UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of hypnosis as an interview technique on eyewitness memory McEwan, N. Hope 1984

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1984_A8 M39.pdf [ 5.8MB ]
Metadata
JSON: 831-1.0096304.json
JSON-LD: 831-1.0096304-ld.json
RDF/XML (Pretty): 831-1.0096304-rdf.xml
RDF/JSON: 831-1.0096304-rdf.json
Turtle: 831-1.0096304-turtle.txt
N-Triples: 831-1.0096304-rdf-ntriples.txt
Original Record: 831-1.0096304-source.json
Full Text
831-1.0096304-fulltext.txt
Citation
831-1.0096304.ris

Full Text

THE EFFECT OF HYPNOSIS AS AN INTERVIEW TECHNIQUE ON EYEWITNESS MEMORY by N. HOPE MCEWAN B.A., University of British Columbia, 1980  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS  in  THE FACULTY OF GRADUATE STUDIES (Department of Psychology)  We accept this thesis as conforming to the required standard  THE UNIVERS ilY OF.fe«lflSH COLUMBIA v  MARCH 1984 ® N. Hope McEwan, 1984  In p r e s e n t i n g  this thesis  requirements  f o r an  of  British  it  freely available  agree t h a t for  Library  shall  for reference  and  study.  I  for extensive  h i s or  her  copying or  f i n a n c i a l gain  be  shall  publication  not  be  Date  DE-6  (3/81)  M a  rch  23,  198^  of  Columbia  make  further this  thesis  head o f  this  my  It is thesis  a l l o w e d w i t h o u t my  Psychology  The U n i v e r s i t y o f B r i t i s h 1956 Main Mall Vancouver, Canada V6T 1Y3  the  representatives.  permission.  Department o f  copying of  g r a n t e d by  the  University  the  s c h o l a r l y p u r p o s e s may  understood that  the  I agree that  permission by  f u l f i l m e n t of  advanced degree a t  Columbia,  department or for  in partial  written  ii  ABSTRACT  Seventy-two hypnotically susceptible students witnessed a videotape of a simulated bank robbery. in  one of three  "states":  One week l a t e r observers were interviewed hypnotized,  relaxed, or waking.  Half the  students i n each of these groups received imagery instructions to  those  employed  instructed  to use  by a  police  similar  hypno-investigators; the remainder  reconstructive memory  technique.  Each  were  witness  answered 47 questions pertaining to the videotaped bank robbery and the "live"  context of the viewing room.  misinformation answer. task. the  intended  to lead  Five of these questions contained  the respondent  After the interview, students attempted  giving  the wrong  a photo i d e n t i f i c a t i o n  Overall r e c a l l accuracy f o r both the video and " l i v e " aspects of  critical  event  was  high.  Hypnosis  had no e f f e c t  d e t a i l s of the event, nor were hypnotized the  into  misleading  information.  The type  on memory f o r  students more susceptible to  of memory  technique  interacted  with r e c a l l state to a f f e c t both the number and the proportion of errors contained  i n responses.  This i n t e r a c t i o n  differed  f o r the video and  " l i v e " aspects of the event, suggesting limited generalization of memory for  recorded  significantly  events  to  related  to  real both  life accuracy  situations. of  choice  Confidence and  amount  was of  information r e c a l l e d during the interview, but no such relationship was found f o r number or proportion of errors.  iii  TABLE OF CONTENTS  Abstract  i i  Table of Contents  i i i  List of Tables  iv  List of Appendix J Tables  v  Acknowledgements  vi  Introduction  1  Method  44  Results  51  Discussion  82  References  89  Appendix A  104  Appendix B  108  Appendix C  111  Appendix D  114  Appendix E  116  Appendix F  119  Appendix G  121  Appendix H  128  Appendix I  130  Appendix J  <  132  iv LIST OF TABLES  Table 1  Means and Standard Deviations for Recall Scores for the Critical Event  Table 2  Means and Standard Deviations for Recall Scores for the Video Event  Table 3  64  Pattern of Choices from Photospread by Interview Method and Memory Technique  Table 7  61  Pattern of Responses to Misleading Questions by Interview Method and Memory Technique  Table 6  59  Means and Standard Deviations for Recall Scores for the "Live" Event  Table 5  57  Percentage of Errors Reported About Three Aspects of the Video Event  Table 4  53  66  Pearson Product Moment Correlation Coefficients Between Confidence on Photo Identification Task and Recall Scores  Table 8  72  Means and Standard Deviations for Amended Version of Field's Inventory of Hypnotic Depth by Interview Method and Memory Technique  Table 9  Average of Two Self-Report Ratings by Interview Method and Memory Technique  Table 10  76  Ratings of Amount of Pressure Experienced by Interview Method and Memory Technique  Table 11  74  80  Ratings of Ability to Use Memory Technique by Interview Method  81  V  LIST OF APPENDIX J TABLES  Table 1  Analysis of Variance for Percentage of Errors i n Total Details Reported  Table 2  133  Analysis-of Variance for Total Number of Descriptive & Sequential Errors Reported About Videotape Event  Table 3  134  Analysis of Variance for Percentage of Errors in Details Recalled About Videotape Event  Table 4  Analysis of Variance for Number of Descriptive Errors Reported About Main Bank Robber  Table 5  135  136  Analysis of Variance for Percentage of Errors of Descriptive Details Reported About Main Bank Robber  Table 6  137  Analysis of Variance for Percentage of Errors i n Descriptive Details About Bank Setting  Table 7  Analysis of Variance for Number of Errors In Descriptions of Laboratory Room  Table 8  138  139  Analysis of Variance for Amended Version of Field's Inventory of Hypnotic Depth  140  vi  ACKNOWLEDGEMENTS  I was assisted in the design and writing of this thesis by several people who deserve special thanks.  First,  I would like  to thank my  advisor, John Yuille, for his many helpful comments and especially for his s k i l l f u l interpretation of the data. Bernard  of the Vancouver  Police  Thanks are also due to Mike  Department,  who  provided  valuable information on the use of hypnosis in the VPD.  me  with  I am most  appreciative of the time and effort expended by both Doreen Kum and Lee Porritt  of U.B.C.  —  the former  was  invaluable  as an assistant  experimenter while the latter helped guide me through many d i f f i c u l t times during data analysis on the UBC computer.  Finally, I would like  to express my sincere appreciation to two friends, B.A. Stanmore and Jim Sorfleet, who were continually supportive and patient with me throughout some trying periods during the writing of this manuscript.  1  THE EFFECT OF HYPNOSIS AS AN INTERVIEW TECHNIQUE ON EYEWITNESS MEMORY  Consistently since the turn of the century, researchers have been demonstrating  the unreliability  Muensterberg,  1908; Stern,  of eyewitness  1939; Whipple,  memory  (Binet, 1900;  1909, 1912).  In a classic  experiment, Muensterberg staged a shooting for an unsuspecting group of students to witness. Two weeks later, they were asked to recall as many of the  details of the incident as they could remember.  were analyzed incorrect.  to determine  proportion  of their  statements were  Muensterberg found that error rates ranged from a low of 26%  to a high of 80% per report. elaborated  what  Their written reports  this  pattern  Subsequent investigators have confirmed and  of results.  Using  a  similar  experimental  paradigm, Buckhout (1974) found that approximately 15% of statements made by witnesses were incorrect,  while  Clifford  (1979) reported  slightly  lower error ratios, between 65 and 70%.  The results of these and other  experiments  view  have  led to the current  among  psychologists  that  eyewitness memory is generally very poor. The fact that eyewitness memory can be f a l l i b l e has practical and theoretical  consequences  further investigation.  of such  importance  that  the issue  warrants  In practical applications, eyewitness f a l l i b i l i t y  poses serious problems for police investigators and for the courts, both of whom are dependent on eyewitness descriptions for the identification of suspects.  In a legal setting, the most serious error i n memory i s the  incorrect  identification  of a  suspect  by an eyewitness.  This i s  2  especially worrisome because "eyewitness identification i s often the most convincing  and decisive source of evidence i n a criminal case" (Leippe,  Wells & Ostrom, 1978, p. 345).  The British Home Office was so concerned  with this problem that i t commissioned Lord Devlin to lead an enquiry into the  area  of eyewitness identification  1976, The Home Office published  and i t s legal  implications.  In  the results of Devlin's investigation;  among other things, he advised: that the t r i a l judge should be required by statute...to direct the jury that i t i s not safe to convict upon eye-witness evidence unless the circumstances of the identification are exceptional or the eye-witness i s supported by substantial evidence of another sort (p. 150). Unfortunately  for the accused, the effect of being wrongly identified by  an eyewitness can be serious.  There are numerous examples in both the  legal and psychological literature of cases where mistaken identification resulted in conviction, imprisonment and even execution of innocent people (see,  for example, Brandon & Davies,  1977;  Loftus,  the  1979;  Wall,  1965;  1973; Buckhout, 1974; Goldstein,  Woocher, 1977; Yarmey, 1979b).  situation improved i n recent  years —  Nor has  eyewitness testimony remains  troublesome for the courts today. In addition to the practical concerns noted above, there theoretical  issues  eyewitness memory.  arising  from  an enquiry  into  are also  the f a l l i b i l i t y of  This i s because any rationale that attempts to explain  errors in memory w i l l also reflect something about the way in which memory is presumed to work.  There are two competing approaches to psychological  models of memory that characterize research in the area: or trace theory, means new:  and the reconstructive theory  i t has been  entertained,  the "exact copy"  Trace theory  in one form  i s by no  or another, for  3  centuries, from Aristotle Paivio  (in De Anima) to modern psychologists such as  (1971) and Tulving  structure of memory.  (1974).  Trace theory  emphasizes the mental  Two aspects of the theory are relevant here:  first,  the assumption that aspects of the event are stored somewhere in the brain and, second, that there i s a one-to-one correspondence between a physical stimulus in the environment and the trace laid down in memory, much like the way in which a videotape portrays real-life actions. described  the process this way:  One writer has  "Because the perceptual apparatus works  in a cybernetic fashion much like a giant computerized videotape recorder, the plethora of information  perceived  by the sensory system i s recorded  and stored in the brain at a subconscious level" (Reiser, 1976, p. Proponents of the theory  40).  offer a variety of reasons for memory errors.  For example, there may be obstacles to retrieving the stored  information;  that i s , the details are there but they are only partially accessible, perhaps blocked  by negative  emotions.  Another possibility i s that  there  may be an actual decay of the memory trace i t s e l f , perhaps due to chemical changes in the brain.  Opposing trace theory i s the reconstructive theory,  which focuses on memory processes rather than structure. memory  i s composed  of abstract  Henderson, 1903) or conceptual of this approach assume that  representations  It proposes that  of the world (cf.  schemas (cf. Bartlett, 1932).  Proponents  remembering i s "not the re-excitation of  innumerable fixed, lifeless and fragmentary traces...(but) an imaginative reconstruction..."  (Bartlett,  1932,  p.  213).  According  perspective, an individual actively reconstructs memories from  to  this  conceptual  schemas i n accordance with his or her prior knowledge and understanding of the event.  In this way, a person's recollection of the details of an  4  incident can be influenced by a variety of factors operating both during and after i n i t i a l observation of the event, and these factors can affect the  accuracy  of the person's memory.  For instance, a witness  could  incorrectly perceive some details of the incident, such as a suspect's facial  features  due to poor  lighting  conditions.  Or, he/she  could  incorporate into memory misinformation contained in interview questions. Thus, these two models of memory predict differences in the way errors can arise in eyewitness reports.  Examining the kinds of errors an eyewitness  makes can yield an insight into which theory is a more accurate reflection of the way in which memory works.  The present research was concerned, in  part, with implications of these two theoretical interpretations. During the last decade, eyewitness memory has received considerable attention previous  from memory researchers, and justifiably investigators have  not looked  so.  Unfortunately,  for ways of eliciting  better  reports; they have chosen, instead, to concentrate on the weaknesses of memory (e.g., Buckhout, 1974; Clifford, 1979) • In particular, Elizabeth Loftus has investigated the extent to which witnesses make  errors  by exposing  them  to either  can be induced to  suggestive  or deliberately  misleading post-event information (Loftus, Altman & Geballe, 1975; Loftus, Miller & Burns, 1978; Loftus & Palmer; 1974, Loftus & Zanni, 1975; Miller & Loftus, 1976). have helped hopelessly  Her conclusions regarding  the malleability  of memory  to promote the widespread belief that eyewitness memory i s inaccurate.  Such  wholesale  skepticism  has recently  been  challenged by some researchers (e.g., Wells, 1978; Yuille, 1980) who feel the negative aspects of eyewitness memory have been overstated. experimental  materials and procedure developed  Using  by Loftus, Yuille (1983)  5  showed  a  slide  sequence  of a  observers.  These  witnesses  description  of the incident.  pedestrian-motor  were  later  asked  Analysis  vehicle  accident  to provide  of the reports  to  a written  yielded  concerning both the quantity and quality of information provided.  data Yuille  reported that, on the whole, eyewitness descriptions of the incident were fairly accurate; that i s , of the total number of details reported by the observers, only 30% of them were incorrect. This i s in marked contrast to the high error ratios of 65 to 75% reported by Buckhout and Clifford. Results such as these suggest that, in some instances, eyewitness memory may not be as f a l l i b l e as once suspected.  This discrepancy regarding the  recall abilities of witnesses  highlights the need for further research  that  The study  resolyes the conflict.  provide  reported  further information on this contentious  here was designed to issue by examining how  well observers perform as eyewitnesses. Although i t may be useful, for theoretical situations  in which  witnesses  are coerced  reasons,  into  to construct  giving  inaccurate  testimony, as much of the previous research has done, too much emphasis on the f a l l i b i l i t y of memory can lead to an unbalanced picture of eyewitness capability.  The current negative view of eyewitness memory i s possibly  more a reflection of the design of the experimental situation than of the capabilities incident.  of witnesses  recall  the details  of an  What i s needed i s research that focuses on specific ways to  improve memory. theoretical value. any technique reports.  to accurately  This type  of research  would have both practical and  The criminal justice system would greatly benefit from  that enhanced memory and produced more accurate eyewitness  Research of this  kind would also be advantageous to memory  6  theorists, since examination of successful techniques would yield insights into  the underlying memorial  researchers  have  perspective.  examined  processes.  eyewitness  Yet, as noted  memory  from  this  earlier, few more  The current research was undertaken to f i l l this gap.  Subjects in this study were recruited from undergraduate classes  positive  at the University  of British  Columbia.  They  psychology  were  shown a  videotape of a simulated bank robbery that i s used by some banks to train new tellers.  In the videotape, two men can be seen entering the bank.  One acts as the principal robber who obtains the money from the bank teller; the other acts as his accomplice.  The main bank robber i s visible  to the viewers for the majority of the 90 second videotape.  One week  after viewing the video, the students were asked a series of questions which measured their memory for details of the incident. one  week  between  observing  the videotape  and being  An interval of interviewed was  considered appropriate since i t approximated real-life situations in which there i s usually some delay, from a few hours to several weeks, between an individual witnessing a crime and being questioned by the police.  One  week was chosen as an average time delay that would normally occur. The questions were patterned after  the type used  Department in their interviews with eyewitnesses. part, fairly objective in that they probed without leading the witness.  by the Vancouver Police They were, for the most  for details of the incident  For example, instead of asking whether the  bank robber had a beard or moustache, the interviewer asked "Did he have any facial hair?"  In this way, this experiment was able to measure how  accurately students could remember the details of a complex event, such as a bank robbery.  It was designed specifically to provide conditions under  7  which eyewitnesses would give the most accurate reports. were examined, the major emphasis of the study was  Although errors to determine  the  optimum method of interviewing eyewitnesses. One memory-improving technique that i s gaining widespread  acceptance  by police forces is the use of hypnosis to question witnesses about the details of a crime. psychologists. research. improve  Two  This practice raises several issues of interest to of these provided the major focus for the present  The f i r s t involves the supposition that hypnosis does, in fact, recall  interrogation  and  that,  tool.  as  As w i l l  a  be  consequence,  it  is  an  discussed in some detail  effective  below, this  interpretation of the existing anecdotal and experimental literature on hypnosis i s highly questionable. this  research  was  eyewitness memory.  to  of  students  whether  hypnosis  actually  does  enhance  After viewing the videotape of the bank robbery, one  group of subjects was groups  test  Thus, one of the primary objectives of  hypnotized  served  as  before being  controls  hypnotized during the interview.  —  interviewed.  these  Two  observers  other  were  not  It was predicted that i f hypnosis i s an  effective memory aid, the students who had been hypnotized would respond more accurately to  the  interview questions  than  those  who  had  not.  Anecdotally, i t is often the case that peripheral information, such as a licence plate number, i s recalled under hypnosis when i t has not otherwise been remembered.  Comparable aspects were included in this study so that  this issue could be investigated. service students.  was  taped At each  onto  the  For example, a phone number for repair  video  session the  machine,  assistant  clearly  experimenter  visible  to  the  made a passing  comment to the students while getting the video machine operational, to  8  the effect that she hoped the machine wouldn't break down.  This was an  attempt to get the subjects to notice the phone number without explicitly directing their attention to i t and so, hopefully, constituted a d i f f i c u l t item to remember. hypnotized  This allowed a comparison to be made of the ability of  subjects to accurately recall the number.  more subjects  in the hypnosis  group were able  number, this would provide support  If significantly  to remember the phone  for the trace theory  of memory and  justify the use of hypnosis as an interrogation tool. A second concern of this study relates to the fact that the use of hypnosis  i s based on the assumption that memory works according to trace  theory.  As Putnam (1979) suggests, the detective who uses hypnosis as an  investigative  tool has, implicitly  memory which  assumes  that  "expressed  information  an exact  i s stored  copy theory of  accurately  and that  forgetting i s the result of an inability to retrieve the information" (p. 439).  Hypnosis i s seen as a way of removing  successful  retrieval  of  the  stored  the obstacles  information,  thereby  suppressed information to return to conscious awareness.  blocking allowing  This belief in  the trace theory of memory has important implications for the way in which interviews with eyewitnesses are conducted.  This was the second concern  of this study.  If a rigorous interpretation of trace theory i s assumed,  there i s l i t t l e  danger that a witness  questions.  would be susceptible to leading  In fact, in cases of memory loss, interview questions would be  worded so as to contain key words capable those temporarily inaccessible details.  of providing associations for  On the other hand, i f memory is a  reconstructive process, the way in which a witness i s questioned can have a marked effect on the elicited response.  If witnesses  showed a tendency  9  to incorporate information  contained  in interview questions  into  their  recall of the incident, this would provide support for the reconstructive theory of memory.  Thus, a second aim of this research was to examine the  way in which interview questions could affect the accuracy of eyewitness memory. There has been a good deal of research into the way in which the wording of a question can influence the respondent's answer. part of this century, posing  Binet  (1900) and Muscio  deliberately leading questions  results.  In the early  (1916) established that  to witnesses  could have hazardous  Since then, researchers have consistently shown that  declines as the structure and specificity of questions  accuracy  increases  (e.g.,  Gardner, 1933; Marquis, Marshall & Oskamp, 1972; Marshall, 1969; Marston, 1924;  Snee & Lush,  1941).  For example, Lipton  completely unstructured testimony than testimony questions. response. either  elicited  that  from observers produced greater accuracy  through open-ended, leading or multiple-choice  Even varying just one word in a sentence can influence the For example, Harris (1973) asked subjects one of two questions,  "How  tall  was the basketball player"  basketball player?"  question was used.  or "How  short  was the  He found that students estimated the player as being  significantly shorter (by approximately  and  (1977) reported  10") when the latter form of the  Loftus & Zanni (1975) varied use of the articles "the"  "a" in questions  posed  to witnesses.  They  found  that  observers  interviewed with "the" questions were more likely to say they had seen something, such as broken glass, when i t had not actually been present. Loftus  & Palmer  numerical  (1974) reported  estimates  of vehicle  a similar speeds.  pattern  What  this  of results body  using  of research  10  indicates i s that the form of the question answer that i s received. using  any information  can, indeed,  influence the  Respondents appear to be answering  available to them, even  questions  i f i t i s embedded i n  interview questions, to reconstruct their memory of the event. During further  by  the last decade, Elizabeth Loftus has explored investigating the issue  of suggestibility.  studies, Loftus has deliberately given observers embedded in a questionnaire, which witnesses event.  this matter In numerous  misleading  information,  completed after watching an  Using this paradigm, Loftus has repeatedly found support  for the  reconstructive position in that the post-event misinformation supplied to witnesses was subsequently  incorporated into their memory of the details  of the incident (Loftus, Miller & Burns, 1978; Loftus & Palmer, 1974). She has suggested that the misinformation  actually changes the person's  memory. Not a l l researchers agree with Loftus on the effects of misleading information,  nor with  her interpretation of the memorial  processes  involved (e.g., Dodd & Bradshaw, 1980; Read, Barnsley, Ankers & Whishaw, 1978; Yuille, 1983).  Dodd & Bradshaw examined the effects of manipulating  the credibility of the source  of the misleading information.  Loftus  that  paradigm,  they  found  erroneous  information  Using the  contained  in  leading questions was "remembered" by students i f no mention was made of the  source  source.  of the presupposition  or i t was attributed to a neutral  However, i f the source of the misleading information were shown  to be biased, as in the case of testimony  regarding a car accident given  by the driver who caused i t , subjects did not incorporate the erroneous information into their recall of the event.  This situation should not  11  arise i f memory i s truly being distorted by the wording of the questions. Dodd & Bradshaw argue that "the wording of a sentence may 'distort the memory' only when the presuppositions are introduced as though they come from heaven, or at least when there i s no source who might be presumed to have  complex  intentions" (p. 696).  laboratory experiments exclude  They  such important  the "speaker" that are usually considered  also  point  out that  most  variables as intentions of  in everyday verbal exchanges,  thus jeopardizing the ecological validity of memory experiments modeled after Loftus paradigm. 1  As noted earlier, Yuille (1983), using the Loftus materials, found that witnesses were fairly accurate in their recall of the incident, in that 70% of the details they remembered were correct. also examined the observer's  In this study, he  degree of susceptibility to misinformation  supplied to them using the Loftus questionnnaire.  Yuille failed to find  as high a degree of susceptibility as Loftus had reported. effect of the misinformation which was very salient corner  t r i v i a l aspect still  of traffic  sign at the  but was more noticeable when i t concerned a  (the colour of a passing car); even so, this effect was  not as pronounced  interpreted  was negligible when i t related to a detail  to the scene (the type  of the accident)  In fact, the  these  results  as previously  reported  by Loftus.  Yuille  as supporting  the reconstructive view  that  witnesses used the information in the questionnaire to f i l l in the gaps in their memory of the incident, but only for details previously unattended. He was able to ascertain whether a subject had attended to the detail by f i r s t obtaining a written descritpion of the incident before introducing the misinformation.  It should be noted that this i s a variation of the  12  usual Loftus procedure, whereby observers do not have an opportunity to give  a  free  recall  misinformation.  of  the  incident  before  being  exposed  to  He found that students who mentioned either the traffic  sign or the colour of the passing car in their written reports were not affected by the misleading information.  While his interpretation differs  from Loftus' contention that observers actually change, or amend, their memory due to the misinformation, i t i s consistent with the general view that memory operates  according to reconstructive processes  rather than  according to trace theory. This theoretical controversy was examined in this study by embedding misinformation in five of the 47 interview questions posed to observers after they had witnessed the videotaped bank robbery. misleading  in that they  presupposed  attribute, such as a hat or a watch. the main bank robber asked:  the  "Was  to be  incorrect; however, i f the  The  incorrect  details  would then ask  It was predicted that, i f the trace theory  would  not  to the erroneous be  association to a memory trace that didn't exist. indeed  witness  other misleading questions were  were correct, subjects would be resistant the  trivial  An affirmative answer to this  unsure, the experimenter  he wearing gloves?"  treated in a similar manner.  because  fairly  "Did you notice i f his gloves were leather?"  answered in the negative or was directly  of a  For example, one question concerning  when, in fact, he was not wearing gloves. question would be considered  presence  The questions were  capable  of  information, providing an  However, i f memory were  a reconstructive process, then subjects should make more errors  when answering these questions than the other more objective ones since they would use the erroneous details in their reconstruction of the event.  13  From this experimental design, i t would also be ascertained whether hypnotized witnesses were more susceptible to the misleading than those who this matter.  were not hypnotized.  There is considerable  information confusion  on  Sheehan & Tilden (1983) found that hypnotized individuals  were no more suggestible than non-hypnotized controls; however, they did find  that  questions.  a l l students  made more errors  when exposed  to  misleading  Other studies have supported the notion that hypnosis greatly  increases suggestibility. For example, Putnam (1979) showed 16 students a videotape of a car-bicycle accident about the incident. used by Loftus  and  later asked them 15  questions  Of these, six were leading questions similar to those  (1974; 19.77; Loftus, Miller & Burns, 1978)  in that  the  definite article "the" was used instead of "a" in a question such as "Did you see a stop sign?"  It is believed that questions using "the" presume  the presence of the object, in this case a stop sign, and witness into believing the Subjects  in the  answering the group.  object  was  present  when i t really wasn't.  hypnosis condition made significantly more errors when  leading  questions  than did  subjects  in the  non-hypnosis  However, there were no differences between groups in the accuracy  of responses to the objective questions. design,  so lead the  Using a similar experimental  Zelig & Beidleman (1981) replicated Putnam's results.  (1982) employed hypnotized  a  task  individuals  identifications,  involving made  memory  more  for  errors  than a non-hypnotized  faces,  and  and  had  control group.  Wagstaff  found  fewer The  that  correct  results of  these studies have lead to the widespread belief that "subjects are more suggestible  in  the  hypnotic  state  and  are,  therefore,  influenced by the leading questions" (Putnam, 1979, p. 444).  more  easily  14  If way  this conclusion is valid, i t has important  implications for the  in which hypno-investigators conduct interviews with  Extreme care w i l l have to be taken completely  objective,  and  do  not  eyewitnesses.  to ensure that their questions lead  the  witness  in  any  are way.  Unfortunately, the issues remains controversial due to some methodological problems in the previously cited subjects, 4 per condition, was be  drawn.  incident  Further,  Putnam's sample size  nearly half and  of the  15  questions  one-third of those  (1981) experiment, were misleading.  misleading  questions  artificial.  makes  of  16  too small for any reliable conclusions to  in the Putnam study,  Beidleman  studies.  the  interview  asked about in the  the  Zelig  &  Such a large number of situation  in  this  study  Finally, the format of the Zelig & Beidleman interview may  have confounded the results.  First, a l l of the  were multiple-choice, in contrast to the  15 objective' questions  5 misleading  questions which  required yes/no responses.  Second, a l l the multiple-choice questions were  asked  a l l the misleading  first,  following by  questions.  correct response for a l l of the misleading questions was  Finally,  "no".  the  Thus, the  fact that hypnotized students made more errors replying to these questions could be explained by their preference for saying "yes" and not by fact that they were more suggestible to the misleading information.  the  These  obvious methodological  d i f f i c u l t i e s casts some doubt on the validity of  Zelig  conclusions.  &  Beidleman's  overcome these methodological  The  present  experiment  sought  to  weaknesses by obtaining information from a  total of 72 students, 24 in each of the three different interview states (hypnotized, relaxed or alert).  The interview consisted of 47 questions,  only 5 of which were misleading, and these were interspersed throughout  15  the entire interview session. The  possibility  that hypnotized witnesses are more susceptible  to  leading questions becomes a matter of some concern when one considers that the situation often arises whereby the hypnotized person is very slow to answer questions monosyllabic  or  and,  when the response is finally made, i t is either  consists of very  few  words.  This  reluctance  of some  witnesses to talk a great deal during hypnosis results in the investigator obtaining a very abbreviated  account of the incident.  problem facing the hypno-investigator: more  complete  description  investigators avoid  from  This is the main  how is he or she going to e l i c i t a the  witness?  deliberately leading  In  questions  practice,  (e.g., was  most  his hair  brown?) i f a more objective form of the question can be used (e.g., what colour was his hair?). eye  colour, but how  would the  presence of a watch? used,  the  This is fine for simple attributes like hair or investigator probe for details about the  If the question  mere mention of  the  word  "was  he wearing any  "jewellry" could  suggestible hypnotized witness to believe there was present.  jewellry?" i s  lead  the  highly  a piece of jewellry  The large body of research investigating the effects of the form  of the question on accuracy of response supports this concern.  Of course,  i f a rigorous interpretation of trace theory is assumed, then probing for more specific details, even through the use of leading questions, not be particularly worrisome. using curious  leading  questions  inconsistency  interviewing  practice.  to  reluctance of hypno-investigators  obtain  between By  The  should  more information  belief  contrast,  in  the  trace  proponents  of  reveals theory the  to  a  rather  and  actual  reconstructive  theory would be concerned that any witness, not just those under hypnosis,  16  would be influenced by misinformation contained in interview questions. Use of the videotaped bank robbery as the c r i t i c a l event for students to  witness  in this  study  provided  an opportunity  aspects of the eyewitness situation to be examined. observers identify  for two additional First, how well could  one of the actors from a photo spread  of coloured  pictures or "mugshots", and, second, how confident were students in the accuracy  of their  eyewitness  choice?  Previous  identifications conducted  investigators  have  reported that  under similar laboratory conditions  are not particularly reliable, although the identification rate i s usually above that expected by pure chance.  For example, Leippe, Wells & Ostrom  (1978) found that following a staged theft, 35% of witnesses made false identifications from a photo spread while 35% made no choice at a l l . There is disagreement between  correct  confidence.  i n the literature as to the strength of the relationship identification  Some  of a "suspect"  researchers  report  a  and a high  positive  rating of  association  (e.g.,  Brigham, Snyder, Spaulding & Maass, 1982; Egan, Pittner & Goldstein, 1977; Lindsay, Wells & Rumpel, 1981; Malpass & Devine, Loftus, 1979; Wells, Lindsay & Ferguson,  1980; Powers, Andriks &  1979) while others report a very  weak or even negative correlation between these measures (e.g., Brown, Deffenbacher  & Sturgill,  1977; Buckhout, 1974; Clifford  & Scott, 1978;  Gorenstein & Ellsworth, 1980; Leippe, Wells & Ostrom, 1978; Wells, Lindsay & Tousignant,  1980; Yarmey,  1979a).  In this  study,  this  issue was  examined by asking the students to point to the main bank robber i f they saw his picture in a photo spread of eight mugshots after they had been interviewed by the experimenter.  If they did not see his picture, they  were to indicate that he wasn't there.  The instructions were worded in  17  this manner because i t has been shown that telling a witness that the suspect i s definitely  in the lineup leads to significantly more false  identifications than saying the suspect may Ostrom, 1975).  or may not be there (Hall &  It was felt that the procedure employed here would reduce  any pressure students might feel to make an identification merely because they  were  asked  to  do  so  by  the  experimenter.  identification task, students were asked to rate how in  the accuracy  opportunity correctly  to  of their  choice.  determine,  first,  Following  the  confident they were  Thus, these procedures  provided an  the  ability  students*  overall  to  identify the main bank robber from a photo spread and, second,  the relationship between rate of correct confidence.  identification  and  degree  of  Another reason for including the identification task in this  study is that past research in hypnosis has failed to examine whether first  describing  a  identification. identification,  "suspect" under  That  is,  hypnosis  would  would  hypnosis  affect  facilitate  subsequent accurate  or maybe increase a subject's confidence that his/her  choice was accurate?  Thus, use of the identification task also allowed an  examination  effects  of  the  of  hypnosis  on  identification  rate  and  confidence. As  noted  earlier,  a  general  objective  of  this  research was  to  investigate ways of improving memory and, more specifically, to ascertain the  efficacy  improves  of hypnosis  as  memory i s , indeed,  a memory aid. contentious.  The  Most  events assert that hypnosis does enhance memory. that  since 1970,  when the Los  Angeles  claim that  accounts  hypnosis  of anecdotal  Reiser (1976) claimed  Police Department began "using  18  hypnosis  as an adjunct  in specific  criminal cases  involving homicide,  kidnap and rape", there has been "an approximate 60% increment of success over traditional interrogation techniques" (p. 36). Douce  (1979) indicated that hypnosis  cases.  Typical anecdotal  accounts  improved  Similarly, Kroger & in 60% of  recall  usually claim  that  under  their  hypnosis  witnesses have been able to describe assailants much better, or remember significant details about vehicles, such as make, colour and licence plate numbers,  than  techniques.  they  were  Raginsky  able  to  using  (1969) reported  sessions, one witness was  conventional  that, after  two  interrogation short  hypnotic  capable of recalling the cause of an aircrash  which had eluded him during two years of questioning using more orthodox methods.  Several  other  questioning a witness information  writers  have  under hypnosis  that helped  solve  Douce, 1979); Reiser, 1974;  the  cited  further  instances  when  led to the emergence of valuable  case  (e.g., Hanley,  1969;  Kroger &  1976; Reiser & Nielson, 1980; Schafer & Rubio,  1978). On  the  other  hand,  the  results  of  hundreds  of  psychological  experiments conducted during the last 60 years, only a few of which were direct investigations of eyewitness memory, do not substantiate this claim that  hypnosis  improves  memory.  In  fact,  researchers  conceptually as to whether a "hypnotic state" exists. is a state of altered consciousness  still  disagree  Some claim hypnosis  (e.g., Hilgard, 1973)  while  others  contend that i t is an artifact of the situation elicited by the motivating pressures of the task (e.g., Barber, 1969) or produced because the subject enters into role-playing with the experimenter (e.g., Sarbin & Coe, 1972).  19  Not  surprisingly,  this  theoretical  conflict  is reflected  in the  research reporting the effects of hypnosis on memory.  There have been  relatively  the  few  experiments  that  directly  hypnosis using an eyewitness paradigm. that hypnosis aids memory.  The  Beidleman  earlier  heightened  mentioned  (1981)  suggestibility  investigated  in  connection  hypnotic  recall  and  (1979)  with also  the  in response  greater accuracy.  to neutral questions, nor  Zelig & issue  revealed  hypnotized witnesses did not recall more information than controls  of  Most have not supported the notion  studies by Putnam  for  effects  of  that  non-hypnotized  did they  answer with  Methodological problems inherent in these studies have  been previously discussed.  Sanders & Simmons  (1983)  presented  both a  lineup recognition as well as a structured recall task to hypnotized and non-hypnotized  students.  They also attempted  to lead the witnesses into  false identification and  incorrect answers.  Five of the ten strucutred  recall questions were misleading.  They found no benefit of hypnosis in  either task; in fact, they reported that hypnosis suppressed accuracy due to  susceptibility  to  misleading  implications.  One  general pattern was the study by DePiano & Salzberg hypnosis  enhanced  recall  of  the  incidental  exception (1981)  aspects  of  to  this  who found that an  event.  A  methodological difficulty with this study should be noted, in that DePiano &  Salzberg failed  to report error  rates.  Thus, while  the  hypnotized  students recalled more information than a control group, i t could be that they also made more errors. accuracy  rates  due  to  the  It is particularly claim  increases errors (e.g., Putnam,  1979;  by  important  some researchers  Wagstaff,  1982).  to compare  that  hypnosis  The inconclusive  20  evidence of the past research investigating hypnotic memory enhancement of eyewitnesses was one reason for the research reported here. By far the majority of previous studies has investigated the effects of  hypnosis  on learning  and memory.  The results  from  this  body of  research are, at best, inconclusive, yet i t i s from this literature that proponents  of hypnosis  eyewitness memory. this  make the interprepation  that  hypnosis  improves  One of the reasons for the confusion i s that much of  research i s characterized  by poor  outlined in some detail below.  methodological procedures, as  The present experiment  designed to avoid these methodological flaws.  was specifically  The following review of  some of the previous experiments w i l l serve the dual purpose of presenting typical findings reported in the literature as well as illustrating the problems of control encountered in hypnosis research. A number of different approaches have been taken to determine whether hypnosis improves memory. subjects  Many of the early experimenters age-regressed  to a previous point  in time, encouraging  them to relive an  experience in the hopes that memory for the event would be more vivid and, hence, more accurate.  In a study typical of this kind of research, True  (19^9) regressed 50 students to Christmas day and their birthday at ages 10, 7 and 4.  Each subject was asked to name on what day of the week these  two specific events f e l l , answers  were  correct  at  for each age. age  10,  with  percentages reported for the other two ages. subject back  He reported that 93% of the similar  higher-than-chance  In another study, As (1962)  regressed  a single  to a time  when he knew a foreign  language.  This subject was tested for his memory of the language during  21  several sessions.  As reported that the subject always had better recall  during hypnosis than he had in a waking state.  Unfortunately, since there  was no control for order effects in that hypnotic recall always followed waking  recall,  methodological those  cited  the results grounds.  have  often  reported  Nevertheless, been  by  As  can  age-regression  interpreted  be  questioned  studies  as providing  on  such as  evidence  that  hypnotic regression facilitates retrieval of memories which subjects had previously been unable to recall. Subsequent attempts to replicate these results have been remarkably unsuccessful.  When Best & Michaels (1954) regressed five subjects to two  past birthdays, nine out of ten times the subject misidentified the day of the week on which the birthday had fallen. Reiff & Scheerer  Using  the same task, both  (1959) and Barber (1961) were unable to obtain the high  percentage of correct responses obtained by True.  Further, Sarbin (1950)  compared the test performance of nine adults regressed to age 8 with their performance on the same intelligence test when they were actually 8 years old.  None of the subjects showed the same mental age as they had on the  original test. Most of the age regression studies reporting memory enhancement have serious methodological  flaws.  often f a i l  to counter-balance  absolutely  necessary  Experimenters using a within-subject design hypnotic  control since  and waking recall.  some researchers  (e.g.,  This i s an Erdelyi &  Becker, 1974; Erdelyi & Kleinbard, 1978) have shown that memory improves with repeated  attempts at recall.  the performance of age-regressed  Further, as Barber (1961) has noted, subjects should be compared to that of  22  subjects  receiving  regression suggestions  without  hypnotic induction.  When Barber (1961) instructed alert subjects to regress to an earlier age, a l l differences in recall ability between the groups disappeared.  Other  researchers who have provided additional methodological controls for the effects of experimenter bias and level of subjects' motivation have found no increase in recall which could be attributed to hypnotic age regression (Leonard,  1965; Wall & Lieberman, 1976).  It i s appropriate to conclude  that the age regression studies do not offer conclusive support for the notion that recall i s improved through the use of hypnosis. Other  researchers  have  attempted  to enhance  memory  by  leaving  subjects with a post-hypnotic suggestion of improved performance. experiment  representative of this  approach,  In an  Hammer (1954) tested nine  subjects on a total of 25 cognitive tasks, including learning meaningful prose and nonsense syllables. state  and twice  after  Each subject was tested twice in a normal  being  hypnotized  and  receiving  suggestions of improved ability.  Subjects performed  more  the post-hypnotic  proficiently  researchers  have  concentration,  after  hearing  reported  study  similar  improvements  habits and acquisition  (Eisele & Higgins, 1962; Illovsky,  most of these tasks suggestion.  in reading  and retention  concluded  Other ability,  of material  1963; Krippner, 1963; Lodato, 1964;  McCord, 1956; McCord & Sherrill, 1961; Sears, 1955, 1956). experimenters  post-hypnotic  A l l of these  that the post-hypnotic suggestion led to better  memory or performance. As with the age regression studies, a l l of the research examining the effects  of  post-hypnotic  suggestions  on  memory  lacks  a  crucial  23  methodological  control.  The  performance of hypnotized  subjects should  have been compared to a control group who had not been hypnotized but who had  received motivating  1965b).  suggestions  for improved  performance  (Barber,  It could be that the suggestions alone, and not the hypnotic  induction, produced  the facilitating effect.  Barber & Calverley (1966)  included such a control condition when they tested 90 subjects for memory of material learned two months previously. subject  (hypnotic  or  waking)  (motivation,  regression  instructions  at  conditions.  These  to  a l l ) and results  Lyon-James (1957) and  and  the  found are  Leonard  the  time no  They varied the state of the type  of  of  recall  original  learning,  memory improvement  in accordance (1963) both  with  instructions  in any  those  or  of the  obtained  of whom employed  a  no  by  similar  experimental design (cited in Barber & Calverley, 1966). Another, more indirect, approach some researchers have taken i s to examine the effects of hypnosis on the acquisition of material. studies,  subjects are  usually  presented  state; recall is then attempted  material while  In these  in a hypnotic  in either a waking or a hypnotic state.  As an example of this type of study, Rosenhan & London (1963) presented nonsense syllables to 16 "good" and 16 "poor" hypnotic male students, in waking and categories scale.  hypnotic on  They  subjects was  conditions.  the basis of their found  that  learning  Students  were separated  scores on a hypnotic performance  of  greatly increased i f they attempted  the recall  into the  two  susceptibility low susceptible in a hypnotic  state while the performance of the highly susceptible subjects in this condition  decreased  slightly.  These  results  were  interpreted  as  24  indicating  that hypnosis facilitates  learning  for the  less  susceptible  subjects but somehow impairs i t for the very susceptible. Barber (1965b) offered another interpretation, suggesting that large gains shown by "poor" subjects under hypnosis may their learning considerably subjects. their  the  have been due to  less in the waking condition than the "good"  Thus, when the poor students attempted recall under hypnosis,  performance reached  supposition  is correct,  the  normal level.  i t is s t i l l  Whether  unclear  why  or  the  not  Barber's  "poor"  subjects  benefitted the most from the induction when they were the ones who,  as  indicated by their performance on the hypnotic susceptibility scale, were supposed to be less affected by hypnosis. The majority of researchers  investigating the effects of hypnosis on  the acquisition of material have not found consistent differences between the recall of waking and acquisition  and  associates)  and  retention  hypnotic subjects. of  slight  meaningful  words  (adjective  the noun  nonsense syllables presented in a normal waking state,  during hypnosis or in a simulated a  both  Young (1925) assessed  improvement  hypnotic condition.  Although there  in memory for meaningful childhood  was  events, Young  concluded that overall there were no clear-cut differences in recall among the  three  conditions.  Eysenck (1941), Das Swiercinsky & Coe  Similar  results were obtained  by  (1961), Schulman & London (1963) and,  Gray  (1934),  more recently,  (1970), a l l of whom reported no differences in recall  between waking and hypnotic subjects. Much of the above cited research  relates only  tangentially to  issue of whether hypnosis improves eyewitness memory, since,  as  the  noted  25  earlier, the majority of these experiments nonsense  syllables  A few  dealt with such material as  studies have examined  this  assumption  more  directly by allowing subjects to observe material in a waking state and obtaining recall under hypnosis or some other state.  Again, results from  these studies are mixed, with methodological shortcomings precluding any firm assertion that  the hypnotic induction  leads to an  improvement in  memory. For example, Sears (1954) allowed 24 subjects a 30 second glimpse of a table containing a large number of objects. They attempted recall of the items in several sessions, each subject recalling f i r s t while awake and then after a hypnotic induction.  Sears found that subjects always  recalled more items when they were in a hypnotic trance than in the normal waking state.  However, since he  did not. counterbalance  the order of  waking and hypnotic recall, his findings remain questionable. In  other, more methodologically sound, studies, improved  sometimes  been  reported  using  hypnosis.  compared waking and hypnotic recall learned several years previously. significant  improvement  Stalnaker  &  recall has  Riddle  (1932)  of poetry and prose that had been  They reported that hypnosis produced a  in memory for both  similar type of study, White, Fox & Harris  types  of passages.  (1940) found  that  In a  hypnosis  facilitated recall for poetry but not for paired-associates or pictures. Similarly, Rosenthal (1944) found a slight  improvement due to hypnosis.  He presented meaningful and nonsense material to 13 students who showed medium to high hypnotic susceptibility. waking state, and  Recall was  f i r s t assessed in a  then in counter-balanced hypnotic and waking states.  The same suggestions for recall were given in both conditions.  In three  26  out of four experiments, no significant  differences were found in the  amount or type of material recalled by subjects in both conditions. one  experiment,  though,  hypnotized subjects recalled  In  more than waking  subjects, but the effect was obtained only for the meaningful material. Finally, Dhanens & Lundy (1975) attempted to control for the effects of degree  of susceptibility,  subject.  Improved  recall  motivating instructions, was found  only  and state  of the  for subjects recalling the  meaningful material in the hypnosis plus motivation condition. On the other hand, a number of researchers comparing hypnotic recall have failed to find  waking and  a significant effect due to hypnosis.  Huse (1930) presented waking subjects with nonsense syllables and tested each person for recall in both waking and hypnotic conditions. no advantage for hypnotic recall.  He found  Similar results have been obtained by  Barber & Calverley (1966), Cooper & London (1973) and Mitchell (1932). In many of the studies reported above, variables other than hypnosis have not been properly controlled and have confounded example, the type of material which ranges  from  sterile  childhood events.  nonsense  the results. For  subjects were required  syllables  to potentially  to recall  emotion-laden  In the majority of cases where hypnosis did not improve  memory, nonsense syllables served as the stimuli to be remembered (e.g., Barber & Calverley, 1966; Cooper & London, 1973; Das, 1961; Huse, 1930; Rosenthal,  1944; Schulman & London,  Young, 1925).  1963; White,  Fox & Harris, 1940;  As stimuli, nonsense syllables are particularly unsuitable  because they are generally poorly remembered.  For example, consider the  Barber & Calverley (1966) experiment in which students were presented with  27  12 nonsense syllables.  When recall was assessed two months later, the  average number of syllables remembered was less than 1. Memory for such t r i v i a l material i s notoriously poor, so perhaps i t i s not too surprising that hypnosis i s not an effective memory aid in these situations. improvement  for hypnotic recall asked  When  has been reported, in most instances  subjects  were  to recall  more meaningful  or better  material,  such as poetry, prose passages, Morse code,  organized  etc. (e.g., As,  1962; Eisele & Higgins, 1962; Dhanens & Lundy, 1975; Hammer, 1954; Sears, 1954; True, 1949; Stalnaker & Riddle, 1932; White, Fox & Harris,  1940).  While we know from previous research that memory i s best for meaningful material (see, for example, Hyde & Jenkins, 1969; Craik & Jacoby,  1975),  what remains in question i s why hypnosis may be a selective enhancer of memory, effective in one situation but not in another. Perhaps  hypnosis  i s effective  component  that  i s added  anecdotal  cases where hypnosis  only  to a meaningful  when event  there  i s an emotional  for the subject.  In  has been reported as improving memory  (e.g., Hanley, 1969; Reiser, 1974, 1976; Reiser & Nielson, 1980; Schafer & Rubio, 1978), the witnesses were emotionally upset and thus the emotional associations may have interferred details.  In such cases, Dorcus  with  attempts  to remember important  (1960) has suggested that hypnosis aids  recall by overcoming the emotional blocking. An alternative interpretation is to suggest that hypnosis allows the subject to achieve a state of deep relaxation, reducing tension and anxiety caused with the event.  by emotions  associated  While i t i s obviously unethical to upset subjects in an  experimental setting to empirically test this assumption, the possiblility  28  s t i l l remains that hypnosis may not be of any aid to memory in cases where there i s l i t t l e emotional involvement on the part of the witness.  This  could explain the confusion in results reported from experiments conducted within a laboratory setting. The  successful  use of hypnosis as a memory aid reported i n the  anecdotal literature cannot be ignored. only when a more r e a l - l i f e ,  Perhaps hypnosis i s beneficial  or meaning-laden,  context i s provided in  contrast to the relatively a r t i f i c i a l laboratory situation.  While.viewing  a simulated crime in a laboratory setting i s not as emotionally involving as  witnessing a  real-life  incident,  the experience  i s probably more  meaningful to subjects than involving them in learning nonsense syllables or  paired-associates.  eyewitness  situations  Further, the laboratory fairly  well:  in both  can simulate  settings,  real-life  people are not  instantly aware of their role as an eyewitness, nor are they entirely certain of what details they w i l l be asked to remember later. reasons, the eyewitness paradigm investigating  For these  was considered to be well-suited for  the effects of hypnosis on memory.  Another advantage of  this paradigm i s the fact that witnesses can be questioned about both the observed event (such as a videotape of a bank robbery) and the real l i f e situation in which the event occurred (such as the appearance experimenter  showing the videotape).  any differences i n recall Approximately  half  ability  The study reported here monitored  between videotaped and "live" events.  of the interview questions concerned  bank robbery; the remainder  of the  the videotaped  dealt with such aspects of the laboratory  setting as the layout of the laboratory room where students observed the  29  critical  event  and  the appearance of the experimenter  students the videotape.  who  showed the  In order to avoid any problems that might come  from the same person showing the videotape, and hence witnessing i t ,  and  subsequently asking questions about the event, two different experimenters were used, one to show the videotape and one to interview students.  This  also provided an opportunity to ask for additional information about the experimenter who  ran the videotape of the bank robbery. In this way, i t  could be determined whether hypnosis enhanced memory for live events only. There i s yet another component of the hypnotic situation that might play  a  role  in  experimentally.  enhancing  recall  that  has  not  been  explored  This is the strategy by which the subject attempts  to  remember the material. As noted earlier, victims or eyewitnesses who have suffered some emotional trauma in connection with a crime often find i t painful  to  recall  details  of  the  event.  Because  of  this,  hypnoinvestigators on police forces commonly employ an imagery technique which,  as  they  see  emotionally detached  i t , allows  the  person  to  observer of the incident.  become  an  Using this  impassive, strategy, a  witness i s asked to imagine a television screen on which he or she can watch the event unfold. While this technique may be useful in defusing an emotional  situation,  improve recall. 1983)  showing  i t also may  capitalize  on the use  of imagery to  A large body of data exists (summarized in Paivio, that  imagery  helps  recall.  In  cases  1971;  where memory i s  enhanced during hypnosis, i t may be that the imagery instructions given to the  person  state.  are  responsible for the  improvement  and  not  the  hypnotic  The present research examined this possibility by providing half  30  the subjects with imagery instructions.  For comparison purposes,  i t was  desirable to give some kind of memory instructions to students that would be comparable to the imagery condition but which did not involve imagery per se.  An obvious choice was the guided memory technique used by Malpass  & Devine (1981).  This method involved asking eyewitnesses to remember not  only the details of the event they had observed but also their feelings and reactions concerning the day in question. this guided memory procedure  Malpass & Devine found that  enhanced accuracy of identification from a  photospread without biasing memory for the c r i t i c a l event. then,  half  the students  were given reconstructive recall  which required them to recreate their mood, clothing, appeared  In this study,  etc., surrounding  instructions  personal circumstances,  the observed  event.  such as  This technique  to be an appropriate motivational and temporal  control for  contrast with the imagery witnesses. It  i s obvious from the foregoing review of research that no clear  decision can be made concerning the efficacy of hypnosis as a memory aid, due to methodological problems of the previous research.  In fact, Barber  (1969) has been extremely c r i t i c a l of much of the previous research which utilized a non-hypnosis group as the appropriate control for the hypnotic group. the  He has suggested  one contained  that the standard hypnotic inductions (such as  in Stanford Hypnotic  Susceptibility  Scale: Form C,  Weitzenhoffer & Hilgard, 1962) include within them a number of variables that are confounded with the hypnotic state.  For example, he and his  colleagues have shown in a number of studies (summarized in Barber, 1969) that  when  controls are provided  for such  factors  as task-attitude,  31  task-expectation attributed  and  task-motivation,  any  to hypnosis completely disappear.  existing research methodological  discernible  differences  Barber's analysis of the  on hypnosis i s particularly important in highlighting  weaknesses  inherent  in earlier  studies.  Clearly, the  control group to which hypnosis i s compared i s of c r i t i c a l importance i n isolating the aspect of hypnotic induction responsible for bringing on the hypnotic state, which in turn may or may not enhance recall.  While i t i s  tempting to conclude (as has Barber, 1969) that methodological weaknesses in much of the past research negate any facilitation of memory due solely to hypnosis, this criticism cannot be extended to a l l studies reporting hypnotic memory improvement (e.g., Dhanens & Lundy, 1975). Examination of a number of popular hypnotic most place  inductions reveals that  a heavy emphasis on suggestions to the subject  to relax.  Rosenthal (1944) and Dorcus (1960) have both proposed that relaxation i s the  key component  studies  seeking  of hypnosis physiological  that  facilitates  differences  consciousness tend to support this notion. between hypnotic  memory.  between  Results  various  from  states of  It i s possible to distinguish  and waking subjects on measures of skin temperature and  peripheral vasodilation (Reid & Curtsinger, 1968; Timney & Barber, 1969). However, no differences have been found between relaxed and hypnotized subjects on such measures as EEG, EMG, heart rate, respiratory rate or GSR (Coleman, 1976; Morse, Martin, Furst & Dubin, 1977; Paul, 1969; Peters & Stern,  1973; Tebecis,  Hamilton,  1975).  These  Provins, findings  Farnback suggest  & Pentony, there  1975; Walrath &  i s an extremely  close  association between hypnosis and relaxation, leading Edmoriston (1972) to  32  advocate  that  separated  the  from  There earlier,  relaxation  the  has  been  Dhanens  significant however,  just  &  improvement  Lundy  relaxation  component experiment  shortcoming.  In  this  were  to  the  compared  were  contrasted  eyewitness  entire content  from  inductions. deep  designed  study,  and  then  relax  the  relaxation  Thus,  i t  in  hypnotic  procedure  was  very  similar  the  subjects  subject  too  closely.  a  this  typical  study  and  no  who  were  be  an  while  of  that  if  the  to be  be  it  inducing it  them  throughout  standard  did  relaxed  expected  be  state  of  mimic  the  procedure  argued  were,  to  word  hypnotic  relaxation could  to  their  and  a  not  turn  standard  requiring  groups  of  the  in  the  in  induction,  would  by  way  the  students  these  contained  supposed  differences  and  procedure  Obviously,  were  hypnotized  in  ensuring  hypnotic  of  both  ideal  for  methodological  typical  different  control  this  a  significant  Consequently,  subjects,  muscle  found  material;  no  a  recall.  relaxed  different  mentioned  design,  showed  provided  group  As  meaningful  abilities  alert  to  "experimentally  this.  overcome  suggestions  appeared  has  relaxed  quite  the  to  hypnotized  is  of  hypnotic  Subjects  do  subjects  to  of  to  recall  recall  waking,  technique  in  the  performance  This  relaxation  fact,  was  be  within-subject  researcher  assessing  paradigm.  tense  body.  other  a  a  relaxed  when  with  memory  alternately  No  should,  attempting  hypnotic .of  and  220).  using  (1975),  group  memory.  can,  (p.  study  for  control  in  element"  one  improvement  their  present  hypnotic  component  in  occur  that  actual between  groups. As  most  mentioned,  a  of  the  brief  major summary  elements is  now  of  the  current  presented  in  experiment  order  to  give  have the  now  been  reader  a  33  clear picture of the design of the study.  This investigation involved  showing 72 students a complex event for them to observe. simulated  of a  seconds.  One week after viewing the videotape, students were interviewed  alert.  which  in one of three "states":  lasted  approximately  90  videotape  by the experimenter  bank robbery,  The event, was a  hypnotized, relaxedj or  Half the students in each group used the reconstructive memory  technique; the remainder used imagery. attempted to mislead the students.  Five of the 47 interview questions  The remainder were more objective and  measured students' memory for details of both the videotaped bank robbery and the laboratory setting.  After the interview, witnesses were requested  to identify the main bank robber from a photo spread of eight coloured photographs and to rate their level of confidence in the accuracy of their choice. The  complexity  illustrated  by  of  the  the  numerous  previous researchers.  hypnotic  experimental  methodological  situation  problems  i s well  encountered  by  Because of this complexity, several other factors  had to be considered in the present study.  One of the most important of  these was the issue of the students' susceptibility to hypnosis.  In this  study, students were randomly assigned to the experimental and  control  groups, in accordance with proper experimental procedure (e.g. Campbell & Stanley, being  1966).  This meant that a l l subjects had  hypnotized;  otherwise,  the  situation  most  to be susceptible to certainly  would have  arisen whereby an individual resistant to hypnosis would be assigned to the hypnosis group. have failed,  Obviously, attempts to hypnotize these students would  making their  results  meaningless.  Thus, a l l prospective  34  subjects had f i r s t to be screened for their hypnotic susceptibility before they could observe the c r i t i c a l event.  The instrument used to do this was  the Creative Imagination Scale, constructed by Wilson & Barber This test  of hypnotic responsiveness was  (1978).  preferred over other existing  scales because i t f i t the constraints of the present situation.  The f i r s t  limitation had to do with the nature of the population from which subjects were drawn.  A l l subjects in this study were university students between  the ages of 18 and 25.  Other susceptibility scales (such as the Harvard  Group Scale of Hypnotic Susceptibility,  Shor & Orne, 1962;  the Barber  Suggestibility Scale, Barber, 1965a) are worded in a very authoritarian manner, in that experimenter's 1978). with  the test  control  items  imply  (cf. Barber  that  the  & Wilson,  These susceptibility scales were felt  young university  experimenter's  students who,  subject i s under the  1979;  Wilson  & Barber,  to be unsuitable for use  i t was  suggestions i f worded in such  feared, might  resist  the  authoritarian  tones.  By  contrast, the CIS consists of ten very permissively worded suggestions that  guide,  but  do  not  control  the  subject's thinking,  feeling  and  imagining along various themes, including arm heaviness, hand levitation, age regression, and so on. The second limitation posed by this particular experimental situation concerned time constraints. hypnotic  susceptibility,  Due to the need for pre-testing students for  subjects for this  study  introductory psychology classes, where testing occur in large groups. such  scales  as  the  were recruited  from  for suggestibility could  This requirement, however, precluded the use of Barber  Suggestibility  Scale,  since  it  can  be  35  administered to only one person at a time; the CIS, on the other hand, can be administered then,  entire  to either individuals or large groups.  classes of up. to  120 students  In this  were screened  study,  for  their  susceptibility to hypnosis using the CIS. Finally, the use of the CIS averted a problem posed by the other screening  instruments,  like  the  Harvard  Group  Scale  of  Hypnotic  Susceptibility, which require a trance induction to be administered subjects before giving them the test suggestions.  Obviously, for ethical  reasons, subjects about to be hypnotized must be fully procedure  and  participation. and  their  consent  must  be  to  obtained  informed prior  of the  to  their  But providing information about hypnosis to the students,  leading them through  a hypnotic  have  affected  induction, before measuring their  susceptibility  might  suggestibility  test.  Further,  the  susceptibility  scales  as  Harvard  the  performance inductions Group  on  the  subsequent  contained Scale  in  of  Hypnotic  Susceptibility take from 45 minutes to one hour to complete. avoids this situation because i t can be administered without  such  The  CIS  the prior  induction of a hypnotic state and requires a maximum of 30 minutes to complete.  Therefore, i t was considered to be an ideal scale with which to  measure the hypnotic susceptibility of students in this experiment. Another problem encountered by researchers of hypnosis i s the extreme difficulty  of ascertaining just how  hypnotized the subject actually i s .  Besides showing a general relaxation of facial muscles, the subject gives no other overt physical clues to indicate to the hypnotist that he or she has entered the hypnotic state nor how deep the trance i s . Investigators  36  have constructed a variety of measures that attempt to determine the depth of the hypnotic trance.  This i s necessary for... two reasons.  First and  most obvious, the researcher must establish that the induction has had the desired effect of leading the individual into a hypnotic state; otherwise, the effects of hypnosis are not being studied. describing  their  trances have alerted  Accounts of individuals  researchers to a second  concern.  This is that the depth of the trance can fluctuate during the hypnotic session (cf. Tart, 1979).  So not only do researchers need to establish at  the outset that the person i s , indeed, in a hypnotic state, but they also should determine periodically throughout  the session how  deep the trance  is. There are two general ways in which hypnotic depth can be measured. First, i t can be ascertained by the experimenter according to how well the subject responds to suggestions, such as arm  levitation or inability to  open  Group  the  eyes.  Items  on  the  Harvard  Scale  of  Hypnotic  Susceptibility, the Stanford Hypnotic Susceptibility Scale and the Barber Suggestibility Scale a l l challenge the subject in this manner. the  subject  is willing  to comply with  the  sensations offered by  hypnotist, the deeper the trance is assumed to be. measuring hypnotic  depth  throughout  the hypnotic  calibrated  scale.  requires the  subject  session, his or her  Numerous  developed for this purpose.  scales  of  The more  Another approach to  to estimate,  depth of  varying  the  at points  trance along  magnitudes  have  a  been  For example, subjects rate themselves from 0  to 100 on the LeCron Scale (LeCron,  1953), from 1 to 10 on the Harvard  Discrete Scale (O'Connell, 1964), and. from 0 to 4 on the Brief Stanford  37  Scale  (Hilgard & Tart, 1966).  While the self-report scales are  quick,  convenient and non-reactive in that they do not appear to alter the depth of  hypnosis  (cf.  verification behavioural  as  Tart,  1979),  they  are  the  objective  measures  are  too  are  not  tests.  open  On  authoritarian for  to  the  other  use  with  independent hand,  potentially  resistant subjects, such as university students, and their use may the  subject's  depth  of  trance.  Despite  their  the  affect  differences, a  high  correlation between behavioural and experential scores has been reported (Tellegen, 1979). The  present  study  used both objective and  ascertain hypnotic depth. the Hand Lock suggestion,  subjective measures to  One item from the Barber Suggestibility Scale, was  used as the objective measure.  In this  test, the experimenter suggests to the subject that his/her intertwined fingers are locked together like steel and that they won't come apart no matter how hard the subject attempts to unlock them. the test i f he or she cannot unlock effort. in  The subject passes  their fingers after 15 seconds of  This test of hypnotic depth was administered only to the students  the hypnosis  group to ascertain whether the hypnotic  induction had  succeeded in inducing a state of hypnosis in these subjects. The  subjective  measure  involved  having  the  students  provide  a  self-report of their "mental" state using a 36-point scale patterned after one used by the Vancouver Police Department.  This scale was chosen over  the  in  previously mentioned  closely  as  possible  the  self-report  scales  methods by  which  hypnosis would be interrogated by the police.  a  order  "real"  to  simulate  as  eyewitness under  Accordingly, the Vancouver  38  Police were consulted and their procedure adapted to f i t this experimental situation.  The self-report technique was used because, as Tart (1979) has  stressed, i t i s necessary to determine how subject has achieved.  deep a trance the hypnotized  In this study, then, the students provided a rating  of their mental state at the beginning of the interview and again half-way through.  If the individual's rating indicated that he or she was  very light trance, the experimenter subject  was  indicated by a rating between 1 and 12, a medium state between 13 and  2U,  a  into  very  deep  a  deeper  spent one or two minutes guiding the trance  and  back  in a  trance  state  between  25  of  hypnosis.  and  36.  A  light  Acceptable  suggested by the police, ranged between 18 and 26.  ratings,  as  Students in both the  relaxed and awake conditions were required to give these self-reports as well, for two reasons.  First, i t was  in keeping with the design of the  experiment, whereby the only thing allowed to vary between conditions was the hypnotic induction or relaxation exericses and the Hand Lock item from the  Barber  Suggestibility  methodological  control.  Scale.  Second,  So  this  procedure  i t also provided  provided  information  on  good how  successful the relaxation exercises had been in inducing a state of deep relaxation in the subjects in that group, and how wide awake subjects in the alert group were. Edmonston (1981) has asserted that no distinction can be made between physiological  and  behavioural  measures  of  relaxed  and  hypnotized  individuals, nor between their self-reports of their experiences.  If this  is true, then there should be l i t t l e difference between these subjects' scores on an inventory measuring typical hypnotic experiences.  Such an  39  inventory  has  been  constructed  by  Field  (1965).  It consists  of  38  statements describing hypnotic sensations (for example, "I felt apart from everything else") to which the subject responds These  38  items  on  Field's  measurements obtained  from  either True  inventory correlate students  using  the  at the Harvard  Hypnotic Susceptibility, Form A (Shor & Orne, 1965). of  this  inventory  identification  was  task.  completed  The  by  all  inventory was  .01  level  Group  with  Scale of  An adapted version  students  reduced  or False.  from  following  the  items  33  38  to  because five of them made direct references to hypnosis (for example, "I was  in a medium hypnotic state, but no deeper"), making their use in the  non-hypnosis groups unsuitable.  Thus, subjects in this  study answered  True or False to a total of 33 items of Field's Inventory of Hypnotic Depth (1965) which deals with feelings of unawareness, automaticity and compulsion. overall depth group.  Inclusion of this task provided another assessment of the of hypnosis  experienced  by  students  in the  experimental  Perhaps more important, i t also allowed a comparison between the  experiences of relaxed subjects to the hypnotized  ones.  If Edmonston  (1981) i s correct in assuming that relaxation i s a large part of hypnosis, then the relaxed and hypnotized subjects should have similar high scores on the inventory while the alert subjects should obtain low scores. Although the use of the videotaped bank robbery as the c r i t i c a l event for students to witness had several advantages, as noted earlier, i t also had a major drawback — mere 90 seconds. seemed  namely, i t was  too brief in duration, lasting a  Subjects observing the videotape during pilot  uncomfortable  that  the  initial  task  was  so  brief.  testing  This  was  40  reflected in such comments as "Is that a l l you response indicated the  need for a f i l l e r  want us  to do?"  task for subjects  This  to perform  after they had witnessed the bank robbery during the second part of the experiment. Battery  The  (Hakstian  activity, for two minutes  to  stretched  Associative Memory test of the Comprehensive Abilities & Cattell, reasons.  administer  the  1975)  of  filler  First, this test requires approximately ten  (including  duration  appeared to be an appropriate  the  a  short  second  part  practice of  session),  this  so  it  experiment; these  sessions lasted anywhere from 20 to 30 minutes, depending on the number of subjects in the group. memory  ability.  abilities, tests:  with  The  Second, the test provided a rough measurement of CAB  "general  in  general  memory  measures  capacity"  20  being  primary  determined  cognitive by  three  the Associative Memory test, the Meaningful Memory test, and  Span Memory test. these tests are  the  Hakstian & Cattell (1974) have demonstrated that a l l of good discriminators  of ability  to  commit material  to  memory. However, the test of memory span was considered inappropriate for the purposes of this study because i t measures digit span and i s presented auditorily.  The  other two tests differ "primarily in the meaningfulness  of the long-term association formed" (Hakstian & Cattell, 1978, The  p.  Associative Memory test involves the rote memorization of 14  663). design  and number pairings, while the Meaningful Memory test involves memorizing 20 noun-descriptor  pairs.  While the latter test, as i t s name suggests, is  more meaningful, i t is possible that the noun-descriptor like  information  in  Loftus'  post-event  questionnaire  pairs might act by  adding  extra  information and confusing subjects about the details of the bank robbery.  41  Therefore, the Associative Memory test seemed the best f i l l e r task for subjects to perform would probably task,  then,  in that i t provides a measure of memory ability but  not interfere with details of the c r i t i c a l provided  an  opportunity  to  determine  event.  This  i f there  is a  relationship between performance on the CAB Associative Memory test and accuracy of eyewitness reports. Despite experiment,  the  various  i t was  still  methodological possible  controls  to have  employed  factors  independent variables influence the results.  other  in  this  than the  Previous researchers have  identified several sources of artifact that can contaminate experimental results, such as the expectations of the experimenter the subject's awareness of the experimenter's  (Rosenthal, 1966),  intent (McGuire,  1969) and  the concern of the subject that he or she i s being evaluated (Rosenberg, 1969).  The hypnotic situation i s particularly  artifact (summarized in Barber,  1969).  sensitive to sources of  In an effort to identify some of  the factors that might influence a subject's performance in this study, a post-experimental completed  questionnaire  consisting  by a l l subjects as the last  of  task  seven  questions  was  of the experiment.  One  question asked the students to rate to what extent they felt pressured by the experimenter  into doing what was asked  during the interview.  Some  researchers have suggested that the behaviour attributed to the "state of hypnosis" Barber,  i s elicited 1969).  by the motivating  Others  pressures  disagree, contending  of the task  that the standard  (e.g.,  hypnotic  instructions are not nearly constraining as some "control" instructions, such as Barber's  (1969)  Task Motivational Instructions (e.g., Sheehan &  42  Dolby, 1974).  Responses to this question, then, would indicate the degree  to which subjects felt  pressured  to comply with  the experimenter and  whether hypnotized students felt more or less pressure than those in the control groups.  A second question asked i f the subject found i t d i f f i c u l t  to speak during the interview.  This was prompted by the experimenter's  personal experiences with hypnosis  as well as the reports of students  participating i n the pilot study, several of whom mentioned they found i t an effort to speak.  This question, then, addressed the possibility that  brief or incomplete  answers may be due to this difficulty  in speaking  rather  memory.  to be made  than  faulty  It also  allowed  a comparison  between the subjective experiences of hypnotized and relaxed students, as Edmonston  (1972;  1981) has suggested.  Two  questions  determined  subject's previous hypnosis and deep relaxation exercises. an  opportunity  hypnosis Another  to determine  whether  or deep relaxation affected question  asked  previous  experience  the  This provided with  either  performance during the interview.  subjects whether  they  had seen  the assistant  experimenter during the week between their observing the videotape she had showed them) and their interview with the experimenter. the assistant experimenter was an undergraduate  (which Because  student at U.B.C., as were  a l l the subjects, i t was quite possible that some subjects might see her during the intervening week. concerned  details  Since several of the interview questions  of her physical  appearance,  i t was  important  to  establish whether some subjects had had a second opportunity to notice her appearance. students  Finally,  had been  two questions  requested  addressed  the strategy  to remember the videotaped  by which  event.  One  43  required students to rate how successful they felt they had been in using the memory technique, while a second question asked for any comments on the memory strategy.  It was hoped that these questions would provide an  insight into whether subjects were actually able to utilize technique as instructed by the experimenter.  the memory  44  METHOD  Subjects: Seventy-two undergraduate  students were recruited as volunteers from  introductory psychology courses at the University, of British The experiment  was  Columbia.  described to potential subjects as one investigating  the relationship between memory and imagination. No reference was made to hypnosis  at  this  susceptibility  time.  in group  Students  were  first  sessions using the  (CIS) constructed by Wilson & Barber (1978). were screened using this procedure.  screened  Creative Imagination Scale Approximately 1,000 students  Only subjects scoring 21 or over  (approximately 50% of the population) were approached this study.  find  to participate in  Scores in this range indicate medium to high susceptibility  to hypnotic induction. to  for hypnotic  any  susceptibility  Since a number of previous researchers have failed  significant (e.g., Barber,  sex  differences  1969;  in  terms  Cooper & London,  of  hypnotic  1966; Kihlstrom,  Evans, Orne & Orne, 1980), no attempt was made to test equal numbers of males and females. the  study.  Altogether, 51 females and 21 males participated in  Immediately  students were informed experiment,  they may  following that  the  preliminary  screening  session,  i f they continued to participate  in this  be hypnotized at some later point.  Approximately  half of these students indicated after the screening session that they did not wish to continue with the experiment. Testing was completed on a total of 76 students; however, the results of  four  students were  subsequently  discarded from  data  analysis for  45  various reasons. carpet  In reply to a question concerning the colour of the  in the laboratory, one  colourblind.  subject indicated  that he was  green/red  Another student mentioned she arrived late for the videotape  session and so had not seen the whole of the bank robbery.  Finally, two  women indicated they had previously worked as bank tellers and had seen the videotaped bank robbery during their training.  Materials: Students observed a videotape of a simulated bank robbery, used by the Royal Bank for training new tellers.  A copy of the film was provided  by the Vancouver Police Department, who  produced  the Bank.  i t in conjunction with  The opening scene shows customers waiting in a crowded bank.  After a few seconds, viewers can see two men enter the bank. proceeds  to push his way  wicket, while entrance.  the  past the waiting customers  other, armed with a  sawed-off  One of them  to the  shotgun,  teller's  guards  the  The principal bank robber aims a revolver at the teller and  passes her a note demanding money. the two men flee the scene.  After the teller hands over the money,  The entire sequence lasts for approximately  90 seconds, with the principal bank robber visible for the majority of the time. A total  of 47  questions were asked  interview (see Appendix A). Vancouver  Police  eyewitnesses.  of the  students during the  They were patterned after a form used by the  Department  to  obtain  suspect  descriptions  from  Forty-two questions were objective; that i s , they were of  the form "What colour was his hair?  What was he wearing?"  However, five  46  of the questions were misleading in that they presupposed  the presence of  an attribute, such as a hat or a watch (e.g., "Did you notice i f his gloves were leather?").  Thirty  of the 47 questions pertained to the  videotaped bank robbery; the remaining questions concerned the laboratory setting and the physical appearance of the assistant experimenter. order in which these conditions videotape  so  questions were asked  that half  first  and  the  the  students  remainder  was  counter-balanced  were asked  were  first  for details questioned  The  across of  the  about  the  laboratory setting. A l l subjects completed a slightly modified version of Field's (1965) Inventory Scale of Hypnotic Depth (see Appendix B). statements  which the subject rated as True or False (e.g., "Time stood Of the 3 8 statements  still").  This consisted of 3 3  listed  by Field as correlating with the  Harvard Group Scale of Hypnotic Susceptibility, Form A (Shor & Orne, 1965) at the . 0 1 level, five were discarded because they were specific to the hypnotic situation and could not be used in the comparison groups (e.g., "During  the  final  'Countdown' to  wake me  up  I  became more  deeply  hypnotized for a moment"). The post-experimental questionnaire consisted of seven questions (see Appendix C). interview; experimenter  One asked i f subjects found i t d i f f i c u l t to speak during the  one  ascertained whether the  subject had  during the intervening week.  seen  the assistant  Another requested subjects to  rate the amount of pressure they felt during the experimental situation, while two determined  the subject's previous experience with hypnosis and  deep relaxation exercises, i f any.  Finally, one question measured  how  47  well the subject felt he or she had been able to use the memory technique. Subjects were asked photospread  consisting  to identify  of  eight  the principal bank robber  photographs  (see  Appendix  D).  from a These  pictures, taken with a Polaroid camera, showed only the head and shoulders of the individual. the bank robber  The seven men chosen to be the foils closely resembled in terms of age,  sex,  texture, facial features, and presence  race, hair  length, colour, and  of facial hair.  Each photograph  was assigned a number from 1 to 8 , which was recorded on the bottom of the picture.  Procedure: The experiment was  conducted  in three stages.  First, subjects were  screened for their hypnotic susceptibility using the CIS. required  approximately  30 minutes  to  administer  and  This procedure score.  Students  scored their responses to the ten test items of the CIS on an answer sheet distributed students  to them by  the experimenter.  were informed  experiment, they may  that i f they  At the end of this session,  continued  be hypnotized later on.  to participate  in this  Those who objected to this  were asked to indicate so on their answer sheets; the others wrote down a phone number where they  could be  contacted.  These sessions continued  until 76 medium to highly susceptibile subjects were' recruited  for the  experiment. Subject volunteers meeting the criterion score of 2 1 or more reported to the laboratory in small groups to observe the videotape. experimenter  briefly  described  the  purpose  of  the  The assistant experiment  as  48  investigating the relationship between memory and power of imagination. Subjects  were  asked  to pay  close  attention  to the videotaped  bank  robbery.  After watching the taped incident, the students participated i n  a f i l l e r activity consisting of the 14-item Associative Memory test of the Cognitive Abilities Battery (Hakstian & Cattell, 1975). One week later, subjects were conducted they were interviewed individually  to a different room where  by the experimenter.  Subjects were  randomly assigned to conditions, with the restriction that the mean CIS scores for the six groups were approximately  equal.  Mean group scores  ranged from 26.750 to 27.000 out of a possible score of 40. A l l students reclined  in a comfortable  chair.  They were informed  that the purpose of the interview was to obtain information about the events of the previous week.  Subjects about to be hypnotized  information sheet on the effects of hypnosis.  read an  A l l students were then  instructed on how to provide self-report ratings of their mental state using a procedure developed by the Vancouver Police Department.  As used  by the police, this procedure requires the individual to rate his or her depth  of hypnosis  on a 36 point scale.  One to 12 indicates a light  hypnotic state, 13 to 24 a medium trance and 25 to 36 a deep state of hypnosis.  The hypnotized subjects in this study used this procedure to  rate their state; however, the relaxed and waking subjects used a somewhat altered scale, in that the end points were re-named in keeping with their situation.  Relaxed students rated their state of relaxation in the same  manner as the hypnotized students, with the words "state of relaxation" substituted for "state of hypnosis".  However, for subjects questioned in  49  an alert state, 1 to 12 represented feeling wide awake, 13 to 24 slightly sleepy and 25 to 36 intense feelings of sleepiness and drowsiness. Instructions  to  the  students  then  varied,  depending  upon group  membership: Hypnosis:  Subjects were hypnotized  induction requiring approximately  by the experimenter using a standard  15 minutes.  feelings of relaxation and heaviness, summary of suggestions  contained  This induction emphasizes  and encourages eye closure (for a  in this  induction, see  Subjects were then asked to rate their depth of hypnosis.  Appendix E).  Following this,  subjects participated in one item of the Barber Suggestibility Scale (see Appendix F).  A l l subjects successfully completed this item.  Relaxation:  Subjects  listened  to 20 minutes of progressive relaxation  exercises provided on a tape made by Dr. Bryan Hiebert. require subjects to alternately tense and throughout the body, beginning feet  (for a partial  with  transcript,  These exercises  relax different muscle groups  the right hand and ending with the  see  Appendix G).  This  tape makes no  reference to the possibility of entering a state of hypnosis or sleep. Waking:  These subjects received no extra instructions.  Before  questioning  the  students,  the  experimenter  instructions regarding which recall strategy they should interview.  Students using  the  reconstructive technique  spend a few minutes recalling the circumstances the previous week, for instance, how wearing, etc.  (see  Appendix  H).  The  surrounding  use  provided during  the  were asked to the events of  they were feeling, what they were imagery  technique  required  subjects imagine the laboratory incident as i f i t were on videotape  that and  50  could be "played" back on a giant television screen inside their heads. It was suggested that the subject would be able to freeze the videotape, enlarging any area of the screen so as to focus on small details that might otherwise  escape detection (see Appendix H).  When the students  indicated they understood these instructions, they were requested to rate their mental state using the self-report technique outlined above. Following events  this,  the experimenter  of the previous  questions.  week  by  obtained  asking  information about the  subjects  the 47 interview  Half way through the interview, a l l students were requested  to, again, give a self-rating of their state. in writing by the Experimenter.  After  A l l answers were recorded  the interview, subjects in the  hypnosis condition were awakened by the experimenter. Finally,  the students  were asked  to identify  the principal bank  robber i f they saw his picture among the photospread of eight mugshots; i f not, they were to let the experimenter  know he wasn't there.  Students  indicated their degree of confidence in the accuracy of their choice on a 7 point scale (1 = not very confident, 7 = very confident). Subjects then completed  the altered  version of Field's  (1965) Inventory  of Hypnotic  Depth and the post-experimental questionnaire, and were de-briefed.  51  RESULTS  During the final phase of this study, information was obtained from each  subject  experiment.  regarding  the  three  main  issues  investigated  by  this  The f i r s t concerned the efficacy of hypnosis as an aid to  eyewitness memory. Responses to the interview questions served as the measure  of  memory  for  the  critical  event.  A  combination  of  both  structured and open-ended questions was used to e l i c i t details from the students. for  One person was hired to score a l l the interviews.  reliability,  interviews.  a  second  Interrater  individual  reliability  was  consistency between independent judges.  randomly .95,  scored  showing  As a check  some  of  the  a high degree of  The responses to the 42 objective  questions were analyzed to reveal the total number of details reported and how  many  of  these  statements  were  incorrect.  Answers  to  the  five  misleading questions were scored differently, and are reported separately below. Student  responses  fell  into  details or sequential statements.  one  of  two  categories:  descriptive  The former refer to objects and their  attributes (such as weapons or notes), descriptions of people, clothing, etc.,  while the latter involve the events of the videotape, what happened  and in what sequence.  A similar procedure for scoring such subjective  data has been employed by Yuille (1983). scores  (total number of details  and  For each participant, then, two  number of errors) as well as an  accuracy rating in terms of percentage incorrect were obtained on both descriptive  and  sequential  dimensions  for a l l responses  to  the  42  52  interview questions. combined with  Responses relating  those referring  to the videotaped  to the "live" event  measure of memory performance.  event  were  to yield an overall  The means and standard  deviations for  these recall scores are presented in Table 1. Analysis of variance tests performed on each of the three sets of scores produced no significant main effects  for interview method  or memory technique,  interaction between these two variables.  nor was there any  That i s , hypnotized, relaxed and  waking students recalled the same amount of information and made the same number of errors, for both descriptive and sequential aspects of the video and  "live"  event.  technique.  Similarly,  This i s surprising  there  was no main  effect  for memory  since the literature would lead  one to  predict that imagery would produce better recall than the reconstructive method. What does emerge from these scores is the finding that memory for both  the videotape  accuracy.  and "live"  event  was fairly  good, averaging 7^.3%  When the misleading questions were included, the accuracy rate  dropped slightly to 72.8$.  This rate i s consistent with that reported by  Yuille (1983) and, as such, i s far higher than those reported previously in the literature (for example, Buckhout, 1974; Clifford, 1979). Although from  no significant  main effects or interaction  the analyses of the total recall  warranted.  were obtained  scores, further analysis seemed  For example, one of the reasons for including questions about  the assistant experimenter  and the laboratory room in this study was to  allow a comparison of memory for the details of the videotape with those pertaining to the "live" event.  This comparison would determine whether  the pattern of recall differed depending on the type of detail remembered.  53  Table 1  Means & Standard Deviations For Total Recall Scores For Both Video & "Live" Events Total Number Details Reported  Total Number Errors  Percentage Errors  Recon  Recon  Imagery  Recon  Imagery  Imagery  Hypnosis Mean  47.500  48.083  11.583  13.583  24.507  28.742  S.D.  8.733  5.213  3.450  2.678  6.474  7.241  Group Mean  47.792  12.583  26.625  Relaxation Mean  48.583  52.417  13.583  12.667  28.360  23.565  S.D.  8.565  7.960  4.889  4.519  10.583  7.124  Group Mean  51.000  13.125  25.963  Waking Mean  48.417  50.917  12.250  11.667  25.762  23.017  S.D.  8.240  6.908  2.768  3.025  6.220  5.113  Group Mean  49.667  11.959  24.390  54  Accordingly,  the  responses  were  concerning videotape details and  divided  into  two  categories:  the other regarding the "live"  one  event.  Results of this comparison are presented below. Comparison of Memory for Video Versus "Live" Details More questions were asked assistant  experimenter  possible comparison  and  concerning the videotape  laboratory  setting.  Therefore,  x 2 x 2 analysis of variance test was performed total  details  the  only  that could be made to test for differences between the  two was proportion of errors to total details reported.  to  than about the  recalled.  No  significant  Accordingly, a 3  on the ratio of mistakes  main effects  for interview  method or memory technique were noted; however, the three-way interaction was significant, F(2,66) = 6.955, p = .002 (see Appendix J, Table A simple error  effects  proportions  test  was  for video  conducted  and  "live"  comparing details.  comparisons showed a significant difference, F(1,66)  1).  the within-subject Only =  one  6.434,  of  these  p C .05.  Relaxed subjects, using the reconstructive memory technique, made a higher proportion of errors when recalling details about the videotape than they did when reporting details about the laboratory setting (X = 31-404 versus X = 22.418). trend  Although other tests did not reach significance, a similar  emerged  technique.  for waking  and  hypnotized  students  using  the  imagery  That i s , these students showed a tendency to make more errors  proportionally on videotape details versus "live" details. the  possibility  that  This pattern  of memory performance  raises  there are  differences in the way  people recall events they see on a videotape and  the manner in which they remember details about their surroundings.  distinct  55  Because such differences emerged in the pattern of recall of video versus "live" events, further analysis of the memory scores within each category of detail type  seemed appropriate.  Results of these tests are  presented below, f i r s t for the video event, then for the "live" event. Recall of Video Event The means and standard deviations for overall performance for recall of the videotaped event are presented in Table 2.  It can be seen from  this table that the hypnotized students did not report a greater number of details (both correct and incorrect) than either their relaxed or waking counterparts. of recall.  Nor did the two memory techniques produce different levels  However, consistent with the results reported earlier, there  was an interaction between interview method and memory instruction in the total number of errors contained in the students' responses, F(2,66) = 4.507, p = .015 made  more  (see Appendix J, Table 2).  errors  instructions.  using  imagery  The reverse was  than  The hypnotized participants the  reconstructive  true for the relaxation group.  memory  That i s ,  relaxed students using reconstruction made more errors than those using imagery. the  Students questioned in the waking state differed very l i t t l e in  number  of  reconstruction.  errors This  they same  made,  pattern  whether of  they  interaction  used  imagery  emerged  when  or the  proportion of errors to total recall was calculated, F(2,66) = 5.779, p = .005  (see  Appendix  J, Table  3).  The  fact  that hypnosis  yields  less  accurate recall in conjunction with imagery i s in direct opposition to the claims of many hypno-investigators. In order to examine the locus of this interaction, the memory scores  56  Table 2  Means & Standard Deviations For Recall Scores For The Video Event  Total Number Details Reported  Total Number Errors  Percentage Errors  Recon  Recon  Imagery  Imagery  Recon  Imagery  30.917  30.750  7.083  9.333  23.209  30.845  5.282  2.989  2.610  2.741  8.520  10.056  32.167  34.917  10.000  7.667  31.404  22.009  5.781  5.807  3.885  2.902  11.943  7.740  31.500  32.833  7.583  8.167  24.872  25.343  5.300  5.289  .793  2.125  5.523  6.983  57  reported  in Table  2 were separated  into  the following six categories  comprising the events of the videotape: 1)  descriptive details of the main thief;  2) sequential details of the main thief; 3) descriptive details of the accomplice; 4) sequential details of the accomplice; 5) descriptive details of the teller; and 6)  descriptive details of the bank setting.  For each of these categories, analysis of variance tests were conducted on the  total  responses,  details  reported, the number  and finally  of errors  the proportion of errors  contained  to the total  in these details  given. In total, then, eighteen separate analysis of variance tests were conducted number  on the component scores.  of details  reported  No differences were noted in total  in any of the groups.  However, the same  pattern of interaction between memory technique and interview method noted earlier  was  obtained  for the number  of errors  contained  in the  descriptions of the main thief, F(2,66) = 3-542, p = .035 (see Appendix J, Table 4). That than  those  i s , hypnotized students using imagery made more errors  using the reconstructive memory technique,  whereas relaxed  students showed the reverse pattern. Since the total amount of information reported by the students was equal in a l l groups but differences appeared  in the number of errors  contained in those statements, i t i s not surprising that the 3arae pattern of interaction was reflected in various error proportions. for:  These occurred  58  1)  descriptions of the principal thief, F(2,66) = 5.748, p = .003 (see Appendix J, Table 5); and  2)  descriptions of the bank, F(2,66) = 3-374, p = .040 (see Appendix J, Table 6).  While  i t appears  responses reports  that the pattern of correct recall was  consistent in  to many of the interview questions, i t failed  to emerge in  of the  sequence of events  for either  the main thief  or his  accomplice, or in descriptions of the teller. Besides  the  interactive  effects  of  interview method  and  memory  technique, one further difference in recall was noted for details of the video event.  There was  a tendency  to make more errors when describing  peripheral details, such as the bank, than when recalling more salient items.  Examples are presented in Table 3.  Compare the average error rate  of 29.174$ for responses concerning the principal bank robber who was the main focus of attention to that of 44.329% for the bank, a more peripheral aspect of the videotape.  Clearly, the more central the detail, the less  likely i t w i l l be recalled incorrectly. Analysis of the responses to the interview questions concerning the telephone  number posted  observation. correctly.  Not  one  on of  the video machine i s consistent with the  72  students  recalled  the  seven  this  digits  In fact, the majority of participants did not even remember  seeing the number, although i t was in plain view to everyone and explicit reference to i t had been made by the experimenter.  Further, of the 26  students who did acknowledge i t s presence but couldn't recall the digits, only 7 of these were hypnotized, compared  to 8 relaxed and  11 waking  59  Table 3  Percentage Of Errors Reported About Three Aspects Of The Video Event  Description of Main Thief  Description of Teller  Description of Bank  Recon  Recon  Imagery  Recon  Imagery  Imagery  Hypnosis  24.280  35.280  40.774  47.173  30.556  57.639  Relaxation  37.096  24.941  36.389  32.956  62.500  39.583  Waking  26.622  26.827  31.944  33.333  34.028  41.667  Mean  29.174  37.095  44.329  60  participants. often  Far  claimed,  recalled  This  this  piece  and  techniques. eyewitness  More  it  of  peripheral  combine  the  details recall  reconstructive  technique.  improves  at  Recall  of  The on  "Live"  event. the  of  "live"  were  no  error  in  number  that  hypnosis  had  detail,  of no  as  students  effect  at  is who  a l l  on  of  the  entirely when  event  different  hypnosis'  crime,  not  video  the  is  as  there  no  to  the  that memory  question  technique  the  is  with  used  memory  imagery,  ways  suggests  that  an  elicits  police  use,  but  the  evidence  that  hypnosis  effects event  the  were  in  total  however,  number  interview not  deviations  presented in  of  Table  amount neither  of  errors  method  observed  and  for  memory  memory  of  for  overall  memory  4.  As  the  video  information,  both  of were  with  there  contained  any  in  the  for  "live"  details event,  reports  of  there  correct  differences  their  technique  and  among  the  or  the  in  ratio. As  event  are  reported;  groups  fewest  However,  standard  differences  incorrect, six  in  video  and  event  the  recall  a  peripheral  Events  the  Means  is  this  information.  in  of  for  a l l .  differences  recall  for  specifically,  accurate  memory  appears  results  hypnosis  about  most  of  memory  produced  Thus,  pattern  relaxation  enhancing  hypnosis  anything.  retrieving  the  from  with  memory  variance  of  score  were  tests  description laboratory  recall  of room  the  video  event,  analyzed  the  separately.  were  conducted  on  the  assistant  experimenter  where  the  components  two  videotape  A  aspects  had  total of  and been  the the  of of  the  total  six  "live"  analysis event:  description  shown.  "live"  Only  of one  of the the test  61  Table 4 Means & Standard Deviations For Recall Scores For The "Live" Event  Total Number Details Reported Recon  Imagery  Total Number Errors Recon  Imagery  Percentage Errors Recon  Imagery  Hypnosis Mean  16.583  17.333  4.500  4.250  26.904  25.477  S.D.  3.801  3-725  1.567  1.215  6.359  8.618  Mean  16.417  18.500  3.583  5.000  22.418  27.075  S.D.  3.528  3.233  1.505  2.486  9.694  12.910  Mean  16.917  18.083  4.667  3-500  27.246  19.124  S.D.  3.175  3.059  2.640  1.679  14.541  7.845  Relaxation  Waking  62  produced a significant interaction, and that occurred for the total number of errors made by students in their descriptions of the laboratory room, F(2,66) = 3.703, p = .030  (see Appendix J, Table 7).  Here, subjects  questioned in the waking state made the least number of errors (X= 1.417) when using imagery, while participants using reconstruction made the most mistakes versus  (X = 2.417).  Relaxed students showed the opposite trend (2.250  1.833), while  hypnotized  participants  mistakes, regardless of memory technique. results  than  that obtained  made the  same number of  This is a different pattern of  for the videotape  details,  and  raises  the  question of the extent to which results from simulated events, such as the videotape of the bank robbery, can be generalized to recall of  "live"  events. A  summary  findings.  of  First,  the  results  hypnosis  did  to  this  not  aid  point  indicates  students  in  the  three  amount  information they recalled about either the video or "live" event. there is a consistent interaction  main of  Second,  between interview method and memory  instruction that affects both number and proportion of errors contained in witnesses'  reports  of  hypnosis  in  produces  more  reversed  when imagery  the  conjunction accurate  simulated  with recall  is used  the  bank  robbery.  It appears  reconstructive memory  than  relaxation,  but  as a memory technique.  this  that  instructions pattern i s  In this  case,  relaxation yields the most accurate memory. This finding has implications for police investigators who routinely combine hypnosis and imagery during their  interrogation  of  eyewitnesses.  Finally,  striking  emerged in the pattern of recall of video versus "live" events.  differences  63  Misleading Questions A second  issue addressed  by this study was the question of whether  hypnotized students would be more susceptible to misleading questions than control posed  subjects. to  each  Recall  student  that embedded  were  five  in the M7  that were  participant  into giving the wrong answer.  differently  than those given  interview questions  designed  to mislead  the  These responses were treated  to the objective questions, because some  students were either hesitant or unwilling to give a simple "yes" or "no" answer, even after continued probing by the experimenter. responses  to these  questions were coded as  "incorrect",  Consequently, "correct", or  "didn't know", the latter two categories indicating that the student had resisted the misinformation in the leading questions.  Table 5 shows the  overall pattern of responses to the five misleading questions. this  table  are  frequency  scores  for  responses  to  Entries in  a l l five  of  the  misleading questions; that i s , each group of 12 students had a total of 60 responses distributed among the three categories. performed  on these data was  not significant  A chi-square analysis  C^OO)  = 9.516, p >.05),  indicating there were no differences in degree of susceptibility between hypnotized,  relaxed or alert  students  in responding  to the misleading  questions.  This finding i s not consistent with those reported by Putnam  (1979), Wagstaff (1982) and Zelig & Beidleman (1981). On the other hand, in accordance  with Sheehan & Tilden (1983), a l l  students in this study were prone to making more errors in answering misleading questions than the neutral ones.  the  When exposed to misleading  information, the error ratios ranged from a low of 32% to a high of 56%,  Table 5 Pattern of Responses to Misleading Questions By Interview Method and Memory Technique  Memory Technique Recon.  Imagery  Total  Correct  30  24  54  Incorrect  22  31  53  Didn't Know  _8  _5  60  60  120  Correct  24  29  53  Incorrect  27  24  51  Didn't Know  _9  _7  _6  60  60  120  Correct  34  26  60  Incorrect  16  23  39  Didn't Know  IP.  11  UL  60  60  120  Hypnosis  J3  Relaxation  Waking  65  averaging 46$ for a l l groups.  This i s almost  twice that reported for  recall of both the video and "live" events elicited by the more objective questions. Photo Identification The ability of students to correctly choose the main bank robber's picture constituted the third aim of this study, and was tested by the identification  task at the end of the interview.  Table  6 l i s t s the  pattern of photograph choices by interview method and memory technique. Of  the 72 students  indicated  that  photospread.  they  completing  this  thought  task, a total  the target  picture  of 30, or 41.67$, was  not  in the  This ratio compares favourably with that reported by Wells,  Lindsay & Ferguson  (1979) who found that only 21$ of their students made  no identification  given  similar  instructions  (see also Yuille,  1983).  Although  the bank robber's picture was present, the high percentage of  students  exhibiting  considered  such  caution  to be a positive  in identifying  trend  in light  the bank  of findings  robber was by previous  researchers. As can be seen  from Table 6, 12 of the remaining 42 students, or  28.57$, chose the correct  photograph.  This figure i s well above that  expected by chance (12.5$) for an 8-item photospread.  It could be argued  that the functional size of the photospread was only 7, since no one chose photograph number 1 (see Wells, picking any one picture to 14.3$. chose the correct chance.  1978).  This increases the chances of  Even so, the students in this study  photograph at twice as high a rate as expected by  From the table i t can be seen that equal numbers of hypnotized,  Table 6  P a t t e r n o f C h o i c e s From Photospread By I n t e r v i e w Method and Memory Technique  Correct  Incorrect  Not There  Hypnosis Reconstruction Imagery  2  6  4  J?  __6  __4  4  12  8  2  4  6  _2  _3  _7  4  7  13  3  4  5  Relaxation Reconstruction Imagery  Waking Reconstruction Imagery  Grand  Totals  JL  JL  Ji  4  11  9  12  30  30  67  relaxed and waking students made the correct photo identification. it  Thus,  appears that recalling a person's appearance under hypnosis does not  have any effect on subsequent identification. In order to determine and  the relationship between level of confidence  accuracy of photo identification,  students were asked  to rate  how  confident they felt in the accuracy of their choice on a 7 point scale, s  with 1 indicating not very confident and 7 representing very confident. These subjective ratings ranged from a low of 1.0 to a high of 6.0, with a mean of 3.75.  Using data from 71 students (1 subject was dropped  due to  missing data), the correlation between degree of confidence and accuracy of  identification  significant.  was  This  very  low,  is consistent  r(7D with  =  .162,  some of  p > .10,  the  and  previous  non  findings  reported in the literature (for a review, see Deffenbacher, 1980). However, coefficient  there  is  a  complication  between confidence and  in  computing  correct identification  a  correlation  rate  in this  study because students had three ways of responding to the task: could  chose  the  "correct"  picture;  they  could  pick  an  they  "incorrect"  photograph; or they could abstain from choosing any picture by saying "not there".  The last two responses, technically speaking, are incorrect since  the picture of the main bank robber was present in the photospread.  The  above non-significant correlation of .162 was derived by using just such a classification procedure. "identifying" the wrong man  But i t could be argued that the consequences of from a photospread are far more serious than  those arising from not choosing anyone, and so perhaps these two responses should  not be  treated  as  similar.  If those  students responding  "not  68  there" are excluded  from the analysis, the correlation coefficient is of  moderate strength and students who  is significant, r(42) = .310,  p < .05.  Thus, for  choose a picture from the photospread, the more confident  ones tend to have made the correct choice. the growing body of evidence  This finding adds weight to  showing that there i s , indeed, a meaningful  relationship between confidence and identification accuracy. One was  further correlation coefficient between accuracy  computed to see i f those students who  and  confidence  chose the wrong picture were i  more confident than those who  refused  to make an identification.  This  correlation was moderately strong and significant, r(59) = .632,  p<.01,  but  reported  the  above.  relationship was That  i s , the  in the  more confident  choosing an incorrect photograph. those who  opposite  direction  to  that  participants tended  to  be  those  That the less confident students were  did not make any identification is not particularly surprising  when one considers that the "not there" response, by i t s nature, provided those who were particularly unsure with the opportunity to avoid making an error. What emerges from these results is the finding that individuals who identify someone from a photospread, either correctly or incorrectly, are fairly confident that their choice is accurate, whereas those refusing to make an identification are less sure that they are correct.  Further, i f  the opportunity to abstain from choosing a picture is explicitly given in the instructions, a large number of students will exercise caution in this task.  69  Several other post-hoc analyses were performed collected from the students. Moment Correlation  on the information  In particular, a number of Pearson Product  coefficients  were computed  to test  for significant  relationships between various components of memory in this study.  The  results of these tests are presented below. Correlation Analysis The test of Associative Memory, measured by the MA portion of the CAB,  provided  an  opportunity  to  assess  on the standard memory t e 3 t  performance  the  relationship  and ability  between  to give accurate  eyewitness descriptions or to correctly identify the principal bank robber from the photospread.  A total of 33 Pearson Product Moment Correlation  coefficients were computed between a subject's score on the CAB and the various  components  interview  questions.  computations, chance. the  of memory  for the c r i t i c a l  It should  be  noted  event  that,  elicited  of this  by the  number of  two correlations are expected to be significant purely by  In fact, three of these did reach significance.  CAB had a small,  measures of memory:  but significant,  relationship  total number of errors recalled  A high score on to the following about  the " l i v e "  event, r(72) = .2064, p = .041; error ratio in statements made concerning appearance of assistant experimenter, r (72) = .2227, p = .030; and error ratio in a l l statements made concerning the "live" event, r (72) = .2267, p = .028.  These associations  expect that good performance  are counter-intuitive,  since  one would  on one measure of memory ability, the CAB  test, would be mirrored by good performance on the other measure, in this case  a low error  ratio  in recalling  details.  That  the direction of  70  relationship is opposite to expectations is puzzling. If i t i s not due to chance, then i t suggests that the standardized test is measuring an aspect of  memory which  is unrelated to the memory process  eyewitness situations.  that operates in  Nor was the CAB score associated with performance  on the identification task.  Two correlation coefficients were calculated,  one using a l l 72 students and the other excluding those who said "not there".  Both of these coefficients were very small, r (72) = .021, and r  (42) = -.011, respectively.  Thus, there i s no meaningful  relationship  between performance on the CAB and ability to perform as an eyewitness. A similar number of Pearson was calculated  to determine  Product Moment Correlation coefficients  whether high  susceptibility to hypnosis,  measured by the CIS, was significantly correlated with eyewitness memory or identification.  As with the CAB, most correlation coefficients were  very small and failed to reach significance.  The only one to do so was  for answers of "don't know" to the misleading question, r (72) = -.3633, p=.002.  Why  high  susceptibility  to hypnosis  i s associated  with  a  disinclination to answer "dont know" to misleading questions i s unclear and i s most likely due to chance. What did emerge from the correlation analyses was the finding that confidence  on the identification task was related  c r i t i c a l event.  to memory for  the  These correlation coefficients are presented in Table 7.  Only the recall scores of those students who made a choice, either correct or incorrect, from the photospread were used in these calculations; those who  said  "not there"  (n =  30) were  excluded  from  the analyses.  Examination of these findings yields some interesting results, in that the  71  TABLE 7 Pearson Product Moment Correlation Coefficients Between Confidence on Photo Identification Task and Recall Scores*  Descriptive Details Main Thief  P  r  Sequential Details Main Thief  P  r  Descriptive Details Accomplice  P  Sequential Details Accomplice  P  Descriptive Details Teller  P  Descriptive Details Bank  P  Des. & Seq. Details Total Video  r r r r r P  Descriptive Details Experimenter  P  Descriptive Details Laboratory Room  P  Des. & Seq. Details "Live" Event  P  Des. & Seq. Details Video & "Live"  r r r r P  Total No. of Details  Total No. of Errors  .3870 (.006)  .0423 (.396)  -.0639 (.346)  .2211 (.082)  -.1466 (.180)  -.1928 (.114)  .2054 (.099)  .0112 (.472)  -.1064 (.254)  .0801 (.309)  -.0383 (.406)  -.1596 (.159)  .2624 (.049)  .2644 (.047)  .1837 (.125)  .1953 (.111)  -.1058 (.255)  -.1061 (.255)  .4137 (.004)  .0730 (.325)  -.1668 (.149)  .3224 (.020)  .0936 (.280)  .0328 (.419)  .4024 (.005)  .1175 (.232)  -.0561 (.364)  .4296 (.003)  .1258 (.217)  -.0414 (.398)  .4781 (.001)  .1222 (.223)  -.1619 (.156)  Percentage Errors  •Includes only those students who made a choice from photospread (n=42)  72  more confident students tended to report more details.  Total recall of  both videotape and "live" details, excluding the misleading questions, was significantly correlated with confidence in the accuracy of choice from the photospread, r(42) = .4781, p = .001.  This association was reflected  in both the video (r(42) = .4137) and "live" (r(42) = .4296) aspects of the c r i t i c a l event, and in the various components of these recall scores (see Table 7).  However, this association did not hold for the number of  errors committed, or in error ratios. errors correlated event.  Thus, confidence and proportion of  -.1668 for video details, and  It appears that students who  -.0414 for the  "live"  provide more information about an  incident, the details of which may not necessarily be correct, are likely to be more confident about their picture choice on the identification task. Hypnotic Depth Hypnotic depth was  assessed by both objective and subject methods.  Students in the hypnotic condition were challenged with the Hand Lock item from  the Barber  Suggestibility  Scale.  Since a l l students passed  this  test, i t was assumed that they were successfully hypnotized. A  second  measure  of  trance  depth  was  obtained  participants complete an amended version of Field's  by  having a l l  (1965) Inventory of  Hypnotic Depth at the end of the interview. Means and standard deviations of scores obtained on the 33-item inventory are presented in Table 8.  As  can be seen from this table, the average score for hypnotized students was marginally versus  higher  than  that  obtained  by  18.250), whereas the mean score  (13.875).  relaxed  participants  for waking students was  (18.625 lower  An analysis of variance test revealed a significant main effect  Table 8  Means and Standard Deviations for Amended Version of Field's Inventory of Hypnotic Depth By Interview Method & Memory Technique  Memory Technique  Hypnosis  Relaxation  Waking  Reconstruction  Imagery  Average  M  16.333  20.917  18.625  S  4.942  4.055  4.421  M  17.888  18.667  18.250  S  3.996  4.830  4.336  M  13.000  14.750  13.875  S  6.135  6.298  6.080  74  for method of interview, 8).  Although  F(2,66)  =  p<  6.377,  (see Appendix J, Table  .01  there was a significant main effect, a post-hoc  comparison  of means by the Tukey method indicated that the differences in scores depended  on memory  hypnotized  technique  as well  as interview  method.  First,  students using imagery and a l l relaxed subjects scored higher  than waking students using either memory technique.  This i s the expected  finding i f the inventory i s , indeed, measuring depth of hypnosis. hypnotized  students using imagery scored higher than hypnotized subjects  using reconstruction. of  hypnosis  differences support  Second,  Why the memory technique appeared to affect depth  i s not immediately arising  between  to Edmonton's  apparent.  relaxed  (1981)  However,  and hypnotized  contention  that  the lack students  of any  does add  relaxation i s a large  component of the hypnotic state.' As  for providing an assessment  experienced  by students  of the actual depth  in the experimental  of hypnosis  group, the mean of  18.625  obtained for a l l hypnotized participants in this study i s not as high as one might have expected  on a 33-item questionnaire.  his subjects obtained an average score of of 9 . 0 5 .  14.54,  Field reported that  with a standard deviation  However, because the inventory administered  in this experiment  consisted of fewer items, only 3 3 , comparison of scores obtained on these two scales is d i f f i c u l t . Twice during the interview a l l participants were requested their "mental" state using a 36-point scale. in Table 9 .  to rate  These ratings are reported  The two self-report scores were averaged to obtain one score  for each participant. For the hypnotized students, this rating provided a  Table 9  Average of Two Self-Report Ratings By Interview Method & Memory Technique  Memory Technique Reconstruction  Imagery  Average  Hypnosis (n = 24)  19-21  20-96  20.08  Relaxation (n = 24)  22.75  22.71  22.78  Waking (n = 24)  15.08  13.64  14.39  76  subjective measure of hypnotic depth. obtained  a  hypnosis.  score  of  20.08,  As a group, hypnotized  which  represented  a  moderate  students state  of  Of these 24 students, 6 scores indicated the subject was in a  light trance; that i s , they reported an average rating of between 10 and 16, which was below the recommended level of 18 suggested by the Vancouver police.  Examination  of the responses  of these students indicated that  their recall of the c r i t i c a l event did not differ from other hypnotized students who  had given higher ratings; in fact, their scores for total  recall of the event tended  to be higher than the group mean.  these students prone to making more errors —  Nor were  i f anything, they made fewer  errors than those students with higher ratings of hypnotic depth. these  six students  self-report  minimized  the  depth  ratings did not predict  to any  amended inventory of hypnotic depth. obtained a score of 24  on  the  of  their  trance.  Perhaps  Their  low  degree a low score on the  In fact, two of these six subjects  inventory, which i s fairly  high.  The  relaxed participants reported a mean state score of 22.73, suggesting that the relaxation exercises had been fairly successful in inducing a state of deep relaxation in these students.  As expected from the labelling of the  end-point of the scale, students in the alert group rated themselves  lower  than hypnotized or relaxed participants.  The average rating of 14.39 for  this  experiencing  group  drowsiness.  indicates  that  they  were  some  feelings  of  It is d i f f i c u l t to make meaningful comparisons between these  group self-report ratings, because each scale was different  state in each group.  The  determine  whether the hypnotic induction and  designed  to measure a  purpose of the self-reports was to relaxation exercises were  77  successful in producing the intended effects, and these results indicate that they were fairly successful. Post-Experimental Questionnaire Responses to the post-experimental questionnaire were examined in an attempt  to identify  some of the factors that might have influenced an  individual's performance in this study, especially in the hypnosis groups. They  are presented  in the order  in which  they  occurred  on the  questionnaire. Question 1: the  student  representing students  Answers to this question were coded as "yes", indicating that had experienced little  some  or no effort  had greater difficulty  difficulty  in speaking. as a group  in speaking,  or "no",  As expected, hypnotized in speaking  during the  interview than waking students ( 9 ^ (5) = 16.936, p < .005).  Curiously,  only those hypnotized students using the reconstructive memory technique showed this tendency; those using imagery were evenly split between the two  responses.  There  technique should affect  i s no suitable ease  explanation for why the memory  of speaking.  While hypnotized and waking  students did differ in their ratings of the amount of effort required to speak, there was no significant correlation between amount of information recalled and effort in speaking (r(72) = -.088, p>.05). that  brief  or incomplete  answers cannot  Thus, i t appears  be attributed  to a person's  difficulty or hesitancy in speaking. Question  2:  experimenter  Seven students thought at some  time  during  they may have seen  the assistant  the intervening week.  They  were  distributed fairly evenly among the six groups, and examination of their  78  descriptions  of  the  assistant  experimenter  did  not  reval  any  major  advantage in amount recalled or errors committed despite their having seen the assistant on a second occasion. Question 3' attempted had  Four students indicated  they had  to be hypnotized, previously;  been hypnotized, or had  however, none of these students  been assigned to the hypnosis group.  A l l of the students in the  hypnosis condition, then, were similar in having had no prior experience with hypnosis. Question 4:  Over half of the students in this study (n = 41) reported  having previous experience with either deep relaxation exercises or  TM.  These were distributed fairly evenly among the three different interview conditions, with both hypnosis and waking groups consisting of 15 students each who had done deep relaxation exercises or TM before, and the relaxed condition consisting of 11 such subjects. Question  5:  experimenter  The  extent to which each  during the interview into  subject felt doing  just  pressured by  what was  asked  the was  assessed on a 5 point scale, with 1 indicating "a lot of pressure" and 5 indicating "no pressure". presented  in Table 10.  Ratings ranged from 3 to 5 and group means are Examination  of this table reveals very  difference between the three groups.  little  The hypnotized subjects experienced  only a slight amount of pressure (X = 4.417), almost as l i t t l e as did the waking group (X = 4.478) and slightly less than the relaxed students (X = 4.25).  Assuming the students were able to accurately assess the amount of  pressure they experienced, i t appears  that any  behavioural differences  exhibited by the hypnotized students is not attributable to the motivating  79  Table 10  Ratings of Amount of Pressure Experienced By Interview Method & Memory Technique  Memory Technique Reconstruction  Hypnosis  4.583 (n = 12)  Relaxation  4.167 (n = 12)  Waking  Imagery  4.250  Mean  4.417  (n = 12) 4.333  4.250  (n = 12)  4.182  4.750  (n = 11)  (n = 12)  4.478  80  pressure of the experimenter, as suggested by Barber (1969). Questions  6 & 7:  Success  in using the memory technique was rated by  students on a 7 point scale, with indicating very well.  1 representing not very well, and 7  Values ranged from a low of 2 to a high of 7, and  group means are reported in Table 11. Most subjects had a moderate amount of  success in employing  4.868.  the memory strategy, with the overall mean of  Those students using imagery reported a slightly  rate, 4.944 versus 4.792 for the resonstructive method.  higher success The very small  difference between these means suggests that students were able to u t i l i z e both memory techniques equally well, and that any differences in recall between strategies could not be attributed to the ineffective utilization of one technique over the other.  Comments from students on the memory  technique elicited by question 7 supported this interpretation.  Table 11  Ratings of Ability to Use Memory Technique By Interview Method  Memory Technique Reconstruction  Imagery  Hypnosis  4.625  4.625  Relaxation  4.792  5.083  Waking  4.958  5.125  Mean  4.792  4.944  82  DISCUSSION  Does hypnosis improve eyewitness memory? One of the main concerns of this study was to address this contentious issue. instance recalling  were  clear  details  —  hypnosis  The results in this  was of no aid at a l l to witnesses  of the video event.  Hypnosis  did not increase the  amount of information that individuals were able to remember, nor did i t increase the accuracy of those reports.  This was not due to a ceiling  effect; the accuracy rate for individuals in the waking groups averaged 74.9%,  fact,  providing ample opportunity for hypnosis to enhance memory. only  those  hypnotized  students  using  the reconstructive memory  instructions were able to match the performance those individuals employing non-hypnotized controls. many  of the waking students;  imagery were slightly less accurate than the  This situation runs counter to the notion among  hypno-investigators  accompanied by imagery  In  that  hypnosis  instructions.  aids  memory,  especially  Thus, there was no evidence  when from  this study to support the forensic use of hypnosis as an interrogation technique with eyewitnesses. Although  the results  of this  study  are consistent  with  past  laboratory investigations of the effects of hypnosis on eyewitness memory, it  could  situations. questioning  be that  hypnosis  enhances  memory  only  in more real  life  This would explain the numerous anecdotal reports whereby under hypnosis  i s claimed  details as licence plate numbers.  to have produced  such valuable  However, the evidence from the present  study does not support this possibility either.  When recalling details  83  about the "live" aspects of this experiment,  hypnotized students showed  the same level of recall as the control students. errors;  in fact,  hypnotized  individuals  Nor did they make fewer  had the highest  error  rate,  averaging 26.2%, compared to an average of 2k.7% for the relaxation groups and  23.2$ for the waking  statistically  significant,  groups. the  While  trend  these  that  differences  emerges  here  were not offers  no  substantiation to the claim that hypnosis aids memory for "live" events. There  i s one explanation which  may  account  hypnosis to aid memory in research of this type. cannot such  for the failure of Real l i f e  be fully re-enacted in the laboratory because the simulation of events,  experiences.  of necessity, lacks the emotional This  poses  a dilemma.  component  Hypno-investigators  of natural believe that  hypnosis, in conjunction with imagery, removes the emotional memory  that  crime.  But i t could be argued  result  from  traumatic  experiences,  such  blocks to  as witnessing a  that hypnosis will prove effective only  when there i s a high degree of emotionality surrounding event.  situations  Yet the ethics of scientific  research prevent  the observed  the investigator  from subjecting volunteers to such emotionally disturbing situations, and rightfully  so.  While  some researchers have attempted  to arouse  their  participants by showing as the c r i t i c a l event stress-provoking films of accidents (e.g., Zelig & Beidleman, 1981) or surgical procedures  (e.g.,  DePiano & Salzberg, 1981), i t can s t i l l be argued that momentary stress created in the comfortable surroundings of the laboratory cannot possibly mimic the emotions felt situations.  during unexpected, potentially  life-threatening  Thus, i t may be that the investigations of the effects of  84  hypnosis  on  eyewitness  memory  undertaken  in controlled  laboratory  conditions w i l l have restricted ecological validity. The possibility that the conclusions drawn from laboratory research may have limited application to real l i f e the present study.  situations gains support from  The pattern of recall noted for videotape details was  not mirrored by memory of the "live" event.  In the case of the video  event, students questioned in the waking groups showed no effect for the type  of memory  technique  they  used,  while  relaxation  and hypnosis  interacted with this variable to produce reliable differences in recall. Imagery instructions increased errors in recall of the video event when used in combination with hypnosis, but lowered errors i n conjunction with relaxation. as  close  One might conclude from these results that there may not be an  association  between  hypnosis  and  relaxation  researchers have speculated (e.g., Edmonston, 1972; 1981). recall was entirely different  as some  The pattern of  for details of the "live" event.  While  there were few differences that were statistically significant, the trend to emerge was the opposite of that observed for the video event.  Why the  pattern of recall would take one form for the video event and a different one  for the "live"  event  i s not immediately  apparent.  Any post-hoc  explanations that could be concocted would offer l i t t l e insight into the phoneomenon observed here.  Further research i s required to examine this  issue more specifically. A  second  major  concern  of this  susceptibility to misinformation. that  hypnosis  heightens  study  addressed  the issue of  No support was found for the suggestion  susceptibility  to  leading  questions.  The  85  hypnotized students in this study did not make more errors in responding to the misleading questions than their non-hypnotized & Bradshaw (1980) have suggested  counterparts.  that questions designed  Dodd  to lead the  responder w i l l be effective only when the source of the misinformation i s clearly non-biased. show  increased  possibility  Could this explain why hypnotized students did not  susceptibility  is highly unlikely  whether hypnotized  to  the  given  misleading  the tendency  questions?  This  for a l l students,  or not, to make nearly twice as many errors when  answering the leading questions versus the more objective ones.  Thus, the  misinformation did have an effect on students' responses; however, i t did not have a differential effect on hypnotized individuals.  This situation  highlights the need for a l l interrogators to use only objective questions during an eyewitness interview. In addition to practical concerns, this situation has implications for theoretical models of memory as well. displayed  a  tendency  to incorporate  The fact  erroneous  information  misleading questions into their memory i s entirely reconstructive model of memory.  that a l l students from the  consistent with the  Trace theory neither predicts, nor offers  a satisfactory explanation for these results.  Further, responses to the  question regarding the posted telephone number offer no support for trace theory. that  Given the claims of some hypno-investigators (e.g., Reiser, 1976)  memory  works  like  a  videotape,  recording  every  detail  at a  subconscious level which can be accessed by hypnosis, i t i s curious why none of the students, particularly those under hypnosis, could correctly recall the telephone number.  One might argue that the students in the  86  experimental groups were not adequately hypnotized, and so the hypnotic enhancement of memory would not be evident.  Yet the various assessments  of hypnotic depth made throughout the experiment do not give credence to this possibility.  Without independent assessments of hypnotic depth, such  as those provided in this research, there i s a danger for this type of argument to f a l l into circularity; that i s , to use hypnotic facilitation of memory as both the expected outcome of the induction and a measurement of i t s success. A third involved  area  the photo  in which  hypnosis  identification  could  task.  have  Again,  affected  performance  no differences  were  observed among the six groups in rate of correct identification of the main bank robber from the photospread. man's  appearance  identification.  under  hypnosis  The corollary  had  of this  hypnosis did not hinder performance  It appears that recalling the no  effect  i s also  on  important  subsequent to note:  on the identification task either.  The hypnotized students were no more likely to select an incorrect picture from the photospread than the non-hypnotized controls.  Thus, no benefits  at a l l could be attributed to hypnosis i n the present study. facilitated  It neither  nor harmed eyewitness memory in terms of susceptibility to  misinformation or identification from a photospread. Much of the previous research on eyewitness memory has emphasized i t s fallibility  (e.g., Buckhout,  (summarized  in Loftus, 1979), resulting in a rather bleak estimation of  eyewitness performance. with this dismal record.  1974; Clifford,  1979) or i t s malleability  The results of this study are not in accordance Accurate recall for the events of the videotape  87  averaged between 69 and 78$ for the six groups, with slightly higher rates for the "live" event, between 73 and improvement,  these  performance,  in  accuracy  rates  contrast  to  performance reported in the  81$.  Although  represent  previous  fairly  reports.  experimental  constraints of the experimental design.  a  there is room for  literature  good  level  of  Much  of  the  poor  can  be  linked  to  For example, use of forced-choice  questions, generally requiring a yes/no response,  restrict  the range of  possible responses and does not completely tap the participant's memory of the c r i t i c a l police  event.  interrogation  In this  study,  procedures  —  the questions were modeled  after  most were open-ended, allowing the  respondents the opportunity to provide as much or as l i t t l e information as they wished, or to qualify their answers.  Especially with the misleading  questions, the interview was designed so that the "yes", and particularly the  "no",  responses  were  followed  by  other  more  explicit  questions  attempting to determine more accurately the students' memory for specific details. importance  Another constraint of the experimental situation relates to the attached to t r i v i a l aspects of an event.  That i s , questions  regarding unimportant details of an observed scene are given equal, i f not more, weight with questions of central concern. questions  asked  during  an  interview  non-salient aspects of an event observed  contain  When roughly half the misinformation  some time ago,  about  i t i s not too  surprising that the memory scores of witnesses are generally quite low. Much of the previous work in eyewitness memory reflects this situation. But is this truly representative of the typical eyewitness?  The results  of the present study suggest that when recall procedures are constructed  88  to facilitate recall and to allow witnesses the freedom to control their responses, memory can be quite good.  Perhaps memory researchers who  are  interested in extending their laboratory findings to practical situations, such as the court room, should take care to construct their experimental designs so as to more accurately reflect real l i f e practices. The correlational analysis revealed some interesting relationships. First,  a small but  between confidence  statistically on  significant  relationship was  the photo identification  task and  observed  accuracy.  As  noted earlier, there i s a good deal of confusion surrounding this issue. Why  there i s such variation among studies is unclear.  of the response the  witness  Perhaps expansion  categories of the photo identification task, permitting  to  make no  uncertain and who,  identification,  removes  those  who  are  most  i f forced to make a choice, would pick an incorrect  picture thereby clouding the association between accuracy and confidence. The  present  findings also provide additional information regarding  confidence ratings.  Those students who  the  were more confident on the photo  identification  task provided more information about the  than those who  were less confident, but accuracy was not associated with  confidence  to any  degree.  Future  research  critical  event  that considers personality  traits may provide some insight into the nature of this relationship.  89  REFERENCES  As, A. age  The r e c o v e r y o f f o r g o t t e n language knowledge through h y p n o t i c regression:  Hypnosis, 1962, Barber, T.X. II.  case  report.  American  Journal  of  Clinical  5, 24-29-  E x p e r i m e n t a l evidence f o r a t h e o r y o f h y p n o t i c b e h a v i o u r : Experimental  International 9,  A  controls  in  Journal of C l i n i c a l  hypnotic  and  age  regression.  E x p e r i m e n t a l Hypnosis,  1961,  181.  Barber, T.X.  Measuring  " h y p n o t i c - l i k e " s u g g e s t i b i l i t y w i t h and  " h y p n o t i c i n d u c t i o n " ; Psychosomatic influencing  response  to  the  properties,  Barber  norms and  Suggestibility  without variables  Scale  (BSS).  P s y c h o l o g i c a l Reports, 1965a, 809-844, Monograph Supplement. Barber, T.X.  The e f f e c t s o f " h y p n o s i s " on l e a r n i n g and r e c a l l :  methodological 21,  critique.  Journal  of  Clinical  A  Psychology,  1965b,  19-25.  Barber, T.X.  Hypnosis:  R e i n h o l d Co., Barber, T.X.,  A Scientific  suggestions,  Abnormal Psychology, Barber, T.X., Creative  1966,  & W i l s o n , S.C. Imagination  E f f e c t s on r e c a l l and  suggested  JJ_(3),  York:  Van  Nostrand  regression.  Journal  of  The Barber S u g g e s t i b i l i t y S c a l e & the  Scale:  Remembering.  of hypnotic induction,  169-180.  Experimental  The American J o u r n a l o f C l i n i c a l Hypnosis, B a r t l e t t , F.C.  New  1969.  & C a l v e r l e y , D.S.  motivational  Approach.  Cambridge:  & clinical 1979,  applications.  2H2-3),  The U n i v e r s i t y P r e s s ,  84-108. 1932.  90  Best, H.L., & Michaels, R.M. hypnosis.  Living out "future" experience under  Science, 1954,120, 1077. <  Binet, A. La Suggestibility, 1900. In Levine, F.J., & Tapp, J.L. The psychology Kirby.  of criminal  identification:  The gap from  Wade to  University of Pennsylvania Law Review, 1973, 121, 1079-1131.  Brandon, R., & Davies, C. Wrongful Imprisonment.  London:  Allen &  Unwin, 1973Brigham, J.C, Snyder, L.D., Spaulding, K., & Maass, A. The accuracy of eyewitness  identification  in  a  field  setting.  Journal  of  Personality & Social Psychology, 1982, 12_, 673-687. Brown, E., Deffenbacher, K., & Sturgill, W. circumstances  of the encounter.  Memory for faces and the  Journal of Applied  Psychology,  1977, 62, 311-318. Buckhout, R.  Eyewitness Testimony.  Scientific American, 1974, 231(6),  23-31. Campbell, D.T., & Stanley, J.C. Experimental & Quasi-Experimental Designs for Research.  Chicago:  Rand McNally  College Publishing  Company, 1966. Clifford, B. In  D.P.  Eyewitness Testimony: Farrington, K.  Psychology,  The Bridging of a Credibility Gap.  Hawkins,  Law & Legal Processes.  &  S.M.  London:  Lloyd-Bostock  (Eds.)  The MacMillan Press  Ltd., 1979. Clifford, B., & Scott, J. Individual and situational factors in eyewitness testimony. 352-359.  Journal of Applied Psychology, 1978, 63(3),  91  Coleman, T.R. A comparative study of certain behavioral, physiological and  phenomenological  progressive  effects  relaxation  of  hypnotic  procedures.  induction  and two  Ph.D. Dissertation,  Brigham  Young University, 1976. Cooper, L.M., & London, P. Sex and hypnotic susceptibility in children. International Journal of Clinical  & Experimental Hypnosis, 1966,  _ 4 ( 1 ) , 55-60.  Cooper, L.M., & London, P. Reactivation of memory by hypnosis & suggestion.  International  Journal  of Clinical  &  Experimental  Hypnosis, 1973, 2 1 ( 4 ) , 3 1 2 - 3 2 3 . Craik, F.I.M., & Jacoby, L.L. In  R. Restle  "A process view of short-term retention."  (Ed.) Cognitive Theory:  Vol. 1.  Potomac, Md.:  Erlbaum Associates, 1975Das, J.P. Learning & recall under hypnosis & in the waking state: A comparison.  Archives of General Psychiatry, 1961, 4, 517-521.  Deffenbacher, K. Eyewitness accuracy & confidence: Can we infer a relationship?  Law & Human Behavior, 1980, 4 ( 4 ) , 243-260.  DePiano, F., & Salzberg, H.C. Hypnosis as an aid to recall of meaningful information presented under three types of arousal.  International  Journal of Clinical & Experimental Hypnosis, 1981, 29(4), 383-400. Devlin Report. of  Report to the Secretary of State for the Home Department  the Departmental  Committee  on Evidence  of Identification in  Criminal Cases, HMSO, London, 1976. Dhanens, T.P., & Lundy, R.M. The 1975,  International 23(1),  68-79.  Hypnotic & waking suggestions & recall.  Journal of Clincial  & Experimental  Hypnosis,  92  Dodd, D.H., & Bradshaw, J.M. Leading questions & memory: constraints.  Pragmatic  Journal of Verbal Learning & Verbal Behavior, 1980,  19, 695-704. Dorcus, R.M.  Recall under hypnosis of amnestic events.  International  Journal of Clinical & Experimental Hypnosis, 1960, 8(1), 57-60. Egan, D., Pittner, M., & Goldstein, A.G. Eyewitness Photographs  vs. live  models.  identification:  Law & Human  Behavior,  1977, 1,  199-206. Edmonston, W.E.  To what should hypnosis be compared?  appropriate experimental controls:  A Symposium.  The problem of The American  Journal of Clinical Hypnosis, 1972, _4(4), 209-235. <  Edmonston, W.E.  Hypnosis & Relaxation. New York:  John Wiley & Sons,  1981. Eisele, G., & Higgins, J.J. Hypnosis in educational & moral problems. American Journal of Clinical Hypnosis, 1962, 4, 259-263. Erdelyi, M.H., & Becker, J. Hypermesia for pictures:  Incremental memory  for pictures but not for words i n multiple recall t r i a l s .  Cognitive  Psychology, 1974, 6(1), 159-171. Erdelyi, M.H., & Kleinbard, J. Has Ebbinghaus decayed with time? The growth of recall (hypermesia) over days. Psychology:  Journal of Experimental  Human Learning & Memory, 1978, 4(9), 275-289.  Eysenck, H.J. An experimental study of the improvement of mental & physical  functions in the hypnotic  state.  British  Journal of  Medical Psychology, 1941, 18, 304-316. Field, P.B. An inventory scale of hypnotic depth.  International  Journal of Clinical & Experimental Hypnosis, 1965, 13(4), 238-249.  93  Gardner, D.  The perception & memory of witnesses.  Cornell Law  Quarterly, 1933, 18, 391-409. Goldstein, A.G.  The f a l l i b i l i t y of the eyewitness:  Psychological  evidence. Psychology in the Legal Process, Spectrum, 1977. Gorenstein, G.W.,  & Ellsworth, P.C.  Effect of choosing an incorrect  photograph on a later identification by an eyewitness.  Journal of  Applied Psychology, 1980, 65(5), 616-622. Gray, W.H.  The effect of hypnosis on learning to spell.  Journal of  Educational Psychology, 1934, 25, 471-473. Hakstian, A.R., & Cattell, R.B.  The checking of primary ability  structure on a broader basis of performance.  British Journal of  Educational Psychology, 1974, 44, 140-154. Hakstian, A.R., & Cattell, R.B.  The Comprehensive Ability Battery.  Champaign, 111.: Institute for Personality & Ability Testing, 1975. Hakstian, A.R., & Cattell, R.B. basis  of  twenty  primary  Higher-stratum ability structures on a abilities.  Journal  of Educational  Psychology, 1978, 70(5), 657-669Hall, D.F., & Ostrom, T.M. biased  or unbiased  Eyewitness Variables.  Accuracy of eyewitness identification after instructions.  & Testimony  Research:  Cited  in Wells, G.  System Variables  Journal of Personality  &  Social  Applied  & Estimator  Psychology,  1978,  36(12), 1546-1557Hammer, E.F. Post-hypnotic suggestion & test performance.  Journal of  Clinical & Experimental Hypnosis, 1954, 2, 178-185. Hanley, F.W.  Hypnosis in the court room.  Canadian Psychiatric  Association Journal, 1969, 14, 351-354.  94  Harris, R.J. Answering questions containing marked & unmarked adjectives & adverbs.  Journal of Experimental Psychology, 1973,  97, 399-401. Henderson, E.N.  A study of memory for connected trains of thoughts.  Psychological Monographs, 1903, 5, 1-94. Hilgard, E.R.  The domain of hypnosis. American Psychologist, 1973, 28,  972-982. Hilgard, E.R., & Tart, CT. waking  &  hypnosis. Huse, B.  Responsiveness to suggestions following  imagination  instructions  &  following  induction  of  Journal of Abnormal Psychology, 1966, 71_(3), 196-208  Does the hypnotic trance favour the recall of faint memories?  Journal of Experimental Psychology, 1930, J_3, 519-529. Hyde, T.S., & Jenkins, J.J. The differential effects of incidental tasks on  the organization  words.  of recall  of a l i s t  of highly  associated  Journal of Experimental Psychology, 1969, 82, 472-481.  Illovsky, J. An experience with group hypnosis in reading disability in primary behaviour disorders.  Journal of Genetic Psychology, 1963,  102, 61-67. Kihlstrom, J.F., Evans, F.J., Orne, E.C., & Orne, M.T.  Attempting to  breach posthypnotic amnesia. Journal of Abnormal Psychology, 1980, 89(5), 603-616. Krippner, S.  Hypnosis & reading improvement among university students.  American Journal of Clinical Hypnosis, 1963, 5_, 187-193. Kroger, W.S., & Douce, R.G.  Hypnosis in criminal investigation.  Inter-  national Journal of Clinical & Experimental Hypnosis, 1979, 27, 358-374.  95  LeCron, L.M. A method of measuring the depth of hypnosis.  Journal of  Clinical & Experimental Hypnosis, 1953, M.2), 4-7. Leippe, M.R., Wells, G.L., & Ostrom, T.M.  Crime seriousness as a  determinant of accuracy i n eyewitness identification.  Journal of  Applied Psychology, 1978, 63, 345-351. Leonard, J.R. Hypnotic age regression: ablation  hypothesis.  A test of the functional  Journal of Abnormal Psychology,  1965, 70,  266-269. Lindsay, R.C.L., Wells, G.L., & Rumpel, CM.  Can people detect  eyewitness-identification accuracy within and across situations? Journal of Applied Psychology, 1981, 66(1), 79-89. Lipton, J.P. On the psychology of eyewitness testimony.  Journal of  Applied Psychology, 1977, 62, 90-95. Lodato, F.J. Hypnosis as an adjunct to test performance.  American  Journal of Clinical Hypnosis, 1964, 6_, 271-273Loftus, E.F. Incredible eyewitness.  Psychology Today, 1974, Dec,  117-119Loftus, E.F. Shifting human color memory. Memory & Cognition, 1977, 5(6), 696-699Loftus, E.F. Eyewitness Testimony.  Cambridge, Mass.:  Harvard  University Press, 1979. Loftus, E.F., Altman, D., & Geballe, R. witness'  later  recollections.  Effects of questioning upon a Journal  of  Police  Science  &  Administration, 1975, 3(2), 162-165. Loftus, E.F., Miller, D.G., & Burns, H.J. Semantic integration of verbal information  into  a  visual  memory.  Journal  of  Experimental  96  Psychology, Human Learning & Memory, 1978, 4_( 1), 19-31. Loftus, E.F., & Palmer, J.C. Reconstruction of automobile destruction: An example of the interaction between language & memory. Journal of Verbal Learning & Verbal Behavior, 1974, 13 > 585-589. Loftus, E.F., & Zanni, G. Eyewitness testimoy: The influence of the wording of a question. Bulletin of the Psychonomic Society, 1975, 5, 86-88. McCord, H. Hypnosis as an aid to the teaching of a severely mentally retarded  teen-age  boy.  Journal  of  Clinical  &  Experimental  Hypnosis, 1956, 4, 21-24. McCord, H., & Sherrill, C.I. post-hypnotically.  A note of increased ability to do calculus  American Journal of Clinical Hypnosis, 1961, 4,  124. McGuire, W.J.  Suspiciousness of experimenter's intent.  In R. Rosenthal  & R.L. Rosnow (Eds.) Artifact in Behavioral Research.  New York:  Academic Press, 1969, 13-57Malpass, R.S., & Devine, P.G. research.  Realism and eyewitness identification  Law & Human Behavior, 1980, 4(4), 347-358.  Malpass, R.S., & Devine, P.G. Guided memory in eyewitness identification.  Journal  of  Applied  Psychology,  1981, 66(3),  343-350. Marquis, K.H., Marshall, J., & Oskamp, S. Testimony validity as a function of question form, atmosphere and item d i f f i c u l t y . Journal of Applied Psychology, 1972, 5j5, 167-186. Marshall, J. Law & Psychology in Conflict. 1969.  New York:  Doubleday-Anchor,  97  Marston, W.  Studies i n testimony.  Journal of the American Institute of  Criminal Law, 1924, 15_, 5 - 3 1 . Miller, D.G., & Loftus, E.F. Influencing memory for people & their actions. Mitchell, M.B.  Bulletin of the Psychonomic Society, 1976, 7(1), 9-11. Retroactive inhibition & hypnosis.  Journal of General  Psychology, 1932, 7, 343-359. Morse, D.R., Martin, J.S., Furst, M.L., & Dublin, L.L. A physiological & subjective  evaluation  of  meditation,  hypnosis  &  relaxation.  Psychosomatic Medicine, 1977, 39, 304-324. Muensterberg, H.  On the Witness Stand.  New York:  Doubleday, Page,  1908. Muscio, B. The influence of the form of a question.  Journal of  Psychology, 1916, VIII, 351-389O'Connell, D.N.  An experimental comparison of hypnotic depth measured by  self-ratings & by an objective scale.  International Journal of  Clinical & Experimental Hypnosis, 1964, 12., 34-46. Paivio, A. Imagery & Verbal Processes.  New York:  Holt, Rinehart &  Winston, 1971. Paivio, A. The empirical case for dual coding. Imagery, Memory & Cognition.  Hillsdale,  In J.C. Yuille (Ed.) N.J.:  Lawrence Erlbaum  Associates, 1983.. Paul, G.L. Physiological effects of relaxation training & hypnotic suggestions.  Journal of Abnormal Psychology, 1969, 74, 425-437.  Peters, J.E., & Stern, R.M. responses  during  Peripheral skin temperature & vasomotor  hypnotic  induction.  International Journal of  Clinical & Experimental Hypnosis, 1973, 21_, 102-108.  98  Powers, P.A., Andriks, J.L., & Loftus, E.F. Eyewitness accounts of females  and males.  Journal of Applied  Psychology,  1979, 64,  339-347. Putnam, W.H.  Hypnosis & distortions in eyewitness memory.  International Journal of Clinical  & Experimental Hypnosis, 1979,  27(4), 437-448. Raginsky, B.B.  Hypnotic recall of aircrash cause.  International  Journal of Clinical & Experimental Hypnosis, 1969, 17(1), 1-19. Read, J.D., Barnsley, R.H., Ankers, K., & Whishaw, I.Q. Variations in severity  of verbs  interpretation.  &  eyewitnesses'  testimony:  Perceptual & Motor Skills,  An  alternative  1978, 46(3, Pt. 1),  795-800. Reid, A.F., & Curtsinger, G. hypnosis:  The effect  Physiological changes associated with of hypnosis on temperature.  International  Journal of Clinical & Experimental Hypnosis, 1968, 16, 111-119. Reiff, R., & Scheerer, M.  Memory & Hypnotic Age Regression.  New York:  International University Press, 1959. Reiser, M.  Hypnosis as an aid in a homicide investigation.  The American  Journal of Clinical Hypnosis, 1974, 27(2), 84-87. Reiser, M.  Hypnosis as a tool in criminal investigation.  The Police  Chief, 1976, 43(Nov). 36-40. Reiser, M., & Nielson, M. speciality.  Investigative hypnosis:  The American  A developing  Journal of Clinical  Hypnosis,  23(2), 75-77. Rosenberg, M.J. The conditions & consequences of evaluation  1980,  99  apprehension.  In R. Rosenthal & R.L.  Behavioral Research. Rosenhan, D., & London, P. rote learning. Rosenthal, B.G.  New York:  Rosnow (Eds.) Artifact in  Academic Press, 1969, 279-349.  Hypnosis in the unhypnotizable: A study in  Journal of Experimental Psychology, 1963, 65, 30-34.  Hypnotic recall of material learned under anxiety- &  non-anxiety-producing  conditions.  Journal  of  Experimental  Psychology, 1944, 34, 369-389. Rosenthal, R.  Experimenter Effects in Behavioral Research.  New York:  Appelton-Century Crofts, 1966. Sanders, G.S.,  & Simmons, W.L.  accuracy:  Use of hypnosis to enhance eyewitness  Does i t work?  Journal of Applied Psychology,  1983,  68(1), 70-77. Sarbin, T.R.  Mental age changes in experimental regression.  Journal of  Personality, 1950, J_9, 221-228. Sarbin, T.R., & Coe, W.C.  Hypnosis:  Influence Communication.  A Social Psychological Analysis of  New York:  Holt, Rinehart & Winston Inc.,  1972. Schafer, D.W.,  & Rubio, R.  Hypnosis to aid the recall of witnesses.  International Journal of Clinical  & Experimental Hypnosis,  1978,  26, 81-91. Schulman, R.E., & London, R.P.  Hypnosis & verbal learning.  Journal of  Abnormal Social Psychology, 1963, 67, 363-370. Sears, A.B.  A comparison of hypnotic & waking recall.  Journal of  Clinical & Experimental Hypnosis, 1954, 2, 296-304. Sears, A.B.  A comparison of hypnotic & waking learning of the  t  \  100  International  Morse  Code.  Journal of Clinical  & Experimental  Hypnosis, 1955, 3, 215-221. Sears, A.B. Hypnosis & recall.  Journal of Clinical & Experimental  Hypnosis, 1956, 4, 165-171. Sheehan, P.W., & Dolby, R.M. logical-empirical  Artifact & Barber's model of hypnosis: A  analysis.  Journal  of  Experimental  Social  Psychology, 1974, JK), 171-187. Sheehan, P.W., & Tilden, J. Effects of suggestibility & hypnosis on accuracte  &  distorted  retrieval  Experimental Psychology:  from  memory.  Journal  of  Learning, Memory & Cognition, 1983, 9(2),  283-293Shor, R.E., & Orne, E.C. The Harvard Group Scale of Hypnotic Susceptibility,  Form  A.  Palo  Alto,  Calif.:  Consulting  Psychologist Press, 1962. Shor, R.E., & Orne, E.C. The Harvard Group Scale of Hypnotic Susceptibility,  Form A:  An Adaptation for Group Administration  with Self-Report Scoring of the Stanford Hypnotic Susceptibility Scale,  Form  A.  Palo  Alto:  Calif.:  Consulting Psychologists  Press, 1965. Snee, T.S., & Lush, D.E. Interaction of narrative & interrogatory methods of obtaining testimony.  Journal of Psychology, 1941, 11,  229-236. Stalnaker, J.M., & Riddle, E.E. The effect of hypnosis on long-delayed recall.  General Psychology, 1932, 6, 429-440.  Stern, L. The psychology of testimony. Psychology, 1939, 34, 3-20.  Journal of Abnormal & Social  101  Swiercinsky, D., & Coe, W.C.  Hypnosis, hypnotic responsiveness &  learning meaningful material. Experimental Hypnosis, 1 9 7 0 ,  Tart, C.T.  International Journal of C l i n i c a l &  _18.C3),  217-222.  Quick & convenient assessment of hypnotic depth:  scales.  The American Journal of C l i n i c a l Hypnosis,  Self-report  1979,  21(2-3),  186-207.  Tebecis, A.K.,  Provins, M.A.,  Farnback, R.W.,  Hypnosis &  Journal of Nervous & Mental Disease, 1 9 7 5 , 1 6 1 , 1 - 1 7 .  EEG.  Tellegen, A.  On measures & conceptions of hypnosis.  Journal of C l i n i c a l Hypnosis, 1 9 7 9 ,  Tellegen, A., & Atkinson, G. experiences  Timney, B.N.,  219-236.  a  trait  related  Journal of Abnormal Psychology,  & Barber, T.X.  International  21(2-3),  Journal  American  Openness to absorbing & s e l f - a l t e r i n g  ("absorption"),  susceptibility.  J7,  & Pentony, P.  1974,  to  hypnotic  83,  268-277.  Hypnotic induction & o r a l temperature. of C l i n i c a l  &  Experimental Hypnosis, 1 9 6 9 ,  121-132.  True, R.M.  Experimental control i n hypnotic age regression states. 1949,  Science,  Tulving, E.  11°»  585-584.  Cue-dependent forgetting.  American S c i e n t i s t , 1 9 7 4 , 6 2 ,  74-82.  Wagstaff, G.F. Skills,  Wall, P.  Hypnosis & recognition of a face.  1982,  55,  Perceptual & Motor  816-818.  Eyewitness I d e n t i f i c a t i o n i n Criminal Cases.  111.: Wall, P.D.,  Springfield,  Charles C. Thomas, 1 9 6 5 . & Lieberman, L.F.  Effects of task motivation & hypnotic  102  induction & hypermnesia.  American Journal of Clinical  Hypnosis,  1976, 1 8 , 250-253. Walrath, L.C., & Hamilton, D.W. hypnosis.  Autonomic correlates of meditation &  American Journal of Clinical Hypnosis, 1975, 17, 190-197.  Weitzenhoffer, A.M., & Hilgard, E.R. Scale:  Form C.  Stanford Hypnotic Susceptibility  Palo Alto, Calif.:  Consulting Psycholists Press,  1962. Wells, G.L.  Applied eyewitness-testimony research: System variables &  estimator variables.  Journal of Personality & Social Psychology,  1978, 36, 1546-1557. Wells, G.L., Lindsay, R.C.L., & Ferguson, T.J. Accuracy, confidence & juror perceptions in eyewitness identification.  Journal of Applied  Psychology, 1979, 64(4), 440-448. Wells, G.L., Lindsay, R.C.L., & Tousignant, J.P. Effects of expert psychological advice on human performance of eyewitness testimony. Whipple, G.  Whipple, G.  Law & Human Behavior, 1980, 4, 275-285.  The observer as reporter:  testimony".  i n judging the validity  A survey of the "psychology of  Psychologyicai Bulletin, 1909, 6, 753-770.  Psychology of testimony & report.  Psychological Bulletin,  1912, 9, 264-269. White, R.W.,  Fox, G.F., & Harris, W.W.  learned material.  Hypnotic hypermnesia for recently  Journal of Abnormal Social Psychology, 1940, 35,  88-103. Wilson, S.C., & Barber, T.X.  The Creative Imagination Scale as a  measure of hypnotic responsiveness:  Applications to experimental &  103  c l i n i c a l hypnosis.  American  Journal of Clinical Hypnosis, 1978,  20(4), 235-243. Woocher, F.D. Did your eyes deceive you? on the unreliability  Expert psychological testimony  of eyewitness identification.  Stanford Law  Review, 1 9 7 7 , 29, 969-1030.  Yarmey, A.D. Through the looking glass: self-facial poses.  Sex differences in memory for  Journal of Research in Personality,  1979a,  13,  450-459. Yarmey, A.D. The Psychology of Eyewitness Testimony.  New York: The  Free Press, 1979b. Young, P.C. An experimental study of mental and physical functions in the  normal and hypnotic states.  American  Journal of Psychology,  1925, 36, 214-232. Yuille, J.C. A c r i t i c a l examination of the psychological and practical implications of eyewitness research.  Law & Human Behavior, 1980,  4(4), 335-345. Yuille, J.C. Research & teaching with police:  A Canadian example.  International Review of Applied Psychology, in press. Zelig, M., & Beidleman, W.B. of caution.  The investigative use of hypnosis:  International Journal of Clinical & Experimental  Hypnosis, 1981, 29(4), 401-412.  A word  104  APPENDIX A  105  List Of Interview Questions Main Bank Robber 1.  How old was  he?  2.  How t a l l was  3.  About how much do you think he weighed?  4.  What kind of build did he have?  5.  What colour was his hair?  6.  How long was his hair?  7.  Can you describe what kind of hair he had?  8.  What colour were his eyes?  9.  How would you describe his face?  he?  10.  What kind of complexion did he have?  11.  Did he have any distinguishing facial features?  12.  What was he wearing?  13.  Did he have anything in his hands? ( i f reply is gun, ask what kind?)  14. *  Did you notice i f his gloves were leather? ( i f reply is no, ask "was he wearing gloves?")  15.  What did he do?  16.  What did he say?  17.  What is the most distinguishing thing you can remember about him?  18.  Is there anything else you can remember about him that I haven't asked?  Accomplice 19.  What did he look like?  106  20.  What kind of clothes was he wearing?  21. *  Did you notice what colour his hat was? ( i f unsure, ask i f he was wearing a hat?)  22.  What did he do?  Bank Teller 23.  What colour hair did she have?  24.  How long was her hair?  25.  How would you describe her face?  26.  What kind of clothes was she wearing?  27. *  Which wrist was she wearing her watch on? ( i f reply is don't know, ask i f she was wearing a watch?)  28.  Were you able to read her name on the nameplate on the counter? What was i t ?  Bank In General 29.  How many people were waiting in line for a teller?  30.  What did the note that the bank robber gave to the teller say?  Laboratory Room 31.  How many other people were in the room with you, besides the experimenter?  32.  Can you describe what the room looked like?  33. *  What was the floor like? ( i f reply is a carpet, ask i f i t was green?)  34.  Do you remember seeing a phone number on the video machine? What was i t ?  35.  About how long did the videotape of the bank robbery last?  Assistant Experimenter 36.  Did she introduce herself? What was her name?  37.  How old was she?  38.  About how t a l l was she?  39-  What colour hair did she have?  40.  How long was i t ?  41.  What type of hair did she have?  42. * Were the frames of her glasses square or round? 43.  How would you describe her face?  44.  Did she have any distinguishing facial features?  45.  What kind of complexion did she have?  46.  What kind of clothes was she wearing?  47.  Was she wearing any jewellery? What?  'Misleading questions.  108  APPENDIX B  109  Amended Version Of Field's Inventory Scale Of Hypnotic Depth  We are interested in the sensations you might have felt session.  during  this  Please read the following statements and answer T i f you think  the statement i s TRUE OR MOSTLY TRUE or F i f you think the statement i s FALSE OR MOSTLY FALSE.  T OR F ______  1>  Time stood s t i l l .  ^^^^  2. My arm trembled or shook when I tried to move i t .  ________  3. I felt dazed. 4. I felt aware of my body only where i t touched the chair.  ^^^^  5. I f e l t I could have tolerated pain more easily during the experiment.  _______  6.  ^^^^  7. I was delighted with the experience.  ^^^^  8. The experimenter's voice seemed to come from very far away.  _______  9. I tried to resist but I could not.  ^^^__  10. Everything happened automatically.  ______  11- Sometimes I did not know where I was.  ^^^^  12. It was like the feeling I have just before waking up.  _____  13. When I came out I was surprised at how much time had gone  m  I could have awakened any time I wanted to.  by. 14.  I was able to overcome some or a l l of the suggestions.  110  15.  During the experiment I felt I understood things better or more deeply.  16.  At times I felt completely unaware of being in an experiment.  17.  I did not lose a l l sense of time.  18.  It seemed completely different from ordinary experience.  19.  Things seemed unreal.  20.  Parts of my body moved without my conscious assistance.  21.  I felt apart from everything else.  22.  It seems as i f i t happened a long time ago.  23.  I felt uninhibited.  24.  At times I felt as i f I had gone to sleep momentarily.  25.  I felt quite conscious of my surroundings a l l the time.  26.  I could not have stopped during the things the experimenter suggested even i f I tried.  27.  It was a very strange experience.  28.  I felt amazed.  29.  From time to time I opened my eyes.  30.  I couldn't stop movements after they got started.  31.  I had trouble keeping my head up a l l during the experiment.  32.  My mind seemed empty.  33.  It seemed mysterious.  111  APPENDIX C  112  Post-Experimental Questionnaire  We  are  interested  experiment.  i n your  reactions  to  this  second  part  Please answer the questions i n the space provided.  of the I f you  require more room, please use the back of the sheet.  1.  Did you find i t was an e f f o r t f o r you to speak or to answer the questions?  2.  I f so, please explain.  Did you see the woman who showed you the videotape at any time during the intervening week?  3.  Have you ever been hypnotized before?  4.  Have you ever done deep relaxation exercises before?  5.  Please rate to what extent you f e l t pressured by the Experimenter during the interview:  ( c i r c l e the appropriate number)  1  very much pressure into doing just what was asked  2  quite a l o t of pressure into doing just what was asked  3  a moderate amount of pressure into doing just what was asked  4  only s l i g h t l y pressured into doing just what was asked  5  not at a l l pressured into doing just what was asked  113  6.  Please rate  how  instructions  1 not  successful  you were  on t h e t e c h n i q u e  2  3  i n following  the  t o use i n remembering  4  5  experimenter's the  event:  6  very  very  well  7.  Do y o u h a v e a n y  7  well  comments on t h e memory t e c h n i q u e y o u  used?  114  APPENDIX D  Black & White R e p r o d u c t i o n  Note:  Photo No.  o f the Coloured  5 i s the t a r g e t  picture.  Photospread  116  APPENDIX E  117  Partial Transcript Of Hypnotic Induction  "Breathe deeply, settle comfortably in the chair... Raise your right arm, look at your thumbnail.  Now direct your gaze past your thumbnail  to a spot on the wall.  You w i l l notice things getting blurry, your eyes  may feel like closing.  When i t i s comfortable to do so, allow your eyes  to close  (keep repeating suggestions of heavy eyes and to close eyes  until subject closes them). "Imagine  that  you  have sandbags  placed  on your arm,  that i t ' s  getting heavier and heavier, i t ' s coming down closer and closer to your lap (keep doing this until arm comes down to lap). to touch your lap until you're comfortable and  Don't allow your arm ready to relax  even  further. "Now  focus your attention on your neck.  in your neck, just release i t .  Imagine the tension flowing down your  arms, and out through your fingers. chest, release i t . tension  to  flow  If you notice any tension  If there i s any tension in your  Let yourself become completely relaxed, allow the out  through  your  toes.  Your  breathing  evenly,  regularly, your body i s free of tension, feeling relaxed and comfortable. "Allow comfortable.  yourself  to  readjust  in your  chair,  to  be completely  If you have to, you can be alert and wide awake.  right now, just enjoy feeling relaxed.  But  If you hear any outside noises,  you can pay attention to them i f you want, or you can ignore them so that you can become more and more relaxed, and you can go deeper and deeper. "Some people find i t helpful to imagine a staircase with 10 steps.  118  They imagine themselves walking down this staircase.  With each step  they take, they become more and more relaxed, more and more comfortable, they go deeper and deeper. go.  The more relaxed you are, the deeper you can  I'm going to count from  1 to 10.  Imagine to yourself such a  staircase, and that you're going to walk down the steps. more relaxed you are, the deeper you can go.  Remember, the  So let's take that f i r s t  step... 1...more and more relaxed...2....more and more comfortable (count up  to 10 with  suggestions of relaxation,  comfortableness and going  deeper after each count). "Now you're at the bottom  of the staircase.  That's fine.  Some  people find i t helpful to imagine to themselves a place where they've been very relaxed and peaceful. the  beach.  It could be skiing, sailing or lying on  Recreate such a place in your mind now.  re-experience the sensation of relaxation — Can you feel the sun on your skin?  Allow yourself to  it's so peaceful and calm.  Maybe the wind?  Enjoy any sights  you can see, or sounds you can hear.  (More suggestions to enjoy this  place and the opportunity of relaxing).  That's fine.  "(Deepening technique) To help you go even deeper, imagine you are gazing at the centre of a whirlpool.  As you focus your attention on the  centre, you can see the water swirling gently at the edges.  As you look  at the centre, feel the water pulling you deeper and deeper.  The longer  you look, the deeper you can go. are  in f u l l control.  That's fine."  Go as deep as you want, knowing you  119  APPENDIX F  120  Hand Lock Suggestibility  Item  From Barber's Suggestibility  "(Student's  name),  tightly, and interlace correctly,  I'd  like  you  your fingers.  say "that's right").  to  Scale  clasp  your  hands  Put them in your lap.  together ( i f done  Now concentrate on your hands and hold  them together as tightly as you can.  (BEGIN TIMING).  Imagine that your  hands are two pieces of s t e l l  that are welded together so that i t ' s  impossible to get them apart.  They're stuck, they're welded, they're  clamped tightly sut. When I ask you to pull your hands apart they'll be stuck and they won't come apart no matter how hard you try. They're stuck together just like two pieces of steel that are welded together. You feel as i f your fingers were clamped hard, solid, rigid! w i l l stick  together!  in a vise.  The harder you try to pull them apart the more they It's impossible to pull your hands apart!  more you try the more d i f f i c u l t i t w i l l become. apart (end 45 seconds) but you'll steel.  Your hands are  The  Try to pull your hands  find you can't because they're like  (WAIT 5 SECONDS) Try harder, you can't.  (WAIT 10 SECONDS).  You can unclasp your hands now."  TO SCORE: Pass i f incomplete separation after 15 seconds.  121  APPENDIX G  122  Transcript Of Taperecorded Deep Relaxation Exercises  "This tape is to teach you deep muscle relaxation.  If you practice,  you can learn to relax at will, to put yourself into a very pleasant and comfortable  state known as deep relaxation.  loosening any tight  I'd like you to start by  clothing and finding a comfortable  then closing your eyes.  position, and  This method works by teaching you to identify  tension in various parts of your body and then to identify the opposite of that tension, which i s deep relaxation. "I'd  like you to clench your right hand into a f i s t , clench your  right hand into a f i s t and just think about the tension in your right hand. relax.  Feel the knuckles becoming white with tension and then let i t Notice  the contrast  between  the tension  and the relaxation.  Once again, clench your right hand into a f i s t and study the tension in your right hand, and then l e t i t relax. between tension and relaxation.  Notice the pleasant  contrast  Now clench your left hand into a f i s t  and study the tension in your left hand, then let i t relax. contrast between tension and relaxation.  Notice the  Once again, clench your left  hand into a f i s t and study the tension in your left hand.  Then let i t  relax, just let i t go loose, and limp, and relaxed. "Now bend your right hand at the wrist and point your fingers up to the ceiling. then  Study the tension in your right wrist and forearm, and  l e t i t relax,  relaxation.  and  feel  the  contrast  between  tension  and  Once again, bend your right hand at the wrist and point  your fingers up to the ceiling, feel the tension in your right wrist and  123  forearm, and then let i t relax, noting the contrast between tension and relaxation.  Now bend your left hand at the wrist, point the fingers up  to the ceiling, and then let i t relax. very relaxed.  Just go loose, and limp, and  Once again, bending your left hand at the wrist, pointing  your fingers up to the ceiling, study the tension in your wrist and forearm, and then let i t relax.  Notice the contrast between tension and  relaxation. "Now  I'd like you to flex both of your bicep muscles by drawing your  hands up to your shoulders. Bring your hands up to your shoulders, flex both of your bicep muscles. let them relax.  Study the tension in your biceps and then  It isn't necessary to tense your muscles so much that  you get a cramp, only just to tense them enough so that you can feel the tension. to  Once again, flexing your bicep muscles, bringing both hands up  the shoulders, then let them relax, just go loose, and limp, and  relaxed. "Now shrug your shoulders up to your ears, study the tension in your shoulders and the base of your neck, and then let your shoulders relax. Notice the pleasant contrast between the tension and the relaxation. Once again, shrug your shoulders up to your ears, study the tension in your shoulders and the base of your neck, and then just let them relax, just sag down loose and limp and very relaxed. "Now wrinkle up your forehead by raising your eyebrows up to the top of  your head.  relax.  Study the tension  in your forehead and  then let i t  Once again, raising your eyebrows up to the top of your head,  study the tension in your forehead and then let i t relax, let your forehead become more and more smooth, more and more relaxed.  124  "Now eyes  close  and  the  tension  and  then  study  and  then  let  "Now  make  between touch then  the  a  tension  your let  it  your them  relax,  and  tightly,  nose,  around them  as  if  mouth,  study  relax,  the  your  Once  again, and  relax  and  just  touch let  Once  tension the  both it  in  the  lightly  ears,  pleasant  very  your  the  big  and  contrast  eyes  of  feeling  mouth  the  nose,  close.  study  a  your  Study  your  bridge  making  your  around  tightly.  squint  relax,  again,  tension  eyes  eyes  then  noticing  the  your  to  relaxation.  study  squint  relax.  let  smile  your  ears,  very  tension  big  and  of  let  them  cheeks  eyes  bridge  tightly,  your  your  tension  the  in  contrast  smile  as  if  to  your  cheeks,  and  between  tension  and  relaxation. "Now mouth, Once  I'd  and  like study  again,  study  the  between  the  your  chest,  then  let  your  in  tongue  study  and  the  relax,  tongue  against  then  let  study  then  in  let  your  the  up  against  mouth,  i t  chin  and  the  roof  then roof  let of  of  your  i t  relax.  your  mouth,  relax. tension  relax. Once  pleasant  and the  the  it  relaxation.  tension  feeling  up  chest,  and the  your  mouth,  your  chin,  your  inside  i n s i d e your  tension  it  your  chin  and  press  tension  pressing  your  neck  to  the  tension  "Bury your  you  in  the  Notice  the  contrast  your  chin  again, the  contrast  bury  front  of  between  front  your  of  in  neck,  tension  and  back  the  relaxation "Now chair  or  neck,  and  the  I'd  like  you  the  bed  or  then  tension  muscles  go  let  in loose  it  the and  to  press  whatever, relax. back limp  of and  your  head  study  the  Once your  again, neck,  relaxed.  back  against  tension pressing then  Feel  let that  in your it  the the  back  head relax,  relaxed  of of  back, let  feeling  your study those now  in  125  your forehead, your forehead i s becoming more and more smooth, more and more relaxed, and that relaxed feeling spreading down from your fact, your eyes relaxed, your cheeks relaxed, your mouth relaxed, your jaw and your chin relaxed, that relaxation flowing down into your neck, down into your shoulders, down into your biceps, your forearms relaxed, that relaxed feeling spreading down through your wrists and into your hands, all  the way  down to the tips of your fingers.  Very warm and very  relaxed. "Now  take a deep breath and hold i t .  Take a deep breath and study  the tension in your chest, and then let i t relax.  Once again taking a  deep breath and holding i t , study the tension in your chest, and then let i t relax.  Let your breathing become more and more regular, more and  more relaxed.  More relaxed with every breath.  "Now  tighten  up  your  tummy muscles,  abdomen, then let those muscles relax.  study the tension  in your  Once again, tensing the stomach  muscles, study the tension in the stomach, then let them relax, feeling that pleasant contrast between tension and relaxation. "Now  tighten up your buttock muscles, study the tension in your  buttocks, and then let them relax.  Once again, tighten up your buttock  muscles, study the tensions and let them relax.  Let that feeling of  deep relaxation spread down into your buttock muslces. "Now  tighten up your thighs, study the tension in your thighs, and  then let them relax.  Once again, tighten up your thighs, study the  tension in your thighs, then let them relax, go loose and limp and relaxed. "Now  point your toes toward your face, study the tension in your  126  lower legs, then let them relax.  Once again, pointing your toes towards  your face, study those tensions, and then let them relax. "Now point your toes away from your face, study the tension in your lower  legs and  your ankles, and  then let them relax.  Once again,  pointing your toes away from your face, study the tension in your ankles and your lower legs, then let them relax, feeling that pleasant contrast between tension and relaxation. "Now  curl up your toes, curl them up inside your shoes or whatever,  and study the tension in your feet and your toes, and then let them relax.  Once again, curl up your toes and study the tension in your feet  and your toes, and then let them relax.  Let that feeling of relaxation  flow down into your feet and down into your toes. "Now,  to help you  relax  even  further,  I'm  going to review the  different muscle groups that we've relaxed and as I mention each one, they will become even more relaxed than they are now. muscle group, i t w i l l  relax even  fingers  hands and  relaxed, your  As I mention each  further than i t already your  wrists  relaxed,  is....Your  your  forearms  relaxed, your biceps relaxed, and that relaxed feeling flowing up into your shoulders, along the back of your neck, your forehead becoming more and  more  smooth, more and  more relaxed,  relaxation  spreading down  through your face, let your eyes relax, your cheeks  and your mouth  relax, your jaws and your chin relax, the front of your neck relaxed, that relaxed feeling spreading down into your chest, your breathing more and  more regular, more and  more relaxed.  Relaxation spreading down  through your stomach, round the sides and up and down your spine, down into your hips and buttocks, flowing down into your thighs and your  127  calves down  relaxed, into  relaxation peaceful,  your  your  shins  feet,  coursing tranquil  and a n k l e s  a l l the  through  feeling  way  your  relaxed, down  veings,  of relaxation."  to  deep the  bathing  relaxation tips  of  your  flowing  your  toes,  whole  body,  128  APPENDIX H  129  Instructions To Participant On Use Of The Reconstructive Memory Technique  "In this part of the experiment, I'm going to ask you some questions about what happened last week.  Before I do this, I'd like you to spend  a few minutes recalling the circumstances surrounding the events of last week. We have found that reconstructing these circumstances can improve your memory of specific  details.  So I would like you now to begin  imagining yourself on the day of the experiment, and how you felt about being  in the experiment.  him/her to close them.)  (If student's  eyes  Think back to the day of the experiment.  Tell yourself what day of the week i t was....  remember  room...  tired....  Imagine how you got to  Think of what the weather was  at what time  experiment... the  day...  instruct  Picture yourself going back in time to when you  saw the videotape last week.  school that  not closed,  like...  of the day you were supposed  Try to  to go to the  Remember how you were feeling just before you went into Were  you  nervous....stressed....happy....or  maybe  Think of the room where you saw the videotape..... When you  can remember these details, I'd like you to nod your head (wait for head nod).  Good.  As you think about the room and remember how you were  feeling, i t w i l l begin to come back to you more and more clearly and you will be able to experience i t as though i t were happening right now; as though you were there right now and you were experiencing i t for the f i r s t time."  130  APPENDIX I  131  Instructions To Participant On Use Of The Imagery Memory Technique  "In this part of the experiment, you're going to be asked questions about what happened last week. image of an event can going to give  you  We  have found that forming a mental  improve your memory of specific  details.  some instructions that w i l l help you  I'm  to remember.  This technique involves (closing your eyes and, FOR AWAKE STUDENTS ONLY) imagining  what you  First, I'd like you  want to remember pictured on a television  to imagine that you are sitting in a comfortable  chair, maybe like the one you are sitting in now, you  is a T.V.  videotape  last  screen.  screen, perhaps like the one week.  Imagine yourself  and that in front of  on which you  turning  on  the  watched the screen.  Now  imagine you can see yourself coming to the lab today, walking with me down the stairs to the basement of this building, then down the hallway to this room. When you can see this on your T.V.  screen I'd like you to  nod your head (WAIT FOR HEAD NOD).  you could see what the  In this way,  walls in the hallway looked like, i f there were anything  on the walls,  and you could see the glass doors we came through to get to this room. This is how  I would like you to remember what happened last week in the  f i r s t part of this experiment.  Just imagine that you can see everything  that happened unfolding on the screen before you. you  can control how  fast or how  And you'll find that  slowly things move on the screen, so  that i f you wanted you could freeze the action on the screen to enlarge the picture to pick up any details you might otherwise have missed."  132  APPENDIX J  133  Table 1  Analysis of Variance For Percentage of Errors in Total Details Reported Interview Type X Memory Technique X Type of Event  Source  Sum of Squares  Interview Method Memory Technique Interview X Memory Technique Error Type of Event Interview X Event Type Memory Technique X Event Type  df  149.484  38.203  1  320.391  F  74.742  0.688  0.506  38.203  0.352  0.555  160.195  1.474  0.236  7171.500  66  108.659  89.063  1  89.063  1.183  0.281  4.875  0.065  0.937  13.031  0.173  0.679  523.688  6.955  0.002  9.750  13.031  1  Interview X Memory Technique X Event Type 1047-375 Error  Mean Squares  4969.875  66  75.301  134  Table 2  Analysis of Variance For Total Number of Descriptive & Sequential Errors Reported About Videotape Event Interview Method X Memory Technique  Source  Sum of Squares  df  Mean Squares  F  p  Interview Method  11.361  2  5.681  0.793  0.457  0.501  1  0.501  0.070  0.792  472.836  66  7.164  64.582  2  32.291  4.507  0.015  Memory Technique Error Interview X Memory Technique  135  Table 3  Analysis of Variance For Percentage of Errors In Details Recalled About Videotape Event Interview Method X Memory Technique  Source  Sum of Squares  Interview Method Memory Technique Interview X Memory Technique Error  df  Mean Squares  F  50.766  25.383  0.334  0.717  3.328  3.328  0.044  0.835  877.500  438.750  5.779  0.005  5010.453  66  75.916  136  Table 4  Analysis of Variance for Number of Descriptive Errors Reported About Main Bank Robber Interview Method X Memory Technique  Source  Sum of Squares  Interview Method  8.361  Memory Technique  0.347  Interview X Memory Technique Error  df  1  17.861 166.417  66  Mean Squares  F  4.181  1.658  0.198  0.347  0.138  0.712  8.931  3.542  0.035  2.521  137  Table 5  Analysis of Variance for Percentage of Errors of Descriptive Details Reported About Main Bank Robber Interview Method X Memory Technique  Source  Sum of Squares  Interview Method  df  Mean Squares  F  234.469  117.234  0.837  0.438  1.828  1.828  0.013  0.909  Interview X Memory Technique  1610.813  805.406  5.748  0.005  Error  9248.500  Memory Technique  66  140.129  138  Table 6  Analysis of Variance For Percentage of Errors In Descriptive Details About Bank Setting Interview Method X Memory Technique  Source  Sum of Squares  Interview Method  df  Mean Squares  F  2091.094  2  1045.547  0.925  0.401  278.766  1  278.766  0.247  0.621  Interview X Memory Technique  7623.422  2  3811.711  3.374  0.040  Error  74565.625  66  Memory Technique  1129.782  139  Table 7  Analysis of Variance For Number of Errors in Descriptions of Laboratory Room Interview Method X Memory Technique  Source  Sum of Squares  df  Mean Squares  F  p  Interview Method  0.194  2  0.097  0.109  0.897  0.500  1  0.500  0.563  0.456  58.667 6.583  66 2  0.889 3.292  3.703  0.030  Memory Technique Interview X Error Memory Technique  140  Table 8  Analysis of Variance For Amended Version of Field's Inventory of Hypnotic Depth Interview Method X Memory Technique  Source  Sum of Squares  Interview Method  334.750  Memory Technique  102.721  Interview X Memory Technique Error  df  1  45.861 1732.168  66  Mean Squares  F  167.375  6.377  0.010  102.721  3.914  0.050  22.931  .874  0.050  26.245  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0096304/manifest

Comment

Related Items