Science & Society

The neuroscience of motivated cognition

Brent L. Hughes and Jamil Zaki
Department of Psychology, Stanford University, Stanford, CA 94305, USA

Goals and needs shape individuals’ thinking, a phenome- mechanisms through which motivation affects cognition
non known as motivated cognition. We highlight research (Box 1), but it has only recently become a focus of neurosci-
from social psychology and cognitive neuroscience that entific research. These recent efforts converge with behav-
provides insight into the structure of motivated cognition. ioral approaches to provide insights into the structure of
In addition to demonstrating its ubiquity, we suggest that motivated cognition (Box 2). Together, these efforts suggest
motivated cognition is often effortless and pervades that motivated cognition pervades information processing at
information processing. various stages, including perception, attention, and deci-
sion-making. Although these processes are intricately relat-
ed and frequently interact, we use them here as an
The pervasive influence of motivation on cognition organizing principle to briefly summarize how motivations
People often believe that their thinking aims squarely at affect what people see, how they think, and what they decide.
gaining an accurate impression of reality. Upon closer
inspection, this assumption collapses. Instead, like the Perception
inhabitants of Garrison Keillor’s Lake Woebegon, individ- A research tradition dating back to the ‘New Look’ move-
uals often see themselves and close others as possessing ment demonstrates that motives influence what people see.
unrealistically high levels of positive attributes such as For example, people tend to see desirable objects as being
likeability, morality, and attractiveness. This bias persists physically closer to them, and tend to imbue ambiguous
among individuals who should know better: over 90% stimuli (e.g., similar shades of a color) with interpretations
college professors believe their work is better than that associated with reward [5]. Recent neuroscientific investi-
of their peers, CIA analysts overestimate the accuracy of gations echo these findings. In one recent study, partici-
their predictions for future events, and doctors overconfi- pants viewed arrays of moving dots whose direction of
dently estimate their medical knowledge [1]. motion was difficult to discern, and decided which direction
These cases exemplify the phenomenon of motivated they were predominantly moving [6]. Before seeing these
cognition, by which the goals and needs of individuals steer arrays, participants learned that they would receive a re-
their thinking towards desired conclusions [1–3]. A variety of ward if the dots predominantly moved in a particular direc-
motivations pervasively shapes cognition (Box 1). For exam- tion. People tended to see the dots moving in the direction
ple, people wish to live in a coherent and consistent world. associated with reward, and did so by biasing their visual
This leads people to recognize patterns where there are search in favor of the desired perception. Moreover, occipital
none, perceive control over random events, and shift their regions involved in perceptual encoding, and prefrontal
attitudes to be consistent with their past behaviors [2,3]. Peo- regions involved in top-down control, were modulated by
ple also need to feel good about themselves and about others these motivational effects on vision (Figure 1 and [6]).
with whom they identify. As such, people often self-enhance, Motivation also shapes people’s visual perception of
evaluating themselves as having more desirable personal- themselves and others. For example, people see their
ities and rosier future prospects than their peers, and taking own faces and those of close others as being more physically
personal credit for successes, but not failures [1–3]. People attractive than they really are [7], and in-group faces as
likewise elevate their relationship partners and in-group being more distinct and likable than out-group faces [4]. In
members (e.g., people who share their political affiliation) in one recent study, participants viewed images of their own
demonstrably unrealistic ways [3,4]. Motivations can also face morphed with positive (i.e., trustworthy) or negative
have the opposite effect, leading people to derogate out- (i.e., untrustworthy) faces, and evaluated the extent to
group members, even when the lines that divide ‘us’ from which each morph looked like them [7]. Participants iden-
‘them’ are defined de novo by researchers [3,4]. tified more-trustworthy morphs as being more similar to
themselves than were less-trustworthy morphs. As the
Motivated cognition across the information processing similarity between the untrustworthy morphs and the self
stream decreased, activation in occipitotemporal structures asso-
Motivation not only shapes what people think, but also how ciated with facial encoding and prefrontal regions involved
they think. Decades of psychological research identify in top-down control increased. Taken together, these data
suggest that motivation influences lower-level perception
Corresponding author: Hughes, B.L. (blhughes@[Link]). towards desired conclusions and away from undesired
Keywords: motivation; self-enhancement; cognitive dissonance; intergroup bias;
cognitive control; automaticity. conclusions. In addition, they suggest that top-down pro-
1364-6613/
cesses (reflected by prefrontal cortical activation) influence
ß 2014 Elsevier Ltd. All rights reserved. [Link] lower-level perception by biasing visual search in the
direction of desired conclusions [8].
62 Trends in Cognitive Sciences, February 2015, Vol. 19, No. 2
Science & Society Trends in Cognitive Sciences February 2015, Vol. 19, No. 2

Box 1. Three principles of motivated cognition

(A) (B)
Classic social psychological research makes several key points
about motivated cognition: LPFC LOFC
It is pervasive: motives affect a wide array of judgments and
perception, including self- and other-enhancement, illusions of
control, confirmation bias, and in-group bias [2–4]. STS
It is goal-directed: as with other motivated processes, biased
cognition tracks the current need of an individual to fulfill a relevant
goal. For example, people maximally self-enhance after being Occipital FFA
STS
ostracized or negatively evaluated – consistent with an increased
need to self-protect – and reduce enhancement following self-
affirmation (e.g., focusing on core values) – consistent with a (C) (D)
reduced need to self-protect [1–3]. Likewise, threats to group-related
motives (e.g., intergroup competition) increase in-group biases, and Key:
reducing group-related motives by highlighting commonalities
between groups reduces these biases [4]. Finally, individual Percepon
differences associated with chronically elevated motives (e.g.,
narcissism) predict bias [3,4]. Aenon
It is impactful: motivated cognition has far-reaching consequences, MOFC
Decision
and its adaptiveness has been debated. On one hand, self-
enhancement predicts positive adjustment in the face of adversity
[1]. On the other hand, motivated biases are often harmful TRENDS in Cognitive Sciences

[1]. People who are overoptimistic about their health fail to take
Figure 1. Neural systems modulated by motivated cognition. Each point
preventative health measures. Overconfident students show poor corresponds to an activation peak that tracks bias, shape-coded in relation to
skill retention and academic disengagement. Overconfident CEOs processing stage. Activations are projected to the surface of the brain for
take unwarranted risks. Finally, people who self-enhance are illustrative purposes. Abbreviations: FFA, fusiform face area; LOFC, lateral
perceived as arrogant and maladjusted. orbitofrontal cortex; LPFC, lateral prefrontal cortex; MOFC, medial orbitofrontal
cortex; STS, superior temporal sulcus. The key indicates which shapes correspond
to motivation’s influence along the information processing stream.
Attention re-evaluated their personal chances of having each event
People often direct attention towards information that befall them [9]. Participants adjusted their personal esti-
supports desired conclusions [2]. For instance, individuals mates when base-rate information was better, but not
quickly accept information that casts them in a favorable worse, than their initial estimate. Crucially, reduced ac-
light (e.g., as being predisposed towards success), but tivity in lateral prefrontal cortex (LPFC) – a region in-
carefully scrutinize threatening information (e.g., which volved in attentional control [10] – during the presentation
suggests they are likely to fail) [2]. Motives likewise influ- of more pessimistic base-rates predicted an optimistic bias
ence neural systems involved in attentional control in their likelihood estimates [9]. Taken together, these
(Figure 1), diverting attention towards desirable infer- findings suggest that people may heed motives by ignoring
ences and away from undesired ones. For example, in a information that threatens their desired conclusions. This
recent study, participants rated the likelihood that they suggests that, in particular situations, tuning attention
would experience various unpleasant future events (e.g., towards undesirable or threatening information may help
robbery, Alzheimer’s disease), then learned the base-rate people to reduce their motivated biases (Box 2).
probability of experiencing those events, and finally
Decision-making
People often (implicitly or explicitly) exploit convenient
Box 2. The influence of motivation on cognition: effortful or heuristics to reach desired conclusions [2]. For example,
effortless? people draw on accessible information in a biased manner
One way in which neuroscience informs the study of motivated (e.g., by recalling a time they acted intelligently rather than
cognition is by helping to uncover whether motivated biases are foolishly) to uphold positive self-evaluations [2]. A recent
effortful or effortless. One possibility is that motivated cognition study examined reliance on such cognitive loopholes when
reflects effortful attempts to distort information and reach desired
conclusions. However, recent evidence suggests that motivation
self-enhancing [11]. Participants evaluated themselves in
may instead effortlessly influence cognition. Motivated biases comparison to their average peer on broad (e.g., ‘talent’,
exhibit features of automaticity, for instance arising quickly [5], ‘intelligence’) versus narrow (e.g., ‘cleanliness’, ‘tidiness’)
and even when cognitive resources are limited [11]. Second, the traits. When making broad inferences, people can self-en-
most consistent pattern of brain activity associated with motivated hance through reliance on goal-relevant availability heur-
bias is reduced neural activation in structures associated with
deeper information processing [10,12]. For instance, reduced LPFC
istics because some evidence for their talent or intelligence
[9] and OFC [11] activation is associated with more positively biased (e.g., math, social skills) will likely be available. Narrower
evaluations of the self, close others, and in-group members. Taken inferences (e.g., concerning tidiness) afford less opportunity
together, the prepotency of biased cognition and its negative for using such heuristics, constraining positivity biases.
relation to activity in these regions suggest that motivated cognition
Consistent with this idea, participants evaluated them-
may be effortless. This is not to say that bias is always effortless. For
instance, when threatened, people protect the self by applying selves as being better-than-average on broad but not narrow
different standards during self-evaluation, an effect accompanied by traits. Further, the degree to which people self-enhanced
increased OFC activation [13–15]. The effortful or effortless nature of was accompanied by reduced activity in medial and lateral
motivated cognition is likely not fixed, but rather depends on the orbitofrontal cortex (OFC) – regions associated with inte-
current motivational state of an individual.
grating affective and contextual information in the service of
63
Science & Society Trends in Cognitive Sciences February 2015, Vol. 19, No. 2

decision-making (Figure 1 and [12,13]). These findings raise accountable to a third party (e.g., by requiring a justification
the possibility that people may arrive at flattering views of for decisions) reduces effortless biases [13]. At other times –
themselves and close others by failing to recruit the OFC, when the self is threatened – affirming core values or
and this may allow goal-relevant heuristics to be easily highlighting shared identity may reduce people’s sense of
deployed (e.g., when considering broad traits). This suggests threat, as well as reactive biases in response to that threat
that a deeper consideration of desirable and undesirable (Box 1). Advances in our understanding of motivated cogni-
information may reduce positivity biases (Box 2 and [13]). tion could provide individuals and policymakers with strat-
At other times, people shift their attitudes to reduce egies to influence motivation in ways that reduce harmful
discomfort with inconsistency [2]. In classic demonstra- biases. These strategies may allow people to gain self-in-
tions, people are asked to make statements or perform sight, maximize adaptive decision-making, as well as to
behaviors that contradict previously held beliefs (e.g., reduce favoritism towards close others and in-group mem-
performing a boring task and subsequently telling another bers, and improve intergroup relations.
participant that the task was enjoyable). Contradictions
between one’s actions and experience threaten the self- Acknowledgments
image of an individual as being moral and intelligent. We thank Jamil Bhanji, Nick Camp, Sylvia Morelli, Takuya Sawaoka,
Motivated to dispel this threat, people shift their beliefs and Melina Uncapher for helpful comments on prior versions of this
manuscript.
to make them more consistent with their behavior (e.g., by
evaluating the boring task more positively [2]). References
Recent neuroscientific evidence sheds light on the pro- 1 Dunning, D. et al. (2004) Flawed self-assessment implications for
cesses underlying these shifts in belief. In one study, health, education, and the workplace. Psychol. Sci. Pub. Int. 5, 69–106
participants were asked to make difficult choices between 2 Kunda, Z. (1990) The case for motivated reasoning. Psychol. Bull. 108,
480
options they found to be equally attractive, and subse- 3 Taylor, S.E. and Brown, J.D. (1988) Illusion and well-being: a social
quently re-evaluated their attitudes towards those options psychological perspective on mental health. Psychol. Bull. 103,
[14]. People evaluated the chosen options as being signifi- 193–210
cantly more desirable than they had before making any 4 Cikara, M. and Van Bavel, J.J. (2014) The neuroscience of intergroup
choices. Crucially, the degree to which people shifted their relations: an integrative review. Perspect. Psychol. Sci. 9, 245–274
5 Dunning, D. and Balcetis, E. (2013) Wishful seeing: how preferences
attitudes was accompanied by increased activation in OFC. shape visual perception. Curr. Dir. Psychol. Sci. 22, 33–37
These findings suggest that shifts in attitude as a way to 6 Mulder, M.J. et al. (2012) Bias in the brain: a diffusion model analysis of
reduce dissonance and threat may be more deliberative prior probability and potential payoff. J. Neurosci. 32, 2335–2343
than other forms of self-enhancement (Box 2 and [13,15]). 7 Verosky, S.C. and Todorov, A. (2010) Differential neural responses to
faces physically similar to the self as a function of their valence.
Neuroimage 49, 1690–1698
Concluding remarks 8 Summerfield, C. and Egner, T. (2009) Expectation (and attention) in
Budding neuroscientific endeavors have begun identifying visual cognition. Trends Cogn. Sci. 13, 403–409
mechanisms through which motivation affects cognition in 9 Sharot, T. et al. (2011) How unrealistic optimism is maintained in the
various domains including self-perception, person percep- face of reality. Nat. Neurosci. 14, 1475–1479
10 Braver, T.S. (2012) The variable nature of cognitive control: a dual
tion, and intergroup relations. Although often considered
mechanisms framework. Trends Cogn. Sci. 16, 106–113
separately, these social contexts all activate goals that bias 11 Beer, J.S. and Hughes, B.L. (2010) Neural systems of social comparison
information processing. These motives also pervade infor- and the ‘above-average’ effect. Neuroimage 49, 2671–2679
mation processing at several stages including visual percep- 12 Roy, M. et al. (2012) Ventromedial prefrontal-subcortical systems
tion, attention, memory, and decision-making. Motivated and the generation of affective meaning. Trends Cogn. Sci. 16,
147–156
cognition highlights the idea that even presumably low-level 13 Hughes, B.L. and Beer, J.S. (2012) Medial orbitofrontal cortex is
perceptions and inferences – which at first blush appear to associated with shifting decision thresholds in self-serving cognition.
be devoted to objective information gathering – are subject Neuroimage 61, 889–898
to motivational forces. 14 Jarcho, J.M. et al. (2011) The neural basis of rationalization: cognitive
Understanding the structure of motivated cognition is dissonance reduction during decision-making. Soc. Cogn. Affect.
Neurosci. 6, 460–467
crucial to reducing the harmful sequelae of biased thinking 15 Hughes, B.L. and Beer, J.S. (2013) Protecting the self: the effect of
(Box 1). Strategies that promote deeper information proces- social-evaluative threat on neural representations of self. J. Cogn.
sing can constrain bias (Box 2). For example, holding people Neurosci. 25, 613–622

64