See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/221772271

The political left rolls with the good and the political right confronts the bad:
Connecting physiology and cognition to preferences

Article  in  Philosophical Transactions of The Royal Society B Biological Sciences · March 2012

DOI: 10.1098/rstb.2011.0268 · Source: PubMed

CITATIONS READS

81 191

6 authors, including:

Michael D Dodd Amanda Friesen

University of Nebraska at Lincoln Indiana University-Purdue University Indianapolis
98 PUBLICATIONS   1,604 CITATIONS    25 PUBLICATIONS   196 CITATIONS   

SEE PROFILE SEE PROFILE

Carly M. Jacobs Mike Gruszczynski

University of Nebraska at Lincoln Indiana University Bloomington
10 PUBLICATIONS   170 CITATIONS    14 PUBLICATIONS   125 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

In the Eye of the Beer Holder: The Effect of Alcohol, Focus, and Appearance vs. Humanizing Target Characteristics on the Objectifying Gaze in Men View project

Gender and Political Participation View project

All content following this page was uploaded by Michael D Dodd on 23 May 2014.

The user has requested enhancement of the downloaded file.

 Phil. Trans. R. Soc. B (2012) 367, 640–649
doi:10.1098/rstb.2011.0268

Research

The political left rolls with the good and the

political right confronts the bad: connecting
physiology and cognition to preferences
Michael D. Dodd1, Amanda Balzer2, Carly M. Jacobs2,
Michael W. Gruszczynski2, Kevin B. Smith2 and John R. Hibbing2, *
1
Department of Psychology, and 2Department of Political Science, University of Nebraska,
Lincoln, USA
We report evidence that individual-level variation in people’s physiological and attentional responses
to aversive and appetitive stimuli are correlated with broad political orientations. Specifically, we
find that greater orientation to aversive stimuli tends to be associated with right-of-centre and
greater orientation to appetitive (pleasing) stimuli with left-of-centre political inclinations. These
findings are consistent with recent evidence that political views are connected to physiological pre-
dispositions but are unique in incorporating findings on variation in directed attention that make it
possible to understand additional aspects of the link between the physiological and the political.
Keywords: physiology; cognition; politics

1. INTRODUCTION in political orientations within a sample of United

The most intense cultural conflicts tend to be disputes States research participants are instantiated in the pat-
over the proper way to structure and maintain mass- terns of each individual’s physiological and cognitive
scale social life. Accordingly, whether within or responses to emotionally laden stimuli.
across national boundaries, disagreements regarding A growing body of research finds that political
politics (and religion, as the other major force that orientations vary with an array of broader constructs
regulates mass-scale social life) are much more likely such as personality traits [4 – 7], moral foundations
to lead to acrimony and even violence than, say, dis- [8,9], core values [10 – 16], baseline neural structures
agreements over preferred personality traits or tastes [17], neural activation in response to unexpected
in art. Politics can affect the lives of others in a way stimuli [18], self-reported sensitivity to threat [19],
that personality and taste do not. It has been a flash- tendency to perceive threat in faces [20], physiological
point over the centuries and serious attempts at response to threat [21], sensitivity to disgust [22,23]
understanding cultural conflict must address the and possibly even genetics [24,25].
reasons for political differences. The focus of this previous research often is on
What is it that leads individuals—even in nearly responses to reasonably narrow categories of stimuli
identical social milieus—to hold such distinct, often and equally narrow political attitudes (e.g. does a
persistent, and potentially explosive political orien- stronger disgust response correlate with opposition to
tations? Traditional social science approaches have gay marriage?) and this approach is perfectly reason-
ignored the role of biology in these differences and able. Human emotion encompasses a wide array of
focused on variables directly relevant to political life. discrete affective states, including fear, anger, sadness
Scholars argued over the relative contributions of and happiness, and each of these affective states
parental socialization and pertinent adult experiences activates unique neurophysiological pathways and pol-
[1 – 3], but until recently were little concerned with itically relevant issue attitudes. To take one example,
whether political differences have biological markers. Neuberg et al. [26] detail the differences between
Consistent with the theme of this issue, we test the the self-protection and disease-avoidance systems,
possibility that differences relevant to cultural conflict with the former closely tied to threat responses and
are embedded in broad biological processes. In par- the latter to disgust. Each seems to engage different
ticular, we investigate the possibility that variations emotions, inferences and behavioural tendencies, and
there is obvious value in studying responses to
particular categories of stimuli seriatim.
* Author for correspondence ([email protected]). Still, even amidst these distinct pathways, downstream
Electronic supplementary material is available at http://dx.doi.org/ commonality is present. Responding to an aversive fea-
10.1098/rstb.2011.0268 or via http://rstb.royalsocietypublishing.org. ture of the environment, whether it is indicative of a
One contribution of 12 to a Theme Issue ‘The biology of cultural violation of order, purity or security, must ultimately
conflict’. work via a sympathetic nervous system that will prejudice

640 This journal is q 2012 The Royal Society

 Physiology, cognition, politics M. D. Dodd et al. 641

a broad category of action such as avoiding or approach- physiological and cognitive biases, according to the
ing [26]. Consequently, in addition to recognizing the approach-avoidant spectrum. For example, those
uniqueness of individual categories of response, many whose physiology responds strongly to violations of
researchers also have found useful a ‘biphasic’ model their preferences for protection, purity and order and
which holds that emotion is a product of varying acti- are known to devote high levels of attention to such
vation of two motivational systems: appetitive and violations, are likely to take steps in their personal
defensive [27,28]. Both of these systems ‘are evolution- lives to avoid situations in which they encounter viola-
arily old, shared across mammalian species and have tions of security, purity and order. In other words,
evolved to mediate the behaviours that sustain and pro- these individuals may be more likely to display the per-
tect life’ [29, p. 30]. Gray [30,31] has described these sonal values of tradition, conformity and security. On
two core systems as the behavioural inhibition system the other hand, those whose physiology responds
(BIS) and the behavioural activation system (BAS), strongly to stimuli portraying desirable situations and
with the BIS typically activated by aversive stimuli and experiences, and/or those who devote relatively high
the BAS activated by appetitive stimuli [32,33]. levels of attention to appetitive stimuli may be more
A conceptualization in which organisms are likely to subscribe to the personal values of hedonism,
expected to approach appetitive and avoid negative stimulation and self-direction [15].
stimuli is simplistic, but has the additional advantage Further, and more to the point of the present study,
of putting an equal emphasis on appetitive (that is, these individual-level physiological and cognitive vari-
positive or pleasant) events and exposures. Previous ations are likely also to be correlated with political
work on the connection of political orientations preferences. After all, political decisions affect the
and physiological responses has concentrated heavily kind of environment in which one exists. Our theoret-
and perhaps exclusively on aversive stimuli (an unex- ical assumption is that individuals will take steps to
pected, disorderly event, a threatening occurrence or shape their environment into one that is as consistent
a disgusting scene) and thus the political implications as possible with their pre-existing physiological and
of variations in response to appetitive situations have cognitive tendencies. They do so by adopting certain
not yet been tested. Doing so is important because personal values and by advocating certain political
variations in physiology connected with approach be- positions. From this perspective, it makes sense that
haviour could have just as much relevance to political people who are more attentive and responsive to hedon-
orientations as variations in physiology subsequent to ic stimuli would support tax dollars being spent on the
exposure to aversive stimuli. arts and national parks, just as it makes sense that
In sum, though individual emotions clearly have people who are more attentive and responsive to aver-
unique neural and physiological characteristics, a sive stimuli would advocate policies promoting moral
number of empirical studies recognize the broader purity and harsh treatment for norm violators.
biphasic organization of two core motivation systems This theory is supported by recent evidence that
(aversive and appetitive) that mediate the actions of individuals’ personal values correlate with their polit-
both the somatic (voluntary) and autonomic (involun- ical values [42] as well as evidence that people’s
tary) nervous systems, which are the proximate causes political values are related to their motivations in
of behaviour [27,34 – 38]. making moral decisions. Those who, in relative
This approach-avoidant behavioural dimension and terms, stress minimizing harm and maximizing equal-
the distinctive responses to appetitive and aversive ity tend to be left-of-centre in their political beliefs and
stimuli that undergird it are likely to be relevant to those who stress purity and authority tend to be right-
political orientations. On the whole, people are risk- of-centre [8,9]. The missing links are (i) evidence that
sensitive in that their physiological responses and cog- individuals to the right-of-centre do indeed respond
nitive attention are heightened by aversive stimuli, a more physiologically to aversive stimuli, while individ-
pattern that makes sense from an evolutionary point uals to the left-of-centre respond more physiologically,
of view [26]. Previous empirical research supports in relative terms, to appetitive stimuli and (ii) evidence
this line of thought and, on average, physiological that individuals to the right-of-centre pay more atten-
[39] and cognitive [40,41] responses to aversive tion to aversive than to appetitive stimuli while those to
images outstrip those to appetitive images. the left-of-centre pay more attention, in relative terms,
The operative phrase in the preceding passage, to appetitive than to aversive stimuli. In the research
however, is ‘on average’ and substantial individual- described in §2, we provide initial evidence on these
level variation probably exists around the averages. very points: first, that physiological responsiveness
Hair-trigger autonomic nervous systems generate varies predictably across the political spectrum and,
rapid and elevated physiological responses to aversive second, that patterns of attention also vary across the
stimuli and chronic sensitivity to violations of security, political spectrum in a fashion that complements the
purity and order may rivet attention on the pro- physiological results.
blematic aspects of the environment. Conversely,
heightened physiological response to appetitive stimuli
and a chronic craving of new experiential pleasures 2. STUDY 1: PHYSIOLOGICAL RESPONSE
may lead an individual to devote more attention to In the summer of 2007, 200 participants were brought
appealing aspects of the environment. Whatever the to a computer laboratory in Lincoln, Nebraska, USA,
source of these biological and psychological predispo- to complete a survey soliciting their political, personal-
sitions, people may accordingly self-select, often ity and demographic information subsequent to their
subconsciously, into situations likely to match their having been contacted by phone at random by a
Phil. Trans. R. Soc. B (2012)
642 M. D. Dodd et al. Physiology, cognition, politics

professional survey organization. Though in no way a using a pair of AgjAgCl electrodes and standard
representative sample, this group has the advantage psychophysiological equipment. Since eccrine glands
of not being restricted to college undergraduates release moisture as part of sympathetic nervous
and, relatedly, having reasonably representative demo- system activation, and since the rate of movement of
graphic characteristics given the target population: electricity across the surface of the skin is a good indi-
mean age ¼ 42; 52 per cent female; mean income in cator of the presence of moisture, electrodermal
the $40 000 – 60 000 range; and mean educational activity has long been accepted as a fairly direct and
level ¼ some college. These 200 were intended to pure representation of sympathetic activity, making it
serve as a pool from which smaller groups could be a good measure of the psychological concepts of
culled for physiological testing. The particular group emotion, arousal and attention. [45]. There are a
employed in the analysis here consisted of 48 individ- number of approaches to measure skin conductance
uals who were called back later that summer. They level (SCL) response to a stimulus; a common
were selected because of availability and because approach is to measure SCL at two different time
they were the individuals most clearly falling on points, which can be reported either as a raw or
either the political left or the political right according adjusted difference, or as a percentage or proportion
to the survey responses provided during their first [45,46]. This approach has the advantage of providing
visit. Participants were paid $50 for each of their two a means to control for wide variation in baseline elec-
separate trips to the laboratory. The data on two par- trodermal activity and is the approach followed here.
ticipants had to be removed, one owing to a health SCLs for each image were measured as a proportion
issue, the other owing to a mechanical problem with of the SCL recorded while the participant was viewing
a sensor. the fixation point prior to image exposure. This creates
To measure political orientation, several variables a standardized measure where 1 denotes no change in
were combined. Since a US sample was used in our SCL between viewing a fixation point and an appeti-
analyses, we used party labels, ideological labels and tive/aversive image, and numbers greater than 1
individual political issues that would be familiar to indicate an SCL increase. Several other measurement
such a group. Thus, participants were asked to approaches were constructed to capture the difference
(i) report their ideological position on a scale running between SCL during fixation point and SCL during
from strong liberal (left) to strong conservative (right), image exposure, including calculating raw first differ-
(ii) report their partisan affiliation, from strong ences and differences in logged means. The resulting
Democrat (left) to strong Republican (right), (iii) variables were all correlated at levels greater than
answer 28 items on their specific policy preferences 0.90, and we report the proportion measure because
presented in the well-known Wilson –Patterson it lends itself to easy and intuitive interpretation.
format [43], and (iv) complete a social principles All of the images used in the present study were
index. The latter presented subjects with 15 forced rated by 126 independent judges (none of whom was
choices between basic principles of social organization. a participant in the studies reported here) who were
As an example of items in this last category, partici- asked to rate image valence on a nine-point scale
pants indicated whether ‘society works best when . . . whether each image gave them ‘happy/positive’ [1] or
those who break the rules are punished . . . or . . . ‘unhappy/negative’ [9] feelings and to rate how
when those who break the rules are forgiven’ ([44]; a strongly they felt an emotional reaction when looking
full listing of these and the Wilson– Patterson items at the image. Based on these ratings, the three most
can be found in electronic supplementary material, negatively valenced and the three most positively
appendices A and B). For both the Wilson –Patterson valenced images were selected for use during the
issue items and the ‘society works best’ items, an addi- physiological session. The negative (aversive) images
tive index was constructed (with the position on the were a spider on a man’s face (mean valence rating
political right always given the higher coding). These 7.65, s.d. ¼ 1.68), an open wound with maggots in it
four diverse measures of political orientation are (mean valence rating 7.94, s.d. ¼ 1.15), and a crowd
fairly strongly related, with bivariate correlations ran- fighting with a man (mean valence rating 7.83,
ging from 0.57 to 0.75 (p , 0.05 in all cases) and s.d. ¼ 1.16). The three images judged to be the most
with a factor analysis confirming that these four positive (appetitive) were of a happy child (mean
measures tap into a single dimension (a principal com- valence rating ¼ 4.94, s.d. ¼ 2.33), a bowl of fruit
ponents analysis yielded a single factor accounting (4.36, s.d. ¼ 2.21) and a cute rabbit (4.62, s.d. ¼
for approx. 75% of the variance, and factor loadings 2.33). Raters were also asked to report the specific
for the individual variables were 0.79 or higher). emotion they felt when looking at each image. The
The four indicators were weighted equally and most frequently reported emotion for the spider
added together to create a broad measure of left – right image was fear (78% of raters reported the image
political orientation. evoked this emotion), for the maggot image, the
In the physiological session, participants were most frequently reported emotion was disgust (96%),
shown a series of 33 still images. Each image was and for the crowd fighting with a man, the most fre-
shown once and was preceded by a fixation point quently evoked emotion was anger (76%). Thus,
that was displayed during an inter-stimulus interval. these images would seem to capture an array of differ-
The order of slides was initially randomized and then ent negative emotional responses. Positive emotions
presented in the same order to all participants. have fewer discrete categories and, according to the
During the slide show, electrodermal activity (in the raters, the most frequently evoked emotion for all
form of skin conductance readings) was collected three appetitive images was happiness.
Phil. Trans. R. Soc. B (2012)
 Physiology, cognition, politics M. D. Dodd et al. 643

1.015 Table 1. Predicting political orientations with differential

skin conductance reactivity to appetitive and aversive
images. Dependent variable is an aggregate of standardized
1.010 scores on the Wilson – Patterson index, society works best
items, a seven-point political ideology scale and a seven-
point party identification scale, scored such that higher
1.005
skin conductance change

values indicate political orientations towards the right of the

political spectrum.
1.000
estimated standard
coefficient error
0.995
skin conductancea 12.17** 4.13
age 20.04 0.02
0.990 gender 20.36 0.30
income 20.09 0.10
0.985 education 20.25* 0.10
constant 5.66 0.59
n ¼ 46
0.980 R 2 ¼ 0.39;
aversive appetitive adj. R 2 ¼ 0.31
a
Figure 1. Mean skin conductance change (in microsiemens) Degree to which skin conductance increases were greater for
aversive than appetitive stimuli.
as a function of political temperament (left versus right) and *p , 0.05; **p , 0.01 (two-tailed tests).
image type (appetitive versus aversive). Triangles with solid
line, right-of-centre; squares with solid line, left-of-centre.

aversive), as well as on four standard demographic

The relation between political temperament and controls: age, gender, income and education. Higher
electordermal increases in response to aversive/appeti- values on the composite measure of political ideology
tive images was initially examined by dividing indicate right-of-centre orientations and higher values
participants at the mean on the composite measure on the physiological measure indicate relatively greater
of political orientations and then plotting separately electrodermal increases to aversive stimuli, meaning
the physiological response for the left-of-centre and that a positive relationship is expected. As can be
right-of-centre groups. The consistent empirical find- seen in table 1, only one of the control variables is sig-
ing in psychophysiology is that, while participants nificantly related to political orientations: increasing
exhibit an enhanced physiological response to both levels of education correlate with left-of-centre political
appetitive and aversive stimuli, the response is typically orientations. Importantly, however, relatively greater
greater for aversive stimuli [39]. Consistent with the electrodermal increases when viewing aversive stimuli
theory outlined above and with previous research on are indeed a strong predictor of right-of-centre political
narrower emotions [18,21], the hypothesis is that indi- beliefs (b ¼ 12.17; p , 0.01).
viduals on the right side of the political spectrum will A parallel but more politically focused test of this
exhibit increased electrodermal activity when viewing hypothesis is afforded by the fact that people who
aversive images while those on the left side will exhibit care about politics (such as the group of participants
increased electrodermal activity, in relative terms, being analysed here) are likely to find visible political
when viewing the appetitive images. This prediction figures to be either appetitive or aversive. Though
is confirmed by a 2 (image type: appetitive versus there could be numerous reasons for a politician to
aversive)  2 (ideology: left versus right) mixed analy- be viewed favourably or unfavourably, an important
sis of variance (ANOVA) as there is a significant factor for most politically attuned individuals is the
interaction between image type and ideology (F ¼ degree of ideological similarity between themselves
5.60; p , 0.05). As can be seen in figure 1, electro- and the politician in question. A politician with an
dermal increases for those on the political right are ideology that is consistent with that of the respondent
greater for aversive relative to appetitive images, is more likely to be viewed as appetitive, whereas a
whereas for those on the political left the opposite politician with an ideology that is inconsistent with
pattern of results is exhibited. that of the respondent is more likely to be viewed as
Though these initial findings are suggestive, political aversive. Given the results in figure 1, we hypothesized
orientations are better characterized as continuous that the electrodermal responses of individuals on the
rather than dichotomous since many individuals are right would be greater, in relative terms, to ideologic-
political moderates rather than ideologues. Moreover, ally dissimilar politicians, whereas the electrodermal
other variables besides physiological patterns are responses of individuals on the left would be greater,
likely to be relevant to political orientations. Therefore, in relative terms, to ideologically similar politicians.
we regressed the continuous measure of political Kaplan et al. [47] report that neural activity in the dor-
orientation on the mean difference in physiological solateral pre-frontal cortex, the anterior cingulate
response depending on stimulus type (skin conduct- cortex and the insula increased when political parti-
ance increase in response to appetitive subtracted sans viewed images of candidates from the opposing
from skin conductance increase in response to party (compared with images of the favoured party),
Phil. Trans. R. Soc. B (2012)
644 M. D. Dodd et al. Physiology, cognition, politics

1.03 Table 2. Predicting political orientations with differential

skin conductance reactivity to ideologically similar and
ideologically dissimilar political images. Dependent variable
1.02 is an aggregate of standardized scores on the Wilson –
Patterson index, society works best items, a seven-point
political ideology scale and a seven-point party
skin conductance change

1.01 identification scale, scored such that higher values indicate

political orientations towards the right of the political
spectrum.
1
estimated standard
coefficient error
0.99
skin conductancea 8.99** 3.63
age 20.03 0.02
gender 20.39 0.31
0.98 income 0.01 0.11
education 20.29** 0.10
constant 5.42 1.14
0.97 n ¼ 46
aversive appetitive R 2 ¼ 0.35;
adj. R 2 ¼ 0.27
Figure 2. Mean skin conductance change (in microsiemens)
a
as a function of political temperament (left versus right) and Degree to which skin conductance increases were greater for
political image type (appetitive versus aversive). Triangles ideologically dissimilar than for ideologically similar politicians.
with solid line, right-of-centre; squares with solid line, *p , 0.05; **p , 0.01 (two-tailed tests).
left-of-centre.

Figure 2 presents the electrodermal response of

but these researchers did not analyse partisan groups participants on the left and participants on the right
separately, meaning it is unknown whether activation to images of politicians either ideologically similar
to the opposing party was more noticeable among to or different from the participant. As expected, the
those on the right than among those on the left. pattern of responses is similar to what was observed in
Images of well-known American political figures figure 1. The increase in electrodermal activity of
were included in the 33 stimuli presented; specifically, right-of-centre participants is greater for politicians
pictures of Ronald Reagan, Bill Clinton, Hillary with whom they are in ideological disagreement
Clinton and George W. Bush. These four seem appro- than for politicians with whom they are in ideological
priate since at the time of the study (summer of 2007), agreement, whereas the electrodermal activity of
Barack Obama, John McCain and Sarah Palin had not left-of-centre participants is greater for politicians
yet arrived on the national political scene and pre-tests with whom they are in ideological agreement than for
indicated that many participants could not identify politicians with whom they are in ideological dis-
pictures of other important national politicians, agreement. This was confirmed by a 2 (image type:
including (then Vice President) Richard Cheney, appetitive versus aversive)  2 (ideology: left versus
(then Speaker of the House) Nancy Pelosi and the right) ANOVA as there is a significant interaction
previous two Democratic presidential nominees: John between image type and ideology (F ¼ 10.86; p ,
Kerry and Al Gore. 0.01) but no other significant effects or interactions.
To test the hypothesis that left-of-centre participants Thus, whether the focus is on generically aversive/
respond more to ideologically similar (appetitive) polit- appetitive stimuli or on specifically political stimuli, the
icians while right-of-centre participants respond more to results suggest that individuals on the left are more
ideologically dissimilar (aversive) politicians, a definition responsive to appetitive relative to aversive stimuli,
of ideological similarity is necessary. The ideology of the while individuals on the right are more responsive to
aforementioned four politicians is relatively easy to cat- aversive relative to appetitive stimuli.
egorize. As of mid-2007, Bill and Hillary Clinton were Converting political orientation from a dichotomous
nationally visible politicians associated with the left, just to a more appropriate continuous form and adding the
as George W. Bush and Ronald Reagan were highly salient same controls as included in table 1 underscores these
touchstones of the right. Indeed, polls at the time conclusions. As can be seen in table 2, political orien-
suggested that George W. Bush and Hillary Clinton tation is strongly predicted by electrodermal response
were the most polarizing political figures in American to ideologically similar and dissimilar political figures.
politics—and Ronald Reagan and Bill Clinton, as The further respondents are to the political right, the
former two-term presidents on quite different sides of more their electrodermal response to negative images
the ideological ledger, were still able to incite passions. tends to outstrip their response to positive images
The ideology of those participating in the physiological (b ¼ 8.99; p , 0.01). The control variables are, again,
exercise was assessed with the same composite measure not significantly related to political orientation with
as before and the central measure of physiological the exception that increasing levels of education are
change was again mean increase in the participants’ associated with left-of-centre political orientations.
SCLs from the preceding inter-stimulus interval to the Further tests should be run in case there is something
images (pictures of politicians) in question. particular about the political images employed here
Phil. Trans. R. Soc. B (2012)
 Physiology, cognition, politics M. D. Dodd et al. 645

but these initial indications are consistent with the the ‘society works best’ battery of broader political
results in table 1 and figure 1. preferences (see electronic supplementary material,
appendices). As before, these indicators are combined
by first weighting them equally and then creating
3. STUDY 2: ATTENTIONAL PATTERNS an additive index. These procedures allowed each
Though physiology constitutes one element of orien- participant to be assigned an overall ideological score,
tation to categories of stimuli, a more complete with higher numbers indicating location further to the
understanding of the manner in which individuals political right. Dividing these scores at the median
are situated with respect to the aversive and the appe- makes it possible to divide participants into those who
titive in life can be obtained by determining if are (in relative terms) on the left and those who are
individuals have attentional biases to particular stimu- on the right. Given that the control variables in the
lus types. Accordingly, we designed and administered previous study were unrelated to political orientation
a freeview eyetracking study in which individuals (except for education, which would be invariant
could direct their gaze towards either appetitive or for this student sample), they were not collected
aversive images when both types are present. here. Seventy-six participants completed both the
Participants in this study were undergraduates eye gaze exercise and all the necessary survey items.
drawn from the psychology student subject pool at The mean dwell time results for both collage types
the University of Nebraska-Lincoln. Using an SR (single aversive image versus single appetitive image)
Research Ltd. EyeLink II system connected to a can be observed in figure 3 separately for the 38 parti-
Pentium IV PC, participants were seated approxi- cipants furthest to the left and then for the 38
mately 44 cm from the computer screen and viewed participants furthest to the right.
a series of collages for 8 s each, during which time A 2 (image type: appetitive versus aversive)  2
participants were free to view the images in any (collage type: single aversive versus single appeti-
manner they desired. At the end of the viewing tive)  2 (ideology: left versus right) mixed ANOVA
period, a fixation point appeared on the screen until revealed a main effect of image type (F ¼ 41.14; p ,
the space bar was pressed by the participant to initiate 0.01) as aversive stimuli are given more attention over-
viewing of the next collage. Each collage was com- all than appetitive stimuli. As mentioned above, from
posed of four equally sized images, most taken from an evolutionary standpoint, this pattern makes sense
the IAPS database of pre-validated images [48] and since aversive stimuli can do harm and therefore
others previously pre-rated from a separate study. merit more attention than stimuli that seem pleasant
Appetitive images were drawn from the top 20 per and probably harmless [26]. There is also a main
cent of positively rated images and aversive images effect of collage type (F ¼ 34.04, p , 0.01), as an
were taken from the top 20 per cent of negatively increase in the number of aversive images led to differ-
rated images. Six of the collages contained three aver- ences in dwell time for the various image types, as also
sive images and one appetitive image, and six other indicated by a significant image type  collage type
collages contained three appetitive images and one interaction (F ¼ 21.36, p , 0.01). Moreover, there is
aversive image. Critically, all participants saw the a main effect of ideology (F ¼ 7.09; p , 0.01), but
same set of collages. the critical test of the hypothesis that individuals on
The eyetracker allows for real-time recording of the political right pay more relative attention than
gaze behaviour, making it possible to document individuals on the political left to aversive stimuli is
where in the collage the participant looks, when, and the interaction between image type and ideology.
for how long. As such, our investigation focused on This interaction is significant (F ¼ 3.75; p ¼ 0.057).
two critical measurements: Dwell time (ms)—the Those on the left devote more attention to aversive
amount of time spent on each image in the collage, than appetitive images; however, as expected, this
and first fixation time (ms)—the amount of time elap- ‘aversion bias’ is much more pronounced for the polit-
sing relative to the onset of the trial before participants ically right half of the sample. For both trial types
look at each image type. To calculate gaze orientation (single aversive versus single appetitive), in relative
to aversive as opposed to appetitive stimuli, the mean terms, individuals on the right spend a greater
total amount of time the participants fixated the aver- amount of time gazing at aversive images while indi-
sive quadrant (for those six trials on which such viduals on the left spend a greater amount of time
an image was the unique quadrant) was determined. gazing at appetitive images.
The same calculation was then made for those six Though total dwell time is an important measure,
trials in which the appetitive image was the unique additional insight can be drawn from analysing first
quadrant. Importantly, eyetracking is the only atten- fixation time—the amount of time that elapses after
tional measure that allows for an examination of not the onset of the trial before participants look at each
only bias towards specific items, but also avoidant be- image type. Dwell time results relate to the total time
haviour. Given that those on the right exhibit greater spent gazing at an image, while first fixation time pro-
increases in electrodermal activity when viewing aver- vides a measure of attentional bias in terms of how
sive stimuli, it is worthwhile to determine whether quickly an aversive or appetitive image is fixated. It is
they are biased towards or away from aversive stimuli important to note that whereas large dwell time
when given a choice of multiple images to view. values are indicative of an attentional bias towards an
To determine political orientation, participants were image, small (rather than large) first fixation time
asked to indicate party identification, in addition to values are also indicative of an attentional bias towards
completing the Wilson– Patterson issue battery and an image as lower values are representative of faster
Phil. Trans. R. Soc. B (2012)
646 M. D. Dodd et al. Physiology, cognition, politics

(a)
(b) 1950
2750

1750
average dwell time (ms)

2250
1550

1750 1350

1150
1250
950

750 750
aversive appetitive aversive appetitive

Figure 3. Mean dwell time values (in milliseconds) as a function of image type and trial type for left-of-centre and right-of-
centre participants. (a) Represents dwell time on trials in which three appetitive and one aversive image are presented (pre-
dominantly appetitive), whereas (b) represents dwell time on trials in which one appetitive and three aversive images are
presented (predominantly aversive). Note that the values reported are the average dwell time for each individual image, so
on a trial with three appetitive images, the total dwell time for all appetitive stimuli would be the reported number multiplied
by 3. Triangles with solid line, right-of-centre; squares with solid line, left-of-centre.

orientation. Consistent with theory and the results and especially cognitive differences of individuals
above, we hypothesize that, relative to those on the with specific ideological leanings is still in its infancy.
left, those on the right will be faster to orient to Our goal here was to further understand the nature
aversive images. of political differences by combining physiological
All of the main effects and interactions observed and attentional measures in the same study. Doing
in the dwell time analyses above are also observed so permits useful interpretational advances.
for the first fixation time analyses, with the exception Our core finding is that, compared with individuals
of there being no main effect of political orientation on the political left, individuals on the right direct
(F ¼ 75.08, p , 0.01 for the main effect of image more of their attention to the aversive despite displaying
type; F ¼ 7.28, p , 0.01 for the main effect of collage greater physiological responsiveness to those stimuli.
type, and F ¼ 8.26, p , 0.01 for the interaction This combination of physiological and attentional
between image type and collage type). Critically, how- data is worth considering further. Previous research
ever, there is again an interaction between image type on the broader bases of political ideology is often inter-
and ideology (F ¼ 10.62, p , 0.01). As can be seen in preted as suggesting that locations on the right of the
figure 4, those on the political left fixate appetitive political spectrum are a deviation from the norm (or
images more quickly than those on the political right even a pathology) in need of explanation [10,51]. For
while those on the political right are faster to fixate aver- example, McClosky [52, p. 40] concludes those on
sive images relative to participants on the political left. the right are ‘distrustful of differences . . . fear change,
As in the physiology study, these findings can be dread disorder, are intolerant of nonconformity, and
enriched by using a continuous rather than dichoto- derogate reason’ while Block & Block [53, p. 395]
mous measure of political orientation. The continuous find that those on the right are ‘easily victimized,
version of the composite measure of political orientation easily offended, indecisive, fearful, rigid, inhibited,
correlates with attentional bias towards aversive stimuli relatively over-controlled and vulnerable’.
as it relates to both dwell time (r ¼ 0.32; p , 0.01) Demonstrating that those on the right not only
and first fixation time (r ¼ 20.19, p , 0.05). These cor- respond more strongly to aversive images but also
relations further confirm that political orientations devote more attention to aversive images suggests a
tending towards the right of the spectrum are associated different and perhaps less value-charged interpretation
with both faster orienting towards, and greater total of those holding right-of-centre political orientations.
time spent attending to, aversive relative to appetitive It appears individuals on the political right are not so
images whereas the opposite is true for participants on much ‘fearful’ and ‘vulnerable’ as attuned and atten-
the left of the ideological spectrum. tive to the aversive in life. This responsiveness and
attentiveness, in turn, is consistent with the fact that
right-of-centre policy positions are often designed to
4. DISCUSSION protect society from out-group threats (e.g. by sup-
Research placing politics in a deeper biological context porting increased defence spending and opposing
is growing, and helpful reviews are available [49,50] immigration) and in-group norm violators (e.g. by
but research focusing directly on the physiological supporting traditional values and stern penalties for
Phil. Trans. R. Soc. B (2012)
 Physiology, cognition, politics M. D. Dodd et al. 647

(a) (b) 2150

1600
1950

1400
1750
first fixation time (ms)

1200 1550

1350
1000
1150

800
950

600 750
aversive appetitive aversive appetitive

Figure 4. First fixation time values (in milliseconds) as a function of image type and trial type for left-of-centre (square with
solid line) and right-of-centre participants (triangle with solid line). (a) Represents first fixation time to image types on trials in
which three appetitive and one aversive image are presented (predominantly appetitive), whereas (b) represents first fixation
time to image types on trials in which one appetitive and three aversive images are presented (predominantly aversive).
Note that unlike dwell time—in which large values represent a greater attentional bias towards an image—small first fixation
time values represent a greater attentional bias towards an image given that the lower the value, the faster the image was fixated.

criminal behaviour). Rather than using colourful experiential differences mean for cultural conflict? At
adjectives, perhaps, the proper approach is simply to first blush, the implications may appear dire; after
state that the aversive in life appears to be more physio- all, the absence of an objective reality would seem to
logically and cognitively tangible to some people and render fruitless any attempt at reasoning towards an
they tend to gravitate to the political right. optimal political solution. On the other hand, for sev-
As such, these results suggest that a simple distinction eral reasons, the normative implications regarding
between approach and avoidant behaviours may be cultural conflict may not be the cause for alarm.
incomplete. Quite apart from whether an aversive situ- First, our physiological study was weighted towards
ation is approached or avoided, it apparently is possible participants with strong political beliefs and therefore
to attend to it even when it generates substantial physio- the results do not reflect the large percentage of indi-
logical response. This is precisely the pattern in evidence viduals who probably are not physio-cognitively
among many of those on the political right. Seen from predisposed towards any political orientation. These
this perspective and given the compelling evolutionary individuals, often found in the political centre, likely
logic for organisms to be particularly sensitive to aversive are open to efforts at political persuasion.
stimuli [26], it may be that those on the political left are Second, many of the individuals who do display
more out of step with adaptive behaviours. The question politically relevant physiological and cognitive predis-
becomes why those on the left display so little aversion positions may be capable of changing. The ultimate
bias either in their physiology or, to a lesser extent, in source of physiological and cognitive predisposition is
their patterns of attention despite the acknowledged still to be determined (our results say little about ulti-
adaptive value of an aversion bias. Of course, the natur- mate sources), but is almost certainly a combination
alistic fallacy reminds us that behaving in an adaptive of genetics, early developmental experiences and
fashion does not necessarily equate with behaving in a more immediate environmental events. Physiological,
desirable fashion. neurological and cognitive patterns, for example, are
Be this as it may, the central message of these known to be relatively stable over time [54] but hardly
findings is not that one political orientation is some- immutable [55]. What is suggested by our results is
how superior to the other but rather that, in light of that for many, but certainly not all, of those individuals
the connection between location on the political professing ideological convictions, their political beliefs
spectrum and physio-cognitive differences, those on have become biologically instantiated in a fashion that
the political right and those on the political left may renders them sticky and slow to change—somewhere
simply experience the world differently. It is probably between wholly static and completely malleable.
because of these differences that some on the right Evidence that political orientations are often physio-
view those on the left as hedonists who ignore pressing logically and cognitively instantiated leads to the
issues while some on the left view those on the right conclusion that change is possible but likely grudging
as doomsayers who obsess over constructed threats and indeed this phrase seems an apt description of the
and problems. lack of fluidity of the political positions of those with
What does the evidence that political orientations strong orientations. This evidence also implies that dis-
are grounded partially in physiological and cognitive tinct campaign and advertising appeals are likely to be
Phil. Trans. R. Soc. B (2012)
648 M. D. Dodd et al. Physiology, cognition, politics

differentially effective for those on the political right 9 Graham, J., Haidt, J. & Nosek, B. A. 2009 Liberals and con-
and those on the political left. servatives rely on different sets of moral foundations. J. Pers.
Finally and most hopefully, the choice available to Soc. Psychol. 96, 1029–1046. (doi:10.1037/a0015141)
society is not between people whose political orien- 10 Altemeyer, R. A. 1996 The authoritarian specter.
Cambridge, MA: Harvard University Press.
tations are either completely changeable or to some
11 Sidanius, J. & Pratto, F. 1999 Social dominance: an inter-
extent biologically predisposed. Rather, the choice is group theory of social hierarchy and oppression. Cambridge,
between recognizing that physiological and cognitive UK: Cambridge University Press.
patterns lead to politically relevant variations in the 12 Jost, J. T., Glaser, J., Krugalanski, A. W. & Sulloway, F.
manner in which the outside world is experienced or, 2003 Political conservatism as motivated social cogni-
alternatively, pretending that political orientations are tion. Psychol. Bull. 129, 339 –375. (doi:10.1037/0033-
rational, free-floating and unencumbered. Given this 2909.129.3.339)
choice set, we suggest that there are real advantages 13 Baker, W. 2005 America’s crisis of values. Princeton, NJ:
to embracing the relevance of these deeper, biological Princeton University Press.
variables. After all, it is far easier to tolerate differences 14 Stenner, K. 2005 The authoritarian dynamic. Cambridge,
if they are recognized to be in part biologically based UK: Cambridge University Press.
15 Schwartz, S. H. 2007 Universalism values and the inclu-
(consider the debate over homosexuality where those
siveness of our moral universe. J. Cross Cult. Psychol. 38,
acknowledging a biological source are typically more 711 –728. (doi:10.1177/0022022107308992)
tolerant than those maintaining sexual preference is 16 Hetherington, M. J. & Weiler, J. D. 2009 Authoritarian-
entirely environmentally determined). Rather than ism and polarization in American politics. Cambridge,
believing those with political views opposing ours are UK: Cambridge University Press.
lazily uninformed or wilfully obtuse, political tolerance 17 Kanai, R., Feilden, T., Firth, C. & Rees, G. 2011 Polit-
could be enhanced and cultural conflict diminished if ical orientations are correlated with brain structure in
it is widely recognized that at least part of our political young adults. Curr. Biol. 21, 1 –4. (doi:10.1016/j.cub.
differences spring from subconscious physiological 2011.03.017)
and cognitive variations that lead people to experience 18 Amodio, D. M., Jost, J. T., Master, S. L. & Yee, C. M. 2007
Neurocognitive correlates of liberalism and conservatism.
the world in fundamentally different ways and there-
Nat. Neurosci. 10, 1246–1247. (doi:10.1038/nn1979)
fore to believe that fundamentally different political 19 Huddy, L., Feldman, S., Taber, C. & Lahav, C. 2005
policies are appropriate. Threat, anxiety, and support of anti-terrorism policies.
This research was supported by National Science Am. J. Polit. Sci. 49, 610 –625. (doi:10.1111/j.1540-
Foundation grant BCS-0826828, John Hibbing, Principal 5907.2005.00144.x)
Investigator. We would like to thank Billy Kuehn for his 20 Vigil, J. M. 2010 Political leanings vary with facial expression
assistance in data collection. processing and psychosocial functioning. Group Process
Interg. 13, 547–558. (doi:10.1177/1368430209356930)
21 Oxley, D. R., Smith, K. B., Alford, J. R., Hibbing, M. V.,
Miller, J. L., Scalora, M., Hatemi, P. K. & Hibbing, J. R.
REFERENCES 2008 Political attitudes vary with physiological traits.
1 Jennings, M. K. & Niemi, R. G. 1968 The transmission Science 321, 1667–1670. (doi:10.1126/science.1157627)
of political values from parent to child. Am. Polit. Sci. 22 Haidt, J. & Hersh, M. 2001 Sexual morality: The cultures
Rev. 62, 169–183. (doi:10.2307/1953332) of conservatives and liberals. J. Appl. Soc. Psychol. 31,
2 Jennings, M. K., Stoker, L. & Bowers, J. 2009 Politics 191 –221. (doi:10.1111/j.1559-1816.2001.tb02489.x)
across generations: family transmission reexamined. 23 Inbar, Y., Pizarro, D. A. & Bloom, P. 2009 Conservatives
J. Polit. 71, 782–799. (doi:10.1017/S0022381609090719) are more easily disgusted than liberals. Cogn. Emotion 23,
3 Zuckerman, A. S., Dasović, J. & Fitzgerald, J. 2000 714 –728. (doi:10.1080/02699930802110007)
Partisan families: the social logic of bounded partisanship in 24 Settle, J. E., Dawes, C. T., Christakis, N. A. & Fowler,
Germany and Britain. Cambridge, UK: Cambridge J. H. 2010 Friendships moderate an association between
University Press. the DRD4 gene and political ideology. J. Polit. 2, 1189–
4 McCrae, R. R. & Costa Jr, P. T. 2003 Personality in 1198. (doi:10.1017/S0022381610000617)
adulthood: a five factor theory perspective. New York, NY: 25 Hatemi, P. K. et al. 2011 A genome-wide analysis of
Guilford Press. liberal and conservative political attitudes. J. Polit. 73,
5 Caprara, G. V., Schwartz, S., Capanna, C., Vecchione, M. 271 –285. (doi:10.1017/S0022381610001015)
& Barbaranelli, C. 2006 Personality and politics: values, 26 Neuberg, S., Kenrick, D. & Schaller, M. 2011 Human
traits, and political choice. Polit. Psychol. 27, 1– 28. threat management systems. Neurosci. Biobehav. Rev.
(doi:10.1111/j.1467-9221.2006.00447.x) 35, 1042 –1051.(doi:10.1016/j.neubiorev.2010.08.011)
6 Carney, D. R., Jost, J. T., Gosling, S. D. & Potter, J. 2008 27 Bradley, M. M. 2000 Emotion and motivation. In Hand-
The secret lives of liberals and conservatives: personality book of psychophysiology (eds J. T. Cacioppo, L. G.
profiles, interaction styles, and the things they leave Tassinary & G. G. Berntson), pp. 602 –642, 2nd edn.
behind. Polit. Psychol. 29, 807 –840. (doi:10.1111/j. Cambridge, UK: Cambridge University Press.
1467-9221.2008.00668.x) 28 Mortensen, C. R., Becker, D. V., Ackerman, J. M.,
7 Mondak, J. J., Hibbing, M. V., Canache, D., Seligson, Neuberg, S. L. & Kenrick, D. T. 2010 Infection breeds reti-
M. A. & Anderson, M. R. 2010 Personality and civic cence: The effects of disease salience on self-perceptions of
engagement: an integrative framework for the study of personality and behavioral avoidance tendencies. Psychol.
trait effects on political behavior. Am. Polit. Sci. Rev. Sci. 21, 440–447. (doi:10.1177/0956797610361706)
104, 85–110. (doi:10.1017/S0003055409990359) 29 Bradley, M. & Lang, P. J. 2007 The International Affec-
8 Haidt, J. & Graham, J. 2007 When morality opposes tive Picture System (IAPS) in the study of emotion and
justice: conservatives have moral intuitions that liberals attention. In Handbook of emotion elicitation and assessment
may not recognize. Soc. Justice Res. 20, 98–116. (eds J. Coan & J. Allen), pp. 29–46. New York, NY:
(doi:10.1007/s11211-007-0034-z) Oxford University Press.

Phil. Trans. R. Soc. B (2012)

 Physiology, cognition, politics M. D. Dodd et al. 649

30 Gray, J. 1972 The psychophysiological basis of introver- 42 Schwartz, S. H., Caprara, G. V. & Vecchione, M. 2010
sion-extraversion: a modification of Eysenck’s theory. Basic personal values, core political values, and voting:
In The biological bases of individual behavior (eds V. D. a longitudinal analysis. Polit. Psychol. 31, 421– 453.
Nebylitsyn & J. A. Gray), pp. 182–205. San Diego, (doi:10.1111/j.1467-9221.2010.00764.x)
CA: Academic Press. 43 Wilson, G. D. & Patterson, J. R. 1968 A new measure of
31 Gray, J. A. 1981 A critique of Eysenck’s theory of person- conservatism. Br. J. Soc. Clin. Psychol. 8, 264 –269.
ality. In A model for personality (ed. H. J. Eysenck), (doi:10.1111/j.2044-8260.1968.tb00568.x)
pp. 246 –277. Berlin, Germany: Springer. 44 Smith, K. B., Oxley, D. R., Hibbing, M. B., Alford, J. R. &
32 Carver, C. S., Sutton, S. K. & Scheier, M. F. 2000 Hibbing, J. R. 2011 Linking genetics and political atti-
Action, emotion, and personality: emerging conceptual tudes: re-conceptualizing political ideology. Polit. Psychol.
integration. Pers. Soc. Psychol. Bull. 26, 741–751. 32. (doi:10.1111/j.1467-9221.2010.00821.x)
(doi:10.1177/0146167200268008) 45 Dawson, M. E., Shell, A. M. & Filion, D. L. 2007 The
33 Marcus, G. E. 2002 The sentimental citizen: emotion in electrodermal system. In Handbook of psychophysiology
democratic politics. Pennsylvania, PA: Pennsylvania State (eds J. T. Cacioppo, L. G. Tassinary & G. G. Berntson),
University Press. pp. 159 –181, 3rd edn. New York, NY: Cambridge
34 Davidson, R., Ekman, P., Saron, C. D., Senulis, J. A. & University Press.
Friesen, W. V. 1990 Approach/withdrawal and cerebral 46 Andreassi, J. L. 2007 Human behavior and physiological
asymmetry. J. Pers. Soc. Psychol. 58, 330– 341. (doi:10. response, pp. 267–268. New York, NY: Psychology Press.
1037/0022-3514.58.2.330) 47 Kaplan, J. T., Freedman, J. & Iacoboni, M. 2007 Us
35 Lang, P. J., Bradley, M. M. & Cuthbert, B. N. 1990 versus them: political attitudes and party affiliation
Emotion, attention, and the startle reflex. Psychol. Rev. influence neural response to faces of presidential candi-
97, 377–398. (doi:10.1037/0033-295X.97.3.377) dates. Neuropsychologica 45, 55– 64. (doi:10.1016/
36 Davis, M. 2000 The role of the amygdala in conditioned j.neuropsychologia.2006.04.024)
and unconditioned fear and anxiety. In The amgydala, 48 Lang, P. J., Bradley, M. M. & Cuthbert, B. N. 2005
vol. 2 (ed. J. P. Aggleton), pp. 213–287. Oxford, UK: International Affective Picture Systems (IAPS): affective
Oxford University Press. ratings of pictures and instruction manual. Technical
37 Davis, M. & Lang, P. J. 2003 Emotion. In Handbook of psy- Report no. A-6. University of Florida, Gainesville, USA.
chology, vol. 3: biological psychology (eds M. Gallagher & 49 Fowler, J. & Schreiber, D. 2008 Biology, politics, and the
R. J. Nelson), pp. 405–439. New York, NY: Wiley. emerging science of human nature. Science 322, 912 –914.
38 Fanselow, M. S. 1994 Neural organization of the defen- (doi:10.1126/science.1158188)
sive behavior system responsible for fear. Psychon. Bull. 50 McDermott, R. 2004 The feeling of rationality. Perspect.
Rev. 1, 429–438. (doi:10.3758/BF03210947) Polit. 2, 691– 706. (doi:10.1017/S1537592704040459)
39 Hamm, A. O., Schupp, H. T. & Weike, A. I. 2003 Moti- 51 Adorno, T., Frenkel-Brunswik, E., Levinson, D. &
vational organization of emotions: autonomic changes, Sanford, R. 1950 The authoritarian personality.
cortical responses, and reflex modulation. In Handbook New York, NY: Harper.
of affective sciences (eds R. J. Davidson, K. R. Scherer & 52 McClosky, H. 1958 Conservatism and personality. Am.
H. H. Goldsmith), pp. 187 –211. New York, NY: Polit. Sci. Rev. 52, 27– 45. (doi:10.2307/1953011)
Oxford University Press. 53 Block, J. & Block, J. H. 2005 Nursery school personality
40 Bannerman, R. L., Milders, M. & Sahraie, A. 2010 and political orientation: two decades later. J. Res.
Attentional bias to brief threat-related faces revealed by Personal 40, 734 –749. (doi:10.1016/j.jrp.2005.09.005)
saccadic eye movements. Emotion 10, 733–738. 54 Schell, A., Dawson, M., Nuechterline, K., Subotnik, K. &
(doi:10.1037/a0019354) Venture, J. 2002 The temporal stability of electrodermal
41 Miltner, W. H. R., Krieschel, S., Hecht, H., Trippe, R. & variables over a one-year period. Psychophysiology 39,
Weiss, T. 2004 Eye movements and behavioral res- 124 –132. (doi:10.1111/1469-8986.3920124)
ponses to threatening and nonthreatening stimuli 55 Maguire, E. A., Woollett, K. & Spiers, H. J. 2006
during visual search in phobic and nonphobic London taxi drivers and bus drivers: a structural MRI
subjects. Emotion 4, 323 –339. (doi:10.1037/1528-3542. and neuropsychologial analysis. Hippocampus 16, 1091–
4.4.323) 1101. (doi:10.1002/hipo.20233)

Phil. Trans. R. Soc. B (2012)

View publication stats