- Email: [email protected]
Virtual environment for body image modification: virtual reality system for the treatment of body image disturbances
Pergamon
Computers in Human Behavior, Vol. 14, No. 3, pp. 477 ± 490, 1998 # 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0747-5632/98/$19.00 + 0.00
PII: S0747-5632(98)00018-1
Virtual Environment for Body Image Modification: Virtual Reality System for the Treatment of Body Image Disturbances Giuseppe Riva Applied Technology for Neuro-psychology Laboratory, Istituto Auxologico Italiano Abstract Ð The paper describes the characteristics and preliminary evaluation of the Virtual Environment for Body Image Modification (VEBIM), a virtual reality environment aimed at treating body image disturbances and body dissatisfaction associated with eating disorders. Two methods are commonly used to treat body image: (a) a cognitive/behavioral therapy to influence patients' feelings of dissatisfaction and (b) a visual/motorial therapy with the aim of influencing the level of bodily awareness. VEBIM tries to integrate these two therapeutic approaches within an immersive virtual environment. This choice would make it possible both to intervene simultaneously on all of the forms of bodily representations, and to use the psycho-physiological effects provoked on the body by the virtual experience for therapeutic purposes. As well as the description of the VEBIM theoretical approach, the paper also presents a study on two preliminary samples (71 normal subjects, uncontrolled study; 48 normal subjects, controlled study) to test the efficacy of VEBIM. # 1998 Elsevier Science Ltd. All rights reserved Keywords Ð virtual reality, body image disturbances, eating disorders Requests for reprints should be addressed to Giuseppe Riva, Applied Technology for Neuropsychology Laboratory, Istituto Auxologico Italiano, PO Box 1, 28044 Verbania, Italy. E-mail: [email protected] 477
478
Riva
The physical world, including our body, is not given directly in our experience but is inferred through observation and critical reasoning. In fact, we experience our bodies through two multidimensional cognitive constructs (Fisher, 1990; Gallagher, 1995): the body schema and the body image. According to Head (1926), the body schema is a model/representation of one's own body that constitutes a standard against which postures and body movements are judged. This representation can be considered the result of comparisons and integration at the cortical level of past sensory experiences (postural, tactile, visual, kinaesthetic, and vestibular) with current sensations. This gives rise to an almost completely unconscious ``plastic'' reference model that makes it possible to move easily in space and to recognize the parts of one's own body in all situations. If the body schema can be considered a perceptual model of the body, the body image is a cognitive/social/emotional model. In fact, body image is not only a cognitive construction but also a reflection of wishes, emotional attitudes, and interactions with other. According to Schilder (1950), the body image can be defined as ``the picture of our own body which we form in our mind, that is to say, the way in which the body appears to ourselves'' (p. 11). Expanding on Schilder's idea, Allamani and Allegranzi (1990) refer to body image as ``a complex psychological organization which develops through the bodily experience of an individual and affects both the schema of behavior and a fundamental nucleus of self-image'' (p. 121). Body experiences have a long and well-reported association with eating and weight related problems (Barrios, Ruff, & York, 1989; Garfinkel & Garner, 1982; Riva, 1997a; Rosen, 1990; Thompson, 1990, 1992, 1995). One index of the importance of body experience disturbance involves its relevance to agreed-on clinical disorders. The Diagnostic and Statistical Manual of Mental Disorders (4th edition; American Psychiatric Association, 1994) contains a body image criterion that is required for the diagnosis of anorexia nervosa or bulimia nervosa. It has also been suggested that, when there is psychological comorbidity with obesity, it may be strongly due to problematic body experience issues (Thompson, 1992). Today, researchers and clinicians agree that including an assessment and evaluation of body experience disturbance is crucial to any treatment program targeting obesity or eating disorders. Probably, the disturbances of body image associated with the eating disorders can be conceptualized as a type of cognitive bias (Vitousek & Hollon, 1990; Williamson, 1996; Williamson, Cubic, & Gleaves, 1993). The essence of this cognitive perspective is that the central psychopathological concerns of an individual bias the manner in which information is processed. In most cases, it is presumed that this biased information processing occurs automatically. Also, it is generally presumed that the process occurs more or less outside the person's awareness unless the person consciously reflects upon his/her thought processes (as in cognitive therapy). Mineka and Sutton (1992)
Virtual reality for the treatment of body image disturbances
479
have identified four common types of cognitive bias in research related to depression and anxiety disorders: attentional bias, memory bias, judgmental bias, and associative bias. Three of these four types of cognitive bias have been the focus of research related to eating disorders: preoccupation with body size, body dysphoria, and related problems. In contrast to the great number of publications on body image, only a few papers focus on the treatment of a disturbed body image in eating disorders. In general, two direct and specific approaches can be distinguished: a cognitive/behavioral approach aimed at influencing patients' feelings of dissatisfaction with different parts of their bodies by means of individual interviews, relaxation, and imaginative techniques (Butters & Cash, 1987), and a visual/motorial approach which makes use of video recordings of particular gestures and movements with the aim of influencing the level of bodily awareness (Wooley & Wooley, 1985). An interesting possibility that we tried to address in the Virtual Environment for Body Image Modification (VEBIM) is the integration of the different methods (cognitive, behavioral, and visual/motorial) commonly used in the treatment of body experience disturbances (Riva, 1997b; Riva & Melis, 1997; Riva, Melis, & Bolzoni, 1997; Schlundt & Johnson, 1990) within a virtual environment. In particular we tried to integrate the cognitive methods of Countering, Alternative Interpretation, Label Shifting and Deactivating, the behavioral method of Temptation Exposure with Response Prevention and the visual motorial methods of Awareness of the Distortion and Body Image Modification (Table 1) using the virtual environment in the same way as images in the well-known method of guided imagery (Leuner, 1969). It is Leuner's belief that the imagery evokes intense latent feelings that are relevant to the patient's problems. Guided imagery has been found to be a powerful tool in treatment approach ranging from psychoanalytic therapy (Reyker, 1977) to behaviorism (Wolpe, 1958). A choice of this type would make it possible not only to evoke latent feelings, but also to use the psychophysiological effects provoked by the experience for therapeutic purposes. In practically all virtual reality (VR) systems the human operator's normal sensorimotor loop is altered by the presence of distortions, time delays, and noise. Such alterations, that are introduced unintentionally and usually degrade performance, affect body perceptions too. The somaesthetic systems has a proprioceptive subsystem that senses the body's internal state, such as the position of limbs and joints and the tension of the muscles and tendons. Mismatches between the signals from the proprioceptive system and the external signals of a virtual environments alter body perceptions and can cause discomfort or simulator sickness (Sadowsky & Massof, 1994). Such effects, attributable to the reorganizational and reconstructive mechanisms necessary to adapt the subjects to the qualitatively distorted world of VR, could be of great help during the course of therapy aimed at
480
Riva
Table 1. Therapeutical Methods Integrated in VEBIM Methods
Procedures
Cognitive
Countering: Once a list of distorted perceptions and cognitions is developed, the process of countering these thoughts and beliefs begins. In countering, the patient is taught to recognise the error in thinking, and substitute more appropriate perceptions and interpretations. Alternative interpretation: The patient learns to stop and consider other interpretations of a situation before proceeding to the decision-making stage. The patient develops a list of problem situations, evoked emotions, and interpretative beliefs. The therapist and patient discuss each interpretation and if possible identify the kind of objective data that would confirm one of them as correct. Label shifting: The patient first tries to identify the kinds of negative words he/she uses to interpret situations in his/her life, such as bad, terrible, obese, inferior, and hateful. The situations in which these labels are used are then listed. The patient and therapist replace each emotional label with two or more descriptive words. Deactivating the illness belief: The therapist first helps the client list his/her beliefs concerning eating disorders. The extent to which the illness model influences each belief is identified. The therapist then teaches the client a cognitive/behavioral approach to interpreting maladaptive behavior and shows how bingeing, purging, and dieting can be understood from this framework.
Behavioral
Temptation exposure with response prevention (TERP): The rationale of temptation exposure with response prevention is to expose the individual to the environmental, cognitive, physiological, and affective stimuli that elicit abnormal behaviors and to prevent them from occurring. The TERP protocol is usually divided into three distinct phases: (a) comprehensive assessment of eliciting stimuli, (b) temptation exposure extinction sessions, and (c) temptation exposure sessions with training in alternative responses.
Visual/motorial
Awareness of the distortion: The patient is instructed to develop an awareness of the distortion. This is approached by a number of techniques including the presentation of feedback regarding one's self-image. Videotape feedback is also usually used. The patient is videotaped engaging in a range of activities. Modification of the body image: The patient is instructed to imagine him/herself as different on several dimensions including size, race, and being larger or smaller in particular areas. The patient is also asked to imagine him/herself as younger, older, what the patient looks and feels like before and after eating, as well as before and after academic/vocational and social successes and failures.
influencing the way the body is experienced, because they lead to a greater awareness of the perceptual and sensory/motorial processes associated with them. When a particular event or stimulus violates the information present in the body schema (as occurs during a virtual experience), the information itself becomes accessible at a conscious level (Baars, 1988). This facilitates the process of modification and, by means of the mediation of the self (which tries to integrate and maintain the consistency of the different representations of the body), also makes it possible to influence body image. However, until now some problems have existed with VR. There are anecdotal reports that immersive VR can lead to symptoms similar to motion sickness symptoms. Visually induced motion sickness is a syndrome that occasionally occurs when physically stationary individuals view compelling
Virtual reality for the treatment of body image disturbances
481
visual representations of self motion. It may also occur when detectable lags are present between head movements and presentation of the visual display in the head-mounted display (HMD). Motion sickness is characterized by a diverse set of symptoms but is primarily exemplified by nausea and vomiting. In the case of VEBIM, with adults and adolescents as possible subjects, both including people with psychological and physical problems, the need to verify the problems associated with operating in a virtual environment has to be addressed. A first study was designed to: (a) verify the effects provoked by VEBIM on blood pressure and heart rate and (b) verify the effects provoked by VEBIM on the body experience. A second controlled study was set-up to confirm the results obtained in the first study. METHOD Subjects In the first study, 71 subjects (47 males and 24 females; mean age: 23.187.54 years; mean weight: 67.3415.42 kg; mean height: 173.37.97 cm) participated. A total of 48 women participated in the second study. The sample was randomly divided into two groups: the experimental group (mean age: 22.65.84 years; mean weight: 55.14.43 kg; mean height: 168.296.20 cm) and the control group (mean age: 23.14.63 years; mean weight: 55.83.98 kg; mean height: 171.015.80 cm). All the 119 subjects were recruited from the participants at a VR conference. Characteristics of the Virtual Reality Systems VEBIM is based on a Pentium-based immersive VR system (166 Mhz; 32 MB RAM; graphic engine: Matrox Millennium with 4 MB Warm) including an HMD subsystem with head-tracking and a two-button joystick-type motion input device. A head mounted display, specially developed by Virtual.sys (Milan, Italy) was used. The HMD displays 624 lines of 210 pixel to each eye (52 H and 41 V field of view) and uses LCD technology (two active matrix color LCDs). A Logitech 3D mouse provided head tracking. In VEBIM we did not use a stereoscopic display. Previous researches regard stereoscopy as important because it provides the user with a good cue of depth (Barham & McAllister, 1991). However, the refresh rate of graphics decreases by 50% for the need of two different images, one for each eye. Consequently,
482
Riva
we decided against implementing a stereoscopic display. To compensate for the lack of depth cue, we included perspective cues (light and shade, relative size, textural gradient, interposition and motion parallax) in the virtual environment. The data glove-type motion input device is very common in virtual environments because of its ability to sense many degrees of freedom simultaneously. However the operator is also frequently confused by the difficulty in using it correctly, especially when there is a time delay contained in the feed-back loop. To provide a easy method of motion in VEBIM we used a two-button joystick-type input device: pressing the upper button the operator moves forward, pressing the lower button the operator moves backwards. The direction of the movement is given by the rotation of operator's head. Characteristics of the Virtual Environment VEBIM is a five-zone virtual environment developed using the Sense 8 World ToolKit for Windows (Version 2.02). VEBIM consists of two parts (Zones 1 ± 2 and Zones 3 ± 5). The first two zones are designed both to give the subject a minimum level of skill in perceiving, moving through, and manipulating objects in VR, and to focus attention on eating and food choice. The two environments can also be used by the therapist for a comprehensive assessment of stimuli that could elicit abnormal eating behavior. This assessment is normally part of the Temptation Exposure with Response Prevention protocol. The next three zones are designed to modify the body experience of the subject. Zone 1. In this zone the subject becomes acquainted with the appropriate control device, the head mounted display, and the recognition of collisions. To move into the next zone subjects have to weigh themselves on a virtual balance. The balance is used for two functions: 1. it is intended to be an inevitable obstacle for the user, who must focus his/ her attention on this object, representing the importance of the weight dimension in the experiences to come thereafter; 2. it can be used, if needed by the therapist, to display the initial weight of the subject, as acquired in the dialog box at the beginning of the Body Image Virtual Reality Scale (BIVRS). Zones 2a and 2b. These zones show a kitchen (2a; Figure 1) and an office (2b). Each of these rooms is furnished with typical items, and contains different foods and drinks. When the user decide to eat or drink something, all he/she has to do is to ``touch'' a specific item. In this way the food is ``eaten'' and the
Virtual reality for the treatment of body image disturbances
483
Figure 1. View of Zone 2a showing the kitchen.
corresponding caloric intake is automatically recorded in a text file, which is used later to calculate the total income of calories. Located at the end of the second zone is a second virtual balance. According to the ``eaten'' food and to the caloric intake inserted at the beginning of BIVRS the balance will show the new weight of the subject (in kilograms). Zone 3. In this zone the subject is exposed to a series of panels textured with pictures of models, in the typical way of the advertising world (Figure 2). The images are used as stimuli to support a cognitive approach: the elicited feelings are subject to an analysis by the therapist according to the Label Shifting and Objective Counters methods. The feelings and their associated beliefs are identified, broken down into their logical components, replaced with two or more descriptive words, and then critically analyzed. Zone 4. In this zone the user finds a large mirror. Standing by it the subject can look at his/her real body, previously digitized using an EPSON Photo PC camera. The vision of her own body usually elicits in the user strong feelings that can be matched using the Counterattacking and the Countering cognitive methods. The mirror is also used, as indicated by Wooley and Wooley (1985), to instruct the user to imagine him/herself as different on several dimensions including size, race, and being larger or smaller in particular areas. The user is also asked to imagine him/herself as younger, older, what they look and feel like before and after eating, as well as before and after academic/vocational and social successes and failures.
484
Riva
Figure 2. View of Zone 3 whose walls show images of female and male models.
Zone 5. This zone consists of a long corridor ending with a room containing four doors of different dimensions. The subject can move into the last zone only by choosing the door corresponding exactly to his/her width. This procedure is the same as indicated by Wooley and Wooley (1985) to improve awareness of the body distortion. Procedures and Measures All the subjects were submitted to VEBIM for no more than 10 min and no less than 8 min. All the subjects included in the study reached Zone 5. Just before entering the virtual environment and just after all subjects completed three scales for assessing body experience: 1. The Figure Rating Scale (FRS; Thompson & Altabe, 1991) a set of 9 male and female figures which vary in size from underweight to overweight. 2. The Contour Drawing Rating Scale (CDRS; Thompson & Gray, 1995), a set of 9 male and female figures with precisely graduated increments between adjacent sizes. 3. The Breast/Chest Rating Scale (BCRS; Thompson & Tantleff, 1992), a set of 5 male and 5 female schematic figures, ranging from small to large upper torso.For these tests, subjects rated the figures based on the following instructional protocol: (a) current size and (b) ideal size. The
Virtual reality for the treatment of body image disturbances
485
difference between the ratings is called the self-ideal discrepancy score and is considered to represent the individual's dissatisfaction. The findings of Keeton, Cash, and Brown (1990), support the usefulness of the self-ideal discrepancy score in the assessment of body image, as it was shown to relate to other body-image indices and other clinically relevant measures. All the scales have good test ± retest reliability. In the first study only, we recorded also blood pressure and heart rate. Before the virtual experience the subjects were fitted with a blood pressure cuff and were asked to sit quietly during a 5-min baseline period. During this period blood pressure and heart rate were recorded. After the virtual experience and after a 10-min posttask period blood pressure and heart rate were recorded again. Statistical Analysis A power calculation was made to verify the opportunity to obtain statistically significant differences between the pre- and post-VR scores. Given the low statistical power, we decided to use in both studies the Exact methods, a series of statistical algorithms developed by the Harvard School of Public Health that enable researchers to make reliable inferences when data are small, sparse, heavily tied, or unbalanced (SPSS, 1995). The Exact method used to compare the scores was the Marginal Homogeneity Test (Agresti, 1990). RESULTS Study 1 Table 2 lists the means and standard deviations for the pressure and heart rate scores, across the three measurement occasions. No significant difference was found before and after the VR experience. Only two of the subjects experienced simulation sickness. Table 3 presents the means and standard deviations for the body image scores obtained before and after the virtual experience. The Marginal Homogeneity Test reported significant differences Table 2. Means and Standard Deviations for the Pressure and Heart Rate Scores Measurement 1
Systolic Diastolic Heart rate
Measurement 2
Measurement 3
M
SD
M
SD
M
SD
134.61 89.23 74.01
20.06 21.09 15.68
133.09 88.2 72.11
20.8 23.72 11.06
124.13 84.6 72.09
18.56 16.38 14.14
486
Riva
in the Ideal FRS and Ideal CDRS scores: both scores were higher after experiencing VEBIM. An analysis of the three discrepancy indexes revealed significant lower values for FRS and CDRS after the experience in the virtual environment. These results mean that VEBIM is able to reduce the body dissatisfaction of the subjects. Study 2 Table 4 presents the means and standard deviations for the ratings of body image. The Marginal Homogeneity Test reported significant differences in the Ideal FRS scores obtained by the experimental group: the scores were higher (p < .005) after experiencing the procedure used. An analysis of the discrepancy indexes showed a significant lower value for FRS (p < .05) after the virtual experience: body dissatisfaction of the subjects was reduced after the virtual experience. No significant differences were found in the control group (Table 5). Table 3. Means and Standard Deviations for the Body Image Scores (First Study) Real FRS Before After
3.81.34 3.841.28
Ideal FRS 3.18+/0.72*** 3.53+/0.96***
Real CDRS
Ideal CDRS
Real BCRS
Ideal BCRS
4.791.49 4.741.53
4.110.95* 4.211.07*
2.791.30 2.821.27
2.911.95 2.940.97
FRS=Figure Rating Scale; CDRS=Contour Drawing Rating Scale; BCRS=Breast/Chest Rating Scale.*p < .05; **p < .01; ***p < .005. Table 4. Means and Standard Deviations for the Body Image Scores in the Experimental Group (Second Study) Real FRS Before After
3.140.93 3.350.81
Ideal FRS 2.610.8*** 2.910.7***
Real CDRS
Ideal CDRS
Real BCRS
Ideal BCRS
41.18 4.211.25
3.50.83 3.521.08
2.981.49 2.931.35
2.751.13 2.521.08
FRS=Figure Rating Scale; CDRS=Contour Drawing Rating Scale; BCRS=Breast/Chest Rating Scale.*p < .05; **p < .01; ***p < .005. Table 5. Means and Standard Deviations for the Body Image Scores in the Control Group (Second Study) Real FRS Before After
3.181.03 3.211.01
Ideal FRS 2.710.8 2.850.74
Real CDRS
Ideal CDRS
Real BCRS
Ideal BCRS
3.81.1 4.011.05
3.480.73 3.490.88
3.011.01 31.12
2.680.95 2.51.09
FRS=Figure Rating Scale; CDRS=Contour Drawing Rating Scale; BCRS=Breast/Chest Rating Scale.
Virtual reality for the treatment of body image disturbances
487
DISCUSSION Although there is much potential for the use of immersive VR environments in clinical psychology, some problems have limited their application in this field. Some users have experienced side-effects, during and after exposure to VR environments. The symptoms experienced by these users are similar to those which have been reported during and after exposures to simulators with wide field-of-view displays (Cobb, Nichols, & Wilson, 1995; Kennedy, Lanham, Drexler, Massey, & Lilienthal, 1995; Regan, 1995; Regan & Price, 1994). These side-effects have been collectively referred to as ``simulator sickness'' (Kennedy, Lilienthal, Baltzley, Allgood, & Gower, 1987) and are characterized by three classes of symptoms: ocular problems, such as eyestrain, blurred vision and fatigue; disorientation and balance disturbances; and nausea. Exposure duration of less than 10 min to immersive VR environments has been shown to result in significant incidences of nausea, disorientation, and ocular problems (Regan & Price, 1993). The first interesting result of this study is the lack of side effects and simulation sickness in our samples after the experience in the virtual environment, confirming the possibility of using VEBIM for therapeutic purposes. This result, confirmed in both studies, is even more interesting given the sample used. In fact, females tend to be more susceptible to motion sickness than males (Griffin, 1990). However it is difficult to generalize this result to the VR systems at large given the peculiar technical characteristics of VEBIM. The other obtained result is the reduction in the body dissatisfaction of the subjects after the virtual experience. Usually body-image treatment involves a cognitive/behavioral or a visuomotor therapy that requires many sessions. The possibility of inducing a change in the body dissatisfaction after a 8 ± 10-min VR session can be useful to improve the efficacy of the existing approaches. As such, the procedure might be helpful as a part of a comprehensive treatment package to break through the ``resistance'' to treatment in clinical subjects (Vandereycken, 1990). Of course these results are preliminary only. From a clinical view point the issues that we have to address in the future are: 1. Further testing of VEBIM. Even if the data obtained in the two preliminary studies using VEBIM are very promising, we have to test the new environments with clinical and nonclinical subjects. 2. A follow-up study to check how long the influence of the virtual environment lasts. In the preliminary studies we have limited the VR experience to just one session, but from a therapeutic viewpoint it seems more reasonable to repeat the procedure.
488
Riva
3. How to integrate VEBIM into the usual cognitive/behavioral treatments. Even if the procedure might be helpful as a part of a comprehensive treatment package to break through the ``resistance'' to treatment in clinical subjects, it is important to define a specific therapeutic protocol to be used by other clinicians and therapists. Acknowledgments Ð This research is part of the Virtual Reality Environments for Psychoneuro-physiological Assessment and Rehabilitation (VREPAR) Projects, two European Commission funded research projects (DGXIIIÐTelematics for Health Care).
REFERENCES Agresti, A. (1990). Categorical data analysis. New York: John Wiley & Sons. Allamani, A., & Allegranzi, P. (1990). Immagine corporea: dimensioni e misure. Una ricerca clinica [Body Image: Dimensions and measures. A clinical research]. Archivio di Psicologia Neurologia Psichiatria, 2, 171 ± 195. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Baars, B. J. (1988). A cognitive theory of consciousness. New York: Cambridge University Press. Barham, P. T., & McAllister, D. F. (1991). A comparison of stereoscopic cursors for the interactive manipulation of B-splines. Proceedings of Stereoscopic Display and Applications, II, 18 ± 26. Barrios, B. A., Ruff, G., & York, C. (1989). Bulimia and body image: Assessment and explication of a promising construct. In W. G. Johnson (Ed.), Advances in eating disorders (Vol. II). New York: Jay Press. Butters, J. W., & Cash, T. F. (1987). Cognitive-behavioral treatment of women's body image satisfaction: A controlled outcome-study. Journal of Consulting and Clinical Psychology, 55, 889 ± 897. Cobb, S., Nichols, S., & Wilson, J. R. (1995). Health and safety implications of virtual reality: In search of an experimental methodology. Paper presented at the Framework for Immersive Virtual Environments Conference, University of London. Fisher, S. (1990). The evolution of psychological concepts about the body. In T. F. Cash & T. Pruzinsky (Eds.), Body images: Development, deviance and change (pp. 3 ± 20). New York: The Guilford Press. Gallagher, S. (1995). Body schema and intentionality. In J. L. Bermdez, A. Marcel, & N. Eilan (Eds.), The body and the self (pp. 24 ± 38). Cambridge, MA: MIT Press. Garfinkel, P. E., & Garner, D. M. (1982). Anorexia nervosa: A multidimensional perspective. New York: Brunner & Mazel. Griffin, M. J. (1990). Handbook of human vibration. London: Academic Press. Head, H. (1926). Aphasia and kindred disorders of speech. London: Cambridge University Press. Keeton, W. P., Cash, T. F., & Brown, T. A. (1990). Body image or body images: Comparative, multidimensional assessment among college students. Journal of Personality Assessment, 54, 213 ± 230. Kennedy, R. S., Lanham, D. S., Drexler, J. M., Massey, C. J. and Lilienthal, M. G. (1995). Cybersickness in several flight simulators and VR devices: A comparison of incidences, symptom profiles, measurement techniques and suggestions for research. Paper presented at the Framework for Immersive Virtual Environments Conference, University of London.
Virtual reality for the treatment of body image disturbances
489
Kennedy, R. S., Lilienthal, M. G., Baltzley, D. R., Allgood, G. O., & Gower, D. W. (1987). Consistency across measures of simulator sickness: Implications for a biocybernetic safety reporting device. Paper presented at the SAFE Association 25th Annual Symposium, Las Vegas. Leuner, H. (1969). Guided affective imagery: A method of intensive psychotherapy. American Journal of Psychotherapy, 23, 4 ± 21. Mineka, S., & Sutton, S. K. (1992). Cognitive biases and the emotional disorders. Psychological Science, 3, 65 ± 69. Regan, E. C. (1995). An investigation into nausea and other side-effects of head-coupled immersive virtual reality. Virtual Reality, 1(1), 17 ± 32. Regan, E. C., & Price, K. R. (1993). Some side-effects of immersion virtual reality: The effects of increasing head movements, of rapid interaction, and of seated subjects. Report 93R022, Army Personnel Research Establishment, Farnborough. Regan, E. C., & Price, K. R. (1994). The frequency of occurrence and severity of side-effects of immersion virtual reality. Aviation, Space and Environmental Medicine, 65, 527 ± 530. Reyker, J. (1977). Spontaneous visual imagery: Implications for psychoanalysis, psychopathology and psychotherapy. Journal of Mental Imagery, 2, 253 ± 274. Riva, G. (1997a). Representations of eating among adolescent Italian girls. The Journal of Social Psychology, 137(2), 205 ± 217. Riva, G. (1997b). The Virtual Environment for Body-Image Modification: Development and preliminary evaluation. Presence, 6(1), 106 ± 117. Riva, G., & Melis, L. (1997). Virtual reality for the treatment of body image disturbances. In G. Riva (Ed.), Virtual reality in neuro-psycho-physiology (pp. 95 ± 111). Amsterdam: IOS Press. Riva, G., Melis, L., & Bolzoni, M. (1997). Treating body-image disturbances. Communications of the ACM, 40(8), 69 ± 71. Rosen, J. C. (1990). Body image disturbances in eating disorders. In T. F. Cash & T. Pruzinsky (Eds.), Body images: Development, deviance and change (pp. 190 ± 214). The Guilford Press: New York. Sadowsky, J., & Massof, R. W. (1994). Sensory engineering: The science of synthetic environments. Johns Hopkins APL Technical Digest, 15, 99 ± 109. Schilder, P. (1950). The image and appearance of the human body. New York: International Universities Press. Schlundt, D. G., & Johnson, W. G. (1990). Eating disorders: Assessment and treatment. Needham Heights, MA: Allyn & Bacon. SPSS. (1995). SPSS Exact tests 6.1 for Windows. Chicago: Author. Thompson, J. K. (1990). Body image disturbance: Assessment and treatment. Elmsford, NJ: Pergamon Press. Thompson, J. K. (1992). Body image: Extent of disturbance, associated features, theoretical models, assessment methodologies, intervention strategies, and a proposal for a new DSM IV diagnostic categoryÐBody Image Disorder. In M. Hersa, R. M. Eisler, & P. M. Miller (Eds.), Progress in behavior modification (pp. 35 ± 51). Sycamore, IL: Sicamore. Thompson, J. K. (1995). Assessment of body image. In D. B. Allison (Ed.), Handbook of assessment methods for eating behaviors and weight related problems (pp. 114 ± 131). Thousand Oaks, CA: Sage. Thompson, J. K., & Altabe, M. N. (1991). Psychometric qualities of the Figure Rating Scale. International Journal of Eating Disorders, 10, 615 ± 619. Thompson, J. K., & Tantleff, S. (1992). Female and male rating of upper torso: Actual, ideal and stereotypical conceptions. Journal of Social Behavior and Personality, 7, 345 ± 354. Thompson, M. A., & Gray, J. J. (1995). Development and validation of a new body-image assessment scale. Journal of Personality Assessment, 2, 258 ± 269. Vandereycken, W. (1990). The relevance of body-image disturbances for the treatment of bulimia. In M. M. Fichter (Ed.), Bulimia nervosa. Basic research, diagnosis and treatment (pp. 85 ± 91). Chichester, England: Wiley. Vitousek, K. B., & Hollon, S. D. (1990). The investigation of schematic content and processing in eating disorders. Cognitive Therapy and Research, 14, 191 ± 214.
490
Riva
Williamson, D. A. (1996). Body image disturbance in eating disorders: A form of cognitive bias? Eating Disorders, 4(1), 47 ± 58. Williamson, D. A., Cubic, B. A., & Gleaves, D. H. (1993). Equivalence of body image disturbances in anorexia and bulimia nervosa. Journal of Abnormal Psychology, 102, 177 ± 180. Wolpe, J. (1958). Psychotherapy for reciprocal inhibition. Palo Alto, CA: Stanford University Press. Wooley, S. C., & Wooley, O.W (1985). Intensive out-patient and residential treatment for bulimia. In D. M. Garner & P. E. Garfinkel (Eds.), Handbook of psychotherapy for anorexia and bulimia (pp. 18 ± 24). New York: Guilford Press.
Computers in Human Behavior, Vol. 14, No. 3, pp. 477 ± 490, 1998 # 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0747-5632/98/$19.00 + 0.00
PII: S0747-5632(98)00018-1
Virtual Environment for Body Image Modification: Virtual Reality System for the Treatment of Body Image Disturbances Giuseppe Riva Applied Technology for Neuro-psychology Laboratory, Istituto Auxologico Italiano Abstract Ð The paper describes the characteristics and preliminary evaluation of the Virtual Environment for Body Image Modification (VEBIM), a virtual reality environment aimed at treating body image disturbances and body dissatisfaction associated with eating disorders. Two methods are commonly used to treat body image: (a) a cognitive/behavioral therapy to influence patients' feelings of dissatisfaction and (b) a visual/motorial therapy with the aim of influencing the level of bodily awareness. VEBIM tries to integrate these two therapeutic approaches within an immersive virtual environment. This choice would make it possible both to intervene simultaneously on all of the forms of bodily representations, and to use the psycho-physiological effects provoked on the body by the virtual experience for therapeutic purposes. As well as the description of the VEBIM theoretical approach, the paper also presents a study on two preliminary samples (71 normal subjects, uncontrolled study; 48 normal subjects, controlled study) to test the efficacy of VEBIM. # 1998 Elsevier Science Ltd. All rights reserved Keywords Ð virtual reality, body image disturbances, eating disorders Requests for reprints should be addressed to Giuseppe Riva, Applied Technology for Neuropsychology Laboratory, Istituto Auxologico Italiano, PO Box 1, 28044 Verbania, Italy. E-mail: [email protected] 477
478
Riva
The physical world, including our body, is not given directly in our experience but is inferred through observation and critical reasoning. In fact, we experience our bodies through two multidimensional cognitive constructs (Fisher, 1990; Gallagher, 1995): the body schema and the body image. According to Head (1926), the body schema is a model/representation of one's own body that constitutes a standard against which postures and body movements are judged. This representation can be considered the result of comparisons and integration at the cortical level of past sensory experiences (postural, tactile, visual, kinaesthetic, and vestibular) with current sensations. This gives rise to an almost completely unconscious ``plastic'' reference model that makes it possible to move easily in space and to recognize the parts of one's own body in all situations. If the body schema can be considered a perceptual model of the body, the body image is a cognitive/social/emotional model. In fact, body image is not only a cognitive construction but also a reflection of wishes, emotional attitudes, and interactions with other. According to Schilder (1950), the body image can be defined as ``the picture of our own body which we form in our mind, that is to say, the way in which the body appears to ourselves'' (p. 11). Expanding on Schilder's idea, Allamani and Allegranzi (1990) refer to body image as ``a complex psychological organization which develops through the bodily experience of an individual and affects both the schema of behavior and a fundamental nucleus of self-image'' (p. 121). Body experiences have a long and well-reported association with eating and weight related problems (Barrios, Ruff, & York, 1989; Garfinkel & Garner, 1982; Riva, 1997a; Rosen, 1990; Thompson, 1990, 1992, 1995). One index of the importance of body experience disturbance involves its relevance to agreed-on clinical disorders. The Diagnostic and Statistical Manual of Mental Disorders (4th edition; American Psychiatric Association, 1994) contains a body image criterion that is required for the diagnosis of anorexia nervosa or bulimia nervosa. It has also been suggested that, when there is psychological comorbidity with obesity, it may be strongly due to problematic body experience issues (Thompson, 1992). Today, researchers and clinicians agree that including an assessment and evaluation of body experience disturbance is crucial to any treatment program targeting obesity or eating disorders. Probably, the disturbances of body image associated with the eating disorders can be conceptualized as a type of cognitive bias (Vitousek & Hollon, 1990; Williamson, 1996; Williamson, Cubic, & Gleaves, 1993). The essence of this cognitive perspective is that the central psychopathological concerns of an individual bias the manner in which information is processed. In most cases, it is presumed that this biased information processing occurs automatically. Also, it is generally presumed that the process occurs more or less outside the person's awareness unless the person consciously reflects upon his/her thought processes (as in cognitive therapy). Mineka and Sutton (1992)
Virtual reality for the treatment of body image disturbances
479
have identified four common types of cognitive bias in research related to depression and anxiety disorders: attentional bias, memory bias, judgmental bias, and associative bias. Three of these four types of cognitive bias have been the focus of research related to eating disorders: preoccupation with body size, body dysphoria, and related problems. In contrast to the great number of publications on body image, only a few papers focus on the treatment of a disturbed body image in eating disorders. In general, two direct and specific approaches can be distinguished: a cognitive/behavioral approach aimed at influencing patients' feelings of dissatisfaction with different parts of their bodies by means of individual interviews, relaxation, and imaginative techniques (Butters & Cash, 1987), and a visual/motorial approach which makes use of video recordings of particular gestures and movements with the aim of influencing the level of bodily awareness (Wooley & Wooley, 1985). An interesting possibility that we tried to address in the Virtual Environment for Body Image Modification (VEBIM) is the integration of the different methods (cognitive, behavioral, and visual/motorial) commonly used in the treatment of body experience disturbances (Riva, 1997b; Riva & Melis, 1997; Riva, Melis, & Bolzoni, 1997; Schlundt & Johnson, 1990) within a virtual environment. In particular we tried to integrate the cognitive methods of Countering, Alternative Interpretation, Label Shifting and Deactivating, the behavioral method of Temptation Exposure with Response Prevention and the visual motorial methods of Awareness of the Distortion and Body Image Modification (Table 1) using the virtual environment in the same way as images in the well-known method of guided imagery (Leuner, 1969). It is Leuner's belief that the imagery evokes intense latent feelings that are relevant to the patient's problems. Guided imagery has been found to be a powerful tool in treatment approach ranging from psychoanalytic therapy (Reyker, 1977) to behaviorism (Wolpe, 1958). A choice of this type would make it possible not only to evoke latent feelings, but also to use the psychophysiological effects provoked by the experience for therapeutic purposes. In practically all virtual reality (VR) systems the human operator's normal sensorimotor loop is altered by the presence of distortions, time delays, and noise. Such alterations, that are introduced unintentionally and usually degrade performance, affect body perceptions too. The somaesthetic systems has a proprioceptive subsystem that senses the body's internal state, such as the position of limbs and joints and the tension of the muscles and tendons. Mismatches between the signals from the proprioceptive system and the external signals of a virtual environments alter body perceptions and can cause discomfort or simulator sickness (Sadowsky & Massof, 1994). Such effects, attributable to the reorganizational and reconstructive mechanisms necessary to adapt the subjects to the qualitatively distorted world of VR, could be of great help during the course of therapy aimed at
480
Riva
Table 1. Therapeutical Methods Integrated in VEBIM Methods
Procedures
Cognitive
Countering: Once a list of distorted perceptions and cognitions is developed, the process of countering these thoughts and beliefs begins. In countering, the patient is taught to recognise the error in thinking, and substitute more appropriate perceptions and interpretations. Alternative interpretation: The patient learns to stop and consider other interpretations of a situation before proceeding to the decision-making stage. The patient develops a list of problem situations, evoked emotions, and interpretative beliefs. The therapist and patient discuss each interpretation and if possible identify the kind of objective data that would confirm one of them as correct. Label shifting: The patient first tries to identify the kinds of negative words he/she uses to interpret situations in his/her life, such as bad, terrible, obese, inferior, and hateful. The situations in which these labels are used are then listed. The patient and therapist replace each emotional label with two or more descriptive words. Deactivating the illness belief: The therapist first helps the client list his/her beliefs concerning eating disorders. The extent to which the illness model influences each belief is identified. The therapist then teaches the client a cognitive/behavioral approach to interpreting maladaptive behavior and shows how bingeing, purging, and dieting can be understood from this framework.
Behavioral
Temptation exposure with response prevention (TERP): The rationale of temptation exposure with response prevention is to expose the individual to the environmental, cognitive, physiological, and affective stimuli that elicit abnormal behaviors and to prevent them from occurring. The TERP protocol is usually divided into three distinct phases: (a) comprehensive assessment of eliciting stimuli, (b) temptation exposure extinction sessions, and (c) temptation exposure sessions with training in alternative responses.
Visual/motorial
Awareness of the distortion: The patient is instructed to develop an awareness of the distortion. This is approached by a number of techniques including the presentation of feedback regarding one's self-image. Videotape feedback is also usually used. The patient is videotaped engaging in a range of activities. Modification of the body image: The patient is instructed to imagine him/herself as different on several dimensions including size, race, and being larger or smaller in particular areas. The patient is also asked to imagine him/herself as younger, older, what the patient looks and feels like before and after eating, as well as before and after academic/vocational and social successes and failures.
influencing the way the body is experienced, because they lead to a greater awareness of the perceptual and sensory/motorial processes associated with them. When a particular event or stimulus violates the information present in the body schema (as occurs during a virtual experience), the information itself becomes accessible at a conscious level (Baars, 1988). This facilitates the process of modification and, by means of the mediation of the self (which tries to integrate and maintain the consistency of the different representations of the body), also makes it possible to influence body image. However, until now some problems have existed with VR. There are anecdotal reports that immersive VR can lead to symptoms similar to motion sickness symptoms. Visually induced motion sickness is a syndrome that occasionally occurs when physically stationary individuals view compelling
Virtual reality for the treatment of body image disturbances
481
visual representations of self motion. It may also occur when detectable lags are present between head movements and presentation of the visual display in the head-mounted display (HMD). Motion sickness is characterized by a diverse set of symptoms but is primarily exemplified by nausea and vomiting. In the case of VEBIM, with adults and adolescents as possible subjects, both including people with psychological and physical problems, the need to verify the problems associated with operating in a virtual environment has to be addressed. A first study was designed to: (a) verify the effects provoked by VEBIM on blood pressure and heart rate and (b) verify the effects provoked by VEBIM on the body experience. A second controlled study was set-up to confirm the results obtained in the first study. METHOD Subjects In the first study, 71 subjects (47 males and 24 females; mean age: 23.187.54 years; mean weight: 67.3415.42 kg; mean height: 173.37.97 cm) participated. A total of 48 women participated in the second study. The sample was randomly divided into two groups: the experimental group (mean age: 22.65.84 years; mean weight: 55.14.43 kg; mean height: 168.296.20 cm) and the control group (mean age: 23.14.63 years; mean weight: 55.83.98 kg; mean height: 171.015.80 cm). All the 119 subjects were recruited from the participants at a VR conference. Characteristics of the Virtual Reality Systems VEBIM is based on a Pentium-based immersive VR system (166 Mhz; 32 MB RAM; graphic engine: Matrox Millennium with 4 MB Warm) including an HMD subsystem with head-tracking and a two-button joystick-type motion input device. A head mounted display, specially developed by Virtual.sys (Milan, Italy) was used. The HMD displays 624 lines of 210 pixel to each eye (52 H and 41 V field of view) and uses LCD technology (two active matrix color LCDs). A Logitech 3D mouse provided head tracking. In VEBIM we did not use a stereoscopic display. Previous researches regard stereoscopy as important because it provides the user with a good cue of depth (Barham & McAllister, 1991). However, the refresh rate of graphics decreases by 50% for the need of two different images, one for each eye. Consequently,
482
Riva
we decided against implementing a stereoscopic display. To compensate for the lack of depth cue, we included perspective cues (light and shade, relative size, textural gradient, interposition and motion parallax) in the virtual environment. The data glove-type motion input device is very common in virtual environments because of its ability to sense many degrees of freedom simultaneously. However the operator is also frequently confused by the difficulty in using it correctly, especially when there is a time delay contained in the feed-back loop. To provide a easy method of motion in VEBIM we used a two-button joystick-type input device: pressing the upper button the operator moves forward, pressing the lower button the operator moves backwards. The direction of the movement is given by the rotation of operator's head. Characteristics of the Virtual Environment VEBIM is a five-zone virtual environment developed using the Sense 8 World ToolKit for Windows (Version 2.02). VEBIM consists of two parts (Zones 1 ± 2 and Zones 3 ± 5). The first two zones are designed both to give the subject a minimum level of skill in perceiving, moving through, and manipulating objects in VR, and to focus attention on eating and food choice. The two environments can also be used by the therapist for a comprehensive assessment of stimuli that could elicit abnormal eating behavior. This assessment is normally part of the Temptation Exposure with Response Prevention protocol. The next three zones are designed to modify the body experience of the subject. Zone 1. In this zone the subject becomes acquainted with the appropriate control device, the head mounted display, and the recognition of collisions. To move into the next zone subjects have to weigh themselves on a virtual balance. The balance is used for two functions: 1. it is intended to be an inevitable obstacle for the user, who must focus his/ her attention on this object, representing the importance of the weight dimension in the experiences to come thereafter; 2. it can be used, if needed by the therapist, to display the initial weight of the subject, as acquired in the dialog box at the beginning of the Body Image Virtual Reality Scale (BIVRS). Zones 2a and 2b. These zones show a kitchen (2a; Figure 1) and an office (2b). Each of these rooms is furnished with typical items, and contains different foods and drinks. When the user decide to eat or drink something, all he/she has to do is to ``touch'' a specific item. In this way the food is ``eaten'' and the
Virtual reality for the treatment of body image disturbances
483
Figure 1. View of Zone 2a showing the kitchen.
corresponding caloric intake is automatically recorded in a text file, which is used later to calculate the total income of calories. Located at the end of the second zone is a second virtual balance. According to the ``eaten'' food and to the caloric intake inserted at the beginning of BIVRS the balance will show the new weight of the subject (in kilograms). Zone 3. In this zone the subject is exposed to a series of panels textured with pictures of models, in the typical way of the advertising world (Figure 2). The images are used as stimuli to support a cognitive approach: the elicited feelings are subject to an analysis by the therapist according to the Label Shifting and Objective Counters methods. The feelings and their associated beliefs are identified, broken down into their logical components, replaced with two or more descriptive words, and then critically analyzed. Zone 4. In this zone the user finds a large mirror. Standing by it the subject can look at his/her real body, previously digitized using an EPSON Photo PC camera. The vision of her own body usually elicits in the user strong feelings that can be matched using the Counterattacking and the Countering cognitive methods. The mirror is also used, as indicated by Wooley and Wooley (1985), to instruct the user to imagine him/herself as different on several dimensions including size, race, and being larger or smaller in particular areas. The user is also asked to imagine him/herself as younger, older, what they look and feel like before and after eating, as well as before and after academic/vocational and social successes and failures.
484
Riva
Figure 2. View of Zone 3 whose walls show images of female and male models.
Zone 5. This zone consists of a long corridor ending with a room containing four doors of different dimensions. The subject can move into the last zone only by choosing the door corresponding exactly to his/her width. This procedure is the same as indicated by Wooley and Wooley (1985) to improve awareness of the body distortion. Procedures and Measures All the subjects were submitted to VEBIM for no more than 10 min and no less than 8 min. All the subjects included in the study reached Zone 5. Just before entering the virtual environment and just after all subjects completed three scales for assessing body experience: 1. The Figure Rating Scale (FRS; Thompson & Altabe, 1991) a set of 9 male and female figures which vary in size from underweight to overweight. 2. The Contour Drawing Rating Scale (CDRS; Thompson & Gray, 1995), a set of 9 male and female figures with precisely graduated increments between adjacent sizes. 3. The Breast/Chest Rating Scale (BCRS; Thompson & Tantleff, 1992), a set of 5 male and 5 female schematic figures, ranging from small to large upper torso.For these tests, subjects rated the figures based on the following instructional protocol: (a) current size and (b) ideal size. The
Virtual reality for the treatment of body image disturbances
485
difference between the ratings is called the self-ideal discrepancy score and is considered to represent the individual's dissatisfaction. The findings of Keeton, Cash, and Brown (1990), support the usefulness of the self-ideal discrepancy score in the assessment of body image, as it was shown to relate to other body-image indices and other clinically relevant measures. All the scales have good test ± retest reliability. In the first study only, we recorded also blood pressure and heart rate. Before the virtual experience the subjects were fitted with a blood pressure cuff and were asked to sit quietly during a 5-min baseline period. During this period blood pressure and heart rate were recorded. After the virtual experience and after a 10-min posttask period blood pressure and heart rate were recorded again. Statistical Analysis A power calculation was made to verify the opportunity to obtain statistically significant differences between the pre- and post-VR scores. Given the low statistical power, we decided to use in both studies the Exact methods, a series of statistical algorithms developed by the Harvard School of Public Health that enable researchers to make reliable inferences when data are small, sparse, heavily tied, or unbalanced (SPSS, 1995). The Exact method used to compare the scores was the Marginal Homogeneity Test (Agresti, 1990). RESULTS Study 1 Table 2 lists the means and standard deviations for the pressure and heart rate scores, across the three measurement occasions. No significant difference was found before and after the VR experience. Only two of the subjects experienced simulation sickness. Table 3 presents the means and standard deviations for the body image scores obtained before and after the virtual experience. The Marginal Homogeneity Test reported significant differences Table 2. Means and Standard Deviations for the Pressure and Heart Rate Scores Measurement 1
Systolic Diastolic Heart rate
Measurement 2
Measurement 3
M
SD
M
SD
M
SD
134.61 89.23 74.01
20.06 21.09 15.68
133.09 88.2 72.11
20.8 23.72 11.06
124.13 84.6 72.09
18.56 16.38 14.14
486
Riva
in the Ideal FRS and Ideal CDRS scores: both scores were higher after experiencing VEBIM. An analysis of the three discrepancy indexes revealed significant lower values for FRS and CDRS after the experience in the virtual environment. These results mean that VEBIM is able to reduce the body dissatisfaction of the subjects. Study 2 Table 4 presents the means and standard deviations for the ratings of body image. The Marginal Homogeneity Test reported significant differences in the Ideal FRS scores obtained by the experimental group: the scores were higher (p < .005) after experiencing the procedure used. An analysis of the discrepancy indexes showed a significant lower value for FRS (p < .05) after the virtual experience: body dissatisfaction of the subjects was reduced after the virtual experience. No significant differences were found in the control group (Table 5). Table 3. Means and Standard Deviations for the Body Image Scores (First Study) Real FRS Before After
3.81.34 3.841.28
Ideal FRS 3.18+/0.72*** 3.53+/0.96***
Real CDRS
Ideal CDRS
Real BCRS
Ideal BCRS
4.791.49 4.741.53
4.110.95* 4.211.07*
2.791.30 2.821.27
2.911.95 2.940.97
FRS=Figure Rating Scale; CDRS=Contour Drawing Rating Scale; BCRS=Breast/Chest Rating Scale.*p < .05; **p < .01; ***p < .005. Table 4. Means and Standard Deviations for the Body Image Scores in the Experimental Group (Second Study) Real FRS Before After
3.140.93 3.350.81
Ideal FRS 2.610.8*** 2.910.7***
Real CDRS
Ideal CDRS
Real BCRS
Ideal BCRS
41.18 4.211.25
3.50.83 3.521.08
2.981.49 2.931.35
2.751.13 2.521.08
FRS=Figure Rating Scale; CDRS=Contour Drawing Rating Scale; BCRS=Breast/Chest Rating Scale.*p < .05; **p < .01; ***p < .005. Table 5. Means and Standard Deviations for the Body Image Scores in the Control Group (Second Study) Real FRS Before After
3.181.03 3.211.01
Ideal FRS 2.710.8 2.850.74
Real CDRS
Ideal CDRS
Real BCRS
Ideal BCRS
3.81.1 4.011.05
3.480.73 3.490.88
3.011.01 31.12
2.680.95 2.51.09
FRS=Figure Rating Scale; CDRS=Contour Drawing Rating Scale; BCRS=Breast/Chest Rating Scale.
Virtual reality for the treatment of body image disturbances
487
DISCUSSION Although there is much potential for the use of immersive VR environments in clinical psychology, some problems have limited their application in this field. Some users have experienced side-effects, during and after exposure to VR environments. The symptoms experienced by these users are similar to those which have been reported during and after exposures to simulators with wide field-of-view displays (Cobb, Nichols, & Wilson, 1995; Kennedy, Lanham, Drexler, Massey, & Lilienthal, 1995; Regan, 1995; Regan & Price, 1994). These side-effects have been collectively referred to as ``simulator sickness'' (Kennedy, Lilienthal, Baltzley, Allgood, & Gower, 1987) and are characterized by three classes of symptoms: ocular problems, such as eyestrain, blurred vision and fatigue; disorientation and balance disturbances; and nausea. Exposure duration of less than 10 min to immersive VR environments has been shown to result in significant incidences of nausea, disorientation, and ocular problems (Regan & Price, 1993). The first interesting result of this study is the lack of side effects and simulation sickness in our samples after the experience in the virtual environment, confirming the possibility of using VEBIM for therapeutic purposes. This result, confirmed in both studies, is even more interesting given the sample used. In fact, females tend to be more susceptible to motion sickness than males (Griffin, 1990). However it is difficult to generalize this result to the VR systems at large given the peculiar technical characteristics of VEBIM. The other obtained result is the reduction in the body dissatisfaction of the subjects after the virtual experience. Usually body-image treatment involves a cognitive/behavioral or a visuomotor therapy that requires many sessions. The possibility of inducing a change in the body dissatisfaction after a 8 ± 10-min VR session can be useful to improve the efficacy of the existing approaches. As such, the procedure might be helpful as a part of a comprehensive treatment package to break through the ``resistance'' to treatment in clinical subjects (Vandereycken, 1990). Of course these results are preliminary only. From a clinical view point the issues that we have to address in the future are: 1. Further testing of VEBIM. Even if the data obtained in the two preliminary studies using VEBIM are very promising, we have to test the new environments with clinical and nonclinical subjects. 2. A follow-up study to check how long the influence of the virtual environment lasts. In the preliminary studies we have limited the VR experience to just one session, but from a therapeutic viewpoint it seems more reasonable to repeat the procedure.
488
Riva
3. How to integrate VEBIM into the usual cognitive/behavioral treatments. Even if the procedure might be helpful as a part of a comprehensive treatment package to break through the ``resistance'' to treatment in clinical subjects, it is important to define a specific therapeutic protocol to be used by other clinicians and therapists. Acknowledgments Ð This research is part of the Virtual Reality Environments for Psychoneuro-physiological Assessment and Rehabilitation (VREPAR) Projects, two European Commission funded research projects (DGXIIIÐTelematics for Health Care).
REFERENCES Agresti, A. (1990). Categorical data analysis. New York: John Wiley & Sons. Allamani, A., & Allegranzi, P. (1990). Immagine corporea: dimensioni e misure. Una ricerca clinica [Body Image: Dimensions and measures. A clinical research]. Archivio di Psicologia Neurologia Psichiatria, 2, 171 ± 195. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Baars, B. J. (1988). A cognitive theory of consciousness. New York: Cambridge University Press. Barham, P. T., & McAllister, D. F. (1991). A comparison of stereoscopic cursors for the interactive manipulation of B-splines. Proceedings of Stereoscopic Display and Applications, II, 18 ± 26. Barrios, B. A., Ruff, G., & York, C. (1989). Bulimia and body image: Assessment and explication of a promising construct. In W. G. Johnson (Ed.), Advances in eating disorders (Vol. II). New York: Jay Press. Butters, J. W., & Cash, T. F. (1987). Cognitive-behavioral treatment of women's body image satisfaction: A controlled outcome-study. Journal of Consulting and Clinical Psychology, 55, 889 ± 897. Cobb, S., Nichols, S., & Wilson, J. R. (1995). Health and safety implications of virtual reality: In search of an experimental methodology. Paper presented at the Framework for Immersive Virtual Environments Conference, University of London. Fisher, S. (1990). The evolution of psychological concepts about the body. In T. F. Cash & T. Pruzinsky (Eds.), Body images: Development, deviance and change (pp. 3 ± 20). New York: The Guilford Press. Gallagher, S. (1995). Body schema and intentionality. In J. L. Bermdez, A. Marcel, & N. Eilan (Eds.), The body and the self (pp. 24 ± 38). Cambridge, MA: MIT Press. Garfinkel, P. E., & Garner, D. M. (1982). Anorexia nervosa: A multidimensional perspective. New York: Brunner & Mazel. Griffin, M. J. (1990). Handbook of human vibration. London: Academic Press. Head, H. (1926). Aphasia and kindred disorders of speech. London: Cambridge University Press. Keeton, W. P., Cash, T. F., & Brown, T. A. (1990). Body image or body images: Comparative, multidimensional assessment among college students. Journal of Personality Assessment, 54, 213 ± 230. Kennedy, R. S., Lanham, D. S., Drexler, J. M., Massey, C. J. and Lilienthal, M. G. (1995). Cybersickness in several flight simulators and VR devices: A comparison of incidences, symptom profiles, measurement techniques and suggestions for research. Paper presented at the Framework for Immersive Virtual Environments Conference, University of London.
Virtual reality for the treatment of body image disturbances
489
Kennedy, R. S., Lilienthal, M. G., Baltzley, D. R., Allgood, G. O., & Gower, D. W. (1987). Consistency across measures of simulator sickness: Implications for a biocybernetic safety reporting device. Paper presented at the SAFE Association 25th Annual Symposium, Las Vegas. Leuner, H. (1969). Guided affective imagery: A method of intensive psychotherapy. American Journal of Psychotherapy, 23, 4 ± 21. Mineka, S., & Sutton, S. K. (1992). Cognitive biases and the emotional disorders. Psychological Science, 3, 65 ± 69. Regan, E. C. (1995). An investigation into nausea and other side-effects of head-coupled immersive virtual reality. Virtual Reality, 1(1), 17 ± 32. Regan, E. C., & Price, K. R. (1993). Some side-effects of immersion virtual reality: The effects of increasing head movements, of rapid interaction, and of seated subjects. Report 93R022, Army Personnel Research Establishment, Farnborough. Regan, E. C., & Price, K. R. (1994). The frequency of occurrence and severity of side-effects of immersion virtual reality. Aviation, Space and Environmental Medicine, 65, 527 ± 530. Reyker, J. (1977). Spontaneous visual imagery: Implications for psychoanalysis, psychopathology and psychotherapy. Journal of Mental Imagery, 2, 253 ± 274. Riva, G. (1997a). Representations of eating among adolescent Italian girls. The Journal of Social Psychology, 137(2), 205 ± 217. Riva, G. (1997b). The Virtual Environment for Body-Image Modification: Development and preliminary evaluation. Presence, 6(1), 106 ± 117. Riva, G., & Melis, L. (1997). Virtual reality for the treatment of body image disturbances. In G. Riva (Ed.), Virtual reality in neuro-psycho-physiology (pp. 95 ± 111). Amsterdam: IOS Press. Riva, G., Melis, L., & Bolzoni, M. (1997). Treating body-image disturbances. Communications of the ACM, 40(8), 69 ± 71. Rosen, J. C. (1990). Body image disturbances in eating disorders. In T. F. Cash & T. Pruzinsky (Eds.), Body images: Development, deviance and change (pp. 190 ± 214). The Guilford Press: New York. Sadowsky, J., & Massof, R. W. (1994). Sensory engineering: The science of synthetic environments. Johns Hopkins APL Technical Digest, 15, 99 ± 109. Schilder, P. (1950). The image and appearance of the human body. New York: International Universities Press. Schlundt, D. G., & Johnson, W. G. (1990). Eating disorders: Assessment and treatment. Needham Heights, MA: Allyn & Bacon. SPSS. (1995). SPSS Exact tests 6.1 for Windows. Chicago: Author. Thompson, J. K. (1990). Body image disturbance: Assessment and treatment. Elmsford, NJ: Pergamon Press. Thompson, J. K. (1992). Body image: Extent of disturbance, associated features, theoretical models, assessment methodologies, intervention strategies, and a proposal for a new DSM IV diagnostic categoryÐBody Image Disorder. In M. Hersa, R. M. Eisler, & P. M. Miller (Eds.), Progress in behavior modification (pp. 35 ± 51). Sycamore, IL: Sicamore. Thompson, J. K. (1995). Assessment of body image. In D. B. Allison (Ed.), Handbook of assessment methods for eating behaviors and weight related problems (pp. 114 ± 131). Thousand Oaks, CA: Sage. Thompson, J. K., & Altabe, M. N. (1991). Psychometric qualities of the Figure Rating Scale. International Journal of Eating Disorders, 10, 615 ± 619. Thompson, J. K., & Tantleff, S. (1992). Female and male rating of upper torso: Actual, ideal and stereotypical conceptions. Journal of Social Behavior and Personality, 7, 345 ± 354. Thompson, M. A., & Gray, J. J. (1995). Development and validation of a new body-image assessment scale. Journal of Personality Assessment, 2, 258 ± 269. Vandereycken, W. (1990). The relevance of body-image disturbances for the treatment of bulimia. In M. M. Fichter (Ed.), Bulimia nervosa. Basic research, diagnosis and treatment (pp. 85 ± 91). Chichester, England: Wiley. Vitousek, K. B., & Hollon, S. D. (1990). The investigation of schematic content and processing in eating disorders. Cognitive Therapy and Research, 14, 191 ± 214.
490
Riva
Williamson, D. A. (1996). Body image disturbance in eating disorders: A form of cognitive bias? Eating Disorders, 4(1), 47 ± 58. Williamson, D. A., Cubic, B. A., & Gleaves, D. H. (1993). Equivalence of body image disturbances in anorexia and bulimia nervosa. Journal of Abnormal Psychology, 102, 177 ± 180. Wolpe, J. (1958). Psychotherapy for reciprocal inhibition. Palo Alto, CA: Stanford University Press. Wooley, S. C., & Wooley, O.W (1985). Intensive out-patient and residential treatment for bulimia. In D. M. Garner & P. E. Garfinkel (Eds.), Handbook of psychotherapy for anorexia and bulimia (pp. 18 ± 24). New York: Guilford Press.