Course:COGS200/2017W1/group3

From UBC Wiki

Abstract

The research of Bailenson et. al suggest that gaze has an effect on the retention and persuasion of a message being communicated[1]. The use of gaze in the effectiveness of group therapy for alcohol addicts will be studied using virtual reality. Participants will be placed in one of three same-gendered groups of seven individuals. The first group will experiment with an augmented gaze, the second with a reduced gaze, and the third with a neutral gaze. This experiment will span a twelve week study period. Each session will be incrementally increased up to an hour in length in order to accustom the participants' eyes to the virtual reality system. There will be three phases to each testing group to allow for the implementation of the validation sets and data analysis. A survey will also be administered at the end of the experiment to test the level of comfort participants felt with other group members during the study. The results of the study will indicate that an augmented gaze, compared to the other two groups, correlates to an increased level of comfort for the participants. Implementations of study and further research are discussed.

Introduction

Our proposed project is to study the effects augmented gazes have on the success of group therapy. Our study pool will consist of alcohol-addiction rehabilitation patients as they are often sharing personal, relatable, stories. As rehabilitation facilities also limit the access to the outside world, this will also enable us to conduct the experiment in a controlled setting. To determine the effectiveness of our study we will conduct surveys with our participants after each phase to determine how comfortable they felt with the other individuals in the group and how well they connected to each individual.

Background Research on Augmented Gaze

Non-verbal cues such as eye contact are strong human signals that are used to regulate face-to-face conversations and have been proven to increase intimacy between speaker and receiver. Direct gaze allows for an intimate connection to pass between the speaker and the receiver which opens up the receiver to be more positively perceptive to the message of the speaker [1]. In a face-to-face conversation humans have physical limitations which allow them to focus their gaze on only one individual concurrently, however using computer mediated communication, such as a virtual environments, removes that limitation since human movements and behaviors are monitored and re-created by avatars in the virtual world. This process is simultaneous and allows for communication that is closely related to face-to-face conversation without being in the same physical location. This environment allows an individual's gaze to be focused on multiple recipients[1]. These findings can be applied to the group therapy component of alcohol addiction rehabilitations, as a large part of rehabilitation is group therapy which involves speaking, sharing and listening.

Uncanny Valley

Avatars use in a virtual world raises important questions about how human-like the avatars should appear. The uncanny valley suggests that people are not comfortable with a robot showing too many human-like features. If the robot appears to be too human, people become uncomfortable and in most cases express their dislike for the robot[2]. Generally, people were comfortable with robots carrying out actions but became uncomfortable when feelings and sensations were introduced[2]. This research can be expanded into the avatars in the VR system - We need to be careful not to make them too human-like. We will need to design each of the elements of our avatars so that they each match in realism, since a shift in one element can cause an unease in the participants[2]. We will also need to refrain from making the avatars too basic, as the group participants will not expect to have human interactions with them if they are too appliance-like[2]. Using research already conducted[2] we are confident that we can find an adequate equilibrium between the structure of the avatars and the comfort each participant will have with it.

Background research on Group Therapy

Group therapy reduces isolation and enables its members to witness the recovery of others which creates a culture of recovery within the group. Since group therapy usually occurs with individuals who are in a time of pain, trouble, or crisis, having the support and comfort from a group can help participants navigate through these difficult times and can enrich members with insights and guidance from individuals sharing common experiences[3]. It is for these reasons that group therapy can be more beneficial than individual therapy. Also, if an individual is able to emotionally become attached to another group member, group leader, or the group as a whole, then the relationship has the potential to change the person and assist them in committing to recovery[3]. It is for this reason that we want to test the use of gaze in group therapy - We want to know if adjusting the gaze using virtual reality can improve the building of these relationships and help the individuals relate to each other in a more effective way.

Importance of Eye Contact

Eye contact is a crucial component in non-verbal communication. This is evident when infant and mother match gaze - The eye contact solidifies the connection between mother and child and helps build a feeling of safety and security between the two [4]. This phenomenon of attachment and bonding illustrates how eye contact influences the emotional connection between individuals. Also, as evidenced by the relation between therapist and client, when sensitive topics arise, eye contact can be difficult to hold. Additionally, when people feel guilt or shame, they avoid eye contact. This can create a feeling of disconnection between therapist and client, and can lead to the client dreading or quitting therapy[4]. Since the individuals in group therapy will be discussing sensitive topics, as well as sharing stories of guilt and shame, it will be difficult for them to hold eye contact. By using the VR system to conduct group therapy, this will reduce the need for the participants in our group therapy sessions to make eye contact, and will not hinder the relationships and bonds created between the participants.

Hypothesis

Extrapolating from prior research we know that augmenting the gaze in a small group setting can alter the perception of the audience [1] - Our hypothesis is that implementing this in a group therapy session will help the audience relate to the other individuals, as well as assist in creating personal bonds between the participants. Previous research conducted between doctors and patients indicates that non-verbal communication, such as eye contact, correlated to the likability and connectedness a patient felt for a doctor[5]. Increased eye contact was also connected to the perception of empathy[5]. This research leads us to believe that increasing the eye contact will create more empathy between group participants, and will create emotional connections within the group. We also believe that decreasing the gaze will create shallower, less emotional connections.

Methods

Setup

The study will consist of two experimental groups and one control group. Our first experimental group will be exposed to a group therapy session in which each member is the product of an increased augmented gaze. This means that each individual group participants will perceive a gaze from the speaker which imitates direct eye contact (ie: 100% gaze all the time). The second experimental group will consist of a reduced gaze. This means that each individual group participants will perceive absolutely no direct eye contact from the speaker (ie: 0% gaze all of the time). The control group will be exposed to neutral gaze which can be defined as the middle ground between the reduced and increased gazes (ie: 50% of the time gaze).

Each of the groups will consist of seven participants and be single gendered. Clinical literature suggests that the ideal group size for group therapy is approximately seven people, because a larger group will result in frustration in the diminished amount of time that each individual is able to participate, and a group smaller will result in diminished group interaction[6]. The results of a study done by Prendergast et al. show that single gender group therapy is more effective in treating substance abuse issues, such as alcohol abuse[7]. The group therapy will take place in a neutral coloured room and the avatar will be setup to reflect each individual’s likeness.

The length of the study will be twelve sessions long, and the sessions will build up to be an hour in length. The first four sessions will be thirty minutes in length, the next four will be forty five minutes in length and the final four will be an hour in length. The buildup is necessary because virtual reality can cause potential long term eye problems if not treated properly [8]. Other symptoms that users experience if they are brought into VR for too long for their first exposure include dizziness, nausea and tension headaches[8]. The research of Burlingame et al. suggests that a trial of twelve sessions is optimal to build cohesion amongst group members through interaction[9]. At the end of the twelve session study, a survey will be provided to the participants of the study with detailed questions regarding their level of comfort they felt with the other group members and if they related to their stories.

At the start of the study, there will be a short calibration session to set up the VR for each participant, but as the participants will be seated and not moving around the environment, this will not be time-consuming. However, the calibration is critical to ensure that the eye tracking software and virtual environment are acting simultaneously.

Sample Survey for Group Therapy Participants

Validation Sets

Phase 1

To allow for implementation of the validation sets and data analysis, we will have 3 phases to our testing in each testing group. The first phase in the Reduced Gaze (RG) test set will contain no eye contact by the speaker in the VR system. The system will be recording data on the amount of eye contact the other group members make with the speaker, when this eye contact is made, and where else the group member looks. This data collection will be conducted by the eye tracking inside the VR helmets, as well as through the VR system. This data set will be used to create our first validation set and will be averaged for all participants. This data will then be used to improve our algorithm for phase 2.

The first phase in the Increased Gaze (IG) test set will consist of direct contact from the speaker to each group member simultaneously. Similar to the RG test set, with the help of the eye-tracking software, we will record the amount of eye contact the group makes with the speaker, when this eye contact occurs, and when the group adjusts their eyes to the speaker’s body. This validation set will be used to improve our algorithm for phase 2.

The Neutral Gaze (NG) will be the control group for both test sets. In the first phase of this control set, we will conduct the group therapy session with a natural gaze. The speaker in the VR will mimic the actions that the speaker actually makes, but will incorporate no body movement since the first phases of our test sets do not incorporate body movement. Using the eye-tracking software we will record when the group participants break eye contact with the speaker as this information will be our first validation set for the NG.

After completing phase 1, each test set will complete a survey outlining their levels of comfort in the group therapy, how connected and empathetic they felt to the speaker, and how relatable they felt the speaker was. This data will be used to determine if our experiment was successful.

We will also reconfigure our algorithms to adjust the gaze according to the data collected. The participant levels of discomfort will come be matched from the survey and the eye-tracking software, whereas the dis-engagement of gaze will solely come from the eye-tracking software. In the RG test set, just prior to when group participants felt uncomfortable, looked away, or became disengaged, we will increase the gaze from the speaker to each individual simultaneously. In the IG test set we will concurrently decrease the gaze prior to the group participants feeling discomfort or dis-engaging in the gaze with the speaker. Our NG test set gaze will not be altered for phase 2.

In all test sets, when the group participants began to look towards the body of the speaker, we will incorporate subtle hand and body movements to increase the likeliness of the VR, as well as to continue to engage the group participants. We will also incorporate facial movements when the group members were focused on the face of the speaker. The eye-tracking data will infer which areas of the face to move and when to move them.

Phase 2

The second phase of the RG will use the data gathered from the first validation set and will adjust the gaze as follows: Three seconds prior to the time mark when the average group member looked away, felt uncomfortable, or felt disconnected, the gaze from the speaker to the audience will be temporarily increased for all audience members - That is, each audience member will receive direct eye contact for 3 seconds. We will also incorporate body movements into the VR in this phase to help individuals engage with the speaker. When the average audience gaze wanders to the body of the speaker the body movement of the speaker will be increased. During this phase, the same information as the first phase will be collected - The amount of eye contact made by the audience, when this eye contact is made, and when the group member looks at the body of the speaker.

The second phase of the IG will also use the improved algorithm from the first validation set. We will decrease the gaze of the speaker to all participants three seconds prior to the time when the average group member looked away from, felt uncomfortable, or felt disconnected from the speaker. We will also incorporate minor body movements when the gaze of the participants strays to the body of the speaker, as well as minor facial expressions when the average participant is looking directly at the face of the speaker. Eye-tracking will be used in this phase to record when the participants gaze shifts, and where it shifts to.

The second phase of the NG will not augment the gaze - It will only use the validation set to augment the body movement and facial expressions of our test case. Since the NG is our control group it is important not to change the gaze in this test set. Again, we will use the eye-tracking to record where and when the participants gaze wanders.

Upon completion of phase 2, we will have the participants fill out the survey inquiring about their levels of comfort, connectedness, and empathy towards the speakers. Based on previous studies[1][5] our hypothesis for phase 2 of this test set is that there will be a higher comfort level and connectedness with the increased eye contact and body movement in the RG. These engagements will hold the interest of the audience longer and create more empathy than phase 1. While no previous studies indicate our expectations for the alterations made to the IG test set, our intuitive prediction is that this too will increase the connectedness and comfort level the group participants felt to the speaker. We also expect an increased engagement between phase 1 and phase 2 of the NG, since we have incorporated body movement and facial expressions.

We will create a second validation from the data collected in phase 2, and will test this in phase 3 of our experiment.

Phase 3

In phase 3, much like phase 2, we will continue to collect data and implement the updated algorithm from the second validation set. We are hopeful that the improved algorithm will yield more idealistic results than in phase 2.

Discussion

Predictions

Referring back to the original hypothesis of the study, we predict that the participants will feel most comfortable with augmented gaze compared to reduced and neutral gaze. Our predictions are extrapolated from the findings of Bailenson et. al, who found that an increased augmented gaze can increase persuasion of messages presented[1]. One critical component of alcohol abuse group therapy is the ability to receive multiple perspectives, encouragement, and empathy. Therefore, we predict that with an increased augmented gaze, the messages presented by group members will be more persuasive and the participants will be able to relate to the messages presented by the other group members. With this level of relatability and empathy, the group members will feel more comfortable and as a result the group members will feel more engaged and willing to open up in a safe environment. Thus, a higher the level of comfort and empathy will lead to a more effective group therapy treatment.

Why This Method

The reason we chose this method is because gaze can be used to influence perspectives and relatability, and by implementing an augmented gaze in VR we hope to improve the experience of alcohol addicts during group therapy sessions and encourage the support the need to commit to recovery. Since it is difficult for participants to maintain eye contact during group therapy [4], we will need to use the VR system to facilitate group therapy and alter the gaze of the speaker since participants will find it extremely difficult to share personal stories while attempting to alter their own gaze. The VR system will enable us to manipulate the speaker’s gaze to be looking at multiple people simultaneously, as well as enabling us to increase or decrease the amount of time spent on an augmented gaze. By using the data collected, and implementing group therapy sessions using VR, we hope to increase the effectiveness of group therapy sessions, leading to an increase in alcohol addict recoveries.

Why Alcohol-Addicted Participants

Alcohol substance abuse damages the ability to create meaningful connections and relationships with other individuals and decreases the ability to function in a social environment. It can also impair the cognitive function and the physical abilities of the individual[10]. The implications of alcohol addiction makes it difficult for the addict to integrate back into society due to the severe alteration of brain activity[11]. Reports of alcohol consumption in Canada and the United states concluded that millions of citizens are consuming enough alcohol to be a risk of serious health and mental effects[12] [13]. Due to alcohol addiction being an issue in society, more individuals are left needing the support of rehabilitation and recovery. Additionally, alcohol-dependent individuals are sharing emotional stories of their relatable struggles during group therapy sessions which will give us an ideal test group. For these reason, we chose alcoholic addicts as our prime participants.

Multi-Disciplinary Approach

While addictions have a physical and biological component, most addictive behaviours are psychologically based[14]. Psychological factors, such as stress or a history of trauma, can lead an individual to desire re-exposure to a substance or behaviour, often resulting in an addiction. This regular exposure can alter the cortical and sub-cortical regions of the brain - the regions of the brain associated with reward, motivation, memory, impulse control, and judgement[14]. These brain alterations can create cravings for the addiction which are often facilitated due to the inability to successfully regulate this impulse[14]. Through group therapy, these addictive behaviours can be put into perspective and the individual gets the reassurance that they are not alone[15]. Other members can offer ideas for improving a difficult situation or struggle, and can also hold the addict accountable[15]. These methods have proven to be successful in many group therapy situations[3]. Our study will assist in creating bonds with other group members. Even though this approach will not directly address the physical alterations made to the brain, it will address the psychological behaviours that these alterations cause.

Our study also incorporates computer science. We will implement three phases with validation sets between the phases. After each study phase, the computer algorithms will be adjusted to incorporate the data from the previous validation set - These algorithms will alter the amount of time the virtual realities spends engaging a gaze, as well as incorporating body language. The computers will need to be programmed to address the changes in gaze, as well as using the eye-tracking software to collect data on when the gaze needs to be altered. We will also be incorporating virtual reality into our experiment. Since one person is not able to physically hold the same gaze for every individual in a group, we require the virtual reality to adjust the gaze, since every participant aside from the speaker will be perceiving the same gaze from the speaker.

Conclusion

The process of conducting a study is one that requires time, patience and funding. Throughout this process it was refreshing to navigate through the world of experimental design and essentially creating a novel research idea using background information from multiple sources and disciplines. With alcohol addiction being a prevalent issue in society, we sought to explore measures that can improve the group therapy component of the rehabilitation process.

Incorporating the fields of psychology and computer science laid the groundwork to begin exploring the advantages of virtual reality and the psychological effects it can have on individuals that are participating. A critical aspect of group therapy session is sharing personal stories and creating bonds with other addicted individuals. Since eye contact is difficult when individuals are sharing feelings of guilt or shame, as many alcoholics do, and eye contact is essential for creating bonds and empathy, using the virtual reality system will enable connections to be formed as the gaze can be digitally altered. The focus of our study explores how eye contact between a speaker and a group can be optimized to create more meaningful connections and bonds. Virtual reality is also advantageous as it allows for a manipulation of direct eye contact, which would otherwise not be achievable. Using virtual reality also brought some complications to our experiment as using avatars to represent humans could create unease in individuals due to the uncanny valley. One of the areas that we were not able to address in our study was personal bias and how it affects the emotional connection between a speaker and group members. This is an area of research that could further be studied.

As with any scientific study, multiple replications of our experiment would be encouraged to increase confidence with our outcomes. If the results of the experiment align with the hypothesis presented, we would encourage rehabilitation centers to implement group therapy sessions using an augmented gaze with a virtual reality system. Our hope is that this technology based group therapy technique can be implemented in hopes of improving the standard of living and recovery rates of alcoholic addicts.

Future Research

For future research we hope to implement real-time updates to the gaze in the VR system using data from previous validation sets as well as the real-time eye-tracking. This area of research was outside of the scope of our experiment, and pending a relationship between our augmented gazes and emotional connectedness, this would be the next step in our research. Also, personalizing the VR to each individuals interactions would be an additional factor to consider. Instead of using averages to run the phases, individual statistics could be analyzed creating a more personalized experience. An area of research that we could not address in our study was the personal biases. In future studies, these personal biases of the group therapy participants could be negated. To reduce the personal biases, un-gendered, generic avatars could be used in the study to determine if that improves or alters the outcome of the study. This would, however, cause issues with the uncanny valley and should be considered further.

References

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 Bailenson, J. N. (2005). Transformed social interaction, augmented gaze, and social influence in immersive virtual environments. Human Communication Research, 31(4), 511-537. doi:10.1093/hcr/31.4.511
  2. ↑ 2.0 2.1 2.2 2.3 2.4 Gray, K., & Wegner, D. (2012). Feeling robots and human zombies: Mind perception and the uncanny valley. Cognition, 125(1), 125-130. doi:10.1016/j.cognition.2012.06.007
  3. ↑ 3.0 3.1 3.2 Treatment, Center for Substance Abuse, & NCBI Bookshelf. (2005). Substance abuse treatment: Group therapy Substance Abuse and Mental Health Services Administration (US).
  4. ↑ 4.0 4.1 4.2 Howes, R. (2012, February 13). Eye Contact in Therapy, Part I. Retrieved November 23, 2017, from https://www.psychologytoday.com/blog/in-therapy/201202/eye-contact-in-therapy-part-i
  5. ↑ 5.0 5.1 5.2 Montague E, Chen P, Xu J, Chewning B, Barrett B. Nonverbal interpersonal interactions in clinical encounters and patient perceptions of empathy. J Participat Med. 2013 Aug 14; 5:e33.
  6. ↑ Cohen, S. L., & Rice, C. A. (1985). maximizing the therapeutic effectiveness of small psychotherapy groups. Group, 9(4), 3-9. doi:10.1007/BF01458260
  7. ↑ Prendergast, M. L., Messina, N. P., Hall, E. A., & Warda, U. S. (2011). The relative effectiveness of women-only and mixed-gender treatment for substance-abusing women. Journal of Substance Abuse Treatment, 40(4), 336-348. doi:10.1016/j.jsat.2010.12.001
  8. ↑ 8.0 8.1 Chester, R. (2017, February 24). Eye sore: Optometrists warning of eye problems with VR headsets. Retrieved November 27, 2017, from http://www.news.com.au/technology/gadgets/wearables/health-warning-as-immersive-virtual-reality-craze-linked-to-vision-problems/news-story/a67849532e82d857be7a3524c91ef11e
  9. ↑ Burlingame, G. M., McClendon, D. T., & Alonso, J. (2011). Cohesion in group therapy. Psychotherapy, 48(1), 34-42. doi:10.1037/a0022063
  10. ↑ What Are the Problems & Effects of Alcoholism on Families & Marriages. (n.d.). Retrieved November 27, 2017, from https://americanaddictioncenters.org/alcoholism-treatment/family-marital-problems/
  11. ↑ Leshner, Alan I. "Addiction is a Brain Disease, and it Matters." Science, vol. 278, no. 5335, 1997, pp. 45-47
  12. ↑ Canada, P. H. (2016, February 03). The Chief Public Health Officer's Report on the State of Public Health in Canada, 2015: Alcohol Consumption in Canada. Retrieved from https://www.canada.ca/en/public-health/services/publications/chief-public-health-officer-reports-state-public-health-canada/2015-alcohol-consumption-canada.html
  13. ↑ Alcohol Facts and Statistics. (n.d.). Retrieved November 27, 2017, from https://www.niaaa.nih.gov/alcohol-health/overview-alcohol-consumption/alcohol-facts-and-statistics
  14. ↑ 14.0 14.1 14.2 Addiction. (n.d.). Retrieved November 24, 2017, from https://www.psychologytoday.com/basics/addiction
  15. ↑ 15.0 15.1 Johnson, B. (n.d.). Retrieved November 24, 2017, from http://www.apa.org/helpcenter/group-therapy.aspx
Retrieved from "https://wiki.ubc.ca/index.php?title=Course:COGS200/2017W1/group3&oldid=482181"