol-
ogy, 91, 975–993. http://dx.doi.org/10.1037/0022-3514.91.5.975
Wilson, G. D. (1973). Development and evaluation of the C-Scale. In G. D.
Wilson (Ed.), The psychology of conservatism (pp. 49 – 69). London,
England: Academic Press.
Wilson, G. D., & Patterson, J. R. (1968). A new measure of conservatism.
British Journal of Social and Clinical Psychology, 7, 264 –269. http://
dx.doi.org/10.1111/j.2044-8260.1968.tb00568.x
Wilson, J. L. (2005). Nostalgia: Sanctuary of meaning. Lewisburg, PA:
Bucknell University Press.
Wolsko, C., Ariceaga, H., & Seiden, J. (2016). Red, white, and blue enough
to be green: Effects of moral framing on climate change attitudes and
conservation behaviors. Journal of Experimental Social Psychology, 65,
7–19. http://dx.doi.org/10.1016/j.jesp.2016.02.005
Zhou, X., Sedikides, C., Wildschut, T., & Gao, D.-G. (2008). Counteract-
ing loneliness: On the restorative function of nostalgia. Psychological
Science, 19, 1023–1029. http://dx.doi.org/10.1111/j.1467-9280.2008
.02194.x
(Appendices follow)
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
20 LAMMERS AND BALDWIN
Appendix 1
German Stimuli Used in Study 4
Past Focus
Einwanderer waren in der Vergangenheit stets eine Bereicherung
für die Wirtschaft und den Arbeitsmarkt und auch heute werden sie
von positiver Bedeutung sein. Der Zustrom ausländischer Menschen
ist etwas, das auch schon in der Vergangenheit stattgefunden hat und
ohne weitere Probleme für die deutsche Bevölkerung verlief. Frühere
Einwanderungswellen nach Deutschland haben gezeigt, dass die Ge-
sellschaft mit den Anforderungen umgehen kann und davon profitiert.
Durch frühere Einwanderungswellen konnte die deutsche Gesell-
schaft viel lernen, was man auf die heutige Situation anwenden kann.
Insbesondere die Aufnahme und Integration geflüchteter Menschen
basiert auf alten Werten, die seit jeher der deutschen Politik zugrunde
liegen. Die Zuwanderung syrischer geflüchteter Menschen stellt ein
Wiederaufleben einer Verbindung zwischen Deutschland und Syrien
dar, die bereits vor Christus Geburt (durch das römische Reich)
bestand.
Future Focus
Einwanderer werden in der Zukunft eine Bereicherung für die
deutsche Wirtschaft und somit von positiver Bedeutung sein. Der
Zustrom ausländischer Menschen ist ein neues Phänomen, das laut
Zukunftsprognosen ohne weitere Probleme für die deutsche Bev-
ölkerung verlaufen wird. Zukunftsprognosen bezüglich der Ein-
wanderung nach Deutschland zeigen, dass die Gesellschaft mit
den Anforderungen umgehen kann und davon profitiert. Durch
die momentane Einwanderung kann die deutsche Gesellschaft
viel lernen und an den kommenden Herausforderungen weiter
wachsen. Insbesondere die Aufnahme und Integration geflüch-
teter Menschen basiert auf einer Neubetonung unserer Werte,
die nun der deutschen Politik zugrunde liegen. Die Zuwander-
ung syrischer geflüchteter Menschen bietet die Chance, eine
Verbindung zwischen Deutschland und Syrien herzustellen, die
es bisher nicht gab.
Appendix 2
Stimuli Used in Study 8
Past Focus—Proleniency Message
Today’s justice system appears to be quite strict, but fortunately
this wasn’t always the case. In the old days, police were more
lenient. If you committed an offense in traffic or you had a drink
in public, the police would not immediately book you but instead
just give you a warning. I would like to go back to how things were
then, when police were more tolerant and were not so strict toward
offenders.
Future Focus—Proleniency Message
Today’s justice system appears to be quite strict, but fortunately
we can change this. Sometime in the near future I can see a shift
happening so that police will be more lenient. If you commit an
offense in traffic or you have a drink in public, the police will not
immediately book you but instead just give you a warning. I’d like
push for this kind of change, so that in the near future police will
be more tolerant and will not be so strict toward offenders.
Past Focus—Proseverity Message
Today’s justice system appears to be quite lenient, but fortu-
nately this wasn’t always the case. In the old days, police were
stricter. If you committed an offense in traffic or you had a drink
in public, the police would immediately book you and not tolerate
anything. I would like to go back to how things were then, when
police were not so tolerant and were stricter toward offenders.
Future Focus—Proseverity Message
Today’s justice system appears to be quite lenient, but fortu-
nately we can change this. Sometime in the near future I can see
a shift happening so that police will be stricter. If you commit an
offense in traffic or you have a drink in public, the police will
immediately book you and not tolerate anything. I’d like push for
this kind of change, so that in the near future police will not be so
tolerant and will be stricter toward offenders.
Received December 10, 2016
Revision received October 13, 2017
Accepted October 14, 2017 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
21
POLITICAL PREFERENCE AND TEMPORAL FOCUS
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Personality and Social Psychology
This content is subject to copyright. Terms and conditions apply.
