DOI: 10.1126/science.1188595 , 47 (2010); 329Science

et al.Ruud Custers,Operates Outside of Conscious AwarenessThe Unconscious Will: How the Pursuit of Goals

This copy is for your personal, non-commercial use only.

. clicking herecolleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others

. herefollowing the guidelines can be obtained byPermission to republish or repurpose articles or portions of articles

(this information is current as of July 1, 2010 ):The following resources related to this article are available online at www.sciencemag.org

http://www.sciencemag.org/cgi/content/full/329/5987/47version of this article at:

including high-resolution figures, can be found in the onlineUpdated information and services,

http://www.sciencemag.org/cgi/content/full/329/5987/47#otherarticles, 12 of which can be accessed for free: cites 45 articlesThis article

http://www.sciencemag.org/cgi/collection/psychologyPsychology

: subject collectionsThis article appears in the following

registered trademark of AAAS. is aScience2010 by the American Association for the Advancement of Science; all rights reserved. The title

CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience

on

July

1, 2

010

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

The Unconscious Will: How thePursuit of Goals Operates Outsideof Conscious AwarenessRuud Custers* and Henk Aarts*

People often act in order to realize desired outcomes, or goals. Although behavioral sciencerecognizes that people can skillfully pursue goals without consciously attending to their behavioronce these goals are set, conscious will is considered to be the starting point of goal pursuit.Indeed, when we decide to work hard on a task, it feels as if that conscious decision is the first andforemost cause of our behavior. That is, we are likely to say, if asked, that the decision to actproduced the actions themselves. Recent discoveries, however, challenge this causal status ofconscious will. They demonstrate that under some conditions, actions are initiated even though weare unconscious of the goals to be attained or their motivating effect on our behavior. Here weanalyze how goal pursuit can possibly operate unconsciously.

As humans, we generally have the feelingthat we decide what we want and whatwe do. These self-reflections remind us

that we are not bound to the present environ-ment for our actions: We can envision ourselvesin different places, in alternative futures, doingdifferent things. We only have to decide to do so,and we can go see a movie tonight or hang outwith friends in a bar. It is up to us. Our behaviorsseem to originate in our conscious decisions topursue desired outcomes, or goals.

Scientific research, though, suggests other-wise. In a remarkable experiment conductedmore than 25 years ago (1), research participantswere instructed to freely choose when to movetheir index fingers while the timing of the actionitself, of its preparation in the brain, and ofwhen the person became aware of the decisionto act were measured. Although the decision didindeed precede the action, the preparation of thefinger movement in the brain was well on its wayby the time people consciously decided to act.Apparently, when people are persuaded toconsciously set a goal to engage in behavior, theirconscious will to act starts out unconsciously.

The finding that the pursuit of the goals thatwe consciously set and adopt is prepared uncon-sciously, at least in the earliest moments before weact on them, is intriguing. Recent research in socialcognition, however, goes even one step further. Thisresearch shows that goals themselves can arise andoperate unconsciously. Social situations and stimu-li in the surroundings activate or prime goals inpeople’s minds outside of their awareness, therebymotivating and guiding them, for example, toworkharder on a task (2), to reach out a helping hand toothers even when facing obstacles (3), or to ensure

that they can socialize and hang out with friends(4). Thus, goals and their pursuit can be influencedby unconscious sources, and these goals do notneed to be consciously set and adopted before theirinfluence begins to operate (Fig. 1).

A Brief ChronologyThe notion that the pursuit of goals can occurunconsciously is reminiscent of Sigmund Freud,

who proposed that our (often sexual) desires aresuppressed and banished to the dark corners ofthe mind but pop up in hysteria and under hyp-nosis. Whereas Freud’s complex theory on theunconsciouswas largely unfalsifiable (5), research-ers in the behaviorist tradition built more-testabletheories, according towhich neither consciousnessnor cognition but rigid responses to environmentalstimuli determine behavior.

Obviously, the environment plays a crucialrole in directing behavior. However, acting onfixed stimulus-response rules—such as smashinga beeping alarm clock in the morning—is not thewhole story. A substantial part of human be-havior can only be explained by assuming thatpeople have goals in mind that direct their be-havior in a dynamic world (6). Cognitive sci-entists indeed proposed that the flexibility toproduce the same desired outcomes under vary-ing circumstances comes from our capacity tomentally represent what we want and do: to build

and store mental representationsof goals. These goal representationsfunction as beacons for behavior,motivating action and guiding itscourse (7).

For a long time, it was generallyassumed that many of the mentalprocesses that make goal pursuitpossible require consciousness. Butin the past decade or so, the sci-entific study of goal pursuit hasdiscovered that these processescan also operate without consciousawareness, and hence, human be-havior may originate in a kind ofunconscious will. This recent evi-dence that goal pursuit can occurwithout people being conscious ofthe active goal or its influence on theirmotivation and behavior has beenmet with resistance and skepticism,perhaps partly due to its far-reachingimplications for our understanding ofconsciousness and for our view ofwhat it is to be human (8). Further-more, scientists have not come togrips with the potential redundancyof consciousness in (seemingly) voli-tional behavior, because the mecha-nism by which the activation of goalrepresentations can produce goalpursuit unconsciously is not fullyunderstood. Understanding thismechanism is especially important

because unconscious goal pursuit is proposed toplay a key role in many aspects of social life, suchas consumer and health behavior, moral behavior,and social discrimination (9).

Here we review research demonstrating thatgoals and the motivation to pursue them can ariseunconsciously, and we propose a mechanism forhow this may happen. This proposed mechanismis based on the idea that, in principle, the mind

REVIEW

Department of Psychology, Utrecht University, Heidelberglaan 1,3584 CS Utrecht, Netherlands.

*To whom correspondence should be addressed. E-mail:r.custers@uu.nl (R.C.); h.aarts@uu.nl (H.A.)

Fig. 1. The painting Achilles Slays Hector by Peter Paul Rubensdepicts a scene from the Iliad in which no reference whatsoeveris made to conscious decisions or intentions (52). Instead, thepursuits of Achilles and the other characters are determined byexternal factors, such as fate or the gods. Here we argue thatalthough people may have the feeling that their behavior is theresult of their conscious decisions, their goal pursuits too areoften directed by external sources of which they are not conscious.

www.sciencemag.org SCIENCE VOL 329 2 JULY 2010 47

on

July

1, 2

010

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

(and the brain in which it resides) is designed foraction, and continuously and largely uncon-sciously processes behavioral-relevant infor-mation to readily “tell” its owner what she wantsand should do to deal with the opportunities andchallenges presented by the environment. Thus,setting, pursuing, and realizing goals can occurwithout conscious interventions.

Evidence for Unconscious Goal PursuitUnconscious goal pursuit was first systematicallyexamined in social psychological experimentsthat made use of so-called “unrelated studies”setups. Bargh and colleagues (2) used such asetup to prime the goal to achieve—a desiredoutcome most people strive to attain—in U.S.students without them becoming aware of beinginfluenced. Students were seated at a table towork on two seemingly unrelated language puz-zles. For some students, the first puzzle includedwords related to achievement (such as win orachieve), and for others it did not. Students whowere exposed to achievement words were foundto outperform the others on the second puzzle.Furthermore, achievement priming was found toprompt behavioral qualities that are characteristicof motivational states or volition, such as persist-ence in solving puzzles and increased flexibilityon the Wisconsin Card Sorting Task (10), a stan-dard measure of flexibility in cognitive process-ing (11). Extensive debriefing revealed that thestudents did not perceive an influence of the firsttask (in which they were exposed to consciouslyvisible achievement-related words) on their re-sponses to the second.Hence, the effect of achieve-ment priming on subsequent performance andcognitive flexibility was likely to be the result ofunconscious processes.

Several experiments using the unrelatedstudies setup have replicated these goal-primingeffects with different goals and different primes.It has been found that reading words related tocooperation causes people to work together ineconomic games (2) and that perceiving wordsdescribing occupations associated with makingmoney (such as stockbroker) or inferring this goalfrom another person’s actions (such as a personoperating a slot machine) makes them workharder when money is at stake (12). Furthermore,it has been shown that people’s pursuits are in-fluenced by subtle cues in the environment out-side their awareness: Upon entering an office,people become more competitive when seeing aleather briefcase placed on the desk (13), talkmore softly when looking at a library picture onthe wall (14), and clean their table more whenthere is a vague scent of cleaning agent in the air(15). Together, these results show that goalpursuit is influenced and controlled unconscious-ly by social features that have become associatedwith goals, either through direct practice or throughsocial norms, communicationwith important others,or the media.

The studies on unconscious goal pursuitalluded to above, however, are sometimes

criticized for allowing participants to be awareof the primes. Even though participants reportbeing unaware of the influence of the goalpriming on their behavior, they still could haveformed conscious intentions at the moment whenthey consciously perceived the goal information.Hence, their goal pursuit may still have beencaused by their conscious will.

To offer even more compelling evidence forunconscious goal pursuit, researchers haverecently resorted to more stringent methods suchas subliminal stimulation, which prevents con-scious perception of the primes. Subliminalstimulation refers to the presentation of stimuliwith an intensity that is too low to reach thethreshold of conscious awareness. Typically,people cannot consciously detect these stimuli,but they are nevertheless influenced by them.Whether subliminal stimulation can convey

meaningful information has been debated forquite some time (16). However, recent findingsprovide compelling evidence that subliminalprimes affect people’s responses (17), activatesemantically related knowledge (18), and eveninfluence cognitive control in tasks (19). Build-ing on these findings, research has demonstratedeffects of subliminal stimulation on goal pursuit,such as increased task performance after primingof achievement-related words (20), enhancedfluid consumption in a taste task after primingof drinking-related words (21), and an increasein instrumental behavior leading to specific goals(such as helping another person by providinguseful comments) after priming of names ofsignificant others (such as a good friend) oroccupations (such as nurse) associated with thesegoals (12, 22).

It is important to note that in most studies onsubliminal goal priming, people are asked inretrospect to indicate whether they were moti-

vated to pursue the primed goal. The generalfinding of these checks is that although people’sreported motivation often correlates with theirbehavior (people who worked harder reportedthat they were more motivated), these reports arenot influenced by the primes. This suggests thatthe reason that subliminal priming of the goalaffects goal pursuit is not that people becomeconscious of their motivation to pursue the goalafter it is primed. Participants may becomeconscious of their motivation after the behavioris performed and when they are explicitly askedto reflect on it. In other words, the reportedconscious experience of pursuing goals may bean inference rather than the cause of goal pursuit(23, 24).

In sum, a large body of research indicates thatthe pursuit of goals can be evoked outside ofawareness. People become motivated to initiateand exhibit behaviors available in their repertoirewhen goals that are represented as desiredoutcomes are primed, even though they are notaware of the primed goal or its effect on theirmotivation and behavior. But how can thishappen?

A Mechanism for Unconscious Goal PursuitTo understand how the pursuit of goals occursunconsciously, it is important to examine thefunctions that make motivated, goal-directedbehavior possible. Fortunately, behavioral scien-tists have worked hard to reduce the processesinvolved in goal pursuit to a few basic principles.Whereas these principles have been modeled indifferent psychological terms, most models ofgoal pursuit share the three following basicfeatures: a person (i) takes a possible outcomeor goal in mind; (ii) considers whether the actionsand resources to attain the outcome are available;and (iii) assesses the value of the outcome (that is,the extent to which it is rewarding or desirable)(25). Thus, whether people set an outcome thatcomes to mind as a goal to pursue depends on itsattainability and desirability.

The assessment of this attainability anddesirability of a goal is considered to requireconsciousness bymost dominant theories on goalpursuit. Yet the evidence for unconscious goalpursuit suggests that this does not have to be thecase. The notion that goal pursuit can operateunconsciously sounds ridiculous to some peoplebut sensible to others. Sensible or not, theincreasing evidence on this issue is fairly silentabout the mechanisms that allow people topursue goals without conscious awareness. Wewill now discuss a mechanism for unconsciousgoal pursuit and demonstrate that people canunconsciously detect the reward value of aprimed goal and prepare feasible actions thatmake the goal attainable. We will then show howthese two processes work together to producegoal pursuit outside people’s conscious aware-ness (Fig. 2).

Unconscious action preparation and execution.People may often become conscious of the actions

Unconscious activationof a goal representation

Preparationof

action

Detectionof positive

reward signal

Goal pursuit

Fig. 2. The proposed mechanism for unconsciousgoal pursuit.

2 JULY 2010 VOL 329 SCIENCE www.sciencemag.org48

REVIEW

on

July

1, 2

010

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

they prepare and execute, but their consciousknowledge of what exactly they do to reach a goalis surprisingly limited (26). Consider the musclecontractions and relaxations that make our armgrab a cup of coffee. Computational models havebeen proposed that describe how motoric pro-cesses and sensory feedback work together tocontrol this behavior (27), but on a conscious levelwe have no idea how we do it. We just think ofgrabbing the cup and it happens.

We are able to initiate actions by thinkingabout their outcomes, because actions and theiroutcomes are associated on a perceptual, sensory,and motor level (28). Through prior learning,certain patterns of muscle contraction and re-laxation have become associated with their ob-servable outcomes (such as grabbing and lifting acup). Because of these associations, bringing tomind the representation of an outcome preparesand controls perception and action to produce theoutcome without much thought (6, 29). This way,action follows from an ideomotor principle (30):Themere activation of the idea of a behavioral actor outcomemoves and programs the human bodywithout a conscious decision to act. Research onsocial cognition and neuroscience has indeedrevealed that merely seeing or reading about abehavioral act or outcome immediately increasesthe tendency to realize it, even when this “idea” istriggered outside of conscious awareness (31, 32).

The ideomotor principle does not hold onlyfor the preparation and execution of simple goal-directed responses such as grabbing a cup, butalso for cognitive control and more complexsocial behavior. For instance, in a recent exper-iment (19), participants had a goal to judge wordsin terms of either sound or meaning, dependingon the visual cue preceding theword. Sometimes,the cue that corresponded to the opposite taskwas subliminally presented before the actual cue.It was found that the subliminally primed cueenhanced brain activity in the cortical areasrelated to the corresponding goal (involved ineither auditory or semantic processing), whereasactivity related to the consciously cued goal wasreduced. The cognitive control system, in otherwords, responded to the subliminal cues byselecting and preparing the execution of thecorresponding goal.

Research on how people perform more com-plex behavior has also demonstrated that when theyfrequently select a goal (such as going to work),they are not only able to orchestrate and executeactions that are instrumental in realizing it withoutconscious attention to their behavior (absentmind-edly driving one’s car to the office). Priming thegoal immediately selects the actions themselves(33). As such, the attainment of many goals is astraightforward affair: People automatically selectand execute behaviors available in their repertoirewhen a goal is primed, and unconsciously adjusttheir behavior based on perceptual input in thecurrent situation to reach it (26, 34).

Unconscious reward processing and motiva-tion. The pursuit of goals does not only involve

the preparation and execution of the properbehavior upon the priming of a goal representa-tion. Because the world is a dynamic place that isfull of opportunities and distractions, peopleshould be both flexible (performing actions innew settings or switching from one goal to theother) and persistent (keeping one’s eye on theselected goal) to optimize goal pursuit. Peopletherefore also take into account the value orrewarding properties of the goal, because this tellsthemwhether it is warranted to invest the effort orrecruit the resources necessary for maintainingtheir behavior, overcoming obstacles, or deviatingfrom routines to attain the goal. Hence, eventhough priming a goal prepares and programsactions unconsciously, whether this goal willflexibly control information processing and be-havior depends on whether it is worth pursuing.This crucial step from preparing actions toactually pursuing a goal is assumed to require anact of conscious will (35, 36). So, can peopledetermine whether it is worth pursuing a givengoal and invest effort in attaining it without theinvolvement of consciousness? Recent studies onthe basic role of the processing of reward cues inhuman motivation suggest that yes, they can.

Neuroimaging research has discovered thatreward cues are processed by limbic structuressuch as the nucleus accumbens and the ventralstriatum. These subcortical areas play a centralrole in determining the rewarding value ofoutcomes and are connected to frontal areas inthe cortex that facilitate goal pursuit (37). Thesereward centers in the brain respond to evolution-arily relevant rewards such as food and sexualstimuli, but also to learned rewards (such asmoney or status), or words (such as good or nice)that are associated with praise or rewards (38).This demonstrates that regardless of their shapeor form, such positive stimuli induce a rewardsignal that is readily picked up by the brain (39).

Other recent research has demonstrated thatsubliminal primes that are specifically related torewards can motivate people to increase the effortthey invest in behaviors. In one study (37),participants could earn money by squeezing ahandgrip. Before each squeeze, the money thatcould be earned was indicated by a 1-pound or 1-penny coin on the screen. Whereas on some trialsthe coin was clearly visible, on others it waspresented subliminally. Thus, effects of consciousand unconscious reward cues could be comparedwithin one experiment. It was found that peoplesqueezed harder on high than on low rewardtrials, regardless of whether the reward wasconsciously visible or not. Moreover, this effectwas accompanied by activation in the brain areasthat play a role in reward processing and therecruitment of effort for action. Similar effects ofunconscious (and conscious) monetary rewardshave been shown in cognitive tasks that requireflexibility and cognitive resources (40, 41). Thesefindings indicate that conscious and unconsciousreward cues have similar effects on effort andflexible cognitive processing, which suggests that

conscious awareness of rewards is not needed forgoal pursuit to occur.

The observation that a variety of reward cuesare encoded by the same brain system to motivatecognition and action and can be processedunconsciously has led to the proposal that apositive reward signal associated with outcomesplays a crucial role in unconscious goal pursuit(42). Specifically, when a desired outcome or goalis primed, activation of the mental representationof this outcome is immediately followed by theactivation of an associated positive affective tag,which acts as a reward signal for pursuing theprimed goal. The positive reward signal attachedto a goal thus unconsciously facilitates the actualselection of the goal and the subsequent mobili-zation of effort and resources to maintain the goal,unless other (more rewarding) goals gain priority(43, 44). This affective-motivational process relieson associations between the representations ofoutcomes and positive reward signals that areshaped by one’s history (for example, when aperson was happy when making money orperforming well). In this case, the goal is said topreexist as a desired state in themind. Priming thisgoal representation not only prepares the appro-priate instrumental actions but also motivatesbehavior, rendering it persistent and flexible,directed at attaining the desired outcome.

We investigated the role of this positivereward signal in the effect of subliminal goal-priming in teenagers and young adults (45). Theywere seated in front of a computer, allegedly totest their computer mouse skills. Before startingon this test, some participants were subliminallyexposed to words related to the goal of socializ-ing on the computer screen, whereas others wereexposed to words unrelated to this goal. At theonset of the mouse-skill test, they were told that ifthere would be enough time left after the test,they could engage in a lottery in which theycouldwin tickets to a popular student party. Thus,spending more effort (by working faster) on themouse-skill test was instrumental in attaining thegoal to socialize. The participants indeed workedharder on the mouse-skill test when the socializ-ing goal was primed, and this effect was strongerwhen socializing evoked a stronger positivereward signal in the minds of the participants(which was assessed in a separate implicitaffective association task). Importantly, checksindicated that priming caused participants topursue the goal independently of their reportedmotivation to attain it. This finding not onlydemonstrates that people invest effort as a resultof subliminal goal priming but also that theresulting behavior is flexible, because peoplepursued an action that was available in theirrepertoire (skillfully using a computer mouse) butwas novel to attain the goal. Similar effects ofreward value have been documented for other,perhaps more consequential, behaviors. Primingan egalitarian goal, for instance, changes people’svoting behavior to the extent that this goal isrepresented as positive or rewarding (46).

www.sciencemag.org SCIENCE VOL 329 2 JULY 2010 49

REVIEW

on

July

1, 2

010

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

The findings discussed above indicate that thepursuit of goals occurs “out of the blue” when agoal representation is primed and followed by theactivation of a positive reward signal due to anestablished association. However, a preexistingassociation between a goal representation and areward signal may not be the only source fromwhich unconscious goal pursuit arises. Undersome circumstances, unconscious goal pursuitemerges when goal representations are primedtogether with positive reward signals. This abilityto respond to the mere coactivation of goalrepresentations and positive affective cues isthought to play a fundamental role in sociallearning (47) and is considered as basic in moti-vational analyses of human behavior (39). Thus,when a child observes his mother smile whenmunching homemade cookies, a student hears ahilarious joke upon entering the classroom, or aperson strolling around a mall hears people laughwhile reading on a billboard “start your holidayhere,” this can cause the goal representations thatare primed by those situations (eating cookies,achieving at school, booking a vacation) to acquirean intrinsic reward value, which prepares andmotivates goal-directed behavior.

A recent study examined the effects ofcoactivating goal representations and positivereward signals on the preparation and motivationof behavior in more detail. In this study, healthyyoung adults had to squeeze a handgrip inresponse to a start sign while the timing andpersistence of their behavior were measured (48).Before this task, words pertaining to the goal ofphysical exertion were subliminally presented ornot, together with positive words that signalrewards (such as good or nice) or not. In linewith the ideomotor principle, participants whowere subliminally primed with the goal ofexertion started to squeeze earlier. However, onlyparticipants for whom the goal was coactivatedwith a positive reward signal recruited moreresources to execute this goal, as was evidencedby more forceful and persistent squeezing. Again,consciously reported motivation did not showany relation to the subliminal goal primingmanipulation. Hence, activating a goal represen-tation gives behavior a head start, whereas theaccompanying reward signal motivates behavioroutside awareness. Other studies have shown thatthis coactivation procedure yields effects that aresimilar to those of conscious goals (induced byconscious goal instructions or by making peopleaware of their current needs) in tasks that requireflexibility and effort in novel situations (42, 49).

ConclusionThe present review and analysis reveal that thebasic processes necessary for goal pursuit—preparing and directing instrumental actions andassessing the reward value of the goal—canoperate outside conscious awareness. Although itis often taken for granted that goal pursuitoriginates in conscious decisions, it can also arisefrom unconscious sources. This remarkable capac-

ity for unconscious goal pursuit results from thedesign and workings of the brain and mind, whichprocess and represent behaviorally relevantinformation in such a way that goal pursuit canbe controlled by the social situation withoutconscious awareness of the activation andoperation of the goal.

Earlier research has shown that action goals,such as moving a finger, that were initially con-sciously set are unconsciously prepared beforethey are acted on (1). The literature reviewed heresuggests that the unconscious nature of the willhas an even more pervasive impact on our life.Goals far more complex than finger movements,can guide behavior without being consciously setfirst, when they themselves are activated outsideconscious awareness. These unconsciously acti-vated goals cause people to invest effort andselect actions available in their repertoire to attainthe goal in novel settings without them beingaware of the goal or its operation. Overall, theevidence on unconscious goal pursuit indicatesthat the control of unconscious goals is flexibleand effortful, suited to meet the dynamics of theenvironment.

Understanding exactly how unconscious goalsflexibly control behavior remains a challenge forfuture research. It has been argued that goals directattention and behavior, even in the absence ofconscious awareness of the goal (44, 50). That is,the operation of higher cognitive processessupporting goal pursuit (also conceptualized asworking memory or executive control) does notcare much about the conscious state of theindividual. This view concurs with recent insightsthat attention and consciousness are distinct (51).

The research discussed here suggests thatconscious goals (often induced by explicit taskinstructions) and unconscious goals (induced bypriming) have similar effects on tasks that rely onexecutive control. However, it is too early toconclude that consciousness is redundant in thepursuit of goals, as we do not yet know whetherthere are special cases in which consciousness(apart from attention) facilitates performance. Infact, we only know that we can become con-sciously aware of the decisions that we make andthe goals we pursue without having a properempirical test telling us how consciousness itselfexactly influences our behavior. Future researchwill have to explore when consciously and un-consciously activated goals direct attention andinformation processing in similar or distinct man-ners to recruit the cognitive functions and brainsystems that translate goals into behavior.

References and Notes1. B. Libet, C. A. Gleason, E. W. Wright, D. K. Pearl, Brain

106, 623 (1983).2. J. A. Bargh, P. M. Gollwitzer, A. Lee-Chai, K. Barndollar,

R. Trötschel, J. Pers. Soc. Psychol. 81, 1014 (2001).3. R. Custers, M. Maas, M. Wildenbeest, H. Aarts,

Eur. J. Soc. Psychol. 38, 1013 (2008).4. P. Sheeran et al., Br. J. Soc. Psychol. 44, 47 (2005).5. R. R. Hassin, in The New Unconscious, R. R. Hassin,

J. S. Uleman, J. A. Bargh, Eds. (Oxford Univ. Press,New York, 2005), pp. 196–224.

6. W. T. Powers, Science 179, 351 (1973).7. S. Monsell, J. Driver, Eds., Control of Cognitive Processes:

Attention and Performance XVIII (MIT Press, Cambridge,MA, 2000).

8. H. M. Gray, K. Gray, D. M. Wegner, Science 315, 619 (2007).9. J. A. Bargh, Eur. J. Soc. Psychol. 36, 147 (2006).

10. R. R. Hassin, J. A. Bargh, S. Zimerman, Soc. Cogn. 27, 20(2009).

11. A. Miyake et al., Cognit. Psychol. 41, 49 (2000).12. H. Aarts et al., Soc. Cogn. 23, 465 (2005).13. A. C. Kay, S. C. Wheeler, J. A. Bargh, L. Ross, Organ.

Behav. Hum. Decis. Process. 95, 83 (2004).14. H. Aarts, A. Dijksterhuis, J. Pers. Soc. Psychol. 84, 18 (2003).15. R. W. Holland, M. Hendriks, H. Aarts, Psychol. Sci. 16,

689 (2005).16. R. L. Abrams, A. G. Greenwald, Psychol. Sci. 11, 118 (2000).17. F. Schlaghecken, M. Eimer, Psychon. Bull. Rev. 11, 463

(2004).18. L. Naccache, S. Dehaene, Cognition 80, 215 (2001).19. H. C. Lau, R. E. Passingham, J. Neurosci. 27, 5805 (2007).20. W. Hart, D. Albarracín, J. Pers. Soc. Psychol. 97, 1129

(2009).21. E. J. Strahan, S. J. Spencer, M. P. Zanna, J. Exp. Soc. Psychol.

38, 556 (2002).22. G. M. Fitzsimons, J. A. Bargh, J. Pers. Soc. Psychol. 84,

148 (2003).23. H. Aarts, R. Custers, H. Marien, J. Pers. Soc. Psychol. 96,

967 (2009).24. D. M. Wegner, The Illusion of Conscious Will (MIT Press,

Cambridge, MA, 2002).25. J. T. Austin, J. B. Vancouver, Psychol. Bull. 120, 338 (1996).26. P. Fourneret, M. Jeannerod, Neuropsychologia 36, 1133

(1998).27. C. D. Frith, S. J. Blakemore, D. M. Wolpert, Philos. Trans.

R. Soc. London Ser. B 355, 1771 (2000).28. B. Hommel, J. Müsseler, G. Aschersleben, W. Prinz,

Behav. Brain Sci. 24, 849, discussion 878 (2001).29. M. A. Goodale, D. A. Westwood, A. D. Milner, Prog. Brain

Res. 144, 131 (2004).30. W. James, Principles of Psychology (Holt, New York, 1890).31. M. Chen, J. A. Bargh, J. Exp. Soc. Psychol. 33, 541 (1997).32. F. Pulvermüller, Nat. Rev. Neurosci. 6, 576 (2005).33. H. Aarts, A. Dijksterhuis, J. Pers. Soc. Psychol. 78, 53 (2000).34. R. Custers, H. Aarts, Pers. Soc. Psychol. Bull. 33, 623

(2007).35. P. M. Gollwitzer, Eur. Rev. Soc. Psychol. 4, 141 (1993).36. E. L. Deci, R. M. Ryan, Intrinsic Motivation and

Self-Determination in Human Behavior (Plenum,New York, 1985).

37. M. Pessiglione et al., Science 316, 904 (2007).38. W. Schultz, Annu. Rev. Psychol. 57, 87 (2006).39. P. Shizgal, Curr. Opin. Neurobiol. 7, 198 (1997).40. E. Bijleveld, R. Custers, H. Aarts, Psychol. Sci. 20, 1313

(2009).41. E. Bijleveld, R. Custers, H. Aarts, Cognition 115, 330

(2010).42. R. Custers, H. Aarts, J. Pers. Soc. Psychol. 89, 129 (2005).43. G. Aston-Jones, J. D. Cohen, Annu. Rev. Neurosci. 28,

403 (2005).44. A. Dijksterhuis, H. Aarts, Annu. Rev. Psychol. 61, 467

(2010).45. R. Custers, H. Aarts, J. Exp. Soc. Psychol. 43, 312 (2007).46. M. J. Ferguson, J. Pers. Soc. Psychol. 92, 596 (2007).47. N. E. Miller, J. Dollard, Social Learning and Imitation

(Yale Univ. Press, New Haven, CT, 1941).48. H. Aarts, R. Custers, H. Marien, Science 319, 1639 (2008).49. M. Veltkamp, H. Aarts, R. Custers, Eur. Rev. Soc. Psychol.

20, 345 (2009).50. J. A. Bargh, M. J. Ferguson, Psychol. Bull. 126, 925 (2000).51. C. Koch, N. Tsuchiya, Trends Cogn. Sci. 11, 16 (2007).52. J. Jaynes, The Origin of Consciousness in the Breakdown

of the Bicameral Mind (Mariner Books, New York, 1976).53. We thank M. Brass, D. T. Gilbert, T. M. Kenter,

W. Stroebe, K. Van den Bos, the Goallab group(www.goallab.nl), and three anonymous reviewers forcomments. This work was financially supported by theNetherlands Organization for Scientific Research(VENI grant 451-06-014 and VICI grant 453-06-002).

10.1126/science.1188595

2 JULY 2010 VOL 329 SCIENCE www.sciencemag.org50

REVIEW

on

July

1, 2

010

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om