Sie sind auf Seite 1von 14

ACTPSY 1241 No.

of Pages 14, Model 5+

4 December 2007 Disk Used

Available online at


Acta Psychologica xxx (2007) xxx–xxx

2 Inducing and modulating intrusive emotional memories: A review

3 of the trauma film paradigm

4 Emily A. Holmes *, Corin Bourne

5 Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford OX3 7JX, UK
6 Received 20 October 2006; received in revised form 1 October 2007; accepted 12 November 2007

8 Abstract

9 Highly affect-laden memory intrusions are a feature of several psychological disorders with intrusive images of trauma especially asso-
10 ciated with post-traumatic stress disorder (PTSD). The trauma film paradigm provides a prospective experimental tool for investigating
11 analogue peri-traumatic cognitive mechanisms underlying intrusion development. We review several historical papers and some more
12 recent key studies that have used the trauma film paradigm. A heuristic diagram is presented, designed to simplify predictions about
13 analogue peri-traumatic processing and intrusion development, which can also be related to the processing elements of recent cognitive
14 models of PTSD. Results show intrusions can be induced in the laboratory and their frequency amplified/attenuated in line with predic-
15 tions. Successful manipulations include competing task type (visuospatial vs. verbal) and use of a cognitive coping strategy. Studies show
16 that spontaneous peri-traumatic dissociation also affects intrusion frequency although attempts to manipulate dissociation have failed. It

17 is hoped that further use of this paradigm may lead to prophylactic training for at risk groups and an improved understanding of intru-
18 sions across psychopathologies.
19 Ó 2007 Published by Elsevier B.V.

20 PsycINFO classification: 2340


21 Keywords: Intrusive memories; Trauma film; Post-traumatic stress disorder; Peri-traumatic processing; Imagery; Cognitive mechanisms; Stressful film;

22 Visuospatial

24 1. Introduction nent such as mental picture or sounds). This focus mirrors 36

both clinical phenomenology and also healthy autobio- 37

25 Intrusive memories, or intrusions, are involuntary recol- graphical memories for emotional events. Several studies 38
26 lections relating to events that appear, apparently sponta- have suggested that emotional memories typically take 39
27 neously, in consciousness (e.g. Brewin & Saunders, 2001; the form of mental images irrespective of whether such 40
28 Davies & Clark, 1998b; Halligan, Clark, & Ehlers, 2002; memories are intrusive or deliberately recalled (Arntz, de 41
29 Holmes, Brewin, & Hennessy, 2004; Schlagman, Kvavi- Groot, & Kindt, 2005; Conway, 2001) and, conversely, 42

30 lashvili, & Schulz, 2006). Intrusions can be contrasted with imagery seems to have a special impact on emotion 43
31 the deliberate recollection of events or repeated verbal (Holmes & Mathews, 2005). 44
32 rumination over such events. Whilst intrusions can take Intrusive memories occur often in everyday life with 45
33 the form of either sensory mental images or verbal studies in non-clinical populations suggesting that their fre- 46
34 thoughts our main interest is to understand mental imagery quency is approximately 2–4 a day (Berntsen, 1996) or 1–5 47
35 based intrusions (i.e. those which have a sensory compo- a day (Mace, 2005), although they occur less frequently 48
than verbal thoughts (Brewin, Christodoulides J., & 49
Corresponding author. Tel.: +44 (0) 1865 223 912; fax: +44 (0) 1865
Hutchinson, 1996). However, these common, unsolicited 50
223 948. recollections typically present no concern for the experienc- 51
E-mail address: (E.A. Holmes). er and can give rise to positive as well as negative affect. In 52

0001-6918/$ - see front matter Ó 2007 Published by Elsevier B.V.


Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

2 E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx

53 contrast, the vivid re-experiencing of excerpts from a trau- 1964; Lazarus & Opton, 1964; Lazarus, Opton, Nomikos, 108
54 matic event can be extremely distressing and form one of & Rankin, 1965; Speisman, Lazarus, Mordkoff, & Davi- 109
55 the three key symptoms for diagnosis of post-traumatic dis- son, 1964). These predominantly focussed on physiological 110
56 tress disorder (PTSD; American Psychological Association stress responses (heart rate and skin conductance) pro- 111
57 [APA], 1994). There is also considerable evidence implicat- duced by viewing the film and clearly demonstrated that 112
58 ing intrusive image-based memories in other psychological marked stress responses were inducible in the laboratory 113
59 disorders than PTSD such as: social phobia (Hackmann, by a variety of film stimuli. More crucially, these studies 114
60 Clark, & McManus, 2000; Hirsch, Clark, Mathews, & Wil- also showed that stress response severity could be experi- 115
61 liams, 2003); depression (Kuyken & Brewin, 1994, 1999; mentally altered by various manipulations such as: prior 116
62 Reynolds & Brewin, 1999); psychosis (Morrison et al., ‘‘cognitive orientation”, i.e. perceiving events as fictional 117

63 2002); agoraphobia (Day, Holmes, & Hackmann, 2004); or being emotionally detached from events (Lazarus 118
64 and cravings in substance misuse (Kavanagh, Andrade, & et al., 1965); assuming an involved or detached viewing 119

65 May, 2005). See Holmes and Hackmann (2004) for further stance (Koriat, Melkman, Averill, & Lazarus, 1972); or 120
66 examples. Steel, Fowler, and Holmes (2005) suggested that by utilising relaxation, desensitisation, and cognitive 121
67 similar cognitive information-processing mechanisms may rehearsal techniques (Folkins, Lawson, Opton, & Lazarus, 122
68 be involved in the creation of intrusive memories irrespec- 1968). 123
69 tive of disorder. Psychopathological intrusions can be A psychophysiological finding of Folkins et al. (1968) 124

70 viewed as an extension of a continuum from our common, seems particularly intriguing. Relative to a control condi- 125
71 everyday intrusions (see Holmes, 2004). tion, use of either relaxation or cognitive rehearsal (desen- 126
72 The factors that determine whether a memory becomes sitisation without the relaxation component) during the 127
73 intrusive need to be understood. The clinical literature indi- film reduced self-reported anxiety scores and skin conduc- 128
74 cates that peri-traumatic factors (i.e. processes during tance. However, such manipulations did not reduce heart 129
75 encoding of trauma), such as dissociation, are the best pre- rate response during film viewing as hypothesised by a tra- 130
76 dictors of later PTSD symptoms compared to other factors ditional ‘fight or flight’ arousal response to stress. As the 131
77 such as demographics or trauma type (see the meta-analy- authors note, a closer inspection of the data suggests that 132
78 sis by Ozer, Best, Lipsey, & Weiss, 2003). However, as the lack of a significant group difference in heart rate 133
79 argued by Candel and Merckelbach (2004) a limitation of may be due to heart rate reductions in the control group 134
80 many ‘‘peri-traumatic” clinical studies is heavy reliance immediately around the traumatic scenes contained in the 135

81 on retrospective reports of reactions during trauma. Such film. A finding of reduced heart rate, or bradycardia, coin- 136
82 methodology has important limitations since people in gen- cident with peaks in trauma content is also found in 137
83 eral, and PTSD patients in particular, find it difficult to Holmes et al. (2004), see Section 3.2. 138
84 give accurate descriptions of past emotional states. Pro- The trauma film paradigm was further developed by 139

85 spective designs are therefore warranted, however it is Horowitz and colleagues in the 1970s (e.g. Horowitz, 140
86 clearly unethical to deliberately expose research partici- 1969, 1975; Horowitz & Becker, 1971a, 1971b, 1971c, 141

87 pants to real trauma. To circumvent this, some studies have 1973; Horowitz, Becker, Moskowitz, & Rashid, 1972; 142
88 adopted ingenious paradigms such as testing trainee fire- Horowitz & Wildner, 1976). A major development was to 143
89 fighters prior to their exposure to a real fire (Bryant & consider the impact of films (with content depicting blood 144
90 Guthrie, 2005), or using analogues of high anxiety situa- and injury, bereavement and separation, or erotic scenes) 145
91 tions such as volunteer sky divers (Sterlini & Bryant, on the frequency of intrusive thoughts generated. Typically 146

92 2002). Another analogue approach, the trauma film para- intrusive thoughts were measured over short periods 147
93 digm, offering laboratory control, has emerged in the quest (within 5-min of the film ending) rather than the week-long 148
94 for prospective methodologies. diary methods typically used today (see Section 2). This 149
95 The trauma film paradigm involves showing non-clinical considerable body of work (for a review see Horowitz, 150
96 participants short films which contain scenes depicting 1975) systematically considered how intrusion frequency 151

97 stressful or traumatic events. In this context, a traumatic was affected by variables such as: nature of stimuli (film 152
98 event is defined as actual or threatened death or serious type and film repetition); sample populations (including 153
99 injury to the body or self (APA, 1994). Strictly speaking, psychiatric patients and servicemen with prior exposure 154
100 these films might best be referred to as ‘‘films with trau- to trauma); and cognitive processing instructions (to vary 155
101 matic content” since they do not necessarily induce an use of imagery, attention, or modify interpretation of the 156
102 ‘‘intense emotional reaction” as required by the diagnostic meaning of intrusions). In summary, Horowitz (1975) sug- 157
103 criteria for trauma. However, for brevity we use the term gested that the tendency to experience intrusive memories 158
104 ‘‘trauma films”. following a stressful event was a general one, present in 159
the population at large and expected to occur following 160
105 1.1. Historical perspective mild as well as severe stress events. 161
Butler, Wells, and Dewick (1995) extended the paradigm 162
106 The use of the trauma film paradigm was pioneered by by monitoring film-related intrusions for a week after film 163
107 Lazarus and colleagues in the 1960s (e.g. Lazarus & Alfert, viewing. They also provided verbal instructions requesting 164

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx 3

165 participants to modify their cognitive processing. This of the situation, organising information, and placing it in 220
166 instructional manipulation occurred in the 4 min after context) and towards data-driven or perceptual processing 221
167 watching the film and thus reflects post-trauma rather than (focussing on sensory impressions). It is this shift in pro- 222
168 peri-traumatic processing. A ‘‘worry” group was instructed cessing balance towards a relative increase in perceptual 223
169 to worry in verbal form about the film’s contents, whilst an processing that can be considered to be ‘faulty’. Similarly, 224
170 ‘‘imagery” group was asked to imagine in pictorial form the most recent version of DRT (Brewin, 2003) proposes 225
171 the distressing elements of the film. If just the first three that trauma can cause a neurophysiological shift in pro- 226
172 days’ intrusion data were examined then, as predicted, cessing balance away from standard, conscious, verbal pro- 227
173 the worry group reported more intrusions than the imagery cessing of information (‘‘verbally accessible memory”) in 228
174 and control (no-task) groups. However, over the full seven favour of a relative increase in sensory processing and stor- 229

175 day diary period there were no group differences in number age (‘‘situationally accessible memory”). Both theories pro- 230
176 of intrusions. pose that the ‘faulty’ trauma related processing leads to a 231

177 Davies and Clark (1998a) investigated how post-trauma lack of the context for a coherent time code, so that intru- 232
178 thought suppression might affect intrusive memories. Using sions can be perceived as being ‘relived’ or ‘happening 233
179 a Horowitz-type methodology, the study measured intru- again’. Broadly speaking, both models suggest that the 234
180 sive memories for only 4 min post-film. During the first shift in processing balance occurs due to an extreme emo- 235
181 2 min, participants were either asked to deliberately sup- tional response to the traumatic event itself. Emotion is 236

182 press thoughts pertaining to the film, or to think freely. specifically thought to promote perceptual memory encod- 237
183 Meanwhile, they recorded their film related intrusions by ing, as found in experimental results using slide stimuli with 238
184 pressing a ‘clicker’. In a second 2-min period, participants non-clinical participants (Arntz et al., 2005). Arntz et al. 239
185 continued to record their intrusions but both groups were also argue this might be related to the prominence of per- 240
186 instructed to think freely. Relative to the control group, ceptual memories in traumatic memory, such as intrusions 241
187 the suppression group reported a reduced number of film and nightmares. 242
188 related intrusions in the initial period but an increased We suggest that a simplified way to understand this 243
189 number in the second period. ‘faulty information processing’ shift may be to consider 244
the nature of a typical traumatic event: trauma events (or 245
190 1.2. Cognitive theories of PTSD and intrusive memories other very emotional events) can unfold very rapidly reduc- 246
ing time available for sufficient verbal or conceptual pro- 247

191 Although intrusive memories appear to be common- cessing (i.e. for the individual to make sense of the 248
192 place and not necessarily indicative of psychopathology, events). Simultaneously, the individual may focus more 249
193 in their more extreme form they are associated with several intently than usual on sensory or visuospatial information 250
194 psychological disorders. One example is the ‘‘flashbacks” as it may be especially valuable in the current survival con- 251

195 seen in PTSD, historically the ‘‘hallmark” intrusive image text (i.e. identifying potential escape routes) as well as pro- 252
196 disorder. In PTSD, intrusions can be extremely vivid, expe- viding a key learning experience for future dangerous 253

197 rienced as if the events are occurring again in the present. occasions. The trauma film paradigm allows investigation 254
198 (e.g. Ehlers, Hackmann, & Michael, 2004; Hackmann, of the variables that may trigger ‘faulty’ processing con- 255
199 Ehlers, Speckens, & Clark, 2004). Holmes, Grey, and temporaneously with encoding. This type of controlled 256
200 Young (2005) provide several examples of PTSD related analogue of clinical trauma can be used to test specific the- 257
201 intrusions including a patient caught in a railway carriage ory-driven predictions. 258

202 fire: ‘‘[I am] sitting with X trying to move his clothing so
203 he can breathe easily and his skin [is] coming away.” (p. 8). 2. Methodology 259
204 Contemporary theories of PTSD conceptualise intrusive
205 memories as instances of ‘faulty information processing’ 2.1. Method 260
206 (Brewin & Holmes, 2003) and thus place memory processes

207 and encoding mechanisms at the centre of PTSD aetiology The basic methodology involved in recent trauma film 261
208 and treatment (see also Conway, Meares, & Standart, 2004; studies is depicted in Fig. 1. Participants typically complete 262
209 Conway & Pleydell-Pearce, 2000). Specific clinical models a raft of baseline measurements (‘‘pre-film measures”) to 263
210 of this ‘faulty processing’ include Ehlers and Clark’s check for pre-existing vulnerabilities or trait biases, as well 264
211 (2000) cognitive theory of PTSD and the dual representa- as state levels of certain variables, before viewing a short 265
212 tion theory of PTSD (DRT; Brewin, 2001, 2003; Brewin, (8–12-min) film depicting traumatic events (e.g. scenes of 266
213 Dalgleish, & Joseph, 1996). While these theories differ in
214 other respects, in relation to intrusion formation they
215 appear to make similar predictions, as outlined by Holmes
216 et al. (2004) and Mathews and Macleod (2005).
217 Ehlers and Clark (2000) suggest that during trauma our
218 everyday balance of processing style is shifted away from
219 conceptual processing of events (focussing on the meaning Fig. 1. Basic procedure for trauma film paradigm. Q16

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

4 E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx

267 injury or death). The films used include industrial accidents duction. In general there have been two main types of 322
268 (Butler et al., 1995), Steil’s (1996) video of real-life road experimental manipulations: (1) providing instructions 323
269 traffic accidents (Brewin & Saunders, 2001; Hagenaars, regarding the form of processing to be employed or coping 324
270 van Minnen, Holmes, Brewin, & Hoogduin, submitted strategy to be used; or (2) undertaking contemporaneous 325
271 for publication; Halligan et al., 2002; Holmes & Steel, dual tasks whilst viewing the film. Experimental manipula- 326
272 2004; Holmes et al., 2004, Holmes, Oakley, Stuart, & Bre- tion studies can also incorporate ‘correlational’ elements. 327
273Q13 win, 2007; Stuart, Holmes, & Brewin, 2006), a fire safety
274 film depicting an office fire (Davies & Clark, 1998a, 2.3. Manipulations of peri-traumatic processing: an overview 328
275 1998b), and real-life footage of a patient with severe inju- of predictions 329
276 ries dying after treatment in a hospital emergency room

277 (Laposa & Alden, 2006). During the film participants can Peri-traumatic processing refers to processing that 330
278 either be asked to view the film as they would naturally occurs during a traumatic event, for which the trauma film 331

279 (‘‘no task”) or be given a ‘‘task”. Such ‘‘tasks” will be acts as an experimental analogue. As discussed, the two 332
280 described in more detail later, but include concurrent cog- main cognitive models of PTSD (Brewin et al., 1996; Ehlers 333
281 nitive tasks and instructions about the mode of processing & Clark, 2000) make similar predictions about peri-trau- 334
282 to adopt. After the film, state measures are repeated to matic processing and intrusion development. 335
283 assess the impact of film viewing and measures given to

284 check compliance with any experimental manipulations 2.3.1. Peri-traumatic processing in normal or control 336
285 (‘‘post-film measures”). Some studies include additional conditions 337
286 variables, such as physiological measures, during the film. A central tenet of both the clinical models of PTSD 338
287 Before leaving session one, participants are instructed on (Brewin et al., 1996; Ehlers & Clark, 2000) is that there 339
288 use of the intrusion diary and asked to record any intrusive are two forms of peri-traumatic cognitive processing that 340
289 memories of scenes from the trauma film spontaneously occur simultaneously for any given event: a verbal or con- 341
290 occurring over the coming week. Participants return a week ceptual form and a sensory visuospatial or perceptual 342
291 later to submit their diaries and complete further tests form. The latter type of processing can include sensory 343
292 (‘‘follow up measures”) such as recall for the film’s content information in any sensory modality (visuospatial, audi- 344
293 and estimation of compliance with completing the diary. tory, olfactory, etc.) though the focus with visual traumatic 345
294 To provide a check for demand characteristics, participants film stimuli is on visuospatial processing. For the sake of 346

295 may be asked to rate their predictions about the results in clarity, the terms verbal (meaning verbal and conceptual) 347
296 relationship to the experimental hypotheses. Prediction and visuospatial (meaning perceptual and sensory) will typ- 348
297 ratings can then be compared with participants’ actual ically be used for the remainder of the paper without imply- 349
298 performance. ing any preference for either cognitive model. This division 350

299 The diary methodology is crucial to the paradigm. In also reflects that used in cognitive models of working mem- 351
300 our studies, the weekly pen-and-paper intrusion diary pro- ory i.e. the visuospatial scratch pad and phonological loop 352

301 vides a reminder for participants as to what constitutes an (e.g. Andrade, Kavanagh, & Baddeley, 1997; Baddeley, 353
302 intrusive memory (i.e. spontaneous not deliberately 1986; Kemps, Tiggemann, Woods, & Soekov, 2004). 354
303 recalled, image based, etc.). Each day is separated into time Clinical models of PTSD propose that the relative bal- 355
304 periods (e.g. morning, afternoon, evening, night). The par- ance of verbal and visuospatial processing at encoding is 356
305 ticipant is asked to enter the number of intrusions experi- a major factor determining whether an event subsequently 357

306 enced for each period together with a description of the becomes intrusive. 358
307 content of each intrusion. This provides a check that Figs. 2a–c provide an overview of such clinical models 359
308 recorded intrusions indeed relate to the trauma film. It also of intrusion development and a heuristic for considering 360
309 allows intrusions to be matched to the corresponding film how dual tasks/mode of processing manipulations might 361
310 section which is useful for peri-traumatic heart rate analy- influence intrusion frequency. Fig. 2a depicts a scenario 362

311 ses or when using a within-subjects design (Holmes et al., where information from the film enters the cognitive sys- 363
312 2004, 2007; Stuart et al., 2006). Information is also col- tem and is processed simultaneously both verbally and 364
313 lected on the nature of the intrusion (image based, thought visuospatially. If there is a sufficient balance of verbal rel- 365
314 based, or both; typically only intrusions including imagery ative to visuospatial processing then the encoded events 366
315 are used in analyses). are unlikely to intrude. However, when emotionally trau- 367
matic events cause a relative increase in visuospatial com- 368
316 2.2. Designs pared to verbal processing, the likelihood that the 369
memory will intrude is increased. Fig. 2a thus illustrates 370
317 Some trauma film studies are correlational in design, a control condition in which ‘‘viewing the film as normal” 371
318 considering the relationship between pre-existing trait lev- produces a baseline number of intrusions. In comparison 372
319 els and subsequent frequency of intrusive memories. Other to this (no task) control condition, the subsequent dia- 373
320 studies attempt to deliberately manipulate cognitive grams (Figs. 2b and c) illustrate the impact of various 374
321 processes to investigate effects on intrusive memory pro- manipulations of peri-traumatic processing. 375

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx 5

trauma film without affecting the amount of verbal process- 396

ing, leaving the level of verbal processing now relatively 397
high. Overall, Fig. 2b represents an ‘‘attenuation of intru- 398
sions” condition which can be achieved via two alternative 399
approaches (increasing verbal-conceptual or decreasing vis- 400
uospatial processing). 401
It is noted that the competition for specific types of cog- 402
nitive resources approach aligns with the working memory 403
rationale forwarded by Baddeley and Andrade (2000) to 404
reduce the emotionality of non-clinical mental images. This 405

approach is also being used in other contexts such as the 406
role of EMDR in trauma therapy (Kavanagh, Freese, And- 407

rade, & May, 2001; Van den Hout, Muris, Salemink, & 408
Kindt, 2001), and the role of imagery and visuospatial 409
tasks in both substance misuse cravings (Kavanagh et al., 410
2005; May, Andrade, Pannaboke, & Kavanagh, 2004) 411
and food cravings (Kemps et al., 2004). 412

2.3.3. Peri-traumatic processing in ‘‘intrusion amplification” 413
conditions 414
Fig. 2c represents an ‘‘amplification of intrusions” con- 415
dition where the balance between the two forms of process- 416
ing is shifted in the opposite direction to that illustrated in 417
Fig. 2b (i.e. in the relative favour of visuospatial process- 418
ing) thereby leading to more intrusions than under control 419
conditions (Fig. 2a). Again, this shift can be generated via 420
two alternative methods: (1) by a relative increase in visu- 421
ospatial processing, depicted as Approach A (top arrow) 422

in Fig. 2c; or (2) by decreasing verbal processing resources 423

Fig. 2. (a) Control condition: intrusion frequency is determined by the allocated to the trauma film via a competition for resources 424
relative balance of verbal and visuospatial processing. (b) Attenuation rationale (depicted as Approach B, bottom arrow, in 425
condition: intrusions decreased by relative increase in Verbal processing. Fig. 2c). Again, irrespective of whether Approach A or B 426

(c) Amplification condition: intrusions increased by relative increase in is employed, our simplified model of PTSD predicts that 427
Visuospatial processing. (a–c) Heuristic diagram for experimental manip-
ulation of intrusive memories in the trauma film paradigm.
a shift in favour of visuospatial processing will lead to a 428

greater number of intrusions, as depicted in Fig. 2c, relative 429

to control conditions (Fig. 2a). 430
While not all studies employ all three experimental con- 431
376 2.3.2. Peri-traumatic processing in ‘‘intrusion attenuation” ditions depicted in Figs. 2a–c, these diagrams aim to pro- 432
377 conditions vide a heuristic representation of the peri-traumatic 433

378 Given that cognitive models of PTSD suggest that intru- mechanisms proposed to lie behind intrusion development 434
379 sive memories are generated by a relative shift towards vis- and how these mechanisms may be manipulated. An inter- 435
380 uospatial processing, it is hypothesised that if a processing esting feature of the overarching model is that we can 436
381 shift can be generated in the opposite direction (i.e. in derive predictions that would lead either an increase or a 437
382 favour of verbal processing) then this may protect against decrease in intrusions. Such bi-directional predictions pro- 438

383 intrusions. This leads to the prediction that if the film is vide ways to experimentally test the counterargument that 439
384 viewed under conditions which create a ‘‘pro-verbal” pro- any peri-trauma film task that might helpfully reduce intru- 440
385 cessing shift then a reduced number of intrusions (relative sions is ‘merely’ working due to distraction from the film. 441
386 to the control condition) would be experienced. This sce-
387 nario is illustrated in Fig. 2b, with the ‘‘pro- verbal” shift 3. Findings 442
388 being generated by either of two possible methods:
389 Approach A (top arrow) requires participants via an expli- 3.1. Types of study 443
390 cit instruction to increase the amount of verbal processing
391 they are undertaking. Alternatively, Approach B (bottom The 10 recent trauma film studies reviewed in this sec- 444
392 arrow) exploits a competition for resources rationale, and tion can be separated into three groups. The first group 445
393 uses a concurrent dual task to compete for visuospatial of studies comprise correlational analyses. These studies 446
394 resources (e.g. a spatial tapping task). Approach B reduces have considered the relationship between a range of indi- 447
395 the visuospatial processing resources allocated to the vidual differences (personality factors and cognitive traits) 448

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

6 E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx

449 and film intrusions. The second group has used experimen- task hypothesised to mimic the divided attention aspect 503
450 tal manipulations of cognitive processing during trauma of dissociation. Contrary to predictions, the dual task led 504
451 film viewing, thereby testing predictions from cognitive the- to significantly fewer intrusions than the control group. 505
452 ories of PTSD (as illustrated in Figs. 2a–c). A third group However, as no measure of dissociation was taken it is 506
453 of studies has specifically considered the role of dissocia- unclear whether the dual-task manipulation stimulated dis- 507
454 tion in intrusive memory production. Many studies incor- sociation as intended. Holmes (2000) argued that the tap- 508
455 porate elements from more than one of these ping task, rather than mimicking dissociation, instead 509
456 classifications but each study has been categorized based competed for visuospatial resources involved in intrusion 510
457 on the nature of its principle hypothesis. formation (see Fig. 2b) thereby reducing the number of 511
intrusions. This suggestion thus lead Holmes et al. (2004) 512

458 3.1.1. Correlational studies to consider thoroughly the effect of dual tasks on intrusive 513
459 Davies and Clark (1998b) used a correlational design to memory formation, as described later. 514

460 examine the relationship between frequency of intrusive Halligan et al. (2002) investigated the effect of cognitive 515
461 memories following a traumatic fire safety film and a variety processing style (following Ehlers & Clark, 2000) on intru- 516
462 of personality traits, cognitive processing styles and mood sion development, in Experiment 1 via an instructional 517
463 measures. Pre-film measures included neuroticism and manipulation and in Experiment 2 via groups divided by 518
464 extraversion (Eysenck Personality Questionnaire; Eysenck naturally occurring individual differences in information 519

465 & Eysenk, 1975), trait anxiety (State Trait Anxiety Inven- processing style. In Experiment 1, two groups were given 520
466 tory – STAI-T; Spielberger, Gorsuch, & Lushene, 1970), either ‘‘Conceptual” or ‘‘Perceptual” instructions for view- 521
467 depression (Beck Depression Inventory – BDI; Beck, Rush, ing the film. Pre-film measures included the State Anxiety 522
468 Shaw, & Emery, 1979), and specifically designed self-report Inventory (STAI-S; Spielberger et al., 1970) and a cognitive 523
469 scales concerning mental imagery, vulnerability to harm by processing questionnaire to assess perceptual and concep- 524
470 fire, use of thought suppression, and intrusion proneness. tual style processing (CPQ; developed by the authors). 525
471 Pre- and post-film mood ratings were also taken (happy, Post-film measures included the STAI-S and a retrospec- 526
472 anxious, depressed and angry). Results showed that the tive state version of the CPQ. The Conceptual group was 527
473 number of intrusive memories experienced in the week fol- instructed to watch the film whilst focusing on the story, 528
474 lowing the trauma film was predicted by: (i) increase in following what was happening to the people and why, 529
475 anger across the film; (ii) self-rating for vulnerability to and thinking about what might happen next. This group 530

476 harm by fire; and (iii) the interaction between these two. was thus encouraged to increase the relative amount of ver- 531
477 These three variables accounted for 30% of the variance bal processing of the film as illustrated in Fig. 2b, 532
478 in intrusive memories frequency and no other variable Approach A. The Perceptual group was instructed to 533
479 reached significance. watch the film whilst being absorbed in the images and 534

480 Interested in intrusions in psychosis as well as PTSD, sounds, viewing the scenes as unconnected snapshots. They 535
481 Holmes and Steel (2004) used a correlational design to were therefore requested to increase the relative amount of 536

482 investigate intrusions in relation to schizotypy (Oxford- visuospatial processing of the film (Fig. 2c, Approach A). 537
483 Liverpool Inventory of Feelings and Experiences – O-LIFE; Both groups experienced significant increases in state anx- 538
484 Mason, Claridge, & Jackson, 1995), trait dissociation iety from pre- to post-film and the experimental manipula- 539
485 (Dissociative Experiences Scale – DES-II; Carlson & Put- tion of processing style appeared successful. Contrary to 540
486 nam, 1993) and peri-traumatic dissociation (Peritraumatic predictions there was no significant difference in intrusions 541

487 Dissociative Experiences Questionnaire – PDEQ; Marmar, between the groups. The authors noted a potential con- 542
488 Weiss, & Metzler, 1997). Results showed a significant posi- found in the processing style manipulation; individual’s 543
489 tive relationship between positive symptom schizotypy innate processing bias may not have been adequately coun- 544
490 (Unusual Experiences subscale of O-LIFE) and number of termanded by the task instructions. Indeed, with data col- 545
491 trauma-related intrusions. Trait dissociation was a univari- lapsed across groups, intrusion frequency showed a 546

492 ate predictor of intrusions but became non-significant with positive correlation with perceptual processing and a nega- 547
493 the inclusion of schizotypy. PDEQ correlated with the tive correlation with conceptual processing. 548
494 Cognitive Disorganisation subscale of O-LIFE but did In Halligan et al. (2002) Experiment 2, a pre-film trait 549
495 not predict intrusions. These findings indicate that trait dis- version of the CPQ was used to divide participants into 550
496 sociation may be a mediating mechanism by which high ‘‘conceptual” or ‘‘perceptual” information processing style 551
497 schizotypes are vulnerable to traumatic intrusions. groups (selecting upper quartile scores). Other pre-film 552
measures included the STAI-T, STAI-S and trait dissocia- 553
498 3.2. Experimental manipulation of peri-traumatic cognitive tion (TDQ; Murray, Ehlers, & Mayou, 2002) with the 554
499 processing STAT-S repeated post-film. Follow-up measures, in addi- 555
tion to intrusions, included other analogue PTSD symp- 556
500 Brewin and Saunders (2001) attempted to investigate the toms (fear, avoidance and arousal) using a purpose- 557
501 effect of peri-traumatic dissociation on intrusions. They designed questionnaire. As predicted, the Perceptual group 558
502 compared a no task condition with a concurrent tapping experienced significantly more intrusions than the Concep- 559

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx 7

560 tual group. The Perceptual group also found their intru- As predicted (see Fig. 2b, Approach B), compared to the 617
561 sions more distressing and reported more fear, avoidance control group, there were fewer intrusions in both the Vis- 618
562 and arousal symptoms. Interestingly, the perceptual group uospatial and the Overpracticed group, but not in the Sin- 619
563 had significantly higher scores of pre-film state anxiety, gle Key group. There was a significant linear trend between 620
564 trait anxiety, and trait dissociation although these con- visuospatial task demand and intrusive memory frequency. 621
565 founds were controlled for in the analysis. Although Exper- Again, spontaneous increases in state dissociation across 622
566 iment 2 did not use an experimental manipulation, the groups were associated with intrusions. These results sup- 623
567 study can still be interpreted within Figs. 2b and c as the port the hypothesis that it is the visuospatial nature of 624
568 pre-selection of individuals with naturally extreme process- the task that may be effective in reducing intrusions and 625
569 ing styles represents an alternative approach to generating that the effectiveness of that attenuation is proportional 626

570 shifts in processing balance. This study also supports the to the cognitive load of the visuospatial task. 627
571 external validity of trauma films as an analogue of PTSD Experiment 3 in Holmes et al. (2004) compared two ver- 628

572 since, in addition to intrusions, other analogue symptoms bal dual tasks with a no task condition. A ‘‘Verbal Interfer- 629
573 of PTSD were induced. ence” group was required to count backwards in threes 630
574 In a series of experiments, Holmes et al. (2004) examined whilst viewing the film. This task utilises verbal compo- 631
575 the effect of various dual tasks (during a trauma film) on nents of working memory (Vallar & Baddeley, 1982). Fol- 632
576 intrusive memory formation. Cognitive tasks were used lowing clinical theories, verbal processing is needed to 633

577 to actively modify processing style rather than instructing helpfully process trauma conceptually. The verbal disrup- 634
578 participants to deliberately shift their processing style. In tion/interference task was therefore predicted to increase 635
579 Holmes et al. Experiment 1, two experimental manipula- intrusions (see Fig. 2c, Approach B). In contrast, a ‘‘Verbal 636
580 tion groups were contrasted with a no-task control condi- Enhancement” group was required to describe aloud 637
581 tion. A ‘‘Visuospatial” group was required to tap details of the scenes whilst they viewed the film with the 638
582 repeatedly a specified sequence of five keys on a 5  5 key- expectation that such helpful verbal description would 639
583 pad without looking (after only 1-min of practice) through- enhance verbal processing of the traumatic scenes rather 640
584 out the film (see Fig. 2b, Approach B). This task was the than divert verbal processing resources. ‘‘Verbal Enhance- 641
585 same as that used by Brewin and Saunders (2001) but in ment” was predicted to reduce intrusions (see Fig. 2b, 642
586 this study was hypothesised to utilise visuospatial resources Approach A). As predicted, the Verbal Interference group 643
587 via recalling and repeating the spatial pattern. A ‘‘Dissoci- reported significantly more intrusions than the Control 644

588 ation” group was required to stare at a small dot on the group. Crucially, this result also indicates that the previous 645
589 film screen during the film (Leonard, Telch, & Harrington, results using the visuospatial task cannot merely be due to 646
590 1999). Measures given pre- and post-film included state dis- ‘‘distraction” away from the film – as this counting task is 647
591 sociation (sub-scale of the Clinician Administered Disso- also ‘‘distracting” but led to increased rather than reduced 648

592 ciative States Scale – DSS; Bremner et al., 1998). The key intrusion levels. However, the Verbal Enhancement group 649
593 result was that, as predicted, the Visuospatial group did not experience the predicted reduction in intrusions. 650

594 reported significantly fewer intrusive memories than other Transcripts of the verbalisations indicated that they pre- 651
595 groups (see Fig. 2b, Approach B). Although the dissocia- dominantly consisted of surface level descriptions of the 652
596 tion manipulation gave the highest change in DSS scores, scenes rather than the emotional meaning of the film. Per- 653
597 contrary to prediction this led to no more intrusions than haps this task was unable to recruit the verbal conceptual 654
598 the control condition. However, across all groups com- processing of the type needed to counteract intrusions 655

599 bined, spontaneous increases in state dissociation over (however, see Laposa and Allen, 2005, for an alternative Q2 656
600 the film were positively correlated with intrusions (even methodology). 657
601 when controlling for experimental condition and trait dis- All experiments in Holmes et al. (2004) measured heart 658
602 sociation). It is possible that such artificial dissociation rate using a blood flow optical sensor. Baseline heart rate 659
603 tasks do not provide an adequate analogue of natural, was calculated over a 6-min rest period pre- film. Peri-trau- 660

604 spontaneous dissociative episodes. matic heart rate was calculated over the 12.5-min film dura- 661
605 Holmes et al. (2004) Experiment 2 investigated the tion, yielding an index of film-induced heart rate change. 662
606 potential active ingredients in the visuospatial dual task. As discussed, participants described the content of each 663
607 Three forms of tapping task were compared with no task: intrusion in their diaries. The experimenter later matched 664
608 standard Visuospatial tapping (as in Experiment 1); Over- diary intrusion descriptions to actual sequences in the film 665
609 practiced visuospatial tapping (increasing pattern practice (e.g. a scene of a fireman carrying a baby). Heart rate 666
610 to 6-min); and Single key tapping i.e. repeatedly pressing recordings were synchronised with the film, allowing a 667
611 one key. The ‘‘Overpracticed” condition was used to mean ‘intrusion-sequence’ heart rate to be calculated per 668
612 reduce the general cognitive load required by task perfor- participant and compared to non-intruding sections. 669
613 mance relative to the Visuospatial group, while still using Experiments 1 and 2 found that peri-traumatic heart rate 670
614 visuospatial resources. The ‘‘Single Key” condition decrease was significantly correlated to increased number 671
615 reduced visuospatial load, checking that any effect of the of intrusions. Additionally, both Experiments 1 and 2 672
616 visuospatial task was not due to simple tapping per se. found that intrusion-sequence heart rate was significantly 673

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

8 E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx

674 lower (by 1.6 and 1.9 bpm, respectively) compared to non- traumatic experience (PDS; Foa, 1995), or intelligence 731
675 intruding film sequences. However, any differences in (WPT; Wonderlic, 1999). A manipulation check confirmed 732
676 Experiment 3 did not reach significance. The significant that the Coping group reported using the required coping 733
677 findings appear to be in conflict with a stress-arousal view strategy more than Controls. As predicted, the Coping 734
678 of the trauma film (see Van Stegeren, this issue; Wolf, this group experienced significantly fewer intrusions than Con- 735
679 issue) but may be consistent with a ‘freeze and surrender’ trols. Whilst this study did not set out especially to manip- 736
680 or orientating response to threat stimuli and trauma (e.g. ulate the balance of processing as proposed by cognitive 737
681 Nijenhuis, Vanderlinden, & Spinhoven, 1998; see also theories of PTSD (Brewin et al., 1996; Ehlers & Clark 738
682 Campbell, Wood, & McBride, 1997; Lang, Bradley, & 2000), the authors note that this manipulation ‘‘echoes” 739
683 Cuthbert, 1997). Interestingly, the results are comparable the conceptual processing results of Halligan et al. (2002). 740

684 to Folkins et al. (1968) which found evidence of bradycar- We assert that the cognitive coping strategies adopted 741
685 dia in the control group immediately around the traumatic may have specifically worked by boosting verbal/concep- 742

686 scenes contained in the film (see Section 1.1). Further inves- tual processing of the film (i.e. Approach A in Fig. 2b). 743
687 tigation is warranted. This suggestion is supported by the finding that scores on 744
688 Q4 Stuart et al. (2005) modified the trauma paradigm to use Halligan’s et al. (2002) Conceptual Processing Question- 745
689 within-subjects rather than between-subjects conditions. A naire (CPQ) were strongly associated with effective use of 746
690 within-subjects design is made possible by again linking the coping strategy. That experimental support was pro- 747

691 intrusion descriptions in the diary to particular segments vided for the effectiveness for coping strategies derived 748
692 of the film (as described in Section 2.1). The experimenter from real-life trauma situations (see Study 1) also lends fur- 749
693 can thus retrospectively determine the number of intrusions ther support for the external validity of the trauma film 750
694 associated with each condition (marked by portion of the paradigm. Finally, it highlights the potential for the trauma 751
695 film). The experiment sought to test the impact of an alter- film paradigm to test experimentally prophylactic treat- 752
696 native, clinically relevant, visuospatial task (see Fig. 2b, ments to prevent PTSD in at risk groups such as emergency 753
697 Approach B) – clay modeling (of pyramids and cubes) – services personnel. 754
698 compared to a no-task control condition (see Fig. 2a). Clay
699 modeling is sometimes used to ‘ground’ traumatised clients 3.3. Experimental studies of peri-traumatic dissociation 755
700 to reduce dissociation during psychological therapy.
701 Results showed that participants experienced significantly Much of the evidence for peri-traumatic dissociation 756

702 fewer intrusions from the section of the film during which being a predictor of later PTSD diagnosis is based upon 757
703 they were modeling than during the no-task sections (the retrospective self-report data for the traumatic event, even 758
704 two sections of film were matched for intrusion production if the period between event experience and self-report is 759
705 potential using data from Holmes et al. (2004) and were becoming increasingly short (e.g. Engelhard, van den 760

706 counterbalanced between participants). Interestingly, the Hout, Kindt, Arntz, & Schouten, 2003; Koopman, Classen, 761
707 reduction in intrusions was not accounted for by post-film & Spiegel, 1994; Murray et al., 2000; Shalev, Peri, Canetti, Q11762

708 distress or peri-traumatic dissociation (PDEQ; Marmar & Schreiber, 1996; Ursano et al., 1999). Indeed, Ozer et al. 763
709 et al., 1997). (2003) showed in a meta-analysis that one of the strongest 764
710 A further use of the trauma film paradigm is provided predictors for a subsequent PTSD diagnosis following a 765
711Q10 by Laposa and Alden (2005, Study 2). In Study 1, Laposa traumatic event was retrospective, self-reported peri-trau- 766
712 and Alden undertook structured interviews of acute care matic dissociation. However, as argued by Candel and 767

713 health workers to provide an analysis of cognitive coping Merckelbach (2004) we also need prospective evidence. 768
714 strategies used in traumatic situations. The most effective The trauma film paradigm has been used to prospectively 769
715 strategies were used in Study 2, the trauma film study, to investigate the role of dissociation in intrusion develop- 770
716 experimentally test their effectiveness in reducing intrusion ment, though attempts to date appear to be frustratingly 771
717 relative to a no-strategy control group. Such strategies inconclusive. 772

718 included: directing attention to mechanical steps of medical Holmes et al. (2004, Expt. 1) found no effect of an active 773
719 treatment; focusing on events and processes occurring in manipulation of dissociation task (dot staring) on intru- 774
720 the ‘here and now’; and recalling prior training and apply- sions, yet found intrusions to be associated with spontane- 775
721 ing it to solve medical problems. ous increases in film-induced state dissociation. This was 776
722 In Study 2, the no-strategy Control group was told sim- replicated in Experiment 2 but not Experiment 3. One 777
723 ply to watch the video of real events in a hospital emer- possibility is that dissociation is a complex construction 778
724 gency room, whilst the Coping group was given requiring a more detailed and precise definition. Holmes 779
725 instructions such as: to focus on the medical procedures et al. (2005) therefore suggested that there are two dis- 780
726 being used by the medical staff; and analyse what the staff tinct types of dissociation: detachment (i.e. being detached 781
727 are trying to accomplish. The groups did not differ on pre- from everyday experience e.g. feeling ‘spaced-out’); and 782
728 film measures of trait anxiety (STAI-T; Spielberger et al., compartmentalisation (i.e. deficit in control of processes 783
729 1970), depression (BDI-II; Beck, Steer, & Brown, 1996), or actions via a division or compartmentalisation of 784
730 trait dissociation (DES-II; Carlson & Putnam, 1993), prior functions). 785

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx 9

786 Holmes et al. (2007) used a within-subjects design (sim- recollections of perceptual aspects of the identified scene. 843
787 ilar to Stuart et al., 2005) to contrast active manipulations Memory fragmentation was assessed via visual analogue 844
788 of detachment dissociation via hypnosis. All participants scales ratings for how fragmented or ‘‘snap-shot” like 845
789 were pre-screened for hypnotic susceptibility (using the memories were. Participants completed the PDEQ (Mar- 846
790 Harvard Group Scale of Hypnotic Susceptibility; Shor & mar et al., 1997). Scores on the memory disturbance tests 847
791 Orne, 1962) and viewed the entire film whilst hypnotised. were not predicted by peri-traumatic dissociation (PDQ). 848
792 For the ‘‘control” condition section of the film participants However, memory fragmentation and short-term intrusion 849
793 were instructed, under hypnosis, to view the film normally frequency correlated with PDEQ score but not to a trait 850
794 and from their own perspective. In the ‘‘Dissociation” con- measure of dissociation (DES; Bernstein & Putnam, 851
795 dition section, participants were given suggestions to view 1986). This study not only provides further evidence for 852

796 the film as if they were disconnected from, or outside of, the role of spontaneous dissociation in intrusion formation 853
797 their bodies. The trauma film significantly increased state but extends the effect of dissociation to problems of mem- 854

798 dissociation (DSS; Bremner et al., 1998) and this increase ory coherence – another key phenomenon of PTSD. 855
799 was greater for the Dissociation condition. However, con-
800 trary to prediction, there were no more intrusions in the 4. Discussion 856
801 Dissociation condition than the control condition. It is
802 possible that detachment dissociation per se is not involved Taken together, the studies discussed above show that 857

803 in intrusive memory formation. However, perhaps sponta- the trauma film paradigm can induce negative mood, dis- 858
804 neous detachment dissociation has some qualities not tress, dissociation, and intrusive memories for film content. 859
805 captured by the hypnotically suggested detachment In addition to looking at intrusions, some studies (e.g. Bre- 860
806 manipulation. win & Saunders, 2001; Halligan et al., 2002; Laposa & 861
807 In contrast, Hagenaars et al. (submitted for publication) Alden, 2006) have shown that the trauma film paradigm 862
808 considered compartmentalisation dissociation and its role can induce other analogue PTSD-like symptoms (e.g. fear, 863
809 in intrusive memory formation. Again, hypnotic suggestion avoidance, and arousal). These findings support the 864
810 was used to induce the desired form of dissociation, in this assumption that the trauma film paradigm provides a use- 865
811 case full body catalepsy (Dissociative Non-Movement or ful analogue to real life trauma. Interestingly, there is still 866
812 DNM group). Two other groups were used, a non-hyp- some debate as to whether PTSD can actually be produced 867
813 notic group that was instructed not to move intentionally from viewing traumatic events via television or video rather 868

814 (NM group) and a non-hypnotic, non-restricted group than first person experience, especially if there is no direct 869
815 (Controls). The groups did not differ in trait dissociation relationship between viewer and some of the victims (e.g. 870
816 (DES-II; Carlson & Putnam, 1993), pre-film state dissocia- Holmes, Creswell, O’Connor, & Saunders, 2003; Pfeffer- 871
817 tion (DSS; Bremner et al., 1998), or peri-traumatic dissoci- baum, Pfefferbaum, North, & Neas, 2002). Further, Mol 872

818 ation (PDEQ; Marmar et al., 1997). The effectiveness of the et al. (2005) note that many individuals meet diagnostic lev- 873
819 compartmentalisation manipulation was supported by a els for PTSD like symptoms without having experienced an 874

820 significantly higher score for the DNM group, relative to event that meets current diagnostic criteria for being ‘‘trau- 875
821 the other groups, on the Catalepsy Questionnaire (CQ; matic” (Criterion A; APA, 1994). It seems possible that 876
822 Hagenaars, Roelofs, Hoogduin, & van Minnen, 2006) intrusions may develop on a continuum of stressfulness. 877
823 and the Somatafrom Dissociation Questionnaire – Peri- Such a continuum may range from, say, viewing fictitious 878
824 traumatic (SDQ-P; Nijenhuis, van Engen, Kusters, & van stressful films, through viewing films of real and self-related 879

825 der Hart, 2001). The two non-movement groups combined trauma, to the spectrum of ‘‘true” traumatic events 880
826 reported significantly more intrusions than Controls. How- (Holmes, 2004). Accordingly, it is clear that ethical consid- 881
827 ever, there was no significant difference between the DNM erations are especially important when using this paradigm. 882
828 group relative to the other two groups combined. These Such ethical safeguards include: non-inclusion of partici- 883
829 results suggest that somatoform compartmentalisation pants with mental health difficulties; clear information to 884

830 per se may not be involved in intrusive memory formation participants about film content prior to their participation; 885
831 but that non-movement may increase intrusion frequency. use of precautionary measures to deal with potentially dis- 886
832 This later finding may relate to ‘‘freezing” during trauma tressed participants (most studies have been conducted 887
833 (e.g. Nijenhuis et al., 1998). under the guidance of clinical psychologists); and provision 888
834 A further use of the trauma film paradigm is provided of contact details to participants in the event of any con- 889
835 by Kindt, van den Hout, and Buck (2005). Memory frag- cerns even after the study has ended. 890
836 mentation and disturbances for trauma film scenes, as well The findings of the reviewed studies, taken as a whole, 891
837 as self-rated frequency of intrusions, was examined over a appear consistent with the heuristic diagram we presented 892
838 4-h period following film viewing. Memory disturbances in Figs. 2a–c. This figure illustrates a simplified account of 893
839 were assessed via a sequential test, where five scenes from the cognitive processes proposed to underlie intrusive mem- 894
840 the film were shown in random order and had to placed ory formation common to recent cognitive models of PTSD 895
841 in sequential order; and a perceptual memory test, where (Brewin et al., 1996, 2001; Ehlers & Clark, 2000). We sug- Q12896
842 participants responded to scene descriptions with detailed gest that in using the Trauma film paradigm, intrusive mem- 897

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

10 E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx

898 ory formation has been shown to be modified by all of the extracted specific methods for experimental participants 955
899 four alternative analogue peri-traumatic approaches shown to employ. These more precise instructions successfully 956
900 in Figs. 2b and c: (1) Fig. 2b, Approach A; Both Laposa and had the predicted impact on intrusions. This well executed 957
901 Alden (2006, Study 2) and Halligan et al. (2002, Expt. 2) study may indicate the benefit to future work of thoroughly 958
902 provide evidence that a relative increase in helpful verbal delineating variables of clinical interest in target clinical 959
903 processing, either via coping strategies or high trait concep- populations before testing them in the laboratory. 960
904 tual processing, is associated with a relative reduction in However, one peri-traumatic factor whose role in intru- 961
905 intrusions; (2) Fig. 2b, Approach B; sion formation remains relatively unclear is that of dissoci- 962
906 Holmes et al. (2004, Expts. 1 and 2) and Stuart et al. ation. Ozer’s et al. (2003) meta-analysis of 68 clinical 963
907 (2006) provide evidence that a visuospatial dual task trauma studies suggested that the strongest predictor for 964

908 reduces intrusions, in line with a competition of visuospa- a subsequent PTSD diagnosis is level of peri-traumatic dis- 965
909 tial cognitive resources rationale; (3) Fig. 2c, Approach sociation. Despite this persuasive clinical link, no trauma 966

910 A; Halligan et al. (2002, Expt. 2) provides evidence that film study has yet been able to manipulate successfully 967
911 high trait visuospatial/perceptual processing may be intrusions by experimentally altering level of dissociation 968
912 related to increased intrusions; and (4) Fig. 2c, Approach and findings seem rather mixed. For example, Holmes 969
913 B; Holmes et al. (2004, Expt. 3) provide evidence that use et al. (2004) found that film-induced spontaneous increases 970
914 of a verbal dual task (competing with ‘helpful’ conceptual in dissociation were correlated with intrusions in Experi- 971

915 processing) rather than being distracting actually increases ments 1 and 2 (although not Experiment 3) as did Kindt 972
916 intrusive memory formation. This latter finding has been et al. (2005). However, Holmes and Steel (2004) found that 973
917 recently replicated in our lab, although current work by a correlation between trait dissociation and intrusions 974
918 Krans, Holmes, Näring, and Becker (2006), with some became non-significant when controlling for schizotypy. 975
919 methodological differences, failed to re-replicate the bi- Halligan et al. (2002) found that whilst trait dissociation 976
920 directional effect, and further work is clearly needed. was associated with intrusions, this relationship was not 977
921 It is noted that whilst the data appears to be consistent significant when controlling for cognitive processing style. 978
922 with the heuristic (derived from contemporary clinical Hagenaars, van Minnen et al.’s (submitted for publication) 979
923 models) presented in Figs. 2a–c, this may not rule out alter- results indicate that compartmentalisation dissociation 980
924 native explanations of the data. Work by Pearson, Sawyer, may be mediating intrusion development via non-move- 981
925 and Holmes (in press) suggests that the modulation of ment rather than dissociation per se. A further alternative 982

926 intrusions using dual tasks may be more strongly associ- is that peri-traumatic dissociation is a correlate of, or mar- 983
927 ated with general attentional load than task modality per ker for, a further psychological or physiological process 984
928 se. In two experiments, Pearson et al. (in press) examined (such as non-movement) which itself is the cause of intru- 985
929 the impact of verbal, visuospatial, and executive concur- sion formation. This would leave peri-traumatic dissocia- 986

930 rent tasks on the occurrence of memory intrusions for posi- tion being highly correlated with intrusions, as is found 987
931 tive and negative stimuli selected from the International clinically, but without a causal role. This would explain 988

932 Affective Picture System (IAPS; Lang, Ohmann, & Vaitl, the null findings of studies attempting to manipulate intru- 989
933 1988). This material involves static slide images rather than sions via inducing dissociation. Clearly work is still 990
934 film presentation. Results indicate that a significant reduc- required to tease apart the role of peri-traumatic 991
935 tion in the occurrence of intrusions during a one-week per- dissociation. 992
936 iod following initial exposure was more strongly associated Other areas of interest raised by the reviewed studies 993

937 with the general attentional load of concurrent tasks than include the psychophysiological investigation of intrusions 994
938 their modality (either verbal or visuospatial). However, a and use of the paradigm to study memory related phenom- 995
939 pure attentional load approach may have difficulty explain- enology of PTSD other than intrusions. Folkins et al. 996
940 ing the bi-directional impact of verbal and visuospatial task (1968) and Holmes et al. (2004) point to trauma-induced 997
941 shown in Holmes et al. (2004). Clearly, further empirical instances of bradycardia. Further work is required to inves- 998

942 work is required to resolve this intriguing debate and to tigate if this spontaneous bradycardia may be related to an 999
943 explore the interface between clinical and experimental orienting or freezing response to intense trauma (e.g. Nijen- 1000
944 accounts of intrusions. huis et al., 1998), or could be a physiological correlate of 1001
945 Even if the theoretical assumptions in Figs. 2a–c prove some other change such as balance of cognitive processing 1002
946 to be correct, there is still a methodological concern over styles. It will be interesting to employ the paradigm to 1003
947 the ability for a balance of processing style to be altered investigate neural substrates of PTSD using neuroimaging 1004
948 via explicit instructions (rather than a task). Halligan techniques (c.f. Elzinga & Bremner, 2002). While some 1005
949 et al. (2002, Expt. 1) and Holmes et al. (2004, Expt. 3) studies (e.g. Halligan et al., 2002; Kindt et al., 2005; Laposa 1006
950 found that instructions intended to facilitate deliberate & Alden, 2006) have used the trauma film paradigm to 1007
951 attempts by participants to consciously control their pro- investigate memory disorganisation and fragmentation, 1008
952 cessing style did not have an effect on intrusions. However, further work is needed in this area. Another area for refine- 1009
953 following a systematic analysis of strategies used by emer- ment might involve diary methodology – we rely on explicit 1010
954 gency personnel, Laposa and Alden (2006) carefully recognition of an intrusion as related to the film, which 1011

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx 11

1012 may be difficult in some cases (e.g. in re-experiencing with- that may be relevant at memory encoding in order for 1069
1013 out explicit awareness or where there are source monitor- intrusive memories to be formed. Considerable further 1070
1014 ing errors, Steel et al., 2005). Other methods to access work using the paradigm is required to advance the under- 107
1015 other types of trauma memories (which may not always standing, prevention, and treatment of post-traumatic 1072
1016 be recognized) will be valuable. symptoms. As we have argued elsewhere (Holmes & Hack- 1073
1017 The trauma film paradigm also offers a smörgåsbord of mann, 2004) affect laden intrusions are a feature not only 1074
1018 possibilities for investigating non-peri-traumatic factors of PTSD but also of many other disorders such as social 1075
1019 involved in intrusive memories. For example, ongoing phobia (Hackmann et al., 2000; Hirsch et al., 2003), depres- 1076
1020 work by Wessel (2006) is investigating the role of deficient sion (Kuyken & Brewin, 1994, 1999; Reynolds & Brewin, 1077
1021 cognitive control in intrusive memory production. Nixon 1999); psychosis (Morrison et al., 2002); and agoraphobia 1078

1022 (2006) is currently using trauma films depicting physical (Day et al., 2004). Given the proposed role of intrusions 1079
1023 and sexual violence to investigate the role of post-event in the maintenance of these disorders, then the trauma film 1080

1024 processing in intrusions. Early findings suggest that factors paradigm could perhaps be adapted for the concerns and 108
1025 such as a participant’s interpretations of experiencing features of a given disorder. This should provide another 1082
1026 intrusions may modulate intrusion frequency. We suggest tool in experimental psychopathology by which to under- 1083
1027 this finding may echo work by Orne (1962) which showed stand distressing intrusive imagery across psychological 1084
1028 that if participants are told that intrusions are evidence disorders. 1085

1029 of psychopathology, then intrusion frequency increases.
1030 Nevertheless, the trauma film paradigm points to at 5. Uncited references 1086
1031 least one clear and valuable finding: namely, that intrusions
1032 can be reduced by peri-traumatic interventions. The results Bremner et al. (1992), Bryant and Harvey (1995), Bywa- 1087
1033 Q3 of Laposa and Alden (2004), Holmes et al. (2004), and Stu- ters et al. (2004). Q1 1088
1034 art et al. (2005) among others, suggest clear potential for
1035 prophylactic measures for at risk groups. Laposa and
1036 Alden show that pre-training individuals with coping strat- Acknowledgements 1089
1037 egies appropriate for members of the emergency services
1038 could well reduce trauma induced symptoms. Additionally, This paper was partially supported by a Royal Society 1090
1039 following Holmes et al., it may prove that contemporane- Dorothy Hodgkin Fellowship and an Economic and Social 109

1040 ous visuospatial tasks (provided such activity did not inter- Research Council (ESRC) Grant Ref RES-061-23-0030 to 1092
1041 fere with emergency personnel’s principle function), may Emily A. Holmes and a Medical Research Council Stu- 1093
1042 also be protective, (see, for example, on line article by Hirs- dentship to Corin Bourne. We are grateful to Chris Brewin 1094
1043 hon, 2004). As many traumas can not be anticipated, it is for his support. 1095

1044 also important to test any effectiveness of such intervention

1045 not only ‘‘peri-traumatically” but in the aftermath of trau- References 1096

1046 matic events, beginning to be explored in our lab’s current

1047 studies (Holmes, 2007). In the context of one of the evi- American Psychiatric Association (1994). Diagnostic and statistical manual 1097
1048 dence-based treatments for PTSD known as ‘‘EMDR” of mental disorders (4th ed.). Washington, DC: American Psychiatric 1098
1049 (eye movement desensitisation and reprocessing”) it has Association. 1099
1050 been argued that the eye movement component may indeed Andrade, J., Kavanagh, D., & Baddeley, A. (1997). Eye-movements and 1100
visual imagery: A working memory approach to the treatment of post- 110

1051 act as a visuospatial task aiding desensitisation to emotive traumatic stress disorder. British Journal of Clinical Psychology, 36, 1102
1052 visual memories (Andrade et al., 1997; Kavanagh et al., 209–223. 1103
1053 2001; Van den Hout et al., 2001). Within a clinically trau- Arntz, A., de Groot, C., & Kindt, M. (2005). Emotional memory is 1104
1054 matised sample, Lilley, Andrade, Turpin, Sabin-Farrell, perceptual. Journal of Behavior Therapy and Experimental Psychiatry, 1105
1055 and Holmes (submitted for publication) have found sup- 36, 19–34. 1106
Baddeley, A. D. (1986). Working memory. Oxford: Oxford University 1107

1056 port for this proposal using a side-to-side eye movement Press. 1108
1057 task while reliving traumatic ‘hotspots’ (Holmes et al., Baddeley, A. D., & Andrade, J. (2000). Working memory and the 1109
1058 2005). However, other visuospatial tasks than eye move- vividness of imagery. Journal of Experimental Psychology-General, 1110
1059 ments (including our pattern tapping and moulding tasks) 129(1), 126–145. 111
1060 may also have benefits. Further, such tasks might be used Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive 1112
therapy of depression. New York: Guilford Press. 1113
1061 as an adjunct to exposure therapy or other therapeutic Beck, A. T., Steer, A., & Brown, K. (1996). Beck Depression Inventory-II. 1114
1062 techniques where images are evoked (rather than just part San Antonio: Harcourt Brace. 1115
1063 of EMDR per se). Such an approach may have application Bernstein, E., & Putnam, F. W. (1986). Development, reliability and 1116
1064 for intrusive images across other psychological disorders validity of a dissociation scale. Journal of Nervous and Mental Disease, 1117
1065 other PTSD such as substance misuse cravings (Kavanagh 174, 727–735. 1118
Berntsen, D. (1996). Involuntary autobiographical memory. Applied 1119
1066 et al., 2005) and food cravings (Kemps et al., 2004). Cognitive Psychology, 10, 435–454. 1120
1067 We believe that the trauma film paradigm provides an Bremner, J. D., Krystal, J. H., Putnam, F. W., Southwick, S. M., Marmar, 112
1068 invaluable tool for insight into the factors and mechanisms C. R., Charney, D. S., et al. (1998). Measurement of dissociative states 1122

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

12 E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx

123 with the clinician-administered dissociative states scales (CADSS). Elzinga, B. M., & Bremner, J. D. (2002). Are the neural substrates of 1191
124 Journal of Traumatic Stress, 11, 125–136. memory the final common pathway in posttraumatic stress disorder 1192
125 Bremner, J. D., Southwick, S. M., Brett, E., Fontana, A., Rosenheck, R., (PTSD)? Journal of Affective Disorders, 70, 1–17. 1193
126 & Charney, D. S. (1992). Dissociation and posttraumatic stress Engelhard, I. M., van den Hout, M. A., Kindt, M., Arntz, A., & Schouten, 1194
127 disorder in Vietnam combat veterans. American Journal of Psychiatry, E. (2003). Peritraumatic dissociation and posttraumatic stress disorder 1195
128 149, 328–332. after pregnancy loss: a prospective study. Behaviour Research and 1196
129 Brewin, C. R. (2001). A cognitive neuroscience account of posttraumatic Therapy, 41, 67–78. 1197
130 stress disorder and its treatment. Behaviour Research and Therapy, 39, Eysenck, H. J., & Eysenk, S. B. G. (1975). Manual of the Eysenck 1198
131 373–393. personality questionnaire (junior and adult). London: Hodder & 1199
132 Brewin, C. R. (2003). Posttraumatic stress disorder: Malady or myth. New Stoughton. 1200
133 Haven: Yale University Press. Foa, E. B. (1995). The posttraumatic diagnostic scale (PDS) manual. 1201
134 Brewin, C. R., Christodoulides, J., & Hutchinson, G. (1996). Intrusive Minneapolis: National Computer Systems, p. 55. 1202

135 thoughts and intrusive memories in a nonclinical sample. Cognition Folkins, C. H., Lawson, K. D., Opton, E. M., & Lazarus, R. S. (1968). 1203
136 and Emotion, 10(1), 107–112, 6. Desensitization and the experimental reduction of threat. Journal of 1204
137 Brewin, C. R., Dalgleish, T., & Joseph, S. (1996). A dual representation Abnormal Psychology, 73(2), 100–113. 1205

138 theory of posttraumatic stress disorder. Psychological Review, 103, Hackmann, A., Clark, D. M., & McManus, F. (2000). Recurrent images 1206
139 670–686. and early memories in social phobia. Behaviour Research and Therapy, 1207
140 Brewin, C. R., & Holmes, E. A. (2003). Psychological theories of 38(6), 601–610. 1208
141 posttraumatic stress disorder. Clinical Psychology Review, 23(3), Hackmann, A., Ehlers, A., Speckens, A., & Clark, D. M. (2004). 1209
142 339–376. Characteristics and content of intrusive memories in PTSD and their 1210
143 Brewin, C. R., & Saunders, J. (2001). The effect of dissociation at encoding changes with treatment. Journal of Traumatic Stress, 17(3), 231–240. 1211

144 on intrusive memories for a stressful film. British Journal of Medical Hagenaars, M. A., van Minnen, A., Holmes, E. A., Brewin, C. R. & 1212
145 Psychology, 74, 467–472. Hoogduin, K. A. L. (submitted for publication). The effect of 1213
146 Bryant, R. A., & Guthrie, R. M. (2005). Maladaptive appraisals as a risk peritraumatic somatoform dissociation on intrusion development: 1214
147 factor for posttraumatic stress: A study of trainee firefighters. An experimental study. Q5 1215
148 Psychological Science, 16(10), 749–752. Hagenaars, M. A., Roelofs, K., Hoogduin, K. A. L., & van Minnen, A. 1216
149 Bryant, R. A., & Harvey, A. G. (1995). Avoidant coping style and post- (2006). Motor and sensory dissociative phenomena associated with 1217
150 traumatic stress following motor vehicle accidents. Behaviour Research induced catalepsy. Journal of Clinical and Experimental Hypnosis, 54, 1218
151 and Therapy, 33, 631–635. 234–244. 1219
152 Butler, G., Wells, A., & Dewick, H. (1995). Differential effects of worry Halligan, S. L., Clark, D. M., & Ehlers, A. (2002). Cognitive processing, 1220
153 and imagery after exposure to a stressful stimulus: A pilot study. memory, and the development of PTSD symptoms: two experimental 1221
154 Behavioural and Cognitive Psychotherapy, 23, 45–56. analogue studies. Journal of Behaviour Therapy and Experimental 1222
155 Bywaters, M., Andrade, J., & Turpin, G. (2004). Determinants of the Psychiatry, 33, 73–89. 1223

156 vividness of visual imagery: The effects of delayed recall, stimulus affect Hirsch, C. R., Clark, D. M., Mathews, A., & Williams, R. (2003). Self- 1224
157 and individual differences. Memory, 12(4), 479–488. images play a causal role in social phobia. Behaviour Research and 1225
158 Campbell, B. A., Wood, G., & McBride, T. (1997). Origins of orienting Therapy, 41(8), 909–921. 1226
159 and defensive responses: An evolutionary perspective. In P. J. Lang, R. Hirshon, R. (2004). Worry beads: Handling stress by handling objects. 1227
160 F. Simons, & M. T. Bababan (Eds.), Attention and orienting: Sensory Science NetLinks, American Association for the Advancement of 1228
161 and motivational processes (pp. 41–68). New Jersey: Lawrence Science (retrieved 20.10.06). 1229

162 Erlbaum. sci_update.cfm?DocID=209. 1230

163 Candel, I., & Merckelbach, H. (2004). Peritraumatic dissociation as a Holmes, E. A. (2000). Peri-traumatic processing and the development of 1231

164 predictor of posttraumatic stress disorder: A critical review. Compre- intrusions. Unpublished Doctoral Thesis. Royal Holloway, University 1232
165 hensive Psychiatry, 45(1), 44–50. of London. 1233
166 Carlson, E. B., & Putnam, F. W. (1993). An update on the Dissociative Holmes, E. A. (2004). Intrusive, emotional mental imagery and trauma: 1234
167 Experiences Scale. Dissociation: Progress in the Dissociative Disorders, Experimental and clinical clues. Imagination, Cognition and Personal- 1235
168 6, 16–27. ity, 23(2&3), 147–154. 1236
169 Conway, M. A. (2001). Sensory-perceptual episodic memory and its Holmes, E. A. (2007). Intrusive memory development in the context of the 1237

170 context: Autobiographical memory. Philosophical Transactions of the traumatic film paradigm. In M. Moulds (chair) From the laboratory to 1238
171 Royal Society, 356, 1375–1384. the clinic: Translating experimental findings into innovative treatments 1239
172 Conway, M. A., Meares, K., & Standart, S. (2004). Images and goals. for clinical disorders. Symposium at the British association of behav- 1240
173 Memory, 12(4), 525–531. ioural and cognitive psychotherapies 35th annual conference. 1241
174 Conway, M. A., & Pleydell-Pearce, C. W. (2000). The construction of Holmes, E. A., Brewin, C. R., & Hennessy, R. G. (2004). Trauma films, 1242
175 autobiographical memories in the self-memory system. Psychological information processing, and intrusive memory development. Journal of 1243

176 Review, 107, 261–288. Experimental Psychology: General, 133(1), 3–22. 1244
177 Davies, M. I., & Clark, D. M. (1998a). Thought suppression produces a Holmes, E. A., Brown, R. J., Mansell, W., Fearon, P. R., Hunter, E. C. 1245
178 rebound effect with analogue post-traumatic intrusions. Behaviour M., Frasquilho, F., et al. (2005). Are there two qualitatively distinct 1246
179 Research and Therapy, 36, 571–582. forms of dissociation? A review and some clinical implications. Clinical 1247
180 Davies, M. I., & Clark, D. M. (1998b). Predictors of analogue post- Psychology Review, 25, 1–23. 1248
181 traumatic intrusive cognitions. Behavioural and Cognitive Psychother- Holmes, E. A., Creswell, C., O’Connor, T. G., & Saunders, A. (2003). 1249
182 apy, 26, 303–314. Intrusive mental images of 11 September 2001 terrorist attacks as 1250
183 Day, S. J., Holmes, E. A., & Hackmann, A. (2004). Occurrence of imagery experienced by London school children. In J. Prasko (Ed.), XXXIII 1251
184 and its link with early memories in agoraphobia. Memory, 12(4), Annual congress of the european association for behavioural and 1252
185 416–427. cognitive psychotherapies (EABCT) (Vol. Book of Abstracts, pp. 46). 1253
186 Ehlers, A., & Clark, D. M. (2000). A cognitive model of posttraumatic Prague: Psychiatrie, Supplement 4. 1254
187 stress disorder. Behaviour Research and Therapy, 38, 319–345. Holmes, E. A., Grey, N., & Young, K. A. D. (2005). Intrusive images and 1255
188 Ehlers, A., Hackmann, A., & Michael, T. (2004). Intrusive re-experiencing ‘‘hotspots” of trauma memories in Posttraumatic stress disorder: An 1256
189 in post-traumatic stress disorder: Phenomenology, theory, and ther- exploratory investigation of emotion and cognitive themes. Journal of 1257
190 apy. Memory, 12(4), 403–415. Behaviour Therapy and Experimental Psychology, 36, 3–17. 1258

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx 13

1259 Holmes, E. A. & Hackmann, A. (Eds.). (2004). Mental imagery and Lazarus, R. S., & Alfert, E. (1964). The short-circuiting of threat by 1327
1260 memory in psychopathology [special issue]. Memory, 12(4). experimentally altering cognitive appraisal. Journal of Abnormal and 1328
1261 Holmes, E. A., & Mathews, A. (2005). Mental imagery and emotion: A Social Psychology, 69, 196–205. 1329
1262 special relationship? Emotion, 5(4), 489–497. Lazarus, R. S., & Opton, E. M. (1964). The study of psychological stress: 1330
1263 Holmes, E. A., Oakley, D. A., Stuart, A., & Brewin, C. R. (2007). A summary of theoretical formulations and experimental findings. In 133
1264 Investigating peri-traumatic dissociation using hypnosis during a C. D. Spielberger (Ed.), Anxiety and Behaviour (pp. 225–262). New 1332
1265 traumatic film. Journal of Trauma and Dissociation, 7, 91–111. York: Academic Press. 1333
1266 Holmes, E. A., & Steel, C. (2004). Schizotypy: A vulnerability factor for Lazarus, R. S., Opton, E. M., Nomikos, M. S., & Rankin, N. O. (1965). 1334
1267 traumatic intrusions. Journal of Nervous and Mental Disease, 192(1), The principle of short-circuiting of threat: Further evidence. Journal of 1335
1268 28–34. Personality, 33(4), 622–635. 1336
1269 Horowitz, M. J. (1969). Psychic trauma: Return of images after a stress Leonard, K. N., Telch, M. J., & Harrington, P. J. (1999). Dissociation in 1337
1270 film. Archives of General Psychiatry, 20, 552–559. the laboratory: A comparison of strategies. Behaviour Research and 1338

1271 Horowitz, M. J. (1975). Intrusive and repetitive thoughts after experi- Therapy, 37(1), 49–61. 1339
1272 mental stress. Archives of General Psychiatry, 32, 1457–1463. Lilley, S. A., Andrade, J., Turpin, G., Sabin-Farrell, R., & Holmes., E. A. 1340
1273 Horowitz, M. J., & Becker, S. S. (1971a). Cognitive response to stress (submitted for publication). Selective visual working memory inter- 134

1274 and experimental demand. Journal of Abnormal Psychology, 78, 86– ference with vividness and emotionality of post-traumatic stress 1342
1275 92. disorder images. Q6 1343
1276 Horowitz, M. J., & Becker, S. S. (1971b). Cognitive response to stressful Mace, J. H. (2005). Priming involuntary autobiographical memories. 1344
1277 stimuli. Archives of General Psychiatry, 25, 419–428. Memory, 13(8), 874–884. 1345
1278 Horowitz, M. J., & Becker, S. S. (1971c). The compulsion to repeat Marmar, C. R., Weiss, D. S., & Metzler, T. J. (1997). Peritraumatic 1346
1279 trauma: Experimental study of intrusive thinking after stress. Journal Dissociative Experiences Questionnaire. In J. P. Wilson & T. M. 1347

1280 of Nervous and Mental Disease, 152, 32–40. Keane (Eds.), Assessing psychological trauma and PTSD (pp. 412–428). 1348
1281 Horowitz, M. J., & Becker, S. S. (1973). Cognitive response to erotic and New York: The Guilford Press. 1349
1282 stressful films. Archives of General Psychiatry, 29, 81–84. Mason, O., Claridge, G., & Jackson, M. (1995). New scales for the 1350
1283 Horowitz, M. J., Becker, S. S., Moskowitz, M. L., & Rashid, K. (1972). assessment of schizotypy. Personality and Individual Differences, 18, 135
1284 Intrusive thinking in psychiatric patients after stress. Psychological 7–13. 1352
1285 Reports, 31, 235–238. Mathews, A., & MacLeod, C. (2005). Cognitive vulnerability to emotional 1353
1286 Horowitz, M. J., & Wildner, N. (1976). Stress films, emotion, and disorders. Annual Review of Clinical Psychology, 1(7), 1–29. 1354
1287 cognitive response. Archives of General Psychiatry, 33, 1339–1344. May, J., Andrade, J., Pannaboke, N., & Kavanagh, D. (2004). Images of 1355
1288 Kavanagh, D. J., Andrade, J., & May, J. (2005). Imaginary relish and desire: Cognitive models of craving. Memory, 12(4), 447–461. 1356
1289 exquisite torture: The elaborated intrusion theory of desire. Psycho- Mol, S. S. L., Arntz, A., Metsemakers, J. F. M., Dinant, G., Vilters-van 1357
1290 logical Review, 112(2), 446–467. Montfort, P. A. P., & Knottnerus, J. A. (2005). Symptoms of post- 1358
1291 Kavanagh, D. J., Freese, S., Andrade, J., & May, J. (2001). Effects of traumatic stress disorder after non-traumatic events: evidence form an 1359

1292 visuospatial tasks on desensitization to emotive memories. British open population study. British Journal of Psychiatry, 186, 494–499. 1360
1293 Journal of Clinical Psychology, 40(3), 267–280. Morrison, A. P., Beck, A. T., Glentworth, D., Dunn, H., Reid, G. S., & 136
1294 Kemps, E., Tiggemann, M., Woods, D., & Soekov, B. (2004). Reduction Larkin, W. Imagery and psychotic symptoms: A preliminary investi- 1362
1295 of food cravings through concurrent visuospatial processing. Interna- gation. Behaviour Research and Therapy, 40(9), 1053–1062. Q8 1363
1296 tional Journal of Eating Disorders, 36(31), 31–40. Murray, J., Ehlers, A., & Mayou, R. A. (2002). Dissociation and 1364
1297 Kindt, M., Van den Hout, M., & Buck, N. (2005). Dissociation related to posttraumatic distress disorder: Two prospective studies of motor 1365

1298 subjective memory fragmentation and intrusions but not to objective vehicle accident survivors. British Journal of Psychiatry; The Journal of 1366
1299 memory disturbances. Journal of Behavior Therapy and Experimental Mental Science, 180, 363–368. 1367

1300 Q7 Psychiatry, 36(1), 43–59. Nijenhuis, E. R. S., van Engen, A., Kusters, I., & van der Hart, O. (2001). 1368
1301 Koopman, C., Classen, C., & Spiegel, D. (1994). Predictors of Peritraumatic somatoform dissociation in relation to recall of child- 1369
1302 posttraumatic stress disorder amoung survivors of the Oakland/ hood sexual abuse. Journal of Trauma and Dissociation, 2, 49–68. 1370
1303 Berkley, Calif., firestorm. American Journal of Psychiatry, 151, Nijenhuis, E. R. S., Vanderlinden, J., & Spinhoven, P. (1998). Animal 137
1304 888–894. defensive reactions as a model for trauma induced dissociative 1372
1305 Koriat, A., Melkman, R., Averill, J. R., & Lazarus, R. S. (1972). The self- reactions. Journal of Traumatic Stress, 11(2), 243–260. 1373

1306 control of emotional reactions to a stressful film. Journal of Person- Nixon, R. D. V. (2006). Understanding the development of intrusive 1374
1307 ality, 40(4), 601–619. memories following trauma [Abstract]. In Proceedings of the 4th 1375
1308 Krans, J., Holmes, E., Näring, G., & Becker, E. (2006). The effect of international conference on memory. University of New South Wales: 1376
1309 attentional deployment on the emotional impact of a trauma film. Sydney, Australia. 1377
1310 Poster session presented at the Consortium of European Research on Orne, M. T. (1962). On the social psychology of the psychological 1378
1311 Emotion (CERE), Louvain-La-Neuve, Belgium. experiment: With particular reference to demand characteristics and 1379

1312 Kuyken, W., & Brewin, C. R. (1994). Intrusive memories of childhood their implications. American Psychologist, 17, 776–783. 1380
1313 abuse during depressive episodes. Behaviour Research and Therapy, 32, Ozer, E. J., Best, S. R., Lipsey, T. L., & Weiss, D. S. (2003). Predictors of 138
1314 525–528. posttraumatic stress disorder and symptoms in adults: A meta- 1382
1315 Kuyken, W., & Brewin, C. R. (1999). The relation of early abuse to analysis. Psychological Bulletin, 129, 52–73. 1383
1316 cognition and coping in Depression. Cognitive Therapy and Research, Pearson, D. G., Sawyer, E. H., & Holmes, E. A. (in press). Intrusive 1384
1317 23(6), 665–677. memories for affective pictures: Effects of differential working memory 1385
1318 Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1997). Motivated load during encoding. Q141386
1319 attention: Affect, activation and action. In P. J. Lang, R. F. Simons, & Pfefferbaum, B., Pfefferbaum, R. L., North, C. S., & Neas, B. R. (2002). 1387
1320 M. T. Bababan (Eds.), Attention and orienting: Sensory and motiva- Commentary on ‘‘Television images and psychological symptoms after 1388
1321 tional processes (pp. 99–135). Mahwah, NJ: Erlbaum. the September 11 terrorist attacks: Does television viewing satisfy 1389
1322 }
Lang, P. J., Ohmann, A., & Vaitl, D. (1988). The International Affective criteria for exposure to posttraumatic stress disorder?”. Psychiatry: 1390
1323 Picture System [photographic slides]. Gainsville, FL: University of Interpersonal and Biological processes, 65, 306–309. 139
1324 Florida, The Centre for Research in Psychopysiology. Reynolds, M., & Brewin, C. R. (1999). Intrusive memories in depression 1392
1325 Laposa, J. M., & Alden, L. E. (2006). An analogue study of intrusions. and posttraumatic stress disorder. Behaviour Research and Therapy, 1393
1326 Behaviour Research and Therapy, 44(7), 925–946. 37(3), 201–215. 1394

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002
ACTPSY 1241 No. of Pages 14, Model 5+
4 December 2007 Disk Used

14 E.A. Holmes, C. Bourne / Acta Psychologica xxx (2007) xxx–xxx

395 Schlagman, S., Kvavilashvili, L., & Schulz, J. (in press). Effects of age on Ursano, R. J., Fullerton, C. S., Epstein, R. S., Crowley, B., Vance, K., 1421
396 involuntary autobiographical memories. In J. Mace (Ed.). Involuntary Kao, T., & Baum, A. (1999). Peritraumatic dissociation and posttrau- 1422
397Q15 memory. Blackwells. matic stress disorder following motor vehicle accidents. American 1423
398 Shalev, A. Y., Peri, T., Canetti, L., & Schreiber, S. (1996). Predictors of Journal of Psychiatry, 156, 1808–1810. 1424
399 PTSD in injured trauma survivors: A prospective study. American Vallar, G., & Baddeley, A. D. (1982). Short-term forgetting and the 1425
400 Journal of Psychiatry, 153, 219–225. articulatory loop. Quarterly Journal of Experimental Psychology: 1426
401 Shor, R. E., & Orne, E. C. (1962). The Harvard group scale of hypnotic Human Experimental Psychology, 34(A), 53–60. 1427
402 susceptibility, Form A.. Palo Alto: Consulting Psychologists Press. Van den Hout, M., Muris, P., Salemink, E., & Kindt, M. (2001). 1428
403 Speisman, J. C., Lazarus, R. S., Mordkoff, A., & Davison, L. (1964). Autobiographical memories become less vivid and emotional after eye 1429
404 Experimental reduction of stress based on ego-defense theory. Journal movements. British Journal of Clinical Psychology, 36, 209–223. 1430
405 of Abnormal and Social Psychology, 68(4), 367–380. Van Stegeren, A. (this issue). The role of the noradrenergic system in 1431
406 Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual for emotional memory. Acta Psychologica. 1432

407 the state-trait anxiety inventory. Paolo Alto: Consulting Psychologists Wessel, I. (2006). Deficient cognitive control as a predictor of intrusive 1433
408 Press. memories of an emotional film [Abstract]. In Proceedings of the 4th 1434
409 Steel, C., Fowler, D., & Holmes, E. A. (2005). Trauma-related intrusions international conference on memory. University of New South Wales: 1435

410 and psychosis: An information processing account. Behavioural and Sydney, Australia. 1436
411 Cognitive Psychotherapy, 33, 139–152. Wonderlic, E. L., & Associates (1999). Wonderlic personnel test and 1437
412 Steil, R. I. (1996). Posttraumatische intrusionen nach verkehrsunfällen scholastic level exam user’s manual. Libertyville: Wonderlic Personnel 1438
413 (posttraumatic intrusions after road traffic accidents). Frankfurt: Peter Test, Inc. 1439
414 Lang. Wolf, O. T. (this issue). The influence of stress hormones on emotional 1440
415 Sterlini, G. L., & Bryant, R. A. (2002). Hyperarousal and dissociation: A memory: Relevance for psychopathology. Acta Psychologica. 1441

416 study of novice skydivers. Behaviour Research and Therapy, 40(4), 1442
417 431–437.
418 Stuart, A. D. P., Holmes, E. A., & Brewin, C. R. (2006). The influence of a
419 visuospatial grounding task on intrusive images of a traumatic film.
420 Behaviour Research and Therapy, 44(4), 611–619.

Please cite this article in press as: Holmes, E. A. , & Bourne, C., Inducing and modulating intrusive emotional memories: A review ...,
Acta Psychologica (2007), doi:10.1016/j.actpsy.2007.11.002