Ego Depletion Induces Mental Passivity: Behavioral Effects Beyond Impulse Control

Motivation Science © 2017 American Psychological Association 2017, Vol. 3, No. 4, 321–336 2333-8113/17/$12.00 http://dx.doi.org/10.1037/mot0000058 Ego Depletion Induces Mental Passivity: Behavioral Effects Beyond Impulse Control

Andrew J. Vonasch Kathleen D. Vohs University of North Carolina at Chapel Hill University of Minnesota

Anastasiya Pocheptsova Ghosh Roy F. Baumeister University of Arizona Florida State University

Three studies supported the hypothesis that people can become mentally and physically passive when resources are depleted by prior acts of self-control. Feeling depleted and recent self-regulatory exertion were associated with preferences for passive behaviors like resting and watching TV. Participants who had to maintain attention in the face of distractions preferred to avoid making consumer decisions as compared with participants who did not. Breaking a habit caused hungry participants to eat more peanuts when doing so was easy (the peanuts lacked shells) and to eat less when eating required minor preparatory action (the peanuts had shells). Mental passivity induced by depletion of self-control causes both passive behavior and impulsive behavior, depending on the situation.

Keywords: self-regulation, passivity, ego depletion

The difference between active and passive status quo. In other words, passive responding responding is one of the most fundamental di- means letting things happen and not intervening mensions among the countless ways that living to change things. creatures can respond to environmental events. The view of self-regulation as an effortful ac- Active responding means that the creature re- tivity that depletes an energy resource has stimu- acts to the external circumstances by seeking to lated much research (for meta-analysis, see Hag- steer the course of events in one rather than ger, Wood, Stiff, & Chatzisarantis, 2010; for another direction. In contrast, passive respond- recent overview of theoretical disputes and issues, ing involves allowing events to unfurl without see Baumeister & Vohs, 2016). The underlying attempting to control the outcome or alter the assumption has been that a particular resource, akin to the folk notion of willpower, is expended in regulating behavior such as by inhibiting impulses. The present paper tests an extension to this This article was published Online First July 13, 2017. theory. Moving beyond resisting temptation, it Andrew J. Vonasch, Department of Psychology and Neu- tests the idea that the difference between active roscience, University of North Carolina at Chapel Hill; and passive responding involves the expenditure Kathleen D. Vohs, Carlson School of Management, Univer- of this same resource. We reasoned that top-down sity of Minnesota; Anastasiya Pocheptsova Ghosh, Eller This document is copyrighted by the American Psychological Association or one of its allied publishers. College of Management, University of Arizona; Roy F. control takes energy and will decline when regu- This article is intended solely for the personal use of the individualBaumeister, user and is not to be disseminated broadly. Department of Psychology, Florida State Uni- latory resources are low. This may result in in- versity. creased inaction insofar as action requires initia- This research was supported in part by Land O’ Lakes tive, but it may also produce increased behavior Excellence in Marketing funds and a grant from the John insofar as inhibitions or stopping the behavior Templeton Foundation. We thank Korey Jones, Katie Lewis, and Marco Romero for data collection and Colin DeYoung for requires top-down control. helpful comments. Correspondence concerning this article should be ad- Passivity dressed to Andrew J. Vonasch, Department of Psychology and Neuroscience, The University of North Carolina at Chapel Hill, 235 East, Cameron Avenue, Chapel Hill, NC Living things vary widely in their capacity 27599. E-mail: [email protected] for active response. It is arguable that plants

321 322 VONASCH, VOHS, POCHEPTSOVA GHOSH, AND BAUMEISTER

have almost no active responses. As a revealing expensive form of control is curtailed, to con- example, they remain inactive even while other serve energy, and this could mean broadly a creatures eat them. Animals, in contrast, move passive attitude toward life and events. If men- to eat and to resist being eaten. Among simple tal passivity is a central aspect of depletion, then animals, passive responding means physical depletion may produce physical passivity, espe- stillness, and active responding means physical cially when there are no automatic impulses to movement. take action, and when action requires exerting In humans, the active versus passive distinc- top-down control (e.g., difficult decisions, or tion occurs at both the physical level (stillness initiative). vs. movement) and at the mental level (relying In sum, our investigation focused on the hy- on automatic vs. controlled processes). Mental pothesis that ego depletion produces mental passivity is the absence of active processing to passivity, which entails an abdication of top- override one’s incipient or automatic responses. down mental control over responses. This could Physical passivity is usually a result of mental produce physical passivity and inaction (as in passivity, such as during rest or sleep. To be Banker, Ainsworth, Baumeister, Ariely, & sure, physical passivity can stem from active Vohs, 2017; Baumeister et al., 1998, Study 4; mental processing, such as when someone tries Giacomantonio, Jordan, Fennis, & Panno, 2014; to hold still while being tickled or injected. Job, Bernecker, Miketta, & Friese, 2015)—but Being mentally passive does not necessarily under some circumstances could yield increases mean one’s body is unmoving. A passive person in physical activity (as in most other depletion might simply follow orders, such as walking to studies summarized in Hagger et al., 2010). a specific place. Mental passivity involves letting one’s automatic responses to the environ- Active Versus Passive Mental Processes ment drive one’s behavior. Depending on one’s and Behaviors automatic impulses, mentally passive behavior can be either physically passive or physically Either action or inaction can result from men- active. tal passivity. Carver, Johnson, and Joormann Mentally active versus passive responding (2008) proposed that lack of executive function seemingly goes to the essence of volition. Many control leaves the person responding reflexively human decisions and experiences depend on the to environmental cues. That sort of a mentally difference between active and passive respond- passive response style can theoretically produce ing. A health warning may be passively ignored responses ranging from physical lassitude to or met with an active change of behavior. Peo- disinhibited violence. ple’s moral judgments depend greatly on this The hypothesis that ego depletion can pro- difference. Most people judge actively killing duce mental passivity fits some previous find- someone to be much more reprehensible than ings. For example, judges on parole boards can passively letting someone die, even if the out- make an easy, default decision (to keep a pris- come and motive are the same (Rachels, 2007). oner in prison) or a risky, difficult decision (to Indeed, basic linguistic structure attests to the grant a parole and allow the convict to rejoin importance of the active-passive dimension: ac- society, where he might commit a crime). Dan- tive versus passive verbs can express the same ziger, Levav, and Avnaim-Pesso (2011) found thought often with substantially different mean- that judges granted parole at reasonably high This document is copyrighted by the American Psychological Association or one of its allied publishers. ings (such as “Mistakes were made”). rates when the judges had been rested or fed, This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Depletion has often been understood as trans- but as the day wore on they became increasingly fer of control from conscious to unconscious, likely to favor the default response of denying from central executive to automatic (Hofmann, parole. Likewise, over the course of making Friese, & Strack, 2009). One way of summariz- many decisions while buying a car, buyers in- ing this diverse literature is to suggest that top- creasingly tend to select the standard or default down mental control is generally reduced or option (Levav, Heitmann, Herrmann, & Iyen- even abdicated during ego depletion. The view gar, 2010). Polman and Vohs (2016) confirmed that depletion brings abdication of control fits that decision fatigue is associated with prefer- many findings, such as increased yielding to ence for the status quo option. Banker, Ain- impulse (Hagger et al., 2010). The resource- sworth, Baumeister, Ariely, and Vohs (2017) EGO DEPLETION INDUCES MENTAL PASSIVITY 323

found that even when people’s decisions di- sive responding, in contrast to the more studied verge from the status quo, they diverge less effect of depletion increasing disinhibition. when people are depleted. In making decisions, Although many published studies support the then, depletion increases the tendency to stick claims of the limited resource model, the exis- close to a safe, default response. tence of the ego depletion effect has recently The limited resource model of self-regulation been disputed (Carter, Kofler, Forster, & Mc- suggests that people conserve resources by re- Cullough, 2015; Hagger et al., 2016). One pur- ducing top-down control after they have ex- pose of the present research was to conceptually pended some effort in self-control or decision- replicate depletion’s effects of increasing both making (Baumeister, Vohs, & Tice, 2007). disinhibited and passive behavior. To improve Recent evidence has linked prior exertion of the replicability of our results, our studies used self-control to a lessening of deliberate, larger sample sizes to achieve higher statistical thoughtful responding. For example, fMRI power than is common in this literature. Addi- studies have found that brain areas involved in tionally, the present studies tested the hypo- the implementation of control—but not the de- thesis that the same cause of depletion could tection of a self-control conflict—became less produce divergent, disinhibited or passive be- active after depletion (Hedgcock, Vohs, & Rao, havior, depending on the situation. 2012). Thus, depletion led to reflexive but not The idea that depletion can produce either reflective responding, which increases reliance action or inaction is consistent with motiva- on situational cues (Carver et al., 2008). tional intensity theory (Brehm & Self, 1989). When top-down control is reduced, responses That theory assumes that people conserve en- depend on automatic processes and external ergy and restrict their efforts to the minimum cues (Hofmann, Friese, & Strack, 2009). Deple- levels required for task success. As tasks be- tion should produce that pattern, alongside the come perceived as more difficult, people exert impulse to conserve energy (Muraven, Shmueli, more effort to meet the perceived effort required & Burkley, 2006). Hence easy access to tempt- to succeed. Notably, however, people stop ex- ing rewards might cause a depleted person to erting effort when they perceive higher effort take disinhibited action, as much work has costs than likely rewards from the task (Ag- found (see Hagger et al., 2010). In contrast, tarap, Wright, Mlynski, Hammad, & Black- depleted people might become inactive if there ledge, 2016). Depletion is assumed to increase are few immediate rewards or other external the perceived effort required for most tasks impetus to act—as well as if it would seemingly (Wright, 2014; Richter, Gendolla, & Wright, take active exertion to pursue rewards. 2016). Therefore, depletion is expected to in- Some existing literature supports the idea that crease effortful action (i.e., disinhibition) in depletion can produce physically passive re- most cases. However, whenever the perceived sponding. Experiment 4 from Baumeister et al. effort becomes so great that the task no longer (1998) found that depleted people passively sat seems worthwhile, passivity would result. longer through a boring movie when skipping Hence we predicted that self-regulatory re- the movie required them to alter their behavior. source depletion could produce either action or More recent studies showed that depletion in- inaction, depending on the situation. Depletion creases passive resting, and thoughts about rest- in the context of easy and immediate rewards ing, among people who implicitly believe that should produce disinhibited activity. In con- This document is copyrighted by the American Psychological Association or one of its allied publishers. their willpower is limited (Job, Bernecker, trast, depletion should produce a reluctance to This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Miketta, & Friese, 2015). One study examined put forth effort to pursue rewards, either be- depletion’s effect on both active and passive cause few rewards are readily available or be- behavior, by testing how much risky reward cause claiming them would require high effort. seeking people would do when depleted versus not. Depleted people sought rewards more when Present Studies they were easy to obtain, but less when seeking rewards involved the effort of pumping a man- Three studies tested the hypothesis that par- ual bicycle pump (Giacomantonio, Jordan, Fen- ticipants would become mentally passive fol- nis, & Panno, 2014). The findings support the lowing self-regulatory exertion. The basic de- idea that depletion can lead to physically pas- sign was to instruct some participants to use 324 VONASCH, VOHS, POCHEPTSOVA GHOSH, AND BAUMEISTER

self-control initially, whereas others not; then the self (Ryan & Frederick, 1997). The main all participants were measured on tasks in which predictions were that high reports of depleting they could respond passively or actively. Study actions and of feeling depleted would be linked 1 tested whether recent self-regulatory exertion to high preferences for passive activities, low and feelings of depletion predicted preferences preference for active pastimes, and low feelings for passive behavior. Studies 2 and 3 were ex- of vitality. periments testing whether self-regulatory exertion caused mental passivity. Study 2 tested whether depletion would produce mental pas- Method sivity, operationalized as whether participants postponed making a decision. Study 3 offered Participants immediate rewards to some participants and manipulated how easily available they were. Three hundred ten participants (182 female) Study 3 tested the moderation hypothesis that completed the study. Six participants were ex- depletion’s effect on mental passivity would cluded for failing an attention check. They were produce either physical inactivity or impulsive recruited from Amazon’s Mechanical Turk and appetitive action, depending on the opportunity were paid a small amount for participating. for immediate reward. Sample size was chosen to have greater than 90% power to find a significant effect if d ϭ .4. Study 1: Self-Reported Depletion Predicts Preferences for Passive Activities and Procedure Weaker Feelings of Vitality Participants completed the subjective vitality Study 1 provided the first evidence linking scale (Ryan & Frederick, 1997; seven items, ego depletion to passive feelings. It relied on ␣ϭ.91, e.g., “I feel energized. I feel alive and correlations among self-report variables (unlike vital”), and depleted feelings (Wright & the experiments reported subsequently). Wright Baumeister, 2015; three items, ␣ϭ.83, i.e., “In and Baumeister (2015) developed brief mea- the past couple of hours, have you felt that: It’s sures (suitable for experience sampling) of feel- hard to make up your mind about even simple ing depleted and of having engaged in activities things? Things are bothering you more than that could deplete regulatory resources. Both they usually would? You have less mental and measures are imperfect proxies for depletion, emotional energy than you normally have?”) each measuring a single aspect of a complex and depleting events scales (Wright & phenomenon. Feeling depleted is presumably Baumeister, 2015; three items, ␣ϭ.67, i.e., “In often the result of doing depleting things, but it the last 60 minutes, have you: Forced yourself could also result from lack of rest, believing that to do something that you didn’t really want to recent events caused you to be depleted (Job, Dweck, & Walton, 2010), or other causes. In do? Used self-control to stop yourself from do- contrast, recent engagement in depleting events ing something that you wanted to do [e.g., have presumably causes depleted feelings much of you resisted desire or temptation]? Exerted the time, but perhaps not always. Participants mental effort to make decisions?”), all on filled out those and then expressed their current 7-point Likert scales. Participants next were This document is copyrighted by the American Psychological Association or one of its allied publishers. degree of inclination to perform various activi- told that the study concerned what people feel This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. ties, which were selected as a diverse assort- “in the mood” to do at times. They reported the ment that would include some highly active and degree to which they would prefer to perform other passive ones. The active activities were various passive activities (six activities: sleep, selected as requiring self-initiated responses sunbathe, watch TV, knit, relax on the couch, (initiative), ongoing decision making and action listen to music; ␣ϭ.52) and active activities control, physical movement, and/or effort ex- (seven activities: play a sport, cook, dance, penditure, whereas the passive ones were cho- write, debate politics, work out, solve a cross- sen based on lacking those features. We also word puzzle; ␣ϭ.74), on 7-point Likert scales, measured vitality, which is the state of feeling presented in a random order. Last, demographic alive and alert and having energy available to information was collected. EGO DEPLETION INDUCES MENTAL PASSIVITY 325

Results and Discussion ences for active ones. We used PROCESS Model 4 (Hayes, 2013) to test for mediation As predicted, the more that participants re- using depleting events as the independent vari- ported depleted feelings, r(294) ϭ .258, p Ͻ able, depleted feelings as the mediator, and par- .001, and the presence of recent depleting ticipants’ relative preferences for active versus ϭ Ͻ events, r(296) .253, p .001, the higher their passive behaviors as the dependent variable. preferences for passive behaviors, see Table 1. The significant indirect effect (–.025, SE ϭ Although depleted participants desired to be .009, 95% CI [–.04, Ϫ.01]) indicated that de- passive, depletion did not produce the expected pleting events were associated with stronger concomitant desire to avoid being active. Re- relative preferences for passive versus active cent depleting events tended to predict stronger activities because depleting events were associ- preferences for active activities, r(296) ϭ .113, ϭ ated with concomitant depleted feelings. The p .051. One possible explanation for this remaining direct effect was nonsignificant (t ϭ counterintuitive finding is that the relationship Ϫ.37, p ϭ .71), suggesting that depleted feel- between depleting events and preferences for ings fully mediated the relationship between active activities may be a selection effect. Peo- depleting events and behavior preferences. ple who particularly enjoy active activities may To compare activity preferences among de- be more likely to have done an active (and depleting) activity recently. Therefore, people pleted and nondepleted participants, we con- who did depleting activities would prefer active ducted a within-participants analysis. For this analysis, depleted feelings and depleting events activities because of self-selection, not neces- ␣ϭ sarily because depletion makes people prefer were combined by averaging their scores ( active activities. In support of this interpreta- .644). Type of measure (active vs. passive) was tion, depleted feelings did not predict prefer- entered as a within-participants variable in a ences for active behaviors, r(294) ϭϪ.012, repeated measures ANOVA, with degree of de- p ϭ .841. Presumably, if depletion caused peo- pletion as a predictor. This within-participants ple to want to be active, the feeling of depletion analysis revealed that, overall, participants pre- would increase activity preferences, but it did ferred passive (M ϭ 3.83, SE ϭ 0.055) to active not. Overall, the impact of depletion seemed to behaviors (M ϭ 2.51, SE ϭ 0.061), F(1, 301) ϭ be an attraction to passivity rather than aversion 365.89, p Ͻ .001. This difference was influ- to activity. enced by the extent to which participants felt A mediation analysis lent further support to depleted, as evidenced by the significant inter- the idea that depletion is associated with a rel- action between the within-participants factor ative preference for passivity rather than activ- (ratings of passive and active activities) and ity. We computed a new variable for relative depletion scores, F(1, 301) ϭ 11.206, p Ͻ .001 preferences by subtracting participants’ prefer- (see Figure 1). Consistent with the correlations ences for passive behaviors from their prefer- reported in Table 1, this additional analysis sup-

Table 1 Correlations Between Depletion, Vitality, and Preferences for Active vs. Passive Behaviors This document is copyrighted by the American Psychological Association or one of its allied publishers.

This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Variable 1 2 3 4 5 6 7 1. Depleted feelings — — ءءDepleting events .475 .2 — ءء855. ءءDepletion combined .865 .3 — ءءϪ.444 ءءϪ.232 ءءVitality Ϫ.524 .4 — ءءActive Ϫ.012 .113† .062 .410 .5 — ءءϪ.107† .301 ءء309. ءء253. ءءPassive .258 .6 — ءءϪ.556 ءء626. ءء444. ءءϪ.109† Ϫ.189 ءءActive – Passive Ϫ.226 .7 M 8.78 10.27 3.19 19.99 17.21 22.79 Ϫ1.32 SD 4.72 4.57 1.34 9.80 7.23 5.84 1.22 .p Ͻ .01 ءء .p Ͻ .10 † 326 VONASCH, VOHS, POCHEPTSOVA GHOSH, AND BAUMEISTER

Figure 1. More depleted participants (one standard deviation above the mean of six depletion items) favored more passive activities and felt less vital than less depleted participants (one standard deviation below the mean of the depletion items).

ported the hypothesis that depletion was asso- p Ͻ .001. Recent depleting events also pre- ciated with stronger preferences for passive be- dicted less vitality, r(296) ϭϪ.232, p Ͻ .001. haviors. Thus, Study 1 found increased preference for Because the measure of preferences for pas- passivity, as well as reduced vitality, among sive activities had low internal consistency, we people who reported feeling depleted or who unpacked the scale into its individual items. reported having engaged in depleting activities Depleted feelings were significantly associated recently. People preferred passive activities to with preferences for the four of the six passive active ones, especially when they were de- activities (sleep, r ϭ .39, p Ͻ .001; sunbathe, ϭ ϭ ϭ ϭ pleted. To be sure, people have many reasons r .13, p .033; knit, r .16, p .009; relax for preferring one activity to another, and de- r ϭ p ϭ r ϭ on the couch, .13, .023; watch TV, pletion may not change an Olympian into a Ϫ.007, p ϭ .90; listen to music, r ϭϪ.046, p ϭ couch potato (as supported by the modest size .43). Meanwhile, depleted feelings were not sig- This document is copyrighted by the American Psychological Association or one of its allied publishers. of correlations found in this study). Nonethe- nificantly associated with preferences for any of This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. the active activities (play a sport, r ϭϪ.086, less, despite people’s intrinsic preferences for p ϭ .14; cooking, r ϭϪ.05, p ϭ .40; dance, r ϭ some activities over others, feeling depleted .04, p ϭ .53; write, r ϭ .07, p ϭ .22, debate was associated with a shift toward preferences politics, r ϭϪ.03, p ϭ .65; work out, r ϭϪ.06, for more passive activities in general. Depletion p ϭ .30; solve a crossword puzzle, r ϭ .02, p ϭ did not generally alter preferences for active .72). Consistent with the aggregate measures, activities. Indeed, the only (marginally signifi- depleted feelings were associated with prefer- cant) finding was that people who had been ences for passive, but not active behaviors. doing strenuous things expressed a mild prefer- As predicted, more depleted feelings were ence for further active activities. Because Study associated with less vitality, r(294) ϭϪ.524, 1 used a correlational design, it cannot address EGO DEPLETION INDUCES MENTAL PASSIVITY 327

causality. Two subsequent experiments were two alternatives. You also have the option of not buy- conducted to test for causality. ing either of them and looking for a digital camera at another store. Study 2: Postponing Decisions Participants were further told that the cameras did not differ on price. Study 2 tested the hypothesis that self-control The cameras were described using seven at- depletion would cause passivity in decision mak- tributes (zoom, display size, memory, picture ing. Prior work has established that after making resolution, weight, battery life, and extras) and decisions, self-control is impaired (Vohs et al., differed on three of these last attributes. Two 2008), and after acts of self-control, decision mak- models of a real camera brand (Canon) were ing shifts toward simpler, easier, and more heu- chosen to enhance the realism of participants’ ristic processes (Pocheptsova et al., 2009). Study 2 choices. The attributes were presented in a 2 ϫ sought to carry that line of work further. We 7 matrix using MouselabWeb software (http:// predicted that after engaging in self-control (and www.mousewelabweb.org, Willemsen & John- diminishing their executive resources), partici- son, 2010). To view the attributes, participants pants would postpone making decisions. moved the computer cursor over a box (an action termed “mousein”) showing the attribute Method name. When the cursor hovered over the box, it displayed information about that attribute. If the Participants cursor moved out of the box (“mouseout”), the information was hidden again. One hundred fifty-nine undergraduates par- While viewing the page with product descrip- ticipated in a two-cell (depletion vs. no- tions, participants chose whether to continue depletion) between-participants design. The searching or choose one of the two options. study was run for a week in the lab and the Three buttons at the end of the page represented sample size was determined by the number of the following choices: “Prefer to continue students who elected to sign up for a study that searching and decide in the future,” “Choose week. The resulting sample size had 80% power Canon PowerShot SX 210IS,” and “Choose to detect an effect size of d ϭ .447. Canon PowerShot S90.” The main dependent measure was decision deferral, operationalized Procedure as whether participants chose either camera versus declined to make a choice (indicating that Participants first were told that the session they would postpone the decision into the in- involved different experiments. Their initial definite future). MouselabWeb recorded the task manipulated attention control and hence time of each box opening and closing so we self-regulatory resources (Schmeichel, Vohs, & could calculate the length of time participants Baumeister, 2003). Participants watched an au- spent examining the available information dioless video of a woman being interviewed. about attributes. We also recorded the number Irrelevant words (e.g., hair, sky) periodically of mousein and mouseout actions. appeared onscreen for 10 s each. In the depletion condition, participants were instructed to Results focus on the woman and avoid the words. Thus, This document is copyrighted by the American Psychological Association or one of its allied publishers. they had to regulate their attention to curtail the As predicted by the passivity hypothesis, de- This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. tendency to look at the novel stimuli. In the pleted participants were significantly more neutral condition, participants were simply told likely than nondepleted controls to fail to come to watch the film. Next, participants were asked to a decision and instead to leave the choice to imagine going shopping for a digital camera. until some unspecified future time (57.9% vs. The screen showed two options (which were 42.2%), ␹2 ϭ 3.93, p ϭ .047; r ϭ .16. Thus, the approximately equally good), along with prod- hypothesis that depletion increased passivity uct names and features. Specifically, partici- was supported in the sphere of decision making. pants were told the following: Exploratory analyses tested the relationship Imagine that you would like to buy a digital camera. At between decision deferral and time spent on the store you normally shop, you find the following various parts of the task. The difference be- 328 VONASCH, VOHS, POCHEPTSOVA GHOSH, AND BAUMEISTER

tween the conditions in decision postponement rather than action, depleted participants who was not due to length of time spent on the web contemplated making a choice between the al- page when they were not looking at the attri- ternatives were more likely (than nondepleted butes (time between mouseout and mousein ac- participants) to postpone making a choice. tions) nor due to the number of times looking at Although depleted participants exhibited pas- the attributes (number of mousein and mouse- sivity or lack of action, they did not simply out actions), because neither duration nor num- avoid relevant choice information. Depleted ber of times looking at the attributes varied by persons did not increase their off-task time (as condition (ts Ͻ 1). However, duration of time indicated by time spent looking at the website looking at each attribute (time between clicking overall). They did, however, spend more time on an attribute and moving the cursor away with each piece of relevant information dis- from the attribute) did vary by condition. De- played onscreen, and at the end of this process pleted participants spent more time looking at indicated that they had failed to reach a decision the attribute information displayed than nonde- and would leave the choice until later. This ϭ ϭ pleted participants (Mdepletion 11.35 s, SD pattern is consistent with past literature in which ϭ 5.42, 95% CI [10.11, 12.59]; Mno-depletion depleted people spent equal or more time mak- 10.26 s, SD ϭ 4.60, 95% CI [9.26, 11.26]), ing decisions and this extra time on the choice t(157) ϭ 2.12, p ϭ .035, d ϭ 0.21. tasks did not lead to better choice outcomes One possible explanation for why depleted (Levav et al., 2010; Pocheptsova et al., 2009, participants spent more time on the attribute Studies 3 and 4). Moreover, it is consistent with information is that depleted participants were the idea that depletion interferes with active thinking carefully to make their decisions and mental processing. thus arrived at no decision. However, further exploratory analyses suggested that careful Study 3: Difficult or Easy Eating thinking did not explain the found effect of depletion on choice deferral. To test this, we ran Eating is a context that is often studied within a bootstrapping mediation analysis (Model 6, self-control paradigms because many people try Hayes, 2013) using 10,000 bootstrap samples, (with limited success) to control their eating, with both time spent looking at the attributes making it a common but difficult self-control and time spent not looking attributes as poten- task (Herman & Polivy, 2011). Eating also has tial mediators. Although the direct effect of the quality of requiring more or less impulse depletion on deferral rate remained significant control, depending on the ease of eating. From (␤ϭϪ.78, 95% CI [Ϫ1.44 to Ϫ.12], none of Schachter and Rodin (1974) to Wansink (2014), the indirect effects were significant (time look- research has shown that people who are over- ing at the attributes: b ϭ .09, 95% CI [Ϫ0.01 to weight make it relatively easy to overeat, such .33], time not looking at the attributes: b ϭ .04, as by using utensils as opposed to chopsticks 95% CI [Ϫ0.03 to .25]). Thus, although de- and choosing a seat near the buffet. pleted participants spent more time looking at Study 3 created a situation wherein eating attribute information, which could imply more was either easy or difficult to enact. Prior re- thinking, that time spent did not mediate the search has shown that depletion causes in- effect of depletion increasing decision deferral. creases in many kinds of activity, including If depleted participants were thinking more, that eating (Kahan, Polivy, & Herman, 2003; Vohs This document is copyrighted by the American Psychological Association or one of its allied publishers. thinking did not significantly aid them in ulti- & Heatherton, 2000). Study 3 aimed to recon- This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. mately making a decision. cile the fact that prior research has shown that depleted people become behaviorally disinhib- Discussion ited (i.e., show more activity, such as overeat- ing) with our thesis that depletion also can make When the choice options are both attractive people passive. Study 3 used a different manip- but differ on several attributes, choice among ulation of depletion (overriding the habit of them requires initiative, while avoidance in en- crossing out occurrences of the letter e on pages gaging in attribute trade-offs leads to choice of text) and measured initiative versus passivity deferral (Dhar, 1997). Consistent with our prop- based on the amount of food consumed. The osition that depletion leads to more inaction dependent measure was adapted from early EGO DEPLETION INDUCES MENTAL PASSIVITY 329

work by Schachter and Rodin (1974): Partici- of this manipulation is that participants in the pants took an ostensible taste test involving depletion condition must repeatedly and effort- peanuts either with or without shells. Eating fully override their automatic response to cross peanuts in shells takes more active participation out every e (Baumeister, Bratslavsky, Muraven, than eating ones without shells. & Tice, 1998). The hypothesis was that depletion increases Next, participants began a taste test. They mental passivity, thereby reducing top-down rated their current hunger level on a 10-point control and increasing automatic responses, Likert scale ranging from not at all hungry to whether action or inaction. Depletion would very hungry before being given bowls of un- reduce eating when eating was relatively effort- salted dry roasted peanuts (115g Ϯ 1), saltine ful (peanuts in shells). In contrast, depletion and crackers (36.2g), and ratings sheets. Saltines the resulting mental passivity would disinhibit were included with the peanuts to reduce pos- eating when it is relatively effortless (no sible suspicion about a taste test with only a shells)—provided, of course, that there is an single type of food, but the a priori predictions impulse to disinhibit. We assessed participants’ were about peanut consumption. Items on the level of hunger, on the assumption that only taste test form were constructed to appear like hungry participants would have the urge to eat. legitimate 5-point Likert style taste test ques- Among such individuals, we predicted that de- tions: for example, “How appealing are these pletion would decrease eating of peanuts with peanuts to you? How much do you like the shells but increase the (easy) eating of peanuts taste/flavor of this food? How do you like the without shells. texture of this food? How salty is this food to you? How likely would you be to buy this food Method in the store? How do you like the feel of the food in your mouth?” The experimenter in- Participants structed participants to rate the foods and, as exiting, casually mentioned that they could eat One hundred forty-three undergraduates (95 as much as they liked. After 6 min, the food was women) participated. We aimed to collect data cleared and participants completed a six-item from at least 120 participants, and continued measure of mental tiredness, the Multidimen- data collection through the end of the semester. sional Fatigue Symptom Inventory Mental Fa- The study had 80% power to detect an effect of tigue Subscale (Smets, Garssen, Bonke, & De size f2 ϭ .10 for each main effect and interac- Haes, 1995). Unbeknownst to participants, a tion. One participant was excluded for ignoring research assistant weighed the bowl of peanuts instructions. Two participants were excluded before and after the taste test, and subtracted the for drinking coffee during the taste test. The post- from the preweight to determine the design was a 2 (depletion vs. no-depletion) ϫ 2 amount eaten. (peanuts: with shells vs. without shells) between-participants design, with a continuous in- Results and Discussion dividual difference factor of hunger. We regressed grams of peanuts eaten on cen- Procedure tered participants’ hunger ratings, depletion condition, shells condition, and all interactions This document is copyrighted by the American Psychological Association or one of its allied publishers. Participants’ first task manipulated self- (R2 ϭ .17). As predicted, there was a significant This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. control resource depletion. All participants interaction among the three predictor variables practiced crossing out instances of the letter e (B ϭϪ4.28), t(133) ϭ 2.96, p ϭ .004. Con- that appeared on a piece of text. Then partici- trolling for the three variable interaction, there pants were given a second page with instruc- also was a significant hunger by depletion in- tions that varied by condition. No-depletion teraction (B ϭ 3.09), t(133) ϭ 3.44, p Ͻ .001, condition participants continued their habit of indicating that depletion had different effects on crossing out every e. Depletion condition par- consumption for hungry versus unhungry participants were instructed to cross out every e ticipants. The other two interactions and main except if it was followed by a vowel or a vowel effects were not significant (i.e., the hunger by appeared two letters before the e. The purpose shells interaction, B ϭ 1.32, t[133] ϭ 1.33, p ϭ 330 VONASCH, VOHS, POCHEPTSOVA GHOSH, AND BAUMEISTER

.185; the depletion by shells interaction, B ϭ We also examined comparisons within the Ϫ2.98, t[133] ϭ 0.996, p ϭ .32; the main effect peanut condition. Hungry participants ate more of hunger, B ϭϪ.46, t[133] ϭ 0.67, p ϭ .51; of the easy-to-eat peanuts when depleted than the main effect of depletion, B ϭ .32, t[133] ϭ when not depleted, consistent with the usual .16, p ϭ .88; the main effect of shells, B ϭ depletion effect of disinhibition (B ϭ 7.06), Ϫ1.02, t[133] ϭ .47, p ϭ .64). Analyses of the 95% CI [1.60, 12.52], t(133) ϭ 2.56, p ϭ .012. simple effects (with high and low hunger levels This effect was however eliminated, indeed al- at Ϯ1 standard deviation beyond the mean) most reversed, when the peanuts were difficult revealed support for the hypotheses (see Figure to eat, as hungry participants ate nonsignifi- 2). For descriptive statistics, see Table 2. cantly fewer difficult-to-eat peanuts when depleted than when nondepleted (B ϭϪ5.25), Hungry Participants 95% CI [Ϫ11.86, 1.36], t(133) ϭ 1.57, p ϭ .12. These findings support the hypothesis that Figure 2 shows the results for hungry partic- depletion reduces top-down control. These par- ipants (one standard deviaition above the hun- ticipants were motivated to eat (i.e., were hunger mean). When depleted, they ate signifi- gry). Depletion disinhibited them, so they ate cantly more when the peanuts were easy to eat more if eating was easy. But depletion also than when difficult (B ϭϪ10.45), 95% CI Ϫ Ϫ ϭ Ͻ produced passivity in the face of external de- [ 16.71, 4.18], t(133) 3.30, p .001. This mands for effort. When eating required effort to supports the main hypothesis that depletion in- break the shells, depletion failed to increase duces mental passivity, causing inaction: Even eating among hungry participants, and indeed when they were hungry, depleted participants the trend suggested it made them eat less. were less willing to open peanut shells in order to get to some food. It is not that depletion Not Hungry Participants reduced hunger, as indicated by the high consumption by depleted participants in the easy- Eating amounts did not vary much among not to-eat condition, and by the lack of difference in hungry participants. The only significant effect hunger ratings between depletion (M ϭ 4.39, was that participants tasting peanuts without the SE ϭ .25, 95% CI [3.89, 4.89]) and no depletion shell ate significantly less when depleted than (M ϭ 4.67, SE ϭ .27, 95% CI [4.15, 5.20]) when not depleted (B ϭϪ6.41), 95% CI [Ϫ12.20, conditions, F(1, 134) ϭ .60, p ϭ .44. When Ϫ.62], t(133) ϭ 2.19, p ϭ .030. All other con- hungry participants were not depleted, they ate trasts (ts Ͻ 1.23). Thus, when motivation to about the same amount regardless of ease of consume was low, depletion and difficulty were eating (t Ͻ 1, ns). Thus, active versus passive irrelevant (see Figure 2). responding was only relevant to the eating of Overall, the results were also consistent with depleted persons. motivational intensity theory (Brehm & Self,

15 This document is copyrighted by the American Psychological Association or one of its allied publishers.

This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. 10 Shells No Shells 5 Grams of peanuts eaten Grams of peanuts

0 Neutral Depleon Neutral Depleon Low Hunger High Hunger

Figure 2. Consumption of peanuts by condition, among hungry and unhungry participants (one standard deviation above and below the mean of hunger). EGO DEPLETION INDUCES MENTAL PASSIVITY 331

Table 2 activities. In Study 2, depleted people devoted Descriptive Statistics for Study 3 plenty of time to reading information about Variable MSDcameras—but without ultimately completing the task of choosing which camera they pre- Peanuts eaten (grams) 8.60 9.51 ferred. This is consistent with past research Crackers eaten (grams) 10.91 7.82 Mental Fatigue Index 2.23 .81 showing that depletion undermines people’s Hunger before taste test (1 to 10 scale) 4.84 2.18 ability to select tasks appropriate for their de- Hunger after taste test (1 to 10 scale) 4.52 2.13 pleted mental state (DeWall, Baumeister, Mead, & Vohs, 2011). Indeed, it fits much evidence that making decisions requires effort expenditure (e.g., Vohs et al., 2008). 1989). According to this theory, motivation in- Study 3 revealed that the same activity (eat- creases with perceived effort until the perceived ing) either can increase or decrease in the de- effort becomes so high that the task is insuffi- pleted state, depending on whether the activity ciently rewarding to merit further effort. Deple- requires active effort and on whether the person tion is assumed to increase perceived effort is motivated to eat. This corroborates and builds (Wright, 2014). Thus, depletion would make on Carver et al.’s (2008) dual process model, participants more motivated to eat peanuts, as which states that behavior can be predicted by long as they were hungry and the effort required the degree of reflection versus impulsivity along to eat the peanuts was relatively low. However, with the reward value of the stimulus. After in the shells condition, participants were hun- following challenging rules requiring them to gry, but the perceived effort required to eat the first form and then break a habit, hungry partic- peanuts was higher, because of the shells. One ipants showed two different patterns in two way of interpreting the results is that partici- different conditions, such that one pattern pants might have perceived that the reward of showed depleted persons to increase eating eating peanuts was lower than the potential ef- while the other showed decreased eating. fort costs of opening the shells. However, we (Meanwhile, participants who were not hungry measured eating, which does not necessarily ate about the same amount regardless of ease track effort. Therefore, although the results are and depletion.) What determined whether de- consistent with the motivational intensity the- pleted, hungry participants ate much or little ory, this experiment more directly tested the was how difficult it was to get access to the limited resource model. food. Even the small requirement of having to crack the shells in order to get at the peanuts General Discussion was enough to discourage depleted participants from getting the food they wanted. But when no The central conclusion is that ego depletion shelling was required, hungry, depleted persons produces a state of mental passivity. This does ate more than nondepleted ones. These findings not necessarily mean that the person physically confirm the effects in the Carver et al. (2008) does nothing. Rather, either physical stillness or theory, both more passive and more impulsive. impulsive hyperactivity can both flow from These findings also corroborate and extend mental passivity, depending on the situation. upon the results of Giacomantonio, Jordan, Fen- The depleted person seems not to exercise top- nis, and Panno (2014). Their study involved This document is copyrighted by the American Psychological Association or one of its allied publishers. down mental control over behavioral responses, either pushing a button or pumping a bicycle This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. as compared with other persons. Lacking initia- pump in order to fill virtual balloons worth tive, the person may sometimes indeed sit still. rewards. The balloons would sometimes pop if But an impulsive depleted person may act vig- pumped too full. Hence, pumping involved orously on those impulses, without the restrain- some risk. They found that depletion made peo- ing influence of inhibitory executive control. ple take greater risks in the pursuit of reward Depletion made people less selective in what when doing so was physically easy (pushing a they chose to do. Although depleted people’s button). Like participants in our peanut study strongest preferences were to do passive, ener- (3), depletion increased reward pursuit when gy-saving activities, in their passive state, they rewards were immediate and easy to attain. In could also be compelled to do energy-intensive these situations, depletion led to action, presum- 332 VONASCH, VOHS, POCHEPTSOVA GHOSH, AND BAUMEISTER

ably because action was the default and auto- though to our knowledge this prediction has not matic response. In the shells condition of the yet been tested. peanut study, eating involved more effortful Both the limited resource model (Baumeister et override of the default passive response. Deple- al., 1998) and motivational intensity theory tion again increased the default response, to be (Brehm & Self, 1989) assume that people con- passive and not open the shells. Similarly, de- serve energy or effort, and this affects how they pletion in the bicycle pump condition led to the regulate their behaviors. Motivational intensity tendency to be passive and pump the balloons theory is primarily concerned with how situations less. In some ways, the peanut study was a more affect effort and is often tested using physiological direct test of the passivity hypothesis. Passive measures (e.g., Agtarap, Wright, Mlynski, Ham- behavior in the balloons study could result from mad, & Blackledge, 2016). The limited resource a propensity to be passive, but it could also model, by contrast, has mostly been tested in result from a propensity to avoid risk. However, terms of depletion’s behavioral consequences. there is little risk to avoid in eating peanuts. The present research suggests that the two models, Thus, one benefit of the present studies was while distinct, are largely compatible in that they their varied methods in demonstrating the effect make similar predictions about the effects of de- of depletion on passive behavior. Across multi- pletion on effort. However, the scope of the mo- ple methods, the present studies provide further tivational intensity model is broader, explaining evidence that depletion increases passive be- why effort is sometimes low even when people havior when immediate rewards are not present having plentiful self-regulatory resources. Addi- or are difficult to obtain. tionally, the motivational intensity model assumes Ego depletion leads to mental passivity, that depletion’s effects on effort result from it which can impair self-control in two ways, de- increasing people’s perceptions of task difficulty, pending on the default response: increase pas- such that their motivations change. This motiva- sive responding or increase impulsive respond- tional explanation is more consistent with the pro- ing. In contrast, counteractive self-control is a cess model (Inzlicht & Schmeichel, 2012, dis- set of strategies people engage in to aid them- cussed below) than the limited resource model. selves in pursuing long term goals, especially in Future research testing predictions that differ be- the face of short term temptations (Trope & tween motivational intensity theory and the lim- Fishbach, 2000). For example, people can im- ited resource model could provide valuable infor- pose fines on themselves if they fail to adhere to mation about the mechanism behind ego a three day fast as part of a medical procedure, depletion, which remains an unsettled research thereby increasing the likelihood that they will question (see Baumeister & Vohs, 2016; Inzlicht successfully complete the fast (Trope & Fish- & Schmeichel, 2012; Job, Dweck, & Walton, bach, 2000). Presumably, one reason why self- 2010). imposed fines are helpful is they change the Although the present article is framed from automatic response people have when they ac- the theoretical perspective of the limited re- tually have to make the decision—the fine sources model (Baumeister, Vohs, & Tice, makes the otherwise tempting option much less 2007), it is worth discussing how the results appealing. For another example of counterac- pertain to alternative models of ego depletion tive self-control, health conscious exercisers de- (i.e., Inzlicht & Schmeichel, 2012; Job, Dweck, value the appeal of chocolates compared to & Walton, 2010). This document is copyrighted by the American Psychological Association or one of its allied publishers. healthy foods when making health-related deci- This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. sions (Myrseth, Fishbach, & Trope, 2009). This Process Model strategy might also work because it shifts the automatic response to the chocolate; chocolate The process model (Inzlicht & Schmeichel, is less tempting if it is less appealing. Ego 2012) holds that ego depletion results from depletion leads to increased reliance on default shifting motivational processes. As a person responses, and counteractive self-control proac- engages in self-control to perform a “have to” tively changes those responses to be better task, over time one’s motivation gradually shifts aligned with one’s long term goals. Therefore, to a “want to” task. Simultaneously, one’s at- counteractive self-control strategies may be par- tention begins to shift away from task-oriented ticularly important when people are depleted, demands to cues of reward and gratification that EGO DEPLETION INDUCES MENTAL PASSIVITY 333

are often associated with other “want to” tasks. icant differences in the direction predicted by In many situations, the process model and the the theory. We stipulate that we also conducted limited resource model make similar predic- additional studies, not reported here, and they tions. For example, the process model would too found significant effects consistent with the also predict that depletion might cause passiv- theory. The continuing accumulation of signif- ity. Many passive activities are enjoyable “want icant findings consistent with ego depletion is to” kinds of tasks. Indeed, in one experience- increasingly difficult to reconcile with the hy- sampling study, the most intense desire people pothesis that there is no such phenomenon. had was to sleep (Hofmann, Vohs, & Baumeis- ter, 2012). The process model might also ex- Limitations and Future Directions plain the appeal of active activities such as sports in the wake of depleting events, as sports The surprising finding of Study 1 was that are often “want to” tasks. The process model is people who recently did depleting things pre- also consistent with the peanut study’s findings ferred active (as well as passive) activities. We that depleted people ate more easy to eat pea- argued that the most likely explanation for this nuts, but fewer difficult to eat peanuts, because pattern is a selection effect—that people who eating peanuts is presumably a “want to” task, prefer active activities are more likely to have whereas opening shells is not. It is less clear recently done something depleting. Future lon- how the process model would predict the results gitudinal examinations of depletion might pro- of Study 2, in which people postponed final vide causal evidence in support of this idea that decision-making about products despite spend- individual differences might account for peo- ing the time to review information about the ple’s preferences. products. It is unlikely that people thought that Study 2 showed that depleted people spent gathering information about products was a more time evaluating decisions, only to fail to “want,” but deciding on a product was a “need.” come to a final decision. This finding is consistent with past findings that depleted people Implicit Theories of Willpower spend equal or more time making decisions as nondepleted people, but that this extra time on Another theoretical perspective on ego deple- the choice tasks does not lead to better choice tion is the idea that people’s implicit theories about whether willpower is limited or nonlim- outcomes (Levav et al., 2010; Pocheptsova et ited affect their behavior in response to deple- al., 2009). Contrarily, shorter rather than longer tion (Job, Dweck, & Walton, 2010; Job, Wal- decision times often leads to simplified decision ton, Bernecker, & Dweck, 2015). According to strategies (Dhar & Nowlis, 1999; Dhar, Nowlis, this model, people who believe their willpower & Sherman, 2000). One intriguing possibility is can be depleted by short tasks behave as though that the mental passivity brought about by de- it has been depleted. One limitation of the pres- pletion may impair people’s ability to monitor ent research is that because it did not measure the strategies they use to make decisions implicit willpower theories, it cannot test the (Carver & Scheier, 2001). Although usually predictions of this model. Past research suggests longer decision times lead to more complex and that believing in limited willpower can increase better decisions, depletion appears to degrade passive resting (Job, Bernecker, Miketta, & Fri- the decision-making process in two ways— This document is copyrighted by the American Psychological Association or one of its allied publishers. ese, 2015) and impulsivity (Job, Dweck, & making it both slower and less effective. If This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Walton, 2010). Future research should test people were strategically approaching the task, whether these effects are also moderated by the one would expect them to realize that they were presence of immediate reward. unlikely to make a decision in their passive state We noted earlier that some writers have ques- of mind, and therefore not spend so much time tioned the existence of ego depletion effects, contemplating their decision. But in the de- based in part on a single failure to replicate pleted state, the mind may be too passive to (Hagger et al., 2016). Such questions make it notice its own impairment. Future research imperative to report further evidence. The pres- should investigate why depleted people seem to ent findings provide further evidence of the lack the appropriate strategic response to deci- reality of ego depletion, as they yielded signif- sion making. 334 VONASCH, VOHS, POCHEPTSOVA GHOSH, AND BAUMEISTER

Conclusion 355. http://dx.doi.org/10.1111/j.1467-8721.2007 .00534.x The present findings extend the concept of Brehm, J. W., & Self, E. A. (1989). The intensity of volition beyond self-regulation and choice, to motivation. Annual Review of Psychology, 40, encompass active responding. Taking the initia- 109–131. http://dx.doi.org/10.1146/annurev.ps.40 tive to choose effortful activities, to make a .020189.000545 decision rather than postpone it, and even to Carter, E. C., Kofler, L. M., Forster, D. E., & Mc- satisfy one’s hunger when eating requires effort, Cullough, M. E. (2015). A series of meta-analytic tests of the depletion effect: Self-control does not is apparently difficult for and unappealing to seem to rely on a limited resource. Journal of people whose self-regulatory resources are low. Experimental Psychology: General, 144, 796– Volition is a key aspect of the human self 815. http://dx.doi.org/10.1037/xge0000083 (e.g., Baumeister, 1998). Decision making and Carver, C. S., Johnson, S. L., & Joormann, J. (2008). self-regulation have long been recognized as Serotonergic function, two-mode models of self- key aspects of volition. Apparently, taking ini- regulation, and vulnerability to depression: What tiative to respond actively is another important depression has in common with impulsive aggres- aspect. Much prior work has shown that when sion. Psychological Bulletin, 134, 912–943. http:// the self’s energy is low, it ceases to regulate dx.doi.org/10.1037/a0013740 behavior effectively and makes decisions based Carver, C. S., & Scheier, M. F. (2001). On the on simple, easy processes. The present findings self-regulation of behavior. New York, NY: Cam- extend that work to suggest yet another conse- bridge University Press. quence of depleted resources: The self starts to Danziger, S., Levav, J., & Avnaim-Pesso, L. (2011). Extraneous factors in judicial decisions. Proceed- favor doing nothing. ings of the National Academy of Sciences of the United States of America, 108, 6889–6892. http:// dx.doi.org/10.1073/pnas.1018033108 References DeWall, C. N., Baumeister, R. F., Mead, N. L., & Vohs, K. D. (2011). How leaders self-regulate Agtarap, S. D., Wright, R. A., Mlynski, C., Hammad, their task performance: Evidence that power pro- R., & Blackledge, S. (2016). Success importance motes diligence, depletion, and disdain. Journal of and urge magnitude as determinants of cardiovas- Personality and Social Psychology, 100, 47–65. cular response to a behavioral restraint challenge. http://dx.doi.org/10.1037/a0020932 International Journal of Psychophysiology, 102, Dhar, R. (1997). Consumer preference for a no- 18–24. http://dx.doi.org/10.1016/j.ijpsycho.2016 choice option. The Journal of Consumer Research, .03.004 24, 215–231. http://dx.doi.org/10.1086/209506 Banker, S., Ainsworth, S. E., Baumeister, R. F., Ari- Dhar, R., & Nowlis, S. M. (1999). The effect of time ely, D., & Vohs, K. D. (2017). The sticky anchor Journal of hypothesis: Ego depletion increases susceptibility pressure on consumer choice deferral. to situational cues. Journal of Behavioral Decision Consumer Research, 25, 369–384. Making. Advance online publication. http://dx.doi Dhar, R., Nowlis, S. M., & Sherman, S. J. (2000). .org/10.1002/bdm.2022 Trying hard or hardly trying: An analysis of con- Baumeister, R. F. (1998). The self. In D. T. Gilbert, text effects in choice. Journal of Consumer Psy- S. T. Fiske, & G. Lindzey (Eds.), Handbook of chology, 9, 189–200. social psychology (4th ed., pp. 680–740). New Fishbach, A., & Shah, J. Y. (2006). Self-control in York, NY: McGraw-Hill. action: Implicit dispositions toward goals and Baumeister, R. F., Bratslavsky, E., Muraven, M., & away from temptations. Journal of Personality and This document is copyrighted by the American Psychological Association or one of its allied publishers. Tice, D. M. (1998). Ego depletion: Is the active Social Psychology, 90, 820–832. http://dx.doi.org/ This article is intended solely for the personal use of the individual user and is notself to be disseminated broadly. a limited resource? Journal of Personality and 10.1037/0022-3514.90.5.820 Social Psychology, 74, 1252–1265. http://dx.doi Giacomantonio, M., Jordan, J., Fennis, B. M., & .org/10.1037/0022-3514.74.5.1252 Panno, A. (2014). When the motivational conse- Baumeister, R. F., & Vohs, K. D. (2016). Strength quences of ego depletion collide: Conservation model of self-regulation as limited resource: As- dominates over reward-seeking. Journal of Exper- sessment, controversies, update. Advances in Ex- imental Social Psychology, 55, 217–220. http://dx perimental Social Psychology, 54, 67–127. http:// .doi.org/10.1016/j.jesp.2014.07.009 dx.doi.org/10.1016/bs.aesp.2016.04.001 Hagger, M. S., Chatzisarantis, N. L. D., Alberts, H., Baumeister, R. F., Vohs, K. D., & Tice, D. M. Anggono, C. O., Batailler, C., Birt, A. R., . . . (2007). The strength model of self-control. Cur- Zwienenberg, M. (2016). A multilab preregistered rent Directions in Psychological Science, 16, 351– replication of the ego-depletion effect. Perspec- EGO DEPLETION INDUCES MENTAL PASSIVITY 335

tives on Psychological Science, 11, 546–573. Levav, J., Heitmann, M., Herrmann, A., & Iyengar, http://dx.doi.org/10.1177/1745691616652873 S. S. (2010). Order in product customization deci- Hagger, M. S., Wood, C., Stiff, C., & Chatzisarantis, sions: Evidence from field experiments. Journal of N. L. (2010). Ego depletion and the strength model Political Economy, 118, 274–299. http://dx.doi of self-control: A meta-analysis. Psychological .org/10.1086/652463 Bulletin, 136, 495–525. http://dx.doi.org/10.1037/ Muraven, M., Shmueli, D., & Burkley, E. (2006). a0019486 Conserving self-control strength. Journal of Per- Hayes, A. F. (2013). Introduction to mediation, modera- sonality and Social Psychology, 91, 524–537. tion, and conditional process analysis: A regression- http://dx.doi.org/10.1037/0022-3514.91.3.524 based approach. New York, NY: Guilford Press. Myrseth, K. O. R., Fishbach, A., & Trope, Y. (2009). Hedgcock, W. M., Vohs, K. D., & Rao, A. R. (2012). Counteractive self-control. Psychological Science, Reducing self-control depletion effects through en- 20, 159–163. http://dx.doi.org/10.1111/j.1467- hanced sensitivity to implementation: Evidence 9280.2009.02268.x from fMRI and behavioral studies. Journal of Con- Pocheptsova, A., Amir, O., Dhar, R., & Baumeister, sumer Psychology, 22, 486–495. http://dx.doi.org/ R. F. (2009). Deciding without resources: Re- 10.1016/j.jcps.2012.05.008 source depletion and choice in context. Journal of Herman, C. P., & Polivy, J. (2011). The self- Marketing Research, 46, 344–355. http://dx.doi regulation of eating: Theoretical and practical .org/10.1509/jmkr.46.3.344 problems. In K. D. Vohs & R. F. Baumeister Polman, E., & Vohs, K. D. (2016). Decision fatigue, (Eds.), Handbook of self-regulation: Research, choosing for others, and self-construal. Social Psy- theory, and applications (pp. 522–536). New chological and Personality Science, 7, 471–478. York, NY: Guilford Press. http://dx.doi.org/10.1177/1948550616639648 Hofmann, W., Friese, M., & Strack, F. (2009). Im- Rachels, J. (2007). Active and passive euthanasia. In pulse and self-control from a dual-systems per- N. S. Jecker, A. R. Jonsen, and R. A. Pearlman spective. Perspectives on Psychological Science, (Eds.), Bioethics: An introduction to the history, 4, 162–176. http://dx.doi.org/10.1111/j.1745-6924 methods, and practice (pp. 64–69). London, UK: .2009.01116.x Jones and Bartlett Publishers, Inc. Hofmann, W., Vohs, K. D., & Baumeister, R. F. Richter, M., Gendolla, G. H. E., & Wright, R. A. (2012). What people desire, feel conflicted about, (2016). Three decades of research on motivational and try to resist in everyday life. Psychological intensity theory: What we have learned about ef- Science, 23, 582–588. http://dx.doi.org/10.1177/ fort and what we still don’t know. Advances in 0956797612437426 Motivation Science, 3, 149–168. Inzlicht, M., & Schmeichel, B. J. (2012). What is ego Ryan, R. M., & Frederick, C. (1997). On energy, depletion? toward a mechanistic revision of the personality, and health: Subjective vitality as a resource model of self-control. Perspectives on dynamic reflection of well-being. Journal of Per- Psychological Science, 7, 450–463. http://dx.doi sonality, 65, 529–565. http://dx.doi.org/10.1111/j .org/10.1177/1745691612454134 .1467-6494.1997.tb00326.x Job, V., Bernecker, K., Miketta, S., & Friese, M. (2015). Schachter, S., & Rodin, J. (1974). Obese humans and Implicit theories about willpower predict the activation rats. Mahwah, NJ: Lawrence Erlbaum. of a rest goal following self-control exertion. Journal of Schmeichel, B. J., Vohs, K. D., & Baumeister, R. F. Personality and Social Psychology, 109, 694–706. (2003). Intellectual performance and ego deple- http://dx.doi.org/10.1037/pspp0000042 tion: Role of the self in logical reasoning and other Job, V., Dweck, C. S., & Walton, G. M. (2010). Ego information processing. Journal of Personality depletion—Is it all in your head? implicit theories and Social Psychology, 85, 33–46. http://dx.doi about willpower affect self-regulation. Psycholog- .org/10.1037/0022-3514.85.1.33 ical Science, 21, 1686–1693. http://dx.doi.org/10 Smets, E. M. A., Garssen, B., Bonke, B., & De Haes, This document is copyrighted by the American Psychological Association or one of its allied publishers. .1177/0956797610384745 J. C. J. M. (1995). The Multidimensional Fatigue This article is intended solely for the personal use of the individualJob, user and is not to be disseminated V., broadly. Walton, G. M., Bernecker, K., & Dweck, Inventory (MFI) psychometric qualities of an in- C. S. (2015). Implicit theories about willpower strument to assess fatigue. Journal of Psychoso- predict self-regulation and grades in everyday life. matic Research, 39, 315–325. http://dx.doi.org/10 Journal of Personality and Social Psychology, .1016/0022-3999(94)00125-O 108, 637–647. http://dx.doi.org/10.1037/ Trope, Y., & Fishbach, A. (2000). Counteractive pspp0000014 self-control in overcoming temptation. Journal of Kahan, D., Polivy, J., & Herman, C. P. (2003). Con- Personality and Social Psychology, 79, 493–506. formity and dietary disinhibition: A test of the http://dx.doi.org/10.1037/0022-3514.79.4.493 ego-strength model of self-regulation. Interna- Vohs, K. D., Baumeister, R. F., Schmeichel, B. J., tional Journal of Eating Disorders, 33, 165–171. Twenge, J. M., Nelson, N. M., & Tice, D. M. (2008). http://dx.doi.org/10.1002/eat.10132 Making choices impairs subsequent self-control: A lim- 336 VONASCH, VOHS, POCHEPTSOVA GHOSH, AND BAUMEISTER

ited-resource account of decision making, self- cess tracing methods for decision making (pp. 21– regulation, and active initiative. Journal of Personal- 42). New York, NY: Taylor & Francis. ity and Social Psychology, 94, 883–898. http://dx Wright, B., & Baumeister, R. F. (2015). Brief mea- .doi.org/10.1037/0022-3514.94.5.883 sure of ego depletion. Manuscript in preparation. Vohs, K. D., & Heatherton, T. F. (2000). Self- Wright, R. A. (2014). Presidential address 2013: Fa- regulatory failure: A resource-depletion approach. tigue inﬂuence on effort—Considering implica- Psychological Science, 11, 249–254. http://dx.doi tions for self-regulatory restraint. Motivation and .org/10.1111/1467-9280.00250 Emotion, 38, 183–195. http://dx.doi.org/10.1007/ Wansink, B. (2014). Slim by design. New York, NY: s11031-014-9406-5 HarperCollins. Willemsen, M. C., & Johnson, E. J. (2010). Visiting the decision factory: Observing cognition with MouselabWEB and other information acquisition Received February 21, 2017 methods. In M. Schulte-Mecklenbeck, A. Küh- Revision received May 18, 2017 berger, & R. Ranyard (Eds.), A handbook of pro- Accepted May 19, 2017 Ⅲ

Members of Underrepresented Groups: Reviewers for Journal Manuscripts Wanted

If you are interested in reviewing manuscripts for APA journals, the APA Publications and Communications Board would like to invite your participation. Manuscript reviewers are vital to the publications process. As a reviewer, you would gain valuable experience in publishing. The P&C Board is particularly interested in encouraging members of underrepresented groups to participate more in this process.

If you are interested in reviewing manuscripts, please write APA Journals at [email protected]. Please note the following important points:

• To be selected as a reviewer, you must have published articles in peer-reviewed journals. The experience of publishing provides a reviewer with the basis for preparing a thorough, objective review.

• To be selected, it is critical to be a regular reader of the ﬁve to six empirical journals that are most central to the area or journal for which you would like to review. Current knowledge of recently published research provides a reviewer with the knowledge base to evaluate a new submission within the context of existing research.

• To select the appropriate reviewers for each manuscript, the editor needs detailed information. Please include with your letter your vita. In the letter, please identify which APA journal(s) you are interested in, and describe your area of expertise. Be as speciﬁc as possible. For example, “social psychology” is not sufﬁcient—you would need to specify “social cognition” or “attitude change” as well. This document is copyrighted by the American Psychological Association or one of its allied publishers.

This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. • Reviewing a manuscript takes time (1–4 hours per manuscript reviewed). If you are selected to review a manuscript, be prepared to invest the necessary time to evaluate the manuscript thoroughly.

APA now has an online video course that provides guidance in reviewing manuscripts. To learn more about the course and to access the video, visit http://www.apa.org/pubs/ authors/review-manuscript-ce-video.aspx.