The differential influences of positive affect, random reward,and performance-contingent reward on cognitive control
Kerstin Frber & Gesine Dreisbach
# Psychonomic Society, Inc. 2014
Abstract Growing evidence suggests that positive affect andreward have differential effects on cognitive control. So far,however, these effects have never been studied together. Here,the authors present one behavioral study investigating theinfluences of positive affect and reward (contingent and non-contingent) on proactive control. A modified version of theAX-continuous performance task, which has repeatedly beenshown to be sensitive to reward and affect manipulations, wasused. In a first phase, two experimental groups received eitherneutral or positive affective pictures before every trial. In asecond phase, the two halves of a given affect group addition-ally received, respectively, performance-contingent or randomrewards. The results replicated the typical affect effect, interms of reduced proactive control under positive as comparedto neutral affect. Also, the typical reward effects associatedwith increased proactive control were replicated. Most inter-estingly, performance-contingent reward counteracted thepositive affect effect, whereas random reward mirrored thateffect. In sum, this study provides first evidence thatperformance-contingent reward, on the one hand, and positiveaffect and performance-noncontingent reward, on the otherhand, have oppositional effects on cognitive control: Onlyperformance-contingent reward showed a motivational effectin terms of a strategy shift toward increased proactive control,whereas positive affect alone and performance-noncontingentreward reduced proactive control. Moreover, the integrativedesign of this study revealed the vulnerability of positiveaffect effects to motivational manipulations. The results are
discussed with respect to current neuroscientific theories ofthe effects of dopamine on affect, reward, and cognitivecontrol.
Keywords Cognitivecontrol .Emotion .Reward .Dopamine
Positive affect arises in a variety of situationsfor instance,being woken up by the sunlight after a long period of foggydays, winning 100 by being the 100,000th customer of thelocal hardware store, or getting a 100 bonus for above-average performance as a piece-worker. These examples showthat exposure to positive stimuli, receiving an unexpectedrandom reward, or earning a performance-contingent rewardis associated with the elicitation of positive affect. Exactly thisobvious relationship between positive affect and reward hasbeen the basis of a very influential theorynamely, the neu-ropsychological theory of positive affect by Ashby, Isen, andTurken (1999; see also Ashby, Valentin, & Turken, 2002).Because reward effects are mediated by dopamine activity(see, e.g., Schultz, 1992), the authors argued that positiveaffect effects might also be mediated by dopamine. Morespecifically, they assumed that positive affect increases dopa-mine release from the ventral tegmental area into anteriorcingulate cortex and prefrontal cortex, and this mechanismsupposedly modulates cognitive control, in terms of enhancedcognitive flexibility, by facilitating selection of and switchingbetween cognitive sets. That is, positive affectirrespectiveof its originhas long been assumed to have an importantinfluence on the regulation of cognitive control. But is it reallyjustified to treat positive affect from different sources as beingfunctionally the same?
On the one side, and on the basis of the intuitional relation-ship between positive affect and reward along with supportingtheories like the neuropsychological theory of positive affect(Ashby et al., 1999; Ashby et al., 2002), it is not surprising that
Electronic supplementary material The online version of this article(doi:10.3758/s13415-014-0259-x) contains supplementary material,which is available to authorized users.
K. Frber (*) :G. DreisbachDepartment of Psychology, University of Regensburg,Universittsstr. 31, 93053 Regensburg, Germanye-mail: firstname.lastname@example.org
Cogn Affect Behav NeurosciDOI 10.3758/s13415-014-0259-x
the existing literature on affective modulations of cognitivecontrol often ignored the difference between affect and rewardmanipulations. On the other side, it seems just as intuitive toassume that positive affect as a result of performance-contingent reward might rather have a motivational (as op-posed to an affective) influence on behavior. And indeed, tworecent reviews (Chiew & Braver, 2011; Dreisbach & Fischer,2012) challenged the idea of a common mechanism of posi-tive affect and reward effects on the basis of a growing numberof diverging results on affectcognition interactions. Onegroup of studies, all investigating processes of proactive andreactive control (Braver, 2012; Braver, Gray, & Burgess,2007), found opposing effects for positive affect and reward(Chiew & Braver, 2013; Dreisbach, 2006; Frber &Dreisbach, 2012; Jimura, Locke, & Braver, 2010; Locke &Braver, 2008; Padmala & Pessoa, 2011; vanWouwe, Band, &Ridderinkhof, 2011). Likewise, another group of studies, in-vestigating processes of sequential conflict adaptation (i.e.,reduced response conflict on trials following conflict trials),as well found opposing effects for positive affect and randomreward on the one side and performance-contingent reward onthe other (Braem, Verguts, Roggeman, Notebaert, &Roggeman, 2012; Strmer, Nigbur, Schacht, & Sommer,2011; van Steenbergen, Band, & Hommel, 2009, 2010,2012; van Wouwe et al., 2011). It thus seems that althoughreward and positive affect are closely related, both have dis-sociable effects on cognitive controlthe former being moti-vational and the latter affective. However, since affect andreward have never been studied together, it is still hard tojudge which results can actually be attributed to the allegeddifference of reward and affect effects, and which are simplydue to other procedural differences between studies andparadigms.
In the present study, we will focus on how positive affect andreward influence cognitive control strategies within a singleparadigm. According to the dual mechanisms of control(DMC) framework (Braver, 2012; Braver et al., 2007), twocontrol strategies, namely proactivethat is, preparatorycon-trol and reactivethat is, just-in-timecontrol, can be differ-entiated. In theDMC, preparatory control is aimed at preventingconflict and optimizing task performance by sustained activa-tion of task relevant information, whereas reactive control is aless effortful strategy that works as a late correction mechanismwhenever a conflict actually occurs. Empirical results so farindicate that reward increases proactive control strategies(Chiew & Braver, 2013; Jimura et al., 2010; Locke & Braver,2008; Padmala & Pessoa, 2011), whereas positive affect isassociated with reduced proactive control and/or increased re-active control (Dreisbach, 2006; Frber &Dreisbach, 2012; vanWouwe et al., 2011). To illustrate, we will now shortly reviewseveral studies all using the AX-continuous performance task(AX-CPT; Servan-Schreiber, Cohen, & Steingard, 1996), aparadigm that we also used in the present study.
The AX-CPT is a context processing task especially suitedto investigate changes in the use of proactive and reactivecontrol (cf. Braver, 2012; Braver et al., 2007). Therein, spe-cific cueprobe sequences require target or nontarget re-sponses: Target trials are specified as cue A followed byprobe X (AX trials), whereas nontarget trials can be AY(target cue, nontarget probe), BX (nontarget cue, target probe),or BY trials (nontarget cue, nontarget probe). AX sequencesappear with a frequency of 70%, resulting in a high expecta-tion for target trials following the cue A and strong associ-ations between the probe X and target responses. A proac-tive control strategy in this task means strong maintenance ofthe cue for in advance response preparation, which shouldresult in costs in AY trials, in which the cue-triggered expec-tation is violated, and benefits in B-cue trials, in which the cueunequivocally predicts the correct response. Conversely, reac-tive control should result in benefits in AY trials, because thereis no misleading expectation to cause response interference,and costs in BX trials, in which a predominant responsetendency triggered by the X probe has to be overcome. Twostudies (Chiew & Braver, 2013; Locke & Braver, 2008)combined the AX-CPT with a reward manipulation: Bothstarted with a baseline condition without monetary incentivesfollowed by a performance-contingent reward block. Rewardswere given for each correct reaction time (RT) faster than theindividual median RT in the baseline block (Locke & Braver,2008) or specifically in trials with an incentive precue andcorrect RTs within the fastest 30th percentile of individualbaseline RTs (Chiew & Braver, 2013). The results showed aclear motivational effect of reward with generally faster RTs inthe reward blocks. More importantly, the introduction of mon-etary incentives lead to the adoption of a proactive controlstrategywith increased error rates specifically onAY trials andRT benefits especially in B-cue trials. Another two studies(Dreisbach, 2006; vanWouwe et al., 2011) investigated purelyaffective influences on performance in the AX-CPT: Affectwas manipulated with affective picturespositive, neutral, ornegativefrom the International Affective Picture System(IAPS; Lang, Bradley, & Cuthbert, 1999) preceding everytrial (Dreisbach, 2006) or with emotional film clipspositiveor neutralprior to the experiment (van Wouwe et al., 2011).As compared to neutral (and negative) affect, positive affectreduced error rates on AY trials (both studies), and resulted incosts specifically in B-cue trials in the Dreisbach study.Dreisbach interpreted this behavioral trade-off as evidencefor reduced maintenance of the cue under positive affect, thatis, reduced proactive control. Supported by additional analy-ses of event-related potentialsno change in cue-related po-tentials, affective modulation of probe-related potentials inAY trialsvan Wouwe et al. came to the conclusion thatpositive affect increased reactive control. To sum up, rewardmanipulations increased proactive control, whereas positiveaffect reduced proactive and increased reactive control in the
Cogn Affect Behav Neurosci
AX-CPT. Taken together, these studies thus demonstrateatodds with assumptions of the neuropsychological theory ofpositive affect (Ashby et al., 1999; Ashby et al., 2002)diverging effects of reward and positive affect on cognitivecontrol, suggesting that the source of positive affect is indeed arelevant factor.
Further studies (Braem et al., 2012; van Steenbergen et al.,2009, 2010, 2012), all using an Eriksen flanker paradigm(Eriksen & Eriksen, 1974), have indicated an even more com-plex picture by revealing diverging effects of performance-contingent and performance-noncontingent rewards. In thesestudies, performance-contingent reward (i.e., reward only forfast and correct responses) enhanced sequential adaptationeffects (Braem et al., 2012; see also Strmer et al., 2011),whereas performance-noncontingent reward (i.e., reward fol-lowing a random subset of trials) or positive affect reducedsequential conflict adaptation (van Steenbergen et al., 2009,2010, 2012). Though these studies used different reward andaffect manipulations than theAX-CPTstudies reviewed above,they again demonstrate oppositional effects of performance-contingent reward and positive affect. In addition, they suggestthat giving random, performance-noncontingent rewardsmight have an effect similar to that of direct positive-affectinduction. Taken together, the findings reviewed so far suggestthat at least performance-contingent reward and positive affectare not interchangeable, since they do not allow the deductionof effects from one another. Thus, the origin of positive af-fectvia direct induction or as a consequence of performance-contingent or -noncontingent rewardseems to be an impor-tant but so far understudied issue. Consequently, recent re-views on the roles of reward and affect in cognitive control(Chiew & Braver, 2011; Dreisbach & Fischer, 2012) havestressed the importance of a clear differentiation between mo-tivational and affective manipulations, called for further re-search to disentangle effects of reward and affect, and request-ed new integrative studies, which could combine these so farlargely independent lines of research.1
The present study was designed to directly compare theeffects of reward and affect manipulations on cognitive con-trol in a single experiment. For this purpose, we used amodified version of the AX-CPT, because of its proven sen-sitivity to reward and positive affect manipulations that wereviewed above (Chiew & Braver, 2013; Dreisbach, 2006;Locke & Braver, 2008; van Wouwe et al., 2011). The exper-iment started with a first phase, which included only an affectmanipulation between participants (positive vs. neutral affect)via affective pictures comparable to those in the study byDreisbach. In the second phase, the affect groups were splitin half and received either performance-contingent or
-noncontingent rewards, resulting in a complete orthogonaldesign with the following groups: NeutralRewardthat is,neutral affect plus performance-contingent reward;PositiveRewardthat is, positive affect plus performance-contingent reward; NeutralRandomRewardthat is, neutralaffect plus performance-noncontingent reward; andPositiveRandomRewardthat is, positive affect plusperformance-noncontingent reward. The performance-contingent reward manipulation was similar to the procedureby Chiew and Braver (2013), with the first experimental phaseserving as a baseline condition. With this integrative design,we would be able to directly investigate the possible interac-tions between positive affect and reward manipulations.Since, as was outlined above, both seem to have divergingeffects on cognitive control (i.e., motivational as opposed toemotional), it is of major interest how the two might interact.In principle, both effects could coexist, or one effect couldpredominate over the other (see the more detailed hypothesesbelow). In any case, the results would be of theoretical andpractical interest, since they might explain the somewhatheterogeneous literature on cognitionemotion interactions.
On the basis of previous results (Dreisbach, 2006; Frber &Dreisbach, 2012; van Wouwe et al., 2011), the first phase wasexpected to replicate positive affect effects, in terms of reducedproactive control and/or increased reactive control, indicatedby a benefit in AY trials and probable costs in B-cue trials.Again, following previous results (Chiew & Braver, 2013;Jimura et al., 2010; Locke & Braver, 2008; Padmala &Pessoa, 2011), performance-contingent reward was expectedto result in performance optimization via proactive cont...