Virtual Reality Training for Aggression Control

Virtual Reality Training for Aggression Control in a Dutch Prison-based Population - a Randomized Controlled Trial

The main goal of this randomized controlled trial is to investigate the effectiveness of VR-TRAC (Virtual Reality TRaining for Aggression Control) for reducing aggression in a prison-based population. The study-design is a single-blind randomized controlled trial, comparing VR-TRAC to waiting-list control condition (WL). 128 male detainees with aggression regulation problems in the last month (measured with the Aggression Questionnaire [AQ]) and a minimum age of 18 years, recruited from the Penitentiary Institution (P.I.) Vught, the Netherlands will participate in this study. They are randomly allocated to VR-TRAC or control condition WL. The treatment group fills in questionnaires, participates in role-plays and follows the VR-TRAC. The control group filles in the same questionnaires as the treatment group and also participates in the role-plays, but does not participate in the VR-TRAC. Participants receive Care As Usual (CAU) when necessary. To measure the effect of VR-TRAC on aggression, three different types of measurements are used: staff observation, self-report and performance-based. Self-report questionnaires will be scored on three different moments during the study: before the treatment starts, at the end of the treatment, and two- months after the treatment ended. Throughout the sessions, participants are also asked to answer questions to evaluate the sessions. Lastly, to measure the effectiveness of the skills trained in the VR-TRAC, performance-based assessments (role-play tests and vignettes) will be conducted before and after the treatment period.

In an earlier pilot study the Virtual Reality Aggression Prevention Training (VRAPT), developed by Klein Tuente, was explored and refined for a prison-based population. Changes were made and the training was adjusted to Virtual Reality Training Aggression Control (VR-TRAC). In this study the effectiveness of VR-TRAC will be explored. The main goal of this randomized controlled trial is to investigate the effectiveness of VR-TRAC (Virtual Reality TRaining for Aggression Control) for reducing aggression in a prison-based population. The study-design is a single-blind randomized controlled trial, comparing VR-TRAC to waiting-list control condition (WL). 128 male detainees with aggression regulation problems in the last month (measured with the Aggression Questionnaire [AQ]) and a minimum age of 18 years, recruited from the Penitentiary Institution (P.I.) Vught, the Netherlands will participate in this study. They are randomly allocated to VR-TRAC or control condition WL. The treatment group fills in questionnaires, participates in role-plays and follows the VR-TRAC. The control group filles in the same questionnaires as the treatment group and also participates in the role-plays, but does not participate in the VR-TRAC. Participants receive Care As Usual (CAU) when necessary. CAU in prison consists of treatment with the main focus to stabilize a disrupted psychological state (such as pharmacological treatment, supportive contact or a transfer to a Penitentiary Psychiatric Centre where necessary interventions are applied to stabilize the disorder). The training consists of 16 twice-weekly sessions, with a duration of 60 minutes each. The first four sessions focus on the early stages of information processing (what is happening and what does it mean). Session five through eight focus on the late information processing stages (what goals am I trying to achieve, what options do I have to react, what am I going to do, and what is the reaction or behavior). Session 10 through 15 combines the early and late stages, as all newly learned behavior will be incorporated in the interactive scenarios. To train the aforementioned stages, different aggressive-inducing situations are practiced in VR. To measure the effect of VR-TRAC on aggression, three different types of measurements are used: staff observation, self-report and performance-based. Firstly, through staff-observation, using weekly scores of the Observation Scale for Aggressive Behaviour (OSAB) throughout the study period. Secondly, individual changes are measured through self-report measurements. The questionnaires will be scored on three different moments during the study: before the treatment starts, at the end of the treatment, and two- months after the treatment ended. Four questionnaires will be used to measure different types of aggression, namely the Aggression Questionnaire (AQ) (which will also be used as a screening measurement), Difficulties in Emotion Regulation (DERS), the Novaco Anger Scale and Provocation Inventory (NAS-PI), and the Reactive-Proactive Questionnaire (RPQ). To measure impulsiveness the Barratt Impulsiveness Scale (BIS-11) will be used. One questionnaire, the Short Anger Measure (SAM), will be used to measure feelings of anger in the last week. It consists of 12 short items measuring feelings of anger. The questionnaire will be scored weekly and starts four weeks before the treatment starts and will end four weeks after the last treatment. Participants are also asked to answer some questions to evaluate the session through the SRS (Session Rating Scale). Two additional self-report questionnaires will be conducted, one on childhood trauma with the Adverse Childhood Experiences (ACE) and one on substance abuse with the Addiction for Triage & Evaluation (MATE), Lastly, to measure the effectiveness of the skills trained in the VR-TRAC, performance-based assessments (role-play tests and vignettes) will be conducted before and after the treatment period.

Participants assigned to the VR-TrAC group will follow the training: Virtual Reality Training for Aggression. The training consists of 16 twice-weekly sessions, with a duration of 60 minutes each. The treatment protocol used in this training is based on the Virtual Reality Aggression Prevention Training (VRAPT), developed by Klein Tuente et al. (2020). The first four sessions focus on the early stages of information processing (what is happening and what does it mean). Session five through eight focus on the late information processing stages (what goals am I trying to achieve, what options do I have to react, what am I going to do, and what is the reaction or behavior). Session 10 through 15 combines the early and late stages, as all newly learned behavior will be incorporated in the interactive scenarios. To train the aforementioned stages, different aggressive-inducing situations are practiced in VR.

Participants receive Care As Usual (CAU) when necessary. CAU in prison consists of treatment with the main focus to stabilize a disrupted psychological state (such as pharmacological treatment, supportive contact or a transfer to a Penitentiary Psychiatric Centre where necessary interventions are applied to stabilize the disorder).

Virtual Reality Training for Aggression. The training consists of 16 twice-weekly sessions, with a duration of 60 minutes each.

The primary outcome will be measured through a self-report measurement, the 'Aggression Questionnaire (AQ)', wich measures four different types of aggression. The Aggression Questionnaire (AQ) consists of 29 items measuring aggression on four different scales: physical aggression, verbal aggression, anger, and hostility. Items are filled in on a 5-point Likert scale. Higher scores indicate more aggression.

Before the treatment, 2 months after baseline (posttreatment), and 2 months after the treatment ended (follow-up).

The Difficulties in emotion regulation (DERS) consists of 36 items measuring difficulties in emotion regulation. The DERS also provides scores on six subscales: nonacceptance of emotional responses, difficulty engaging in goal-directed behavior when distressed, impulse control difficulties when distressed, lack of awareness of emotions, limited access to strategies for regulation, and lack of emotional clarity. Items are filled in on a 5-point Likert scale. Higher scores indicate more problems with emotion regulation.

Before the treatment, 2 months after baseline (posttreatment), and 2 months after the treatment ended (follow-up).

The Novaco Anger Scale and Provocation Inventory (NAS-PI) consists of two parts, The NAS part contains 48 questions and measures three factors; cognitive, arousal and behavior. The PI part contains 25 items that refer to anger-eliciting situations, rated on a 4-point Likert scale. Higher scores indicate more anger and issues with provocation.

Before the treatment, 2 months after baseline (posttreatment), and 2 months after the treatment ended (follow-up).

The Reactive-Proactive Questionnaire (RPQ) consists of two 23 items (11 items on reactive aggression and 12 items on proactive aggression), rated on a 3-point Likert scale. Higher scores indicate more reactive and proactive aggression.

Before the treatment, 2 months after baseline (posttreatment), and 2 months after the treatment ended (follow-up).

The Short Anger Measure (SAM) is a self-report and consists of 12 items measuring angry feelings and aggressive impulses over the last week, rated on a 5-point Likert scale. Higher scores indicate more aggression.

The Barratt Impulsiveness Scale (BIS-11) measures the personality/behavioral construct of impulsiveness. The questionnaire consists of 30 items, which provide a total score; six first-order factors: attention, motor, self-control, cognitive complexity, perseverance, cognitive instability, and three second-order factors: attentional, motor, non-planning, rated on a 4-point Likert scale. Higher scores indicate more impulsiveness.

Before the treatment, 2 months after baseline (posttreatment), and 2 months after the treatment ended (follow-up).

The IGroup Presence Questionnaire (IPQ) will measure the experience of presence in VR. It consists of 14 items. Items are rated on a 7-point Likert scale. Higher scores indicate a better experience in VR.

For the staff observation, the Observation Scale for Aggressive Behaviour (OSAB) will be used. The OSAB consists of 40 items measuring emotions/mood, aggressive behavior, the reason for the aggressive behavior, sanction for the patient, and socially competent behavior. Items are measured on a 4-point Likert scale. Higher scores indicate more aggressive behavior or less competent behavior.

Every week, four weeks before the start of the treatment and end at four weeks after the last treatment ended.

To measure the effectiveness of the skills trained in the VR-TRAC, scripted role-play assessments will be conducted. Participants in the controlgroup as well in the experimental group will participate in scripted role plays. The scripted role plays consists of aggresive role-plays. Participants must react as they normally would do. The role-plays are filmed and scored. Higher scores indicate less skills or more aggressive reactions.

To measure the effectiveness of the skills trained in the VR-TRAC, vignettes will be conducted. Participants in the controlgroup as well in the experimental group will fill in vignettes. They consist of aggressive scenarios. Participants are asked to write down how they would react normally on the scenario.Higher scores indicate less skills or more aggressive reactions.

Inclusion Criteria: Detainees who are imprisoned in P.I. Vught, The Netherlands, for at least 20 weeks. Detainees with aggression regulation problems in the last month, as measured with the AQ (a minimum score of 70). Minimum age of 18 years old. Exclusion Criteria: SCIL score of 14 or lower. (SCIL: a screener for IQ and is scored for every detainee as part of the reintegration plan inside detention). Acute suicidal behavior or current psychotic episode. Insufficient command and understanding of the Dutch language. Photosensitive epileptic seizure in the past year. An estimated stay of 5 months or shorter

Klein Tuente S, Bogaerts S, Bulten E, Keulen-de Vos M, Vos M, Bokern H, IJzendoorn SV, Geraets CNW, Veling W. Virtual Reality Aggression Prevention Therapy (VRAPT) versus Waiting List Control for Forensic Psychiatric Inpatients: A Multicenter Randomized Controlled Trial. J Clin Med. 2020 Jul 16;9(7):2258. doi: 10.3390/jcm9072258.