Elsevier

NeuroImage

Volume 72, 15 May 2013, Pages 174-182
NeuroImage

Automatic processing of political preferences in the human brain

https://doi.org/10.1016/j.neuroimage.2013.01.020Get rights and content

Abstract

Individual political preferences as expressed, for instance, in votes or donations are fundamental to democratic societies. However, the relevance of deliberative processing for political preferences has been highly debated, putting automatic processes in the focus of attention. Based on this notion, the present study tested whether brain responses reflect participants' preferences for politicians and their associated political parties in the absence of explicit deliberation and attention. Participants were instructed to perform a demanding visual fixation task while their brain responses were measured using fMRI. Occasionally, task-irrelevant images of German politicians from two major competing parties were presented in the background while the distraction task was continued. Subsequent to scanning, participants' political preferences for these politicians and their affiliated parties were obtained. Brain responses in distinct brain areas predicted automatic political preferences at the different levels of abstraction: activation in the ventral striatum was positively correlated with preference ranks for unattended politicians, whereas participants' preferences for the affiliated political parties were reflected in activity in the insula and the cingulate cortex. Using an additional donation task, we showed that the automatic preference-related processing in the brain extended to real-world behavior that involved actual financial loss to participants. Together, these findings indicate that brain responses triggered by unattended and task-irrelevant political images reflect individual political preferences at different levels of abstraction.

Highlights

► Brain responses reflect political preferences without deliberation and attention. ► Activation in ventral striatum encodes preferences for task-irrelevant politicians. ► Activation in insula and cingulate cortex reflects preference for associated parties. ► Activation in insula and cingulate cortex co-varies with future donations to parties.

Introduction

In times of elections, huge budgets are spent on campaigns to inform political preferences and convince people to vote for particular candidates and their affiliated political parties. Recent findings indicate, however, that political preferences are by no means a prime exemplar for deliberate decisions but are considerably shaped by fast, automatic processes. Rapid judgments of competence based solely on the facial appearance of candidates were shown to reliably predict the outcome of elections (Ballew and Todorov, 2007, Todorov et al., 2005). Moreover, implicit measures of attitudes that assess automatic evaluative associations (Greenwald et al., 1998) were found to improve the prediction of supposedly deliberate behavior such as political voting (Friese et al., 2007, Galdi et al., 2008, Karpinski et al., 2005). Furthermore, incidental exposure to environmental cues and irrelevant events has been suggested to shape political choices without participants' awareness (Berger et al., 2008, Carter et al., 2011, Hassin et al., 2007, Healy et al., 2010).

Such automatic processing – guiding human judgments and choices in the absence of conscious deliberation – has previously been found to be reflected in brain responses for non-political stimuli. Neural activation has been shown to reflect preferences for paintings, houses and unknown faces when participants evaluated stimuli with respect to other, non-preference-related aspects (Kim et al., 2007, Lebreton et al., 2009). Activation patterns obtained in the absence of conscious deliberation were also reported to predict subsequent preferences for cars even when attention was diverted from potential choice options (Tusche et al., 2010).

Based on this evidence, we investigated whether brain responses track political preferences when political stimuli (i.e., images of national politicians) are presented to participants outside the focus of attention. We hypothesized that preferences for politicians might be encoded in brain areas such as the ventral striatum (VS), the medial prefrontal cortex (mPFC), the anterior cingulate cortex (ACC), and the insula that have previously been shown to be involved in automatic valuation and incidental processing of popularity of socially tagged stimuli (Kim et al., 2007, Lebreton et al., 2009, Mason et al., 2009, Tusche et al., 2010). Following up on findings that preferences for politicians can be predicted based on rapid inferences from viewing their faces, we assumed that judgments based on the visual appearance of political candidates might mediate these preference judgments (Antonakis and Dalgas, 2009, Ballew and Todorov, 2007, Spezio et al., 2008, Todorov et al., 2005). Given the human capacity of rapid face recognition and automatic retrieval of person knowledge (Gobbini and Haxby, 2007, Todorov et al., 2007), we further hypothesized that task-irrelevant images of prominent politicians might automatically activate mental representations of affiliated parties. Hence, in a second step, we examined whether brain responses obtained during automatic processing of images of national politicians also reflect preferences for associated political parties. Taking advantage of the fact that preferences for a number of German politicians and for their affiliated political parties differ significantly, we used behavioral pretests to identify national politicians who were valued and appreciated, independent of participants' attitudes towards the associated parties. Likewise, we were able to determine several politicians who were consistently judged as rather unpopular — even if they belonged to the preferred political party (Fig. 1A). This allowed us to select politicians such that participants' valuations of politicians were matched across parties and permitted us to disentangle preferences for associated parties from politician-specific processing.

Participants were instructed to perform a demanding visual fixation task while their brain responses were measured using fMRI. At unpredictable intervals, task-irrelevant images of politicians were passively presented in the background while the fixation task continued. Subsequent to scanning, participants' political preferences were measured both for passively viewed politicians and for affiliated parties. Importantly, during the acquisition of brain responses, participants were not aware that political preference judgments would be required later on. We then investigated whether brain responses reflect participants' preferences for the unattended politicians as well as for the associated political parties. Finally, we tested whether automatic preference-related processing in the brain extends to real-world behavior such as voluntary donations.

Section snippets

Participants

Twenty healthy volunteers (aged between 22 and 33 years, 7 female) participated in the fMRI session and the behavioral posttest. Both sections of the experiment were approved by the local ethics committee. All participants were German native speakers, free of psychiatric or neurological history, had normal or corrected-to-normal vision, were right-handed and gave written informed consent. Participants were paid a fixed amount of €12 to take part in the study plus 20% of the remainder of an

Party preference

Self-reported party preference (Likert scale of − 5 to + 5) was determined by subtracting individual liking ratings for PA (Mean ± SD: − 1.56 ± 3.05, range of − 5 to + 4) from those for PB (1.11 ± 1.81, range of − 3 to + 4; at p < 0.05, two tailed paired t-test). Positive values of the preference score (2.67 ± 4.00, range of − 4 to + 9) indicated that PB was favored over PA. Implicit measures of party preference (D scores of IAT, Greenwald et al., 2003) confirmed these self-report assessments for all but one

Discussion

Research has begun to investigate neural activation that underlies the processing of political attitudes and preferences (Amodio et al., 2007, Zamboni et al., 2009), including studies on deliberative processing of political statements and simulated voting for political candidates (Bruneau and Saxe, 2010, Gozzi et al., 2010, Rule et al., 2010, Spezio et al., 2008, Westen et al., 2006). Yet the neural substrate underlying automatic processing of political preferences (Ballew and Todorov, 2007,

References (47)

  • M. van't Wout et al.

    Friend or foe: the effect of implicit trustworthiness judgments in social decision-making

    Cognition

    (2008)
  • D.M. Amodio et al.

    Neurocognitive correlates of liberalism and conservatism

    Nat. Neurosci.

    (2007)
  • J. Antonakis et al.

    Predicting elections: child's play!

    Science

    (2009)
  • C.C. Ballew et al.

    Predicting political elections from rapid and unreflective face judgments

    Proc. Natl. Acad. Sci. U. S. A.

    (2007)
  • J. Berger et al.

    Contextual priming: where people vote affects how they vote

    Proc. Natl. Acad. Sci. U. S. A.

    (2008)
  • T.J. Carter et al.

    A single exposure to the American flag shifts support toward Republicanism up to 8 month later

    Psychol. Sci.

    (2011)
  • A.D. Engell et al.

    Implicit trustworthiness decisions: automatic coding of face properties in the human amygdala

    J. Cogn. Neurosci.

    (2007)
  • E.B. Falk et al.

    Predicting persuasion-induced behavior change from the brain

    J. Neurosci.

    (2010)
  • E.B. Falk et al.

    Neural activity during health messaging predicts reductions in smoking above and beyond self-report

    Health Psychol.

    (2011)
  • M. Friese et al.

    Predicting voting behavior with implicit attitude measures. The 2002 German parliamentary election

    Exp. Psychol.

    (2007)
  • K.J. Friston et al.

    Statistical parametric maps in functional imaging: a general linear approach

    Hum. Brain Mapp.

    (1995)
  • S. Galdi et al.

    Automatic mental associations predict future choices of undecided decision-makers

    Science

    (2008)
  • M. Gozzi et al.

    Interest in politics modulates neural activity in the amygdala and ventral striatum

    Hum. Brain Mapp.

    (2010)
  • Cited by (29)

    • Mind your privacy: Privacy leakage through BCI applications using machine learning methods

      2020, Knowledge-Based Systems
      Citation Excerpt :

      Although the researchers claim that additional research is needed, their results represent a step toward proving that, with the right stimuli, the activation level in different parts of the brain can be indicative of the subject’s racial and gender beliefs. Other studies using both fMRI [39] (which will not be elaborated upon here) and EEG [38] to measure brain activity have shown that a subject’s neural activity changes when reacting to different political candidates. Those changes are correlated with the subject’s opinion about them.

    • A decade of decoding reward-related fMRI signals and where we go from here

      2018, NeuroImage
      Citation Excerpt :

      Other studies have focused on decoding the subjective value of offers (Krajbich et al., 2009; Wang et al., 2014), the value of predicted outcomes (Kahnt et al., 2010, 2011b), consumer choices made inside the scanner (Grosenick et al., 2008), and differences between types of valuation and reward (Clithero et al., 2009, 2011). Moreover, several studies used brain activity acquired during passive viewing (Levy et al., 2011), or even outside the focus of attention (Tusche et al., 2010, 2013), to predict preference-based choices made outside the scanner (Smith et al., 2014). For instance, Tusche et al. (2010) presented a series of images of automobiles in the background while subjects were engaged in a demanding visual fixation task.

    • Multivariate pattern analysis of event-related potentials predicts the subjective relevance of everyday objects

      2017, Consciousness and Cognition
      Citation Excerpt :

      Others have shown that the processing of stimuli on a category level occurred rapidly following the presentation of a natural scene, even though the scene was unattended and irrelevant to the experimental task (Peelen, Fei-Fei, & Kastner, 2009). Brain activity has also been found to be systematically modulated according to individual preferences for specific stimuli, even when attention was directed away from the stimuli (Lebreton, Jorge, Michel, Thirion, & Pessiglione, 2009; Tusche, Bode, & Haynes, 2010; Tusche, Kahnt, Wisniewski, & Haynes, 2013). However, given the poor temporal resolution of fMRI, it remains unclear how rapidly the suggested processes took place.

    • Political Neuroscience

      2016, Neuroimaging Personality, Social Cognition, and Character
    • Pre-existing brain states predict risky choices

      2014, NeuroImage
      Citation Excerpt :

      In sum, our findings suggest that areas in the basal ganglia both predict and reflect gamble choices. Previous studies have shown that information within brain regions can be used to predict subsequent choices about economic risk, consumer products, and even social targets (Christopoulos et al., 2009; Hampton and O'Doherty, 2007; Knutson and Bossaerts, 2007; Levy et al., 2011; Tusche et al., 2010, 2013). Our study differs from these studies in that they predicted upcoming decisions using brain activation after options were presented (Knutson and Bossaerts, 2007; Knutson et al., 2007; Tusche et al., 2010), or after reward feedback for the prior trial (Hampton and O'Doherty, 2007).

    • Nonpolitical images evoke neural predictors of political ideology

      2014, Current Biology
      Citation Excerpt :

      Our results have important implications for the links between biology, emotions, political ideology, and human nature more fundamentally. While previous studies using skin conductance response [9–11], neuroimaging [21–24], and questionnaire [25, 26] measures suggested the role of emotions in political attitudes, to our knowledge, this is the first fMRI study revealing multivariate patterns of brain activity that differ between liberals and conservatives during emotional processing of sensory stimuli. Accumulating evidence suggests that cognition and emotion are deeply intertwined [27], and a view of segregating cognition and emotion is becoming obsolete [28].

    View all citing articles on Scopus
    View full text