'"

_.",:;', ,.""

A1TlTUDES AND SOCIAL COGNITION

Paradoxical Effects of Thought Suppression

Daniel M. Wegner David J. SchneiderTrinity University University of Texas at San Antonio

Samuel R. Carter III and Teri L. WhiteTrinity University

In a first experiment, subjects verbalizing the stream of consciousness for a 5-min period were askedto try not to think of a white bear, but to ring a bell in case they did. As indicated both by mentionsand by bell rings, they were unable to suppress the thought as instructed. On being asked after thissuppression task to think about the white bear for a 5-min period, these subjects sh~ significantlymore tokens of thought about the bear than did subjects who were asked to think about a white bearfrom the outset. These observations suggest that attempted thought suppression has paradoxicaletfects as a self-control strategy, perhaps even producing the very o~on or preoccupation that itis directed against. A second experiment replicated these findings and sh~ that subjects given aspecific thought to use as a distractcr during suppression were less likely to exhibit later preoccupa-tion with the thought to be suppressed.

Consciousness cannot produce a negation except in the form of the most vexing problem of thought suppression: the self-refer-consciousness of negation. ent quality of the plan to suppress. To suppress a thought re-

Sanre, Being and Nothingness (1956, p. 43) quires that one (a) plan to suppress a thought and (b) carry outthat plan by suppressing all manifestations of the thought, in-

It is sometimes tempting to wish one's thoughts away. Un- cluding the original plan. Thought suppression thus seems topleasant thoughts, ideas that are inappropriate to the moment, entail a state of knowing and not knowing at once. Freud ( 1915/or images that may instigate unwanted behaviors each can be- 1957) made this strange dissociated state theoretically possiblecome the focus of a desire for avoidance. Whether one is trying by postulating the unconscious and by further specifying thatnot to think of a traumatic event, however, or is merely attempt- the unconscious was capable of performing the thought sup-ing to avoid the thought of food while on a diet, it seems that pression for consciousness. So, although the unconscious couldthought suppression is not easy. It is said, for instance, that not remove the thought from itself, and consciousness alsowhen the young Dostoyevski challenged his brother not to think could not remove the thought from itself, the unconscious couldof a white bear, the child was perplexed for a long while. Con- perform this housecleaning for the separate, conscious part oftemporary psychology has not focused much inquiry on such the mind.puzzling yet important phenomena, and our research was de- The psychoanalytic emphasis on such unconscious repres-signed to initiate such investigation. sion has resulted in a longstanding bias against the examination

of consciousness during processes of thought suppression.The Problem of Thought Suppression Rather, the process of suppression has been expected to be ob-

servable only after the fact, leaving its mark on memory. Thus.The idea that people may have unwanted thoughts was one of even contemporary research investigates directed forgetting

Freud's fundamental insights. and his notion that people re- (e.g., Geiselman, Bjork, & FlShman, 1983) and posthypnoticpress such thoughts has long served as a theoretical rallying amnesia (e.g., Kihlstrom, 1983) rather than directed inatten-point in the study of psychopathology (Erdelyi & Goldberg, tion or directed conscious avoidance. These lines of investiga-1979; Hart, 1934). Still, classical psychoanalytic theory skirts tion do not attempt to explain how or with what effect people

go about the conscious task of suppressing a thought. Yet thereare multiple instances in everyday life when this is precisely

, what people try to do. Trying not to think about an upcomingWe thank Oaudia Se:ran°' Susan Sh~ckelford, DebbIe S~~ and stressful event, avoiding thoughts of smoking while trying to

Sharon Thoms for help In conducting this research, and -r:°rn GI~I~O, quit, or putting persistent thoughts of a lost love out of mindPaul Paulus, James ~. Pennebaker, Thomas Pyszczynski. and William are common experiences for many. Worries of every kind are

B. Swann, Jr., for their suggesnons. , 'I I 'th ghts th t I ess the desire not, this ' I h uld be addressed to Dan SIIn1 ar y conSCIous ou a peop e exprCorrespondence concerning arnc e s 0 -, '

, I M ,,' Departm ent of Psyc holoav Trinity University, San AntI>- to have. What happens when people make a COnSCIOUS effort toIe ...egner. OJ' .' ular th gh ?nio, Texas 78284. avoId a partlc ou t.

}ournaIofPeroonalityand~ ~oIocy. 1987, Vol. 53. No. I. ~-13C~t 1987 by tbeAmerian ~caI Asoociatioa.lnc. 0022.3514/87/SOO.75

5

"' " '",,' ,Iii,,"

~-.~ -

6 WEGNER. SCHNEIDER. C-\R:rER, AND WHITE

Although the evidence is sketchy, there is some indication from food or addictive substances. Given the assumption thatthat the ~k of conscious thought suppression can be di.fficult. the attempt to avoid a habitual behavior is commonly preceded~y ~dies by McGranahan (1940) and.Sears and Virshup by attempts to suppress or avoid habit-related thoughts, the pat-(C1t~ m Sears, 194~) ~howed m.at ~le instructed to avoid tern of behavior following self-control attempts is informativemaking ~lo~ assoCIattOns to StImulus words often reported about the pattern of thinking that may take place. In the case ofsuch ~atlons nonetheless, even when threatened with shock. abstinence from food. for instance, Polivy and Herman (1985)for doing so. In ~ cases, of course. ~le did not know in indicated that dieting generally causes subsequent overeating.advance the specific thought they were to suppress. only the gen- They cite several converging sources of evidence suggesting thateral category of colors. More recently, Logan (1983) examined the restraint of eating is a reliable precursor of binge eating andpatterns of reaction time to st~ signals given as subjects per- overweight. It seems, then, that the attempt to avoid thoughtsformed brief tasks, and found that although actions can be of food may lead to a later preoccupation with such thoughts.st~ped in midcourse, thoughts seem to run to their conclusion The more general abstinence-violation effect observed by Mar-as long as the stimuli activating them are present. ConSistent Ian and Parks (1982) suggests that the State of abstinence is awith this, but only broadly relevant, are the outcome studies of precarious one, in that relapse to an addictive behavior can bethe thoUght-st~ping therapy technique. This therapy, suggested trigg&:red by a single, seemingly minor violation of the prohibi-by Wolpe and I...aza.rus (1966) for obsessional disorders, appears tion. This, too, is consistent with the idea that an initial attemptgenerally no more successful than no intervention at all (Reed, to suppress thoughts can be followed by an unusual preoccupa-1985). tion with the suppressed thought domain. Their observation

Another line of evidence arises in research that calls for pea- suggests further that the event that serves to halt suppressionpIe to ignore information that is relevant to a judgment they and trigger relapse may be only a single occurrence that drawsmust make. Whether ~le are instructed to ignore the infor- the person's attention to the originally suppressed thought.mation before they encounter it (e.g., Wegner, Coulton, & The tentative conclusions suggested by past findings are two-Wenzlaff, 1985) or are told to disregard it afterwards (e.g., Ross, fold. First, it seems that thought suppression is difficult for pea-Lepper, & Hubbard. 1975), they tend to incorporate it into sub- pIe to do; the conscious avoidance of a thought may be perplex-sequent judgments nonetheless. Jurors are influenced by infor- ing and even time consuming. Second. there is some evidencemati on they have been instructed to disregard (Thompson, to suggest that even when thoughts can be suppressed, they mayFong, & Rosenhan, 1981), media audiences are influenced by return to consciousness with minimal prompting, perhaps tonews they are told is untrue (Wegner, Wenzlaff, Kerker. & Beat- become obsessive preoccupations. These general expectationstie, 1981), and people judging odds are influenced by informa- were explored here in two experiments through the expedienttion even when they have been offered money to ignore it (Tver- of asking people to s;lppress a thought while they deliveredsky & Kahneman, 1974). These effects would seem unlikely if stream-of-consciousness reports in a laboratory setting.

~le could will away their conscious experience of thethoughts they were instructed to ignore (cf. Sherman & CoTty, .1984). The hypothesis suggested by a diverse array of findings ExperIment I

to date, then, is that conscious thought suppression is not a cog- Mi th dnitive transformation that people perform with great facility. e 0

Subjects and design. Trinity University undergraduates (14 men andConsequences of Thought Suppression 20 women) in introductory ~hology cl~ volunteered to partici-

pate in return for extra class credit. Each was randomly assigned to oneThe mental state produced by an attempt at thought suppres- of two experimental conditions, an initial suppression condition or an

sion seems to differ in several ways from that accompanying initial expression condition. These conditions differed only in the ordersimple inattention or unintended distraction. The hypothesis ~f two experimental tasks. For initial suppression, the .subj~ ~ firstsuggested b several theorists is that attempts to suppress Instructed to s~pprcss ~ thought, and ~ to express It; for Imtial ex.

y.. pressiOn, these InstruCUOns ~ gIVen In reverse order.thoughts (or emottons) can result m a subsequent rebound of Pr ed Each b. ect ~rtici pa tcd individuallv stani ng byreadingud .thi DC UTe. su ~ ..-~ "absorpuon WIth those t~lcs. The prototypic st y m s area a set of instructions on how to report one's stream of consciousness.(Janis, 1958) showed that individuals who are personally in- Theinstructions~adaptcdfromthoseuscdby~(1978)and~clined to avoid thinking about an upcoming surgery subse- fashioned to encourage continuous verbalization. The instructionsquently exhibit more anxious reactions to it. Although the asked only for subjects to describe what they ~ thinking; there wasmeaning of this finding still is in debate (e.g., Janis, 1983; Laza- no special appeal for the subject to explain or justify the thought (cf.

rus, 1983), there is a degree of theoretical unanimity in the con- Ericcson & Simon, 1984). .clusion that avoiding a stressful thought can lead to subsequent The participant then was asked and gave Informed con~t to ~dintrusions of that thought (e.g., Horowitz, 1975). Even recent several 5-min Pe:"ods alone repo~ng.to a tape ~rder everythin.8

that comes to mind:' For each perIod. It was explained that the expcn-resear~h m thiS domain contlnues to depend on the p~eselectlo~ menter would say '.begin ,. and then leave the room for the duration ofof subjects who are prone to suppress. (e.g., Burstem & Mel- the period. After one such practice period. the experimenter returnedchenbaum, 1979), however, and for this reason the processes to issue additional instructions. Participants assigned to the initial sup-underlying any transition from suppression to absorption re- pression group were told the following:

main unexamined.The possibility that thought suppression leads to absorption In the next five minutes. please verbalize your thoughts as you didL

can also be found in the reactions people have to abstinence before, with one exception. This time. try not to think of a white

,m",',-

..

PARADOXIC\l EFFEcrs OF THOUGHT SUPPRESSION 7

Table I .od . .slon appears to pr uce a rebound effect, a surge In the fre-

~lment 1.. Measures of Thought by Group and Peri~- quency .of s~uent thought about the white bear during the

Period expr~on penod.EVIdence of a rebound following suP~on also arose in

Group Suppression Expression correlational analyses conduCted within groups. As would be~- --, uu prediCted on the basis of a straightforward individual differ-

Initial expression ..

theBell with mention 2.75 6 96 enc.es In.terpretatIOn,. to~ number of thought tokens for aBell only 4.09 4:86 subject ~ the two Ume perIods (expression and suppression)Mention only 0.43 4.56 was POSlbvely correlated among subjects in the initial expres-

Initial su.PP~o:n sion group, r( I 7) = .55, p < .02. This was not the case, however.Bell WIth mention 1.36 8.00 in the initial suppression group. Here, a zero-order correlationBell only 4.71 7 71 .ndi ted " la Mention only 0.23 6'35 I ca an antagoDlSUc re bon between think;ng of a white

-._~ bear during suppression and thinking of it during the subse-NOte. 11 = 17 for each group. quent ex~on period. This value, r(17) = -.10, was not sig-

nificantly different from zero but was significantly smaller thanthe .55 value observed in the other condition (p < .01). In the

bear. Every time you say "white !Jear" or have "white !Jear" come i~tiaI suppression group, then, success at the initial suppres-to mind, though. please ring the bell on the table before you. Slon task created a readiness for later expression, one that sig-

.nificantly attenuated the more usual tendency for ~le whoFollOWIng this, these participants were given expression instructions for express their thoughts at one time to express them at another.a ~ubsequent .perlod; they ~ asked for an additional 5 min to "try to A further indication of the rebound effect was observed inthink of a white ~" Their task description was otherwise identical to the cowse of thought occurrences over the 5-min periods. Thisthe ~ In that they ~ asked to continue ringing the bell to indicate .each instance of the thought. The participants ~gned to the initial pheno~enon was encountered when a ~YSIS o~ ~e tapes,exp~on group ~ given these instructions in ~ ord~ they by I-~n segm.ents. ~ made for ~otal bell nngs. (~ data~ asked in the initial period to try to think of a white bear and in the for this analYSIS resulting from slightly short final I-lDln seg-subsequent period to try not to think of a white bear. ments for 6 subjects, 3 in each group, were replaced by each

subject's total for the prior I-min segment.) As shown in FigureResults I, bell rings per tninute increased over time during the expres-

sion period in the initial suppression group. This would be un-Analysis of the tape recordings was made for (a) bell rings remarkable except that bell rings per minute decreased over

occurring simultaneously with audible mentions of "white time in every other measurement period: in the initial expres-bear." (b) bell rings occurring alone, and (c) mentions occurring sion group for both expression and suppression periods and inalone. Intercoder reliability between a pair of coders averaged the initial suppression group for the suppression period..94 over the three measures. Table I shows the means for each A 2 (initial instruction group) X 2 (period) x 5 (I-min seg-measure by condition. ment) ANOV A on square root transfonned total bell rings indi-

A 2 (initial suppression vs. initial expression group) x 2 (sup- cated that this trend interaction was reliable. There was a sig-pression period vs. expression period) X 3 (thought measure) nificant three-way interaction of group, period. and linear or-analySis of variance (ANOV A) was conducted on transfonned thogonal polynomial trend over time segments. F( I, 31) = 7.02,thought token scores. A square root transformation was made p < .02. Participants who were allowed to express a thought theyto achieve homogeneity of error variance in this analysis (Kirk. had recently suppressed devel~ an accelerating tendency to1968, pp. 63-65), but for clarity, the reported means are not report the thought. overcotning the more usual progressive ills-transfonned. interest or fatigue displayed by participants in other periods.

Tokens of thought were more prevalent in the expression pc- The stream-of-<:onsciousness protocols revealed some inter-riods than in the suppression periods. The mean of the summed esting facets of the thought-suppression process that are not evi-thought measures during expression was 19.22, and this was dent in the counting of thought tokens. The protocols made itsignificantly greater than the suppression mean of 6.78, F(I, clear. for example, that many of the subjects saw the thought-32) = 41.0 I, p < .000 I. Despite this difference, it is noteworthy suppression task as difficult or puzzling from the outset. Manythat suppression was never complete. Participants indicated also verbalized a strategy soon after encountering the task. say-thinking about a white bear (through a bell ring. mention, or ing to themselves. in effect, "Okay, so I'll think of somethingboth) more than once per tninute even when directly instruCted else:' Perhaps most interesting. then, was the manner in whichto try not to think of a white bear. thoughts of the white bear returned to subjects when they were

Tokens of thought were more frequent in the expression pe- reciting their various replacements for the thought. White-bearriod following initial suppression than in the initial expression mentions and bell rings almost invariably occurred. when theperiod. A significant interaction of initial instruction and pe- subject had finished a sentence or a thought and was silent. Dur-riod. F(l, 32) = 4.79, P < .05, was comprised of a significant ingsuppression, subjects were generally able to keep fromsimple main effect of initial instruction group in the expression thinking about the tar&et thought. or at least ':"'e.re able to keepperiod. F(I, 32) = 5.05, p < .05. A sitnilar effect was not ob- from reporting it. as long as they were v~aIizi~ the thoughtserved in the suppression period following initial expression, of something else. These sorts of thou.ght In~ons were notF( I, 32) < I, indicating an unusual asymmetry: Initial suppres- common, however, during the expressIon penods.. Here, men-

I ..,~_.

-..

8 WEGNER. SCHNEIDER, CARTER. AND WHITE

helps people set aside tlteir immediate concerns in tlte setting5.0 PRESSION and get involved in the experimental task.

SUPPRESSION

Discussion

The paradox.ical effect of thought suppression is that it pro-4.0 du~ a preoccupation with tlte suppressed tltought. These

fi~dings su~ that the task of suppressing a tltoUght is itselfdifficult. leading ~Ie to hold the thought in consciousn~

~ repeatedly even as they try to eliminate it. When they are then~ 3,0 released from the supp~on task and asked instead to go~ ahead and express the thought. they do so at an accelerated rateffi mentioning it more often than if they had simply been asked ~~ express the thought from the start. There are thus both immedi-~ ~ORE ate and delayed tendencies toward conscious preoccupationcr 2.0 with the very thought that is being suppressed.

~ Potential explanations for these two effects can be derived~ from several psychological theories. As noted earlier. psychoan-

alytic theory is not clear regarding the process or consequence1.0 SlFPRESSION of conscious supp~on. and although some ideas might be

AFTER EXPRESSION gl cd fr theF udian .ean om re VerSIon of unconscious repression.~SSlON they are not strictly applicable to the present case. This is parc

BEFORE EXPRESSK)N .cuJariy .thetI true given nature of the thought that subjects were

asked to suppr~ in this study. Psychoanalytic interpretations1 2 3 4 5 are commonly extended only to cases of the f~ or avoid-

MNUTE an~ of threatening or stressful thoughts (Holmes. 1974), ~dFigure 1. Bell rings per miD over the 5-rnin periods. white bears are generally nonthreatening this side of the Arctic

Circle.It may be possible to 3-wmilate these findings to the theory

of psychological reactance (Brehm. 1966)0 Because tlte persontions and bell rings typically occurred in tlte context of subjects' in tlte rebound period is doing the ~te of what the experi-meaningful narratives about white bears. menter requested in tlte suppression period, one might argue

The observed effects have been examined in subsequent stud- that reactance was aroused by the suppression task. By thisies varying several parameters of the experimental design. In logic, tlte experimenter's instruction to avoid thinking of aone such investigation, Shackelford, Wegner, and Schneider white bear restricted the subject's freedom to do so, and this(1987) arranged for three groups of 10 subjects each to partici- restriction made the thought more attractive. Thus, the subjectpate in experimental conditions like those of the initial suppres- tends to dwell on the thought in the suppression period, andsion group in this study. One group participated in an exact once given the freedom to return to tlte thought in the expres-replication, whereas the other groups participated with either sion period, becomes preoccupied with it.(a) no bell, and verbalization serving as the thought measure, or The difficulty with this interpretation comes wilen ~ try to(b) no verbalization required. and the bell alone serving as tlte understand why a negative injunction should create more reac-thought measure. The different metrics used in these groups tance than a positive one. After all, the subjects in the initialmake mean comparisons between groups problematic, but the expression condition also had their freedom restricted; theyzero-order correlation between thought tokens measured dur- were told to think of a white bear. Reactance predicts that theying suppression and subsequent expression was replicated for would avoid thinking of a white bear during the expression pe-all groups (each r < .07). These findings thus cast doubt on the riod, as they did (as compared with subjects in the expressionpossibility that the bell-ringing requirement in the original period of the initial suppression condition). But it would alsostudy had any special influence on the occurrence of the re- predict that initial expression subjects would be motivated tobound effect. avoid the tltought during the subsequent suppression period

One oilier follow-up study by Carter, Wegner, and Schneider and would do so more than the subjects in the initial suppres-(1987) is worth noting in that it showed what may be an impor- sion condition. Such an effect should be of roughly the sametant limiting condition for the finding that ~le have trouble magnitude as the rebound phenomenon observed in the initialsuppressing tltoughts. In this research, the tendency to report suppression condition. This did not happen. and the reactancethinking of tlte to-be-suppressed thought during the suppres- explanation thus falls short of a full explication of these find-

sion period was reliably reduced when subjects were given no ings.chance to practice verbalizing the stream of consciousness be- Another framework that could be applied here is self-percep-fore they began the suppression task. It may be that subjects tion theory (Hem, 1972). This th.eory does n,ot seem relevant tomust be comfortable making such reports before they will ad- the initial difficulty ~le have In sUppfCSSlng thoughts, but itmit to suppression difficulty, or perhaps the initial reporting does offer a prediction in line with some of the rebound evi-

-', --~ , '-

..

PARADOXICAL EFFECTS OF THOUGHT SUPPRESSION 9

~nce. The th~ry says that a ~n w~o perfonns a behavior. in lated to the negative cue. These things are not white bears. Such

presence 0 external COnStral~ts will perform that behavior a connection to all ongoing thoughts is not produced b themore often later when the conStralnts are removed. The experi- complementary insttuction to think of a white bear. h Y

menter's request to suppress thoughts of a white bear is indeed because the ~ ~ thisPOSl.u..1:"-1 oweYer.

lternal ~ b"'~~ ve cue IS ~ Y to exp ore

an. e~ co.nStralnt. and. beca.u5e ~~e find themselves only a limited range of possible thought t~cs: those .thinking of white .bears ~lte ~ COnStralnt. they might de- relevant to white bean. During expression. subjects ~

ve~~ .a self-~lved POS1~ve. attttude towa:d the activity of zoos, the North Pole, big black noses, and the like. Thinkinthinking of whit.e bean. This Idea also predicts, h~ that about a white bear can be carried out successfully merel b g

~~e woul~ think about a white bear in ~ subsequent ex- rehearsing ideas found in memory that are already associa:ve~

pressIon ~od to the degree that they had disobeyed the exter- linked to white bean. And indeed. thinking bo t tb .nal ~nstral~t and ~ought about a white bear in the prior sup.. in the room or entertaining other ongoing ~~ts° d~~

~ress1on perIod. This was not the case, as a zero:or~ correla- pr=on would probably be identified by the subject as a failuretlO~ was o~~. between ~ought .frequenCIes 10 the ~ t.o ~e at the experimental task. Negative cuing, in this

perIods ~ong 1O1tiaJ su~press1on subjects. So self-perception light, brings many more new ~cs of thought into the class of

theory fails to capture .an Important facet of these data. things that are contrasted with (and thereby linked associatively

The pr~ explanation of the thought-suppression effects ap.. to) the thought to be suppressed than does a process of simple

pears to reqUIre the devel~ment of some new theoretical ideas. concentration or directed thinking.

We have devised a framework consistent ~th several key f~- Negatively cued associations are not likely to be very strong,

tures of th~ observed e~ects. and we prOVIde one test of this however. and may not be sufficient to remind the person of

framework 10 the follOWIng study. white bears very frequently (cf. Semin & Rosch. 1981). Because

the person is not consciously trying to learn this association. an

Experiment 2 implicit rather than an explicit association is formed (Graf &

Schacter, 1985). Later. however, when the injunction to avoidThe results of the first experiment may be understood white-bear thoughts is lifted and the person is invited to think

through an analysis of the process of self-distraction. Essen- of white bears, the prior negative<uing facilitates the task of

tially, the individual who is faced with the task of suppressing continuing to think about white bears. Everything in the room,

a thought is given a negative cue for subsequent thinking, an and in recent memory, is now imbued with a slight degree of not

instruction of what.to avoid rather than an instruction of what white bearness, and thoughts of white bears are thus implicitly

to approach. This means that there is no single item for atten- primed by many ongoing conscious thoughts. The negative-cu-

tion; there is, for instance, no one such thing as not white beal: ing explanation of thought-suppression effects, in sum, indi-

This general idea can be helpful for understanding both the ini- cates that the task of stopping a thought has the effect of produc-

tial difficulty of suppression and the source of the subsequent ing associations of that thought with many other thoughts im-

, rebound. mediately available to the person, and that these associations

Suppression is difficult because thinking without focus is function to make the thought rebound when the injunction to

;.j difficult. The person engaging in self-distraction turns to avoid the thought is no longer in effect.

thoughts of many things. The person's attention may range over The negative-cuing interpretation of thought suppression

a wide expanse of mental territory, but in considering such provides some further predictions on the nature of suppression

items the person may find little that is quickly absorbing and effects. It suggests, for instance, that rebound effects should bethat suggests a new line of thought. For example, a person using lar&ely eliminated if the person uses a positive cue for self-dis- .

white bear as a negative cue might say, 'Tll think about the light traction during suppression. The attempt to think of just one

switch instead." The light switch is the focal point for a mo- replacement for the avoided thought rather than many might

ment. but it turns out on examination to be less than intriguing. not keep the person deeply interested. and so might not reduce

The person's attention may drift to something else. or perhaps the initial difficulty of suppression-but it could very well

more commonly, the person consults in memory the most re- short-circuit the rebound. With a single distracter for the to-be-

cently occurring definition of the task at hand-in effect, asking suppressed thought. the person would not later be reminded of

"What am I doing?" When this happens. of course, the person the thought by many other cues and so would not be as likely

is often reminded of a white bear and must begin again. This to become preoccupied with it during the expression period.

circular process may underlie the difficulty of self-distraction. This idea was tested in this experiment.

Using a negative cue for thought tends to distract the person

from the current mental task toward anything else, and the per- Method

son who has so much of nothing in particular to consider then ...consults memory about what should be done to find a reminder Subjects. Undcrgraduatcs3sfrom Tnmty lu~ty and fro~ San ~nth- ..tonio Coll* (16 men and women) vo un~QO to partICIpate, WI

and thus begin again. ...the Trinity students receiving extra credit in introductory ~hology

Although the negative CUIng task does not pr~uce effectIve classes for their participation. Tape recordings for 5 subjects ~ not

suppression, it is effective in producing another Important cog- clearly audible and their data ~ not included.

nitive change. All the things the person is thinking about during Design and procedure. The subjects ~ randomly assigned to three

this task become associated with the negative cue. That is, the experimental conditions. Two of these ~ exact replications of the

person comes to see present stimuli (e.g., the room, experi- initial expression and initial suppression conditions of Experiment I.menter, etc.) as well as thoughts retrieved from memory as re- The third was identical to the initial suppression condition with one

., .

10WEGNER. SCHNEIDER. CARTER. AND WHITE

Table 2 .E.~periment 2: .Veasures of Thought by Group and Period pressJ°Ken group, ovex:aI1 F(2, 46.> =.4.54, p < .02, with the New-

--, ---y ~"VW man- u~ com~son for thiS difference at p < .05. This did

Period not occur In the pnor study, and it is not clear why such a result.-appeared ~ere. The overall tendency for white-bear thoughts to

Group Supp~on Expression OCCUr dunng suppression was not reliably reduced ' th "cused-dis .In e 10-Thought occurrence traction group. So, although these subjects mentioned

.red Volkswagcns often (M = 3.00), they stilI kept tl1in.king of a~m~ expression 4.13. 15.47b white bear during suppression at a rate equivalent to that ofnit! sup!'ressI«:,n 9.17. 34.05~ subjects in the other groups.Focused distraction 5.94 21 00 .

-"--e The ~~ught-duratlon measure indicates that subjects acr~Thought duration all COnditl°n.s ~ about a white bear for a mean of 8.65 s in

In!t!aI expressi~n 9.07 64.87 : :~~resslon .PCrtod. Th~ it seems ~t the ~ces ofInitial suppression 7.71 115.53 ght d~ SUPPression were relatively brief, averagingFocused distraction 9.24 65.18 1.41 s .each. This duration is consistent with our observation in

the pnor study that thoughts of the item to be supp~ in-Note. Means WIth a common SU~pt are SIgnificantly different by a !rude on the verbal stream in only an abbreviated duriNewman-Keulstest,p<.05.ForlmtialexPresslon,n= 15; for the other suppression.Therewasno,,;-;~ t " .way .nggroups, n = 17 each. .~can vanatlon In the duration

of white-bear thoughts acr~ conditions (F < I). By this mea-sure, then, suppression in the three groups was ~erally equiva-lent.

exception: Subjects in this focused distraction condition ~ asked to The mean .durati,:>n of thought about a red Vo~ indistract themselves during initial suppression by means of a single posi- the fOCUsed-distraction condition was 20.7650 the equivalent oftive cue, the thought of a red Volkswagen. They ~ told after the sup- 6.92 s per mention. This sUggests that subjects in this conditionpression instruction, ..Also, if you do happen to think of a white bear. took seriously their task. of focusing on this diStracter.please try to think of a red Volkswagen instead." No further mention of Expression period. Thought occurrence during expressionthe red Volkswagen was made during the remainder of the experiment (as shown in Table 2) was significantly influenced by group, F(2,for these subjects. 46) = 4.62, p < .02. Newman-Keuls comparisons indicated

that the initial suppression group mean was reliably greaterResults than the mean for the initial expression group, p < .05. Thus,

the general finding of a rebound effect in Experiment I was rep..Analysis of the tape recordings was made for the same vari- licated. (The temporal trends observed in that study, shown in

abIes examined in the first study (bell rings occurring with men- Figure I, were not tested here as the result of the heterogeneitytions of a white bear, bell rings occurring alone, and mentions of error-variance problem. Inspection of means suggested thatoccurring alone), and in addition for (a) duration in seconds of the trend for increasing thought over time during expression indiscussion about a white bear. (b) mentions of a red Volkswagen, the initial suppression group was not as pronounced as in theand (c) duration in seconds of discussion of a red Volkswagen. prior study.) The mean level of thought occurrence during ex-Intercoder reliability between a pair of coders averaged .96 for pression in the focused distraction group was significantly lessthe thought-occurrence measures (i.e., bells and mentions) and than the comparable mean in the initial suppression group (p <.86 for the thought-duration measures. .05). This indicates that the rebound effect for thought occur-

An initial attempt was made to use repeated measures AN- rences was reliably reduced in the focused-distraction group.ov AS paralleling those of Experiment I. In that study, we were Thought duration during expression exhibited a marginallyable to solve the problem of significant heterogeneity of error significant group effect. F(2, 46) = 3.0 I, p < .06. The differencesvariance in such analyses by the use of a square root data trans- between means reflected the same pattern as that observed forformation. The same transformation did not achieve homage- the thoUght-occurrence measure: The initial suppression groupneity of error variance in this experiment, and all repeated mea- showed a tendency toward a rebound effect. in that its level dur-sures analyses were therefore disallowed. With the square root ing expression was elevated over that of the initial expressiontransformation, however. homogeneity was obtained in three- group; the focused-distraction group, in turn, showed a ten-group between-subjects ANOV As, and these were thus con- dency for the rebound to disappear, in that its level during ex-ducted for each of two thought measures: an overall measure of pression resembled that of the initial expression group.thought occurrence and the measure of thought duration. Table Correia/ions within groups. As in the prior study, corre-2 shows raw scores for mean number of thought occurrences lations were calculated between thought-«currence scores for(sum of bell rings, mentions of a white bear. and bells with si- subjects in the expression and suppression periods. These cor-multaneous mentions) and the mean thought duration in the relations revealed a pattern of individual variation consistentsuppression and expression periods for each group. with the prior study. Subjects' inclination to note thought oc-

Suppression period. As in Experiment I, subjects in this currences was significantly correlated between the expressionstudy found suppression difficult. Across all conditions, sub- and suppression periods for subjects in the initial expressionjects indicated thinking of a white bear 6.15 times in the 5-min group (r = .50, p < .02). This correlation was nonsignificant insuppression period. Subjects' preoccupation was significantly the initial suppression group (r = .23, p > .18). Although thesegreater in the initial suppression group than in the initial ex- correlations are not reliably different, their relative magnitudes

", ..,...

---o_o.,""""c"""'-",..-."

PARADOXICAL EFFECTS OF THOUGHT SUPPRESSIO N 11

replicate the pattern observed in Experiment I and reveal a ten I a bal th.~ency toward rebound only under conditions of initial suppres: the ~ st:c

ye'were;' Itlicantedbe.c°thisn~lUded that general trends of

slon. 0 orep ca In one..0 The negatIve-ewng hypothesis devised t fi

thBy this correlatIonal measure of the rebound effect, fOCused servatio f th first stud 0 ~unt or e ob-distraction eliminat~ the rebound. The correlation between that SUb~ inea focused~=:~ here. ~ account holdsthough! occurrences.1n the suppression and expression periods CUrrent inclination to think about an~P m:t defeat a :e-for subjects performIng the focused distraction was significant bear b

y focusing ontheSl°ngi A:__- ng ~deder than whi~

( -55 02 ) Th the usual e UQUd1;1CJ; prOVl them. Thisr -., p < ..us, tendency for thought-occur- should, in turn, reduce the degree to which (negatively cued)

rence measures to be correlated was found here even when sub- associations are formed ' ~ the hi bear.ects gaged.. be" .~..~ w te and other cur-J en In suppression lore expression. Apparently, the rent thoughts and so reduce the r:t_-lihoodtha subseqt h . f t . t distra ..~ t uent con-ec nlque 0 UInlng 0 one cter-ln thIS case, the red tinuations of these CUrrent thoughts uld . th . Vi IJ..r..-~ . uffi 0 eradi .0 .0 wo pnme e Idea of the0 ~~~u-1S S Clent to. cate this indicatIon of a sup- white bear during the later expression period. As predicted b

pressIon-induced rebound In thought. Parallel correlational this hypoth-ic fr-'~"" diStJ'acti. d yal .0 ""~ on unng suppresSIon In thisan yses that were conducted WIth the duratIon measures ex- experiment produced a significantatten h f th bo dh ObOted omi}ar b .0 .0 uauon 0 e re unI I ..a Sl pattern ut no Slgnlficant correlatIons In any effect. Subjects in the condition receiving these special instruc-

conditIon. o. ..tions exhibited patterns of thinking largely indistinguishableA~ong subJ~.1n the focused-distractlon group during sup- from those of subjects in the initial expression group and sig-

pressIon, mentIoning a red Volkswagen tended to be correlated nificantly different from those of sub!J' ects .n the.' tiaJ0h 0di . f 0 I 1m suppres-WIt In catIons 0 whIte-bear thought OCCUrrence (r = ,4 I, P < sion group..06). A ~al correlation taking into a.cc°unt the rela~on be- This is but a first outing for the negative-cuing hypothesis, oftween whIte-bear thought occurrences In the suppression and course, and further inquiry will be needed to detennine whetherexpressi.on conditio~s showed that ~entioning a red Volkswa- it serves as an adequate account of the rebound phenomenon.gen. dunng suppressIon was not reliably related to the level of The hypothesis also predicts. for example, that the reboundwhlt~-bear thought occurrence during subsequent expression might be eliminated by dissociating the contexts in which sup-(partial r = .33, ns). The negative-cuing hypothesis predicts a pression and expression are performed. People who try not to

negative correlation here, but it is difficult to discern whether think about a white bear in one context would form negativelythis finding represents an anomafy for the hypothesis or an effect cued associations to a white bear only in that context. On mov-of insufficient within-group sample size for statistical inference. ing to a different context. their Constellation of ongoing

thoughts would change, with far fewer thoughts now priming awhite bear. and the rebound might be defeated. The relief from

Discussion old worries one sometimes experiences on traveling to a newenvironment might be an example of this.

The results of this study are consistent with those of Experi- Focused distraction also has its everyday equivalents. Morement 1 in suggesting that suppression is difficult and that it has than one person has attempted to find comfort in the face ofreliable effects on subsequent thinking. The ability of subjects unwanted thoughts by appealing to a single distracter. be it ato suppress the thought of a white bear was not in strong evi- bare light bulb, a religious icon, or perhaps even a red Volkswa-dence here, as subjects tended to note the thought's occurrence gen. The results of this study suggest that there may be a certainmore than once a minute even as they were attempting to sup- use in turning to a familiar talisman in the pursuit of mentalpress it. This was true also among subjects given a focused dis- peace. Although the distracter may not ease the current tasktraction, the instruction to think of a red Volkswagen whenever of suppression, it could block negative cuing and so serve thea white bear came to mind. welcome purpose of reducing later resurgence of the suppressed

The rebound phenomenon observed in Experiment I was thought.found in this study as well. Subjects who initially suppressedthe white-bear thought and then were allowed to express it General Discussionshowed an elevated tendency to report its occurrence duringtheir expression ~portunity. Subjects in this initial suppression The results of these experiments suggest that the portrayal ofgroup, as it happened. also exhibited a significant elevation in suppression as the parent of obsession may contain a degreetheir degree of thought occurrence during suppression relative of truth. The process begins when a person attempts to put ato the other groups. and this finding raises the possibility that particular thought out of mind. This need not be an especiallysubjects in this group were merely more inclined to report obnoxious or unnerving thought-even the thought of a whitethought occurrences than were subjects in the other groups. The bear will do. The person finds the thought hard to suppress andmeasure of thought duration, however. showed no such pattern, may soon wonder why this particular thought is so insistent.indicating instead a relative but nonsignificant lack of concern Continued suppression may eventually remove theo thoughtwith the white bear during suppression for initial suppression from mind, for the present. Then, however. some remInder oc-subjects. Yet. by this measure, a marginally significant tendency curs. an? inoa moment of weakness the .~n gives license totoward a rebound was observed in the initial suppression group the rumination. Our results suggest that In this moment. an un-during the expression period. In addition, the correlation pat- usual preoccupation with the formerly suppressed th.oUght maytern between thought occurrences in the initial suppression and begin. This preoccupation may grow and prosper. In the per-initial expression groups paralleled the pattern of Experiment son's mind. And quite ironically, the person who IS first most

~,

I ~ WEG:-.lER. SCHNEIDER. CARTER. A:-ID ~'HITE

successl"ul in carrying out the suppression may eventually be dictions crushes and obsessions (see Ra. h S1most susceptible to the resulting obsession. 1978). " .e,g., c man & de I va,

.The momentum imparted to thought by an ~ct of suppres- .The observed processes. though fairly tame in the laboratory,slon has long been know~, or at least sUS~ed, m ~any q~- mIght conceivably create powerful mental preoccupations inters of psychology. The Idea that there IS danger m keepmg natural setti

ngs. This is beca"- in dail lifi '-h. bottled ' 'd has -" ed , y e, suppressIon att mgs up Insl e SUI lac m a varIety ?f ~orms. Lin- tempts and subsequent rebounds of thinking could occur. re-

de.mann. (1944), for example, s~ggested that gn~ng foll.ows peatedly, escalating in response to each other. and so yield dra-

this rule, he argued that suppressIon can produce an msuffiClent maticalIy

magn ified effects. A person mightbegi ha _If .

f k d " .n, per ps OUJy

amount 0 gne wor an so can Impair copmg by the bereaved. on a whim to su

ppress a certain thou ght. The Su 'S' 'I ti ul ' '. , ppre5Slon pro-ImI ar orm .auons promot,e e~ouonal cathars~ of many cess might be difficult, but the person could probably arrive at

sorts, from l~tUng.out a~CSS1v~ Impulses to working through successful suppression in a relatively brief period. Later on,

pent-up feelmgs m relauonshlps. Such .an .antisuppression however. some tri~ for the rebound OCCUrs and the personmodel appears to be an exaggerated generalizaUon ofpsychoan- becomes involved in an excessive level of rumination. It is at

alytic ideas. (cf. Freud, 1914/1958), and alth.ough it is widely this point that the person becomes alarmed, noticing that anadvocated m popular psychology channels, It has to date re- unusual degree of preoccupation is underway. This might pro-mained largely unformulated in psychological theory and un- duce a newly energized attempt at suppression, only to restartsupported in laboratory settings (e.g., Geen & Quanty, 1977). the cycle. Suppression might be yet more difficult at this time,

This picture of neglect and negative evidence is now chang- but it could seem to be the only solution. Eventually, pathologi-

ing, however, suggesting the beginnings of a new psychology of cal levels of obsessive concern could result. A similar analysissuppression. The work of Pennebaker ( 1985) is notable in this might be made of the processes of addiction. Attempts to con-regard, showing in several field investigations that the suppres- trol a habit such as smoking could take the form of thoughtsion or inhibition of emotional and cognitive reactions to trau- suppression, and these in turn could prompt rebounds of exces-malic events may yield physiological changes and subsequent sive attention to the act of smoking. Cycles of suppression andhealth problems. A related line of inquiry by Silver, Boon, and preoccupation might then be standard fare for ~le who at-Stones (1983) suggests that suppression may block a natural tempt to control an addiction in this way, resulting in repetitivetendency to find meaning in traumatic events and that this can abstinence and relapse.hamper effective coping processes. These investigations focus On the practical side, then, perhaps it is fortunate that ouron individuals responding in vivo to profound traumas, and al- findings signal at least one possibility for relief from the ironic

though Pennebaker reported some success in the laboratory complications of thought suppression. Quite simply, it appearssimulation of certain key aspects of this process, it is still true that when suppression is transformed into an active interest inthat phenomena observed in vivid field demonstrations remain a single distracter, the longer term dangers of a rebounding pre-elusive in controlled settings. This may be one way in which the occupation with the suppressed thought may be prevented. Re-laboratory paradigm can contribute. Our findings support the turning to a particular idea whenever one worries might providegeneral idea that suppression can backfire, and at the same time, some reduction in the eventual extent of the worrying. This pro-the results make this point without recourse to the major emo- cedure might prove to be of some use in the elimination of ob-tional traumas usually associated with such effects (cf. Rach- sessional thinking or addictive preoccupation. To be sure, how-man, 1980). The suppression results we have observed suggest ever, the dimensions of this effect are only incompletely graspeda straightforward cognitive mechanism mirroring, and perhaps at this time, and the form of an adequate theory of successfulunderlying, a wide array of psychological phenomena: emo- suppression is not imaginable yet. Much more needs to betional, cognitive, and behavioral as well. learned about the parameters ofth~ paradoxical effects we have

It is yet an open question, of course, whether the suppression observed before we can suggest WIth any.confidence that theyeffects we have observed will generalize to items other than offer a proper analog of naturally OCCurrIng processes. In thewhite bears. The nature of the thought being suppressed- meantime, though, it seems clear that there is little to be gainedwhether it is emotional or not, easily imagined or not, familiar in trying not to think about it.or not, complex or not, and the like-would seem to be an im-portant detertninant of suppression effects. The finding that an Referencesitem as unremarkable as a white bear can yield suppressionproblems, however, suggests that the effects are at least not tied Bem, D, J: (1972). Self-perception theory. In L Berkowitz (Ed.), AIi-to any obvious stimulus qualities. Other points of concern for vances In exp~nmenlai SOCIal ps~hology (Vol. 6, pp. 1-62). Newthe validity of these findings center on the degree to which the York: Academic Press.observed effects might be dependent on the imposition of the BTeAcadehm, J. .W..:.!!_66). A theory of ps~hologicai reactance, New York:

b .'th ght mlC '-1~trappings of the laboratory on su ~ects ou pr~. Burstein, S" & Meichenbaum, D. ( 1979). The work of worrying in chi)-Whether people do attempt by themselves to stop thmki~ drcn undergoing surgery, Journal of Abnormal Child PS}ChologJ( 7,

(without our instruction to do so) and whether they then expen- 127-132.

ence difficulty (when they are not being asked to verbalize or Carter, S. R., III, Wegner, D, M., & Schneider, D, J. (1987), [Limitinggive another signal of their thought) at this point must re~n conditions of thought suppression]. Unpublished raw ~unresolved issues. We can only remark on the many striking Ericsson, K. A., & Simon, H. A. (1984). Protocol analysIS. Verbal reoresemblances between the observed effects and the everyday ob- pons as data. Cambridge: MIT Press. .servations people make about the tenacity of their worries, ad- Erdelyi, M, H., & Goldberg, B. (1979), Let's not sweep repression under~

It-r;"~

.

PARADOXICAL EFFECTS OF THOUGHT SUPPRESSION 13

the rug: Toward a cognitive psychology of repression. In J. F. Kihl- Pennebaker, J. W. (1985). Inhibition and cognition: Toward an under-strom & F. J. Evans (Eds.), Funclional disorders of memory (pp. 355- standing of trauma and disease. Canadian Psychology. 26. 82-95.402). HillsdaIe, NJ: Erlbaum. Polivy, J., & Herman. C. P. (1985). Dictingand binging: A causalanaly-

Freud. S. (1957). Repression. In J. Strachey (Ed.), The standard edition sis. American Psychologist. 40. 193-20 I.of the complete psychological ~rk.r of Sigmund Freud (Vol. 14, pp. P~, K. S. (1978). How gender. solitude. and posture influence the146-158). London: Hogarth. (Original work published 1915) stream of consciousness. In K. s. ~ & J. L. SinF (Eds.), The

Freud, S. (1958). Remembering, repeating, and working-through. In J. stream of consciolLSness (pp. 259-299). New York: Plenum ~Strachey (Ed.), The standard edition of the complete psychological Rachman. S. (1980). Emotional processing. Behavioural Research and~rks of Sigmund Freud (Vol. 12, pp. 145-150). London: Hogarth. Therap}( 18. 51~.(Original work published 1914) Rachman. S., & de Silva, P. (1978). Abnormal and nonnal obsessions.

Gecn, R. G., & Quanty, M. B. (1977). The catharsis ofaggrcssion: An Behavioural Research and Therapy. 16.233-248.__I ' f h th ' I L n--L. . (Ed.) .-' Reed, G. F. (1985). Obsessional experience and compuisi~ behallior:,"VdJuatlon 0 a ypo CSIs. n .~..OWItz ,/IUvunc;es In ~- 0rIa000, F1.: Aademic ~=al SOCIal psychology (Vol. 10, pp. 1-37). New York: Academic Rou, L, Lepper, M. R., ~ Hubbard, M. (1975). ~ce in self-

" .perception and social perception: Biased attributionaJ processes inGeisel~ .R. ~., Bjork, R. ~., & F~~, D. L (1983): Disru~ the debriefing paradigm. Jouma/ of Personality and Social Psycho/-

retneval m directed fof!CtUng: A lInk WIth posthypnotic amnCSla. ogy. 32. 880-892. \ :~Journal of Experimental Psychology: General, 112. 58- 72. ~ J. (1956). Being and nothingness. New York: Phil~hical li-

Graf, P., & Schactcr, D. L (1985). Implicit and explicit memory for new brary.associations in nonnal and amnesic subjects. Journal of Experimen- Scars, R. R. (1943). Swwy of objecti~ studies of pS)f:hoanalyric con-tal Psychology: Leaming, .\.{emo~ and Cognition, 11. 501-518. cepts. New York: Socia! Science Rcsearcl1 Council

Hart, B. ( 1934). The psychology of insanil}! New York: Macmillan. Semin, G. R., & Rosch, E. (1981). Activation of bipolar prototypes inHolmes, D. S. (1974). Investigation of rcp~on: Differential rccalI of attribute inferen~ Jouma/ of Experimental Social Psychology. 17,

material experimentally or naturally associated with ego threat. Psy- 472-484.chological Bulletin. 81. 632-653. Shackelford, S., WegnCl; D. M., &Schncidcr, D. J. (1987). [The effect of

Horowitz, M. (1975). Instrusive and repetitive thoughts after experi- reporting technique on thought su~onJ. Un,published raw data.mental strcss.Archi~sofGeneral PS)f:hial~ 32. 1457-1463. Sherman. S. J., & Corty, E. (1984). Cogm~ heunstl~ In R. S. Wycr,

J I L (1958) P h I al N '" k.W' 1 Jr., & T. K. Srull (EdSo), Handbook of SOCial cognition (Vol. I, pp.anls, ..syc o.oglc stress.. ew .lor. ley. , 189-286). HillsdaIe, NJ: Erlbaum.

Janis, I. L (1983). Prcventlng pathogemc denial by means ofstrcssmoc- Silver, R. L., Boon, C., & Stones, M. H. (1983). Searching for meaningulation. In S. Brcznitz (Ed.), The denial of stress (pp. 35-76). New in misfonunc: Making sense of incest. Journal of Social Issues. 39.York: International Universities ~ 81-102.

Kihlstrom, J. F. (1983). Instructed forgetting: Hypnotic and nonhyp- Thompson, W. C., Fong,G. T.,&Rosenhan,D.L(1981).Inad~blenotic. Journal of Experimental Psychology: Genera/. 112. 73-79. evidence and juror verdicts. Journal of Personality and Social Psy.

Kirk, R. E. (1968). Experimenlal design. Belmont, CA: Brooks/Cole. chology. 40. 453-463.Lazarus, R. S. (1983). The costs and benefits of denial. In S. Brcznitz Tvcr5ky, A., & Kahneman, D. (1974). Judgment under uncertainty:

(Ed.), The denial of stress (pp. 1-30). New York: International Uni- Heuristics and biases. Science. 185. 1124-1131.vcrsities Press. Wegner, D. M., Coulton, G., & Wcnzlalf, R. (1985). The transparency

lindemann, E. (1944). Symptomatology and management of acute of denial: Briefing in the debriefing paradigm. Journal of Personalityal .r p h ' 101 141 148 and SOCial PS)f:hoiogy. 49. 338-346.

grief. American Journ oJ s}Y:. lalry: .'. .-.W CI; D. M., WcnzIaff, R., Kerker, R. M., & Beattie, A. E. (1981).Logan, G. D. (19.83). On ~e abilIty to InhIbIt sImple thoughts and ac- ~crimination through innuendo: Can media questions become pub-

tions: I. St~Slgnal studies of deciSIon and memory. Journal of Ex- tic answers? Journal of Personality and Social PS)f:hology. 40. 822-perimenlal Psychology: Leamlng, Memo~ and Cognition. 9, 585- 832.

606. .Wolpe, J., & Lazarus, A. A. (1966). Behallior therapy techniques. NewMarlatt, G. P., & Parks, G. A. (1982). Self-management of addICtIve York: Pergamon Press.

behaviors. In P. ~oly & F. H. Kanfer (Eds), Selfmanagemenl and .b ha h ( 443-488) N-v "'ork. o--.mon Press. ReceIved September 12, 1985e 1IIor c ange pp. .~- l' .""&a ...9 986

McGranahan, D. V. ( 1940). A critical and experimental stUdy of rcprcs- ReVIsIon receIved October, 1sion. Journal of Abnomlal and Social Psychology, 35. 212-225. Accepted November 3, 1986 .

i, 'i"t.