This article was downloaded by: [UQ Library]On: 04 September 2013, At: 09:19Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal of Clinicaland Experimental HypnosisPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/nhyp20

Are High Hypnotizables EspeciallyVulnerable to False MemoryEffects? A SociocognitivePerspectiveGraham F. Wagstaff a , Jacqueline M. Wheatcroft a &Anna Christina Jones aa University of Liverpool, United KingdomPublished online: 01 Jun 2011.

To cite this article: Graham F. Wagstaff , Jacqueline M. Wheatcroft & Anna ChristinaJones (2011) Are High Hypnotizables Especially Vulnerable to False Memory Effects? ASociocognitive Perspective , International Journal of Clinical and Experimental Hypnosis,59:3, 310-326, DOI: 10.1080/00207144.2011.570658

To link to this article: http://dx.doi.org/10.1080/00207144.2011.570658

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information(the “Content”) contained in the publications on our platform. However, Taylor& Francis, our agents, and our licensors make no representations or warrantieswhatsoever as to the accuracy, completeness, or suitability for any purposeof the Content. Any opinions and views expressed in this publication are theopinions and views of the authors, and are not the views of or endorsed byTaylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor andFrancis shall not be liable for any losses, actions, claims, proceedings, demands,costs, expenses, damages, and other liabilities whatsoever or howsoever causedarising directly or indirectly in connection with, in relation to or arising out of theuse of the Content.

This article may be used for research, teaching, and private study purposes.Any substantial or systematic reproduction, redistribution, reselling, loan, sub-

licensing, systematic supply, or distribution in any form to anyone is expresslyforbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

Intl. Journal of Clinical and Experimental Hypnosis, 59(3): 310–326, 2011Copyright © International Journal of Clinical and Experimental HypnosisISSN: 0020-7144 print / 1744-5183 onlineDOI: 10.1080/00207144.2011.570658

ARE HIGH HYPNOTIZABLES ESPECIALLYVULNERABLE TO FALSE MEMORY EFFECTS?

A Sociocognitive Perspective1

Graham F. Wagstaff, Jacqueline M. Wheatcroft, andAnna Christina Jones2

University of Liverpool, United Kingdom

Abstract: This article examines issues raised by a recent UK legalcase in which the defense argued that the accusations made by thehighly hypnotizable plaintiff were likely based on false memories.The authors argue that the evidence related to hypnotizability andfalse memory production is inconsistent but may be illuminated bya sociocognitive perspective. They present 2 preliminary studies thatindicate that when the instructions imply that accurate reporting is afeature of hypnosis, higher hypnotizables may actually be more resis-tant than low or medium hypnotizables to false memories arisingfrom misleading information given during hypnosis. They concludethat, when memory accuracy is emphasized rather than productivity,there is little evidence to link high hypnotizability with a propensityto produce false memories.

The impetus for this article arose from three recent legal cases in theUnited Kingdom in which the first author was involved as an expertwitness (R. v. Dyer, 2008; R. v. Knight, 2008; and R. v. Barker, 2009). Eachinvolved allegations that a hypnotist had sexually assaulted clientswhile they were undergoing hypnotherapy. In each case, the defen-dant was subsequently found guilty of multiple offenses but, notably,in none of these cases did the complainants claim hypnotic automatismhad taken place; all reported that they had been fully aware of whathad happened to them and were subsequently able to give detailedaccounts to the police. However, one case, in particular, stood out byvirtue of the argument put forward by the defense. In the case of R. v.Steven Dyer, the defense argued that the main allegations made by oneof the complainants were based on false memories. Although no formal

Manuscript submitted July 7, 2010; final revision accepted November 1, 2010.1The authors would like to express their thanks to Kelly Campbell for her help in data

collection and collation.2Address correspondence to Graham F. Wagstaff, School of Psychology, University of

Liverpool, Eleanor Rathbone Building, Bedford Street South, Liverpool, L69 7ZA, UnitedKingdom. E-mail: gwf@Liverpool.ac.uk

310

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

HYPNOTIZABILITY AND FALSE MEMORIES 311

attempt was made to assess the complainant’s hypnotizability, therationale for this argument was that, given how she had responded tohypnosis, the complainant could be judged to be “more than averagelysusceptible to hypnosis,” and people who are susceptible to hypno-sis are more likely to produce false memories. Naturally, no attemptwas made to argue that, during hypnosis, the defendant had giventhe complainant suggestions to imagine that he had sexually assaultedher; rather the argument was based on the premise that, because of herhigh hypnotizability, the complainant would have been more willing toaccept false memories or fantasies of sexual assault that came into herimagination, either during hypnosis or after she had been hypnotized.

The arguments made by the defense in this case raise a number ofimportant issues about the relationship between hypnosis, hypnotiz-ability, and false memories, which will be examined in this article.

Hypnosis, Hypnotizability, and False MemoryProduction

Certainly, a variety of evidence appears to support the view thathypnotic procedures may encourage the production of false memo-ries (for reviews, see Erdelyi, 1994; Krauss, Kinoshita, & McConkey,1989; Mazzoni, Heap, & Scoboria, 2010; Wagstaff, 1999a, 1999b; Webert,2003). For example, in a seminal review of the area, Erdelyi con-cludes that memory improvements associated with hypnosis, whenthey occur, tend to be accompanied by an increase in incorrect falsepositive information (false memories); as a result, overall accuracy, asdetermined by the proportion of correct to incorrect responses is notimproved and may even deteriorate (Dywan & Bowers, 1983; Orne,Whitehouse, Dinges, & Orne, 1988). Research also indicates that hyp-nosis may often encourage witnesses to incorporate more misleadinginformation into their reports (Krauss et al., 1989; Scoboria, Mazzoni,Kirsch, & Milling, 2002; Zelig & Beidleman, 1981) and/or be moreconfident in their incorrect reports (Erdelyi, 1994; Krauss et al., 1989;Nogrady, McConkey, & Perry, 1985; Wagstaff et al., 2004).

However, the exact processes responsible for producing these effectshave been the subject of some conjecture. For example, Dywan (1995)has argued that, in hypnosis, false memories are more likely to be expe-rienced with the same intensity and vividness as true memories andhence more source-monitoring errors are found during hypnosis. Onfirst consideration, therefore, from a state perspective on hypnosis, ifhypnosis induces an altered state of consciousness in which partici-pants are more vulnerable to false memories, and highly hypnotizableindividuals are most likely to be susceptible to this state, it should

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

312 GRAHAM F. WAGSTAFF ET AL.

presumably follow that, under hypnosis at least, highly hypnotizableindividuals will be more likely to display false memories. However,the evidence for a relationship between the induction of hypnosis,hypnotizability, and vulnerability to false memory effects is some-what inconsistent. For example, Register and Kihlstrom (1988) foundno relationship between hypnotizability and susceptibility to mislead-ing information administered during hypnosis. Scoboria et al. (2002)examined both the effects of hypnotic induction and hypnotizabil-ity on misleading information errors and found an effect of hypnoticinduction, but, again, hypnotizability was not related to the tendencyto produce false memory reports. In contrast, other studies haveshown that, in response to misleading information, high hypnotiz-ables are more likely to make false memory errors than low hyp-notizables under both hypnotic and nonhypnotic conditions equally(Barnier & McConkey, 1995; Sheehan, Garnett, & Robertson, 1993;Spanos, Gwynn, Comer, Baltruweit, & de Groh, 1989). Yet, further stud-ies have shown that high hypnotizables report more false memoriesthan low hypnotizables under both hypnotic and nonhypnotic con-ditions but more under the former (Whitehouse et al., 2005) or onlywhen misleading information is presented under hypnotic conditions(Krauss et al., 1989).

The finding of a relationship between hypnotizability and suscep-tibility to false memory errors outside of the context of hypnosis isparticularly problematic. Not only does it suggest that other factorsbesides the induction of hypnosis may be operating to explain therelationship but it also runs counter to how highly hypnotizable indi-viduals, and individuals who are generally vulnerable to false memoryeffects, have been characterized in the literature. For example, in theforensic literature, one of the main tools for assessing susceptibilityto suggested false memories is the Gudjonsson Suggestibility Scale(GSS). The GSS consists of a story followed by a memory test thatconsists mainly of questions containing misleading information. If par-ticipants yield to the misleading information and further pressure tochange their responses, they are judged to be high on interrogativesuggestibility (Gudjonsson, 1984, 1987, 1997).

A number of researchers have argued that individuals who scorehigh on the GSS typically do so because they have a negative mindsettoward the memory-testing situation. That is, they find the situationdistressing and have low expectations of success and high uncertaintyas to the correct answer; hence, to alleviate the distress they accept whatis suggested to them (see, for example, Drake, 2010; Gudjonsson &Clarke, 1986). In contrast, most highly hypnotizable individuals arenot typically characterized by a negative mindset toward the hypnotictesting context; indeed, some researchers have argued that positive atti-tudes and expectancies towards the hypnotic situation are among the

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

HYPNOTIZABILITY AND FALSE MEMORIES 313

most important predictors of high hypnotizability and are fundamentalto the clinical efficacy of hypnotherapy (see, for example, Kirsch, 1991;Silva, Bridges, & Metzger, 2005; Spanos, 1986a). Moreover, there is noevidence that, as general characteristics, high hypnotizables have pooror problematic memories; on the contrary, a number of reports indicatethat, in nonhypnotic contexts, they may show better performance ontests of frontal memory functioning than low hypnotizables (Wagstaffet al., 2004; Wagstaff, Cole, & Brunas-Wagstaff, 2007). Similarly, there isno evidence to suggest that high hypnotizables lack confidence in theirmemories; if anything, they are more likely to be overconfident (Krausset al., 1989; Nogrady et al., 1985). Given these considerations, ratherthan attempting to account for the various findings regarding hypno-tizability and susceptibility to false memories by speculating about thecharacteristics of the hypnotic state, and the vulnerability of hypnotiz-able individuals to its influence, it might be more useful to examine theissues from a sociocognitive perspective.

Hypnosis, Hypnotizability, and False MemoryProduction: A Sociocognitive Perspective

From a sociocognitive perspective, false memory reports that occuras a consequence of hypnotic or related procedures, and any propen-sity high hypnotizables may have to be vulnerable to such reports,arise not because of some intrinsic feature of the hypnotic state, orsusceptibility to this state, but because of the contextual demand char-acteristics of the particular situation and participants’ related expectan-cies and beliefs (Wagstaff, 1999a, 1999b; Wagstaff, Cole, Wheatcroft,Anderton, & Madden, 2008). For example, in line with this view, avariety of researchers of different theoretical persuasions have arguedfor a “criterion shift” interpretation of false memory reports with hyp-nosis. That is, because the expectancies associated with the hypnosiscontext, and instructions that reinforce those expectancies (i.e., such asthe expectation that hypnosis has special properties that enable indi-viduals to generate additional detailed and vivid material), hypnotizedwitnesses may sometimes adopt a more lax criterion for report whenreporting memories, providing details that they might otherwise haverejected on grounds of uncertainty. In addition, they may report vaguedetails or imaginings, including those suggested to them, as confidentmemories on the assumption that, because they are created or pro-vided in the hypnotic context, they will be accurate (see, e.g., Lynn &Nash, 1994; Orne et al., 1988; Wagstaff, 1999a, 1999b; Webert, 2003;Whitehouse, Dinges, Orne, & Orne, 1988; Whitehouse et al., 2005). Theymay also incorporate more misleading information into their testimony

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

314 GRAHAM F. WAGSTAFF ET AL.

simply because the demand characteristics of the situation indicate thatthis is the appropriate response (Wagstaff, Cole, Wheatcroft, et al., 2008;Wagstaff & Frost, 1996).

In support of these kinds of explanations, a variety of evidencesuggests that hypnotically created false memories can be reduced sig-nificantly under conditions that encourage more cautious and truthfulreporting; such as when told that a “hidden-part” of them can describetheir “real” memories, or when cross-examined under oath (Spanoset al., 1989), or when given a financial incentive for accurate report-ing (Murray, Cross, & Whipple, 1992), or when given an opportunity todeny being in a “trance” (Wagstaff & Frost, 1996). False memory reportswith hypnosis are also reduced when it is implied that the experimentis terminated (Barnier & McConkey, 1995), when subjects are contactedby telephone at their home after the experiment (McConkey, Labelle,Bibb, & Bryant, 1990), and when rapport with the hypnotist is down-graded (Sheehan, Green, & Truesdale, 1992). It should be emphasizedthat false memory effects are rarely eliminated entirely using such pro-cedures; however, the issue here is not whether all false memory reportsresult from response bias, but why they tend to be exaggerated whenhypnosis is involved.

These findings suggest that whether or not hypnotic induction andhypnotizability are found to relate to false memory production may behighly dependent on the contextual demands of the particular situationand how both low and hypnotizable individuals interpret them. Forexample, a variety of evidence suggests that, in general, high hypnoti-zables seem especially attuned to respond to the contextual demandsof hypnosis situations in ways that low hypnotizables are not; thatis, high hypnotizables appear to be highly committed to playing therole of the good hypnotic subject as defined by previous expectationsand cues conveyed in the situational context (see, e.g., Spanos, 1992,1986a, 1986b; Perlini, Spanos, & Jones, 1996; Wagstaff, 2004; Wagstaffet al., 2004). High hypnotizables may, therefore, respond to the contex-tual demands of ostensibly “nonhypnotic” situations so long as theyperceive that an association with the behavior of a good hypnotic sub-ject is expected or desired (see, e.g., Kirsch, 1991; Milling, Kirsch, &Burgess, 2000). In support of this interpretation with regard to falsememory production, Barnier and McConkey (1995) found that highhypnotizables reported more false memories than lows in both hyp-notic and nonhypnotic conditions, but the number of false memoryreports dropped significantly when participants were tested after theexperimenter appeared to end the session such that differences betweenhighs and lows were no longer apparent. Similarly, Spanos et al. (1989)found that false memory errors were more often reported by highhypnotizables in both hypnotic and nonhypnotic conditions, but theyfell significantly after they were cross-examined and instructed to be

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

HYPNOTIZABILITY AND FALSE MEMORIES 315

accurate and truthful, such that, again, differences between high andlow hypnotizables were no longer apparent.

Another implication of this line of reasoning is that correlationsbetween hypnotizability and memory performance in nonhypnoticconditions are likely to be influenced by the resemblance of the latterto hypnotic procedures, and the similarity between the kinds of sug-gestions given in hypnotic and nonhypnotic situations. It is notable,for example, that studies that have found that hypnotizability to beassociated with susceptibility to misleading information and falsememory production in both hypnotic and nonhypnotic conditions havetended to use instructions or suggestions designed to produce more, orchanges in, memory details in both conditions (see, e.g., Spanos et al.,1989; Whitehouse et al., 2005).

Also, relationships between hypnotizability and measures of falsememory production may vary depending on whether instructionsor suggestions delineating the role of a “good” subject are presentor absent. For example, in a large study, involving 227 participants,Malinowski and Lynn (1999) looked at factors influencing potentiallyspurious reports of early memories from childhood. Although hyp-notizability and memory reports were measured in different contextsand on different occasions, Malinowski and Lynn still found that twomeasures of hypnotizability showed low but significant correlations(.22 and .27) with reports of early memories in childhood, followinginstructions designed to evoke such memories. However, although notspecifically labeled as “hypnosis,” these procedures were similar tothose found in some hypnotic memory enhancement procedures; thatis, they included eye-closure with fairly emphatic, repetitive sugges-tions to visualize more memories, together with a clear indication toparticipants that this would be successful. In contrast, Malinowski andLynn found no significant relationships between hypnotizability andany of the GSS measures of susceptibility to misleading information,which, again, were administered without hypnosis, but also withoutequivalent repetitive suggestions or instructions for memory enhance-ment (the correlations ranged from .05 to .08). Interestingly, theyalso reported no relationship between fantasy/imaginative involve-ment and either early childhood memories or the GSS, indicating thatinvolvement with fantasy and imagination was not a moderating factor.

However, if it is the case that associations between high hypno-tizability and false memory production, when they occur, occur notbecause of some intrinsic characteristic of high hypnotizables that ren-ders them generally vulnerable to false memory effects, but ratherbecause they have a proclivity to conform with the contextual demandsof situations perceived as hypnosis, or resembling hypnosis, then bychanging the contextual demands specifically towards accurate report-ing, it might actually be possible to reverse the usual direction of

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

316 GRAHAM F. WAGSTAFF ET AL.

the relationship between hypnotizability and susceptibility to falsememory errors, such that high hypnotizables show more resistanceto them.

For example, in a related study, Wagstaff, Cole, Wheatcroft, et al.(2008) reasoned that if expectancy and demand characteristics play arole in increasing false memory responses with hypnosis, it might bepossible to use the same factors to reduce them. So, if subjects aregiven misleading information followed by an instruction suggestingthat hypnosis will reduce the influence of misinformation, hypno-sis may act to reduce rather than increase false memory responses,including spurious confidence in errors. They report two studies insupport of this hypothesis, showing that effects of misleading infor-mation can be eliminated or significantly reduced to below nonhyp-notic levels when participants are given hypnotic induction togetherwith a suggestion that it will help them discriminate between cor-rect and incorrect information (this finding also accords with theview that nonhypnotic baseline levels of susceptibility to misleadinginformation may often be inflated by demand characteristics effects;see Wagstaff, Cole, Wheatcroft, et al., 2008). There was no evidenceof inflated confidence with hypnosis in either study, but, also, inneither study was any relationship found between hypnotizability(measured by hypnotic depth reports) and misleading informationerrors. However, Wagstaff, Cole, Wheatcroft, et al. argued that thismight have been a consequence of the limited range of hypnotiz-ability scores in the samples. Moreover, in these studies the mislead-ing information was presented before the introduction of hypnosis,rather than during hypnosis; and arguably the latter would providea more sensitive indicator of the effects of hypnosis and instruc-tions on vulnerability to misleading information. Indeed, Krauss et al.(1989) have argued that studies in which the misleading informa-tion is presented before hypnotic induction typically show no effectsof hypnotizability. To investigate the possibility, therefore, that, byintroducing the misleading information during, rather than before,hypnosis, and emphasizing accuracy, it may be possible to reverse thetypical relationship between hypnotizability and susceptibility to mis-leading information errors, the following two preliminary studies wereconducted.

Experiment I: Method

ParticipantsThe participants were 30 students (6 males and 24 females) from the

University of Liverpool (age: M = 18.90, range = 18–24, SD = 1.48).None had previously had any experience of hypnosis.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

HYPNOTIZABILITY AND FALSE MEMORIES 317

Materials and ProcedureAll participants received a 2-minute audio recording of a conversa-

tion between two men planning a robbery. They were then given thefollowing instruction stressing the importance of accurate reporting:

People have been shown to be capable of filling in gaps in memory,of distorting information, and being influenced in what is rememberedby leading questions or suggestions. However, hypnosis techniques canimprove memory by focusing your concentration and enabling youto distinguish better between correct information and any misleadinginformation you may have heard.

All participants were then given an audiotape of a standard hyp-notic relaxation induction procedure modified from Barber (1969);hypnotic depth was then measured using the Long Stanford Scale ofHypnotic Depth (LSS; Tart, 1970), which requires subjects to rate theirdegree of experienced depth on a scale from 0 through 10 (with 0 =awake and alert, as you normally are, 1 = borderline state, between sleepingand waking, 2 = lightly hypnotized, 5 = quite strongly and deeply hypno-tized, 8–9 = very hypnotized, and 10 = very deeply hypnotized). It can benoted here that the LSS has been shown to be as reliable and valid ameasure of hypnotizability as standard suggestion-based measures (seeWagstaff, Cole, & Brunas-Wagstaff, 2008).

All participants then received a procedure designed to elicit thestandard misinformation effect; this consisted of a further 2-minuteaudio recording of a woman recounting the first conversation to herfemale friend; however, a number of critical details were changed.Following this, they received a repeat warning: “I am now goingto test your memory for the information contained in the first tape.Remember, that while hypnotized you may find that you may beable to distinguish better between correct information and any mis-leading information you may have heard.” They were then askedto open their eyes, while remaining hypnotized, and to complete aquestionnaire that contained 17 questions concerning details of theplanned robbery; 11 of these asked for details that were discrepantbetween the two conversations, such as “What is the name of theshop?” Participants were specifically instructed to base their answerson the first conversation, to be as accurate as possible, and to pro-vide an answer to every question. After each question, they wererequired to rate their confidence in their answer on a scale of 1 to 5,where 1 indicates not at all confident, and 5 indicates extremely confi-dent. Upon completion of the questionnaire, participants were told toclose their eyes again and the hypnosis termination instructions weregiven.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

318 GRAHAM F. WAGSTAFF ET AL.

Experiment I: Results and Discussion

Preliminary analyses showed no significant effects for age or sex.LSS depth scores ranged from 1 to 7 (M = 3.67, SD = 1.47). Only 2participants scored 1, and none scored 0, so participants were retrospec-tively categorized as either low/medium hypnotizability (3 or below,n = 16) or high hypnotizability (4 and above, n = 14). The 17 items ofthe questionnaire were scored as misled errors (errors containing mis-leading information from the second conversation), nonmisled errors(errors not containing misleading information), and correct responses.The means and standard deviations are shown in Table 1.

A series of F tests on the three dependent variables showed only onesignificant result; as predicted, high hypnotizables showed a small butconsistent trend to produce fewer errors involving misleading informa-tion, F(1, 28) = 3.67, p < .035 (one tailed), d = 0.70. Neither of the othercomparisons was significant (F ≤ 1).

The data for mean confidence in correct scores and mean confi-dence in errors (misled and not misled combined) were then ana-lyzed using a 2 × 2 mixed analysis of variance (ANOVA; High/LowHypnotizability × Mean Confidence in Correct/Errors) with repeatedmeasures on the second factor. Together, both groups showed greaterconfidence in correct responses than errors, F(1, 28) = 8.57, p < .008; themeans were 3.90 (SD = 0.47) and 3.41 (SD = 0.62) for correct responsesand errors, respectively. However, the interaction was not significant(F < 1).

The finding that high hypnotizables integrated less misleading infor-mation into their answers than low/medium hypnotizables produceda medium-to-large effect size (d = 0.70), but the difference was onlysignificant on a one-tailed test and could be considered, therefore,to provide only very tentative support for the hypothesis that if themisleading material is introduced during hypnosis, and accuracy isemphasized, high hypnotizables may show less vulnerability to mis-leading information effects than low hypnotizables. Of course, a moresignificant effect might have been achieved with a larger sample size,

Table 1Means and SDs (in Parentheses) for Responses to Memory Items According toHypnotizability

Hypnotizability Misled Errors Nonmisled Errors Correct

High 2.38 (1.31) 2.19 (1.60) 12.44 (2.19)Low/Medium 3.36 (1.50) 1.86 (1.70) 11.79 (2.46)

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

HYPNOTIZABILITY AND FALSE MEMORIES 319

but if the effect is only evident when large samples are employed itis unlikely to have much forensic relevance. There was no evidence ofhypnotizables showing more errors or greater confidence in errors.

Experiment 2

As the first study looked at the relationship between hypnotizabil-ity and resistance to misleading information only within the context ofhypnosis, a second study was conducted that compared performancebefore and during hypnosis procedures. The materials for this studywere taken from the parallel form of the Gudjonsson SuggestibilityScale (GSS2; Gudjonsson, 1987, 1997), which contains a larger propor-tion of misleading information. It also includes an instruction that givesnegative feedback and encourages accurate recall. It was hypothesized,therefore, that if this instruction is given in a context defined as hyp-nosis, high hypnotizables would be more likely to show resistance tomisleading information.

Experiment 2: Method

ParticipantsThe participants were 21 members of the general public (11 males

and 10 females) from the Liverpool area (age: M = 26.76, range = 16–45,SD = 8.57). None had any previous experience of hypnosis.

Materials and ProcedureAll participants were played an audio recording of a short descrip-

tive story about a bicycle accident from the GSS2. All participants werethen asked the standard 20 questions from the GSS2. Fifteen of these areleading questions that include reference to material not included in thestory, such as “Did the boy commonly ride the bicycle to school?” when“school” was not mentioned in the story. The remaining five questionswere nonleading, requiring a simple “yes” or “no” response, such as“Were the couple called Anna and John?” (which they were). All par-ticipants were then told that they were to receive a hypnotic inductionprocedure to allow them to clear their thoughts and to enhance con-centration. An audiotape of a standard hypnotic relaxation inductionprocedure modified from Barber (1969) was then administered, follow-ing which hypnotic depth was measured using the LSS. All participantswere given the standard GSS instruction that emphasizes the impor-tance of accurate recall: “You have made a number of errors. It is there-fore necessary to go through the questions once more, and this time try

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

320 GRAHAM F. WAGSTAFF ET AL.

to be more accurate.” The 20 questions were then asked a second time(Questionnaire 2). Following this, hypnosis was terminated.

Experiment 2: Results and Discussion

Preliminary analyses showed no significant effects for age or sex.LSS depth scores ranged from 0 to 8 (M = 2.67; SD = 2.05), and par-ticipants were divided into three groups on the basis of these scores,low hypnotizability (0–1, n = 7), medium hypnotizability (2–3, n = 8),and high hypnotizability (4–8, n = 6). For the purposes of the presentexperiment, the most critical variables are the number of times the par-ticipants responded affirmatively to the leading information conveyedin Questionnaires 1 and 2. In terms of the GSS, these measures aretermed Yield 1 and Yield 2 (Gudjonsson, 1984, 1987, 1997). The resultsare summarized in Table 2.

A 3 × 2 mixed ANOVA (Low/Medium/High Hypnotizability ×Questionnaire 1/2) on the yield scores showed a significant main effectfor yield; Yield 1 scores (M = 7.19, SD = 2.73) were significantly higherthan Yield 2 scores (M = 6.38, SD = 2.26), F(1, 18) = 5.97, p < .026.More important, however, the Hypnotizability × Questionnaire inter-action was also significant, F(2, 18) = 5.31, p < .016. (post hoc F testsshowed that only the high hypnotizables produced significantly lower

Table 2Means and SDs (in Parentheses) for Yields 1 and 2 and Correct Responses on GSSaccording to Hypnotizability

Hypnotizability Yield Correct Responses(Misled Errors) (Noncritical Questions)

Questionnaire 1(Yield 1 and Correct Responses before Hypnosis)

Low 7.86 (2.67) 3.71 (1.11)Medium 7.25 (2.31) 4.00 (1.07)High 6.33 (3.50) 4.50 (0.54)

Questionnaire 2(Yield 2 and Correct Responses during Hypnosis)

Low 6.86 (1.57) 4.14 (1.21)Medium 7.87 (1.23) 4.25 (1.04)High 3.83 (2.04) 4.17 (1.16)

Questionnaires 1 and 2Mean Total

Low 7.36 (1.97) 7.86 (1.86)Medium 7.56 (1.56) 8.25 (1.98)High 5.08 (2.76) 8.24 (1.67)

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

HYPNOTIZABILITY AND FALSE MEMORIES 321

Yield 2 scores, p < .011). As a consequence, the Yield 2 scores for highhypnotizables were lower than those for either of the other groups(p < .031, the effect sizes d were 0.99 and 2.40, for the lows and medi-ums, respectively). There were no differences between groups in theYield 1 scores. There was also no significant overall main effect forlevels of hypnotizability on the combined yield, F(2, 18) = 2.69, p < .09.

In addition to the yield scores, the GSS also provides two other mea-sures of the extent to which participants yield to suggestive pressure;these are Shift, which is the number of times the participant changes aresponse to an item having received the negative feedback (regardlessof the direction of the change), and total suggestibility, which is calcu-lated by summing the Yield 1 and shift scores. Scores on two measureswere also calculated and analyzed using one-way ANOVAs. Neither ofthe effects approached significance (F ≤ 1); the mean shift scores were5.86 (2.61), 4.00 (2.13), and 5.33 (3.20), and the mean total suggestibil-ity scores were 13.71 (4.68), 11.25 (2.31), and 11.67 (6.21) for the low,medium, and high hypnotizability groups, respectively.

It can be noted that, unlike in the first study, because in the GSS themisleading information is conveyed in the way the question is asked,participants who correctly fail to accept the premise in the leadingquestions do not tend to offer alternative answers that are “in error.”Nonmisled errors were, therefore, negligible and no significant differ-ences between conditions were found (F < 1). Moreover, in accordancewith standard GSS scoring, a wide variety of answers is accept-able as indicating nonyielding, including “no” and “can’t remember”(Gudjonsson, 1997). Although not in error, technically such answers donecessarily indicate a “correct” answer, in the sense of an accurate rec-ollection of events. It was, however, possible to perform a further 3 × 2ANOVA on the data for correct responses to the five noncritical ques-tions for Questionnaires 1 and 2. Again, the data are summarized inTable 2. None of the effects approached significance (F < 0.1).

In summary, these results indicate that during hypnosis, the highsshowed significantly less susceptibility to influence from misleadinginformation than the other groups; the differences showing large effectsizes. Moreover, the highs were the only group to show a significantdecline in susceptibility to misleading information following hypnoticinduction.

General Conclusion

Notwithstanding the reasonably large effect sizes, given the smallsample sizes involved these two studies, further replication is obvi-ously needed to establish their reliability. Nevertheless, taken together,the results of these studies and the literature reviewed in this article

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

322 GRAHAM F. WAGSTAFF ET AL.

could be considered to lend some support to the view that highlyhypnotizable individuals are not inherently more vulnerable to falsememory errors either within or outside the context of hypnosis. Rather,in accordance with previous sociocognitive conceptions, high hypnoti-zables may be especially sensitive to the contextual demands of how a“good” subject should behave in situations defined as, or perceived bythem to be similar to, “hypnosis” in terms of their procedures (Perliniet al., 1996; Spanos, 1986a, 1992; Wagstaff et al., 2004). Consequently,when the instructions convey the message that the role of the hypno-tized individual is to report accurately, rather than increase productiv-ity, there are indications that both hypnotic induction (Wagstaff, Cole,Wheatcroft, et al. 2008) and high hypnotizability may be associatedwith greater resistance to false memory errors.

The arguments presented in the present article are not intended, inany way, as an endorsement of the use of hypnosis as a memory facil-itation device by either police or clinicians, as there remain substantialproblems with its use in this role (Wagstaff, 1999a, 1999b; Webert, 2003).Nevertheless, if valid, they may be relevant to court cases such as theDyer case, where hypnosis has already been used, and the vulnera-bility of the previously hypnotized witness to false memory effects isan issue. Given the available evidence, it seems reasonable to concludethat if, no suggestions for false memory are given at any stage, and theemphasis is placed on accuracy rather than productivity, there is littlereason to assume that high hypnotizability will be associated with anyspecial propensity to produce false memories.

References

Barber, T. X. (1969). Hypnosis: A scientific approach. New York, NY: Van Nostrand.Barnier, A. J., & McConkey, K. M. (1995). Reports of real and false memories: The rele-

vance of hypnosis, hypnotizability, and the context of memory test. Journal of AbnormalPsychology, 101, 521–527.

Drake, K. (2010). Interrogative suggestibility: Life adversity, neuroticism and compliance.Personality and Individual Differences, 48, 493–498.

Dywan, J. (1995). The illusion of familiarity: An alternative to the report-criterion accountof hypnotic recall. International Journal of Clinical and Experimental Hypnosis, 43, 194–211.

Dywan, J., & Bowers, K. (1983). The use of hypnosis to enhance recall. Science, 22, 184–185.Erdelyi, M. W. (1994). The empty set of hypermnesia. International Journal of Clinical and

Experimental Hypnosis, 42, 379–390.Gudjonsson, G. H. (1984). A new scale of interrogative suggestibility. Personality and

Individual Differences, 5, 303–314.Gudjonsson, G. H. (1987). A parallel form of the Gudjonsson Suggestibility Scale. British

Journal of Clinical Psychology, 26, 215–221.Gudjonsson, G. H. (1997). The Gudjonsson Suggestibility Scales Manual. Hove, United

Kingdom: Psychology Press.Gudjonsson, G. H., & Clarke, N. K. (1986). Suggestibility in police interrogation. Social

Behavior, 1, 83–104.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

HYPNOTIZABILITY AND FALSE MEMORIES 323

Kirsch, I. (1991). The social learning theory of hypnosis. In S. J. Lynn and J. W. Rhue(Eds.), Theories of hypnosis: Current models and perspectives (pp. 439–466). New York,NY: Guilford.

Krauss, J., Kinoshita, S., & McConkey, K. M. (1989). Hypnotic memory and confidentreporting. Applied Cognitive Psychology, 3, 35–51.

Lynn, S. J., & Nash, M. R. (1994). Truth in memory: Ramifications for psychotherapy andhypnotherapy. American Journal of Clinical Hypnosis, 36, 194–208.

Mazzoni, G., Heap, M., & Scoboria, A. (2010). Hypnosis and memory: Theory, labora-tory research, and applications. In S. J. Lynn, J. W. Rhue, & I. Kirsch (Eds.), Handbookof clinical hypnosis (2nd ed., pp. 709–742). Washington, DC: American PsychologicalAssociation.

Malinowski, P. T., & Lynn, S. J. (1999). The plasticity of early memory reports: Social pres-sure, hypnotizability, compliance and interrogative suggestibility. International Journalof Clinical and Experimental Hypnosis, 47, 320–345.

McConkey, K. M., Labelle, L., Bibb, B. C., & Bryant, R. A. (1990). Hypnosis andpseudomemory: The relevance of text context. Australian Journal of Psychology, 42,197–206.

Milling, L. S., Kirsch, I., & Burgess, C. (2000). Hypnotic suggestibility and absorption:Revisiting the context effect. Contemporary Hypnosis, 17, 32–41.

Murray, G. J., Cross, H. J., & Whipple, J. (1992). Hypnotically created pseudomemories:Further investigation into the “memory distortion or response bias” question. Journalof Abnormal Psychology, 101, 75–77.

Nogrady, H., McConkey, K. M., & Perry, C. (1985). Enhancing visual memory: Tryinghypnosis, trying imagination, and trying again. Journal of Abnormal Psychology, 2, 194–204.

Orne, M. T., Whitehouse, W. G., Dinges, D. F., Orne, E. C. (1988). Reconstructing memorythrough hypnosis: Forensic and clinical implications. In H. M. Pettinati (Ed.), Hypnosisand memory (pp. 21–63). New York, NY: Guilford.

Perlini, A. H., Spanos, N. P., & Jones, W. (1996). Hypnotic negative hallucinations: Areview of subjective, behavioral and physiological methods. In R. G. Kunzendorf, N.P. Spanos, & B. J. Wallace (Eds.), Hypnosis and imagination (pp. 199–222). New York,NY: Baywood.

R. v. Barker (Cambridge Crown Court 2009).R. v. Dyer (Canterbury Crown Court 2008).R. v. Knight (Birmingham Crown Court 2008).Register, P. A., & Kihlstrom, J. F. (1988). Hypnosis and interrogative suggestibility.

Personality and Individual Differences, 9, 549–558.Scoboria, A., Mazzoni, G., Kirsch, I., & Milling, L. S. (2002). Immediate and persist-

ing effects of misleading questions and hypnosis on memory reports. Journal ofExperimental Psychology-Applied, 8, 26–32.

Sheehan, P. W., Garnett, M. S., & Robertson, R. (1993). The effects of cue level, hypnotiz-ability and state instruction on responses to leading questions. International Journal ofClinical and Experimental Hypnosis, 41, 287–304.

Sheehan, P. W., Green, V., & Truesdale, P. (1992). Influence of rapport on hypnoticallyinduced testimony. Journal of Abnormal Psychology, 101, 690–700.

Silva, C., Bridges, K. R., & Metzger, M. (2005). Personality, expectancy & hypnotizability.Personality and Individual Differences, 39, 131–142.

Spanos, N. P. (1986a). Hypnosis and the modification of hypnotic susceptibility: A socialpsychological perspective. In P. L. N. Naish (Ed.), What is hypnosis? (pp. 85–120).Milton Keynes, United Kingdom: Open University Press.

Spanos, N. P. (1986b). Hypnotic behavior: A social psychological interpretation ofamnesia, analgesia, and “trance logic.” The Behavioral and Brain Sciences, 9, 449–502.

Spanos, N. P. (1992). Compliance and reinterpretation in hypnotic responding.Contemporary Hypnosis, 9, 7–14.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

324 GRAHAM F. WAGSTAFF ET AL.

Spanos, N. P., Gwynn, M. I., Comer, S. L., Baltruweit, W. J., & de Groh, M. (1989). Are hyp-notically induced pseudomemories resistant to cross-examination? Law and HumanBehavior, 13, 271–289.

Tart, C. T. (1970). Self-report scales of hypnotic depth. International Journal of Clinical andExperimental Hypnosis, 18, 105–125.

Wagstaff, G. F. (1999a). Forensic hypnosis. In I. Kirsch, A. Capafons, E. Cardena-Buela,& S. Amigo (Eds.), Clinical hypnosis and self-regulation therapy: A cognitive-behavioralperspective (pp. 277–310). Washington, DC: American Psychological Association.

Wagstaff, G. F. (1999b). Hypnotically elicited testimony. In A. Heaton Armstrong, E.Shepherd, & D. Wolchover (Eds.), Analysing witness testimony (pp. 277–310). London,United Kingdom: Blackstone.

Wagstaff, G. F. (2004). High hypnotizability in a sociocognitive framework. In M. Heap,D. Oakley, & R. Brown (Eds.), The highly hypnotizable person (pp. 85–114). London,United Kingdom: Brunner-Routledge.

Wagstaff, G. F., Brunas-Wagstaff, J., Knapton, L., Winterbottom, J., Crean, V., Cole, J., &Wheatcroft, J. (2004). Facilitating memory with hypnosis, focused meditation and eyeclosure. International Journal of Clinical and Experimental Hypnosis, 52, 434–455.

Wagstaff, G. F., Cole, J. C., & Brunas-Wagstaff, J. (2007). Effects of hypnotic inductionand hypnotic depth on phonemic fluency: A test of the frontal inhibition account ofhypnosis. International Journal of Psychology and Psychological Therapy, 7, 27–40.

Wagstaff, G. F., Cole, J., & Brunas-Wagstaff, J. (2008). Measuring hypnotizability: The casefor self-report scales and normative data for the Long Stanford Scale. InternationalJournal of Clinical and Experimental Hypnosis, 56, 119–142.

Wagstaff, G. F., Cole, J., Wheatcroft, J., Anderton, A., & Madden, H. (2008). Reducing andreversing pseudomemories with hypnosis. Contemporary Hypnosis, 25, 178–191.

Wagstaff, G. F., & Frost, R. (1996). Reversing and breaching posthypnotic amnesia andhypnotically created pseudomemories, Contemporary Hypnosis, 13, 191–197.

Webert, D. R. (2003). Are the courts in a trance? Approaches to the admissibility of hyp-notically enhanced witness testimony in the light of empirical evidence. AmericanCriminal Law Review, 40, 1301–1327.

Whitehouse, W. G., Dinges, D. F., Orne, E. C., & Orne, M. T. (1988). Hypnotic hypermne-sia: Enhanced memory accessibility or report bias? Journal of Abnormal Psychology, 97,289–295.

Whitehouse, W. G., Orne, E. C., Dinges, D. F., Bates, B. L., Nadon, R., & Orne, M. T. (2005).The cognitive interview: Does it successfully avoid the dangers of forensic hypnosis?American Journal of Psychology, 118, 213–234.

Zelig, M., & Beidleman, W. B. (1981). The investigative use of hypnosis: A word ofcaution. International Journal of Clinical and Experimental Hypnosis, 29, 401–412.

Sind hoch-hypnotisierbare Personen besonders anfällig für FalseMemory-Effekte? Eine soziokognitive Perspektive

Graham F. Wagstaff, Jacqueline M. Wheatcroft und Anna Christina JonesAbstract: Der vorliegende Artikel befasst sich mit Fragestellungen, welchedurch einen aktuellen britischen Rechtsfall ausgelöst wurden, in dem dieVerteidigung argumentierte, dass die Anschuldigungen eines hoch hypno-tisierbaren Klägers wahrscheinlich auf der Grundlage falscher Erinnerungenbasieren würden. Die Autoren argumentieren, dass die Evidenz bezüglichder Hypnotisierbarkeit und der Produktion falscher Erinnerungen inkonsis-tent ist, diese jedoch durch eine soziokognitive Perspektive näher beleuchtet

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

HYPNOTIZABILITY AND FALSE MEMORIES 325

werden könnte. Sie präsentieren 2 Vorstudien in denen gezeigt wird, dasswenn gegebene Instruktionen eine genaue Berichterstattung als Merkmalvon Hypnose implizieren, hoch hypnotisierbare Probanden sogar wider-standsfähiger gegenüber falschen Erinnerungen sein können, welche durchirreführende Informationen während der Hypnose entstehen können, alsmoderat hypnotisierbare Probanden. Die Autoren schließen daraus, dass eskaum Hinweise darauf gibt, hohe Hypnotisierbarkeit und die Neigung zurProduktion falscher Erinnerungen miteinander zu verknüpfen, wenn dieGedächtnisgenauigkeit stärker betont wird als die Produktivität.

Jan MikulicaUniversity of Konstanz, Germany

Les sujets hautement hypnotisables sont-ils particulièrement vulnérables àl’effet des faux souvenirs? Une perspective sociocognitive

Graham F. Wagstaff, Jacqueline M. Wheatcroft et Anna Christina JonesRésumé: Cet article porte sur des questions soulevées dans le cadre d’unepoursuite en justice menée au Royaume-Uni, au cours de laquelle la défensesoutenait que les accusations faites par un plaignant hautement hypnotisableétaient probablement fondées sur de faux souvenirs. Les auteurs affirmentque la preuve liée à la susceptibilité hypnotique et à la production de fauxsouvenirs est contradictoire, mais qu’elle peut être expliquée par la perspec-tive sociocognitive. Ils présentent deux études préliminaires indiquant quelorsque les instructions hypnotiques demandent un témoignage exact, lespersonnes hautement hypnotisables peuvent en fait mieux résister aux fauxsouvenirs découlant d’une information erronée fournie durant l’hypnose,que les personnes faiblement ou moyennement hypnotisables. Les auteursen concluent que lorsque l’accent est mis sur l’exactitude des souvenirs etnon sur la capacité de production de ceux-ci, il existe peu de preuves perme-ttant de lier une haute susceptibilité hypnotique à la propension à produirede faux souvenirs.

Johanne ReynaultC. Tr. (STIBC)

Serán las personas altamente hipnotizables vulnerables a memorias falsas?Una perspectiva sociocognitiva

Graham F. Wagstaff, Jacqueline M. Wheatcroft, y Anna Christina JonesResumen: Este artículo examina cuestiones que surgieron de un caso legalen el Reino Unido en donde la defensa argumentaba que las acusacionesdel demandante altamente hipnotizable probablemente estaban basadas enmemorias falsas. Los autores argumentan que la evidencia relacionada conhabilidad hipnótica y la producción de memorias falsas es inconsistente,pero que puede evaluarse bajo la perspectiva sociocognitiva. Presentan 2estudios preliminares que indican que cuando las instrucciones implicanque el reportaje certero es una de las características de la hipnosis, los

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013

326 GRAHAM F. WAGSTAFF ET AL.

participantes altamente hipnotizables pueden ser más resistentes, que aque-llos poco o medianamente hipnotizables, a que surjan memorias falsas deinformación desorientadora durante hipnosis. Concluyen que, cuando seenfatiza la certeza de la memoria en vez de productividad, hay poca evidenciaque asocie la alta habilidad hipnótica con la propensión a producir memoriasfalsas.

Omar Sánchez-Armáss CappelloAutonomous University of San Luis Potosi,Mexico

Dow

nloa

ded

by [

UQ

Lib

rary

] at

09:

19 0

4 Se

ptem

ber

2013