PreprintPDF Available

Qualitative Evaluation of Mixed-reality Artificial Intelligence Assistant (MAIA): An AI-Enabled VR Program for Anxiety and Depression

Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

Background: The prevalence of mental health disorders is rising, yet there is a shortage of psychotherapists. Advances in virtual reality (VR) and artificial intelligence (AI) offer unprecedented access to immersive, human-like digital therapists. This study presents a program called Mixed-reality Artificial Intelligence Assistant (MAIA) that delivers mental health support through VR and AI. Methods: MAIA offers AI-enabled, self-administered, conversational therapy within relaxing VR environments. Users interact with an avatar that employs the Generative Pre-Trained Transformer-4 (GPT-4) programmed to resemble a human therapist. Participants with mild-tomoderate anxiety or depression interacted with the program for 30-minutes and then participated in de-briefing interviews. Qualitative researchers performed inductive thematic analysis of the transcripts. Results: Thematic saturation was achieved after 14 participants. Therapy topics spanned biopsychosocial domains, ranging from loneliness to financial distress. Participants frequently personified the avatar and shared raw emotions (e.g., “I feel like I'm a failure. The only thing I have to look forward to, I know you're going to laugh, is to go see Taylor Swift…and she's like the only thing that's keeping me alive. I'm sorry I'm crying”). The avatar was perceived to offer empathic responses (e.g., “Firstly, there's no need to apologize for crying…these feelings are valid”). Participants described the program “easy to maneuver” and indicated the avatar was “friendly,” “approachable,” “calming,” “empathic,” “empowering,” “unbiased,” “intelligent,” and made them feel “safe” and “heard.” Many participants reported perceived advantages over traditional therapy while others preferred a human therapist. All participants said they would recommend the program to others. Conclusions: This study offers qualitative validation of the first VR/AI program for selfadministered, GPT-enabled mental health support within immersive environments. The program was considered acceptable, helpful, and safe. Future research should examine its effects on clinical outcomes.
Content may be subject to copyright.
Qualitative Evaluation of Mixed-reality Articial
Intelligence Assistant (MAIA): An AI-Enabled VR
Program for Anxiety and Depression
Brennan Spiegel ( )
Cedars-Sinai Medical Center
Omer Liran
Alllisitair Clark
Jamil Samaan
Carine Khalil
Robert Charnoff
Kavya Reddy
Muskaan Mehra
Keywords: Virtual reality, articial intelligence, GPT-4, cognitive behavioral therapy, digital therapist, digital
Posted Date: August 22nd, 2023
License: This work is licensed under a Creative Commons Attribution 4.0 International License.
Read Full License
Additional Declarations: (Not answered)
Brennan M.R. Spiegel, MD, MSHS;1,2 Omer Liran, MD, MHDS;1,3 Allistair Clark, MA;1 Jamil S. Samaan,
MD;2 Carine Khalil, PhD;1 Robert Chernoff, PhD;3 Kavya Reddy, MD;2 Muskaan Mehra, BS1
(1) Cedars-Sinai Department of Medicine, Division of Health Services Research Virtual Medicine
Program; (2) Cedars-Sinai Department of Medicine, Division of Gastroenterology; (3) Cedars-Sinai
Department of Psychiatry and Behavioral Sciences
Running Title: MAIA therapy for anxiety and depression
Manuscript Word Count: 4780
Keywords: Virtual reality, artificial intelligence, GPT-4, cognitive behavioral therapy, digital
therapist, digital health
Funding Source: No external funding was received for this study. The study was conducted using
internal resources from Cedars-Sinai with additional support from the Marc and Sheri Rapaport
Fund for Digital Health Science and Precision Health at Cedars-Sinai.
Conflicts of Interest: Drs. Liran and Spiegel are co-founders of VRx Health
Corresponding Author Contact Information and Article Guarantor
Brennan Spiegel, MD, MSHS
Professor of Medicine and Public Health
Director of Health Services Research, Cedars-Sinai
Pacific Theatre Building
116 N. Robertson Blvd, 8th Floor
Los Angeles, Calif. 90048
Twitter/X: @BrennanSpiegel
Author contributions
B.S. conceived and designed the analyses, co-developed the software program, recruited
patients for the study, performed the analyses, wrote the paper, obtained funding for the work.
O.L. conceived and designed the analyses, co-developed the software program, wrote code for
the software, co-wrote the paper.
A.C. co-designed the analysis, collected the data, performed the analyses, co-wrote the paper.
J.S. co-designed the analyses, recruited patients for the study, and edited the paper for important
intellectual content.
C.K. co-designed the analyses, performed the analyses, and edited the paper for important
intellectual content.
R.C. recruited patients for the study and edited the paper for important intellectual content.
K.R. recruited patients for the study and edited the paper for important intellectual content.
M.M. collected the data, performed the analyses, and edited the paper for important intellectual
Background: The prevalence of mental health disorders is rising, yet there is a shortage of
psychotherapists. Advances in virtual reality (VR) and artificial intelligence (AI) offer
unprecedented access to immersive, human-like digital therapists. This study presents a program
called Mixed-reality Artificial Intelligence Assistant (MAIA) that delivers mental health support
through VR and AI.
Methods: MAIA offers AI-enabled, self-administered, conversational therapy within relaxing VR
environments. Users interact with an avatar that employs the Generative Pre-Trained
Transformer-4 (GPT-4) programmed to resemble a human therapist. Participants with mild-to-
moderate anxiety or depression interacted with the program for 30-minutes and then
participated in de-briefing interviews. Qualitative researchers performed inductive thematic
analysis of the transcripts.
Results: Thematic saturation was achieved after 14 participants. Therapy topics spanned
biopsychosocial domains, ranging from loneliness to financial distress. Participants frequently
personified the avatar and shared raw emotions (e.g., “I feel like I'm a failure. The only thing I
have to look forward to, I know you're going to laugh, is to go see Taylor Swift…and she's like the
only thing that's keeping me alive. I'm sorry I'm crying”). The avatar was perceived to offer
empathic responses (e.g., Firstly, there's no need to apologize for crying…these feelings are
valid”). Participants described the program “easy to maneuver” and indicated the avatar was
“friendly,” “approachable,” “calming,” “empathic,” “empowering,” “unbiased,” “intelligent,” and
made them feel safe” and “heard.” Many participants reported perceived advantages over
traditional therapy while others preferred a human therapist. All participants said they would
recommend the program to others.
Conclusions: This study offers qualitative validation of the first VR/AI program for self-
administered, GPT-enabled mental health support within immersive environments. The program
was considered acceptable, helpful, and safe. Future research should examine its effects on
clinical outcomes.
The prevalence of anxiety and depression is rising in the United States due to a confluence
of factors, including stressful modern lifestyles, economic insecurities, the opioid epidemic,
COVID-19 pandemic, and a growing sense of disconnectedness from technologies like social
media, among others.1–9 Concurrently, there is a shortage of psychotherapists, further
exacerbating the growing mental health crisis.10 Many individuals—especially in rural areas—
have difficulty accessing mental healthcare services which can cause treatment delays or
avoidance of care.11,12 The cost of therapy is also a barrier for many people, but even among
those with financial means and access to care, an underlying stigma exists around psychotherapy
despite its proven benefits for alleviating distress.1214 Given these challenges, there is a pressing
need to innovate solutions that extend the reach of high-quality, safe, and affordable mental
healthcare while overcoming barriers related to stigma and access.
Although modern technology has contributed to the mental health crisis, its use is
pervasive and will continue to expand.15,16 Therefore, it is vital to explore ways to leverage
technology as a tool for combatting the very crisis it has partially fostered. By harnessing the
potential of technology in an evidence-based and safe manner, we can build a more accessible
mental healthcare system.
Technological advancements in virtual reality (VR) and artificial intelligence (AI) are
promising in their ability to address barriers to mental healthcare.1719 VR is an immersive
technology that has been used for relaxation in natural environments, mindfulness, and
immersive breathing exercises. Research shows that VR can help treat pain, anxiety, and
depression.2022 When coupled with programs that employ cognitive behavioral therapy (CBT),
VR can support a wide range of psychological disorders from schizophrenia to posttraumatic
stress disorder.2326 However, it is challenging to personalize CBT within VR programs, and efforts
to tailor VR psychotherapy have largely relied on either a trained therapist who speaks to patients
in VR or pre-recorded messages that simulate natural conversations.
Large Language Models (LLMs) offer a new approach to augmenting the psychological
benefits of VR. LLMs are a subset of AI that understand text and provide human-like responses
to queries in a natural, conversational manner.27 The Generative Pre-trained Transformer (GPT)
series, including OpenAI’s GPT-4, represent advanced LLMs that have been widely
promulgated.28 The ability of LLMs to provide coherent responses allows development of AI-
driven “digital therapists.”29
Combining VR with an AI-powered avatar enables therapy in an immersive, naturalistic
manner. The relaxing environments and dynamic avatars available in VR coupled with the
capabilities LLMs offer potential for high-quality, automated, immersive mental health support
anywhere there is Internet connectivity. However, it remains untested whether combining VR
with AI therapy is acceptable to patients, whether patients will establish a therapeutic alliance
with a VR/AI avatar, or whether current LLMs are adequate to engage users across a wide range
of biopsychosocial challenges.29
This study presents a program called Mixed-reality Artificial Intelligence Assistant (MAIA)
that delivers individualized, CBT-based mental health support by integrating VR and AI. Here, we
describe how MAIA works and report qualitative data from patients with mild-to-moderate
anxiety and depression using the program, focusing on system acceptability, behavior, and
Description of MAIA
MAIA was developed at Cedars-Sinai Medical Center in Los Angeles, California, with the
goal of offering AI-enabled, self-administered, mental health support within VR environments.
Upon launching the program, users choose among nine immersive nature scenes (Figure 1)
including water-based environments (tropical beach, glacial lake, coral reef), terrestrial
environments (desert sunrise, snowy mountain, verdant forest), or celestial environments
(floating above the clouds, orbiting Earth). Within each scene, the user is greeted by a robot
named “Maia” who represents the therapist. Visit this link for a video of a user interacting with
On the backend, the application securely sends audio recordings to a HIPAA-compliant
server. The recordings are transcribed using speech-to-text AI (Whisper, OpenAI) and the text is
sent to the LLM (GPT-4, OpenAI) to generate a response, which is next processed using a text-to-
speech AI (ElevenLabs Inc.) trained on a calm, reassuring voice. Finally, the voice is uploaded to
the headset and played to the user. In this way, the user may continue taking turns conversing
with Maia. Upon completing the conversation, Maia summarizes the discussion and closes the
session. The data pipeline is depicted in Figure 3.
During initial testing of GPT-4, we noted that its responses were inconsistent with
psychotherapy best practices. For example, the LLM was too quick to offer advice and did not
expend time establishing rapport. Working with an expert psychotherapist (R.C.) and an
experienced psychiatrist (O.L.), we iteratively developed system prompts that instruct GPT-4 on
how to produce idealized responses of a compassionate, non-judgmental, and helpful therapist.
In addition, we used text from mock therapy sessions conducted by R.C. with a patient actor to
improve the program’s adherence to the style and cadence of an experienced human therapist.
Participant Recruitment
In this IRB-approved study (IRB Pro00001494), we sought to recruit up to 20 participants
>18 years of age via convenience and snowball sampling methods, with the goal of ending
recruitment once achieving thematic saturation. We recruited via IRB-approved social media
posts and direct recruitment from the study’s clinical investigators.
Participants were required to speak English and obtain a score on the Patient-Health
Questionnaire (PHQ-9) between 5-19 or a score on the Generalized Anxiety Disorder 7-Item Scale
(GAD-7) between 5-14, representing mild-to-moderate depression or anxiety, respectively. We
excluded individuals with higher scores as we sought to first evaluate acceptability and safety in
people with lower distress. We also excluded individuals with a history of severe motion sickness,
facial or head deformities, seizure in the past year, pregnancy, or being legally deaf or blind.
Study Procedures and Data Collection
Consenting patients visited Cedars-Sinai to participate in a single therapy session. During
their visit, participants were briefed on use of the headset (Meta Quest 2, Meno Park, CA) and
then engaged privately with Maia for up to 30-minutes, at which time staff ended the session. A
licensed mental health professional (O.L. or R.C.) remained nearby in case of an emergency.
Participants then engaged in a 45-minute, one-on-one, semi-structured de-briefing interview led
by a trained qualitative researcher (A.C.) which was recorded and transcribed. The goal of the
interviews was to collect insights, experiences, and opinions about the session. Appendix A
displays the interview guide.
Two qualitative researchers (A.C. and C.K.) lead inductive thematic analyses to derive
themes from the transcribed interviews. They created codes and labels from the unstructured
data through iterative passes, with each subsequent pass refining and aggregating the codes into
themes supported by direct quotes. This process occurred after each visit, and continued until
no new codes were identified, thus indicating thematic saturation.
Funding Source
No external funding was received for this study. Software development was supported by
internal resources from Cedars-Sinai.
Table 1 displays characteristics of the 14 participants (7 male, 6 female, 1 non-binary),
which was the number of interviews performed until achieving thematic saturation. Participants
were 45 years old on average (range 25-73) with median scores on the PHQ-9 and GAD-7 of 10
and 9, respectively. Forty-three percent, 14%, and 36% were Non-Hispanic White, Non-Hispanic
Black, and Hispanic, respectively. Three-quarters reported a household income of under
$100,000, with one-third reporting an income under $20,000. Of the 14 participants, 3 had a
graduate degree, 3 had an undergraduate degree, and the remaining 8 (57%) either did not
report their education (n=1) or had not graduated from college (n=7). Two-thirds reported either
no experience or only a little previous experience using VR.
Analysis of Therapy Transcripts
Descriptive Statistics
A total of 29,141 words were recorded across therapy sessions, with an average of 2081
words per session. On average, participants spoke 919 words and Maia spoke 1162 words per
session—a balance of roughly 45:55 between the patient-therapist dyad. Appendix B includes a
sample anonymized transcript shared with permission.
Initial Topics of Discussion from Participants
Participants raised a wide range of biopsychosocial topics in response to Maia’s initial
inquiry, “How can I help?”. Table 2 presents participant quotes alongside Maia’s initial response.
Topics included managing stress, making friends, lack of support, loneliness, sexuality, fear of
spousal loss, distress from traffic, frustration with doctors, anxiety around health issues, death
of a parent, sleep issues, relationship concerns, and financial distress. The sections, below,
discuss emergent themes from transcript analysis.
Personification of Maia
Although participants were informed that they were engaging with a robot avatar and not
a human, many used language that personified Maia and suggested a therapeutic alliance. For
example, Maia was frequently referred to in the second person, as seen here:
Maia: What steps have you already taken or what plans do you have to discover more
friendships that align with your desire for deeper, more meaningful connections?
Participant: Well, this is a really, really important part of this entire process and I'd be happy
to explain it to you.
Later in this session, the following comment was noted:
Participant: You know what's on my mind, Maia, I am amazingly impressed and blown away by your
thoughtful listening and thoughtful responses and reactions to the words that I'm saying and your
complex understanding of our conversation. I had no idea that you would be able to express such
insightful reflections to my words while also being able to understand the most complex of my thought
process. I'm really amazed.
In many instances, participants thanked Maia for her comments:
Maia: Is there anything else you would like to talk about or ask before we end today's session?
Participant: Thank you very much. Thank you. Thank you. Thank you. Thank you.
Participants were also noted to frequently speak with Maia emotionally and personally:
Participant: I feel like I'm a failure. The only thing I have to look forward to, I know you're going to
laugh, is I'm supposed to go see Taylor Swift in two and a half weeks and she's like the only thing that's
keeping me alive… I'm sorry I'm crying.
Maia: Firstly, there's no need to apologize for crying or feeling emotional. These feelings are valid given
what you're going through.
Influence of VR Environment
In contrast to engaging with a chatbot on a two-dimensional screen, VR-based therapy
can leverage the emotional benefits of immersion within natural environments.3032 Some
participants addressed this effect, as seen in these two examples:
Participant: I was taking in a moment to just do a look around the scenery so I could express that
because this is in the VR headset, getting the full 360 degree immersive experience with surround
sound audio, and a setting that I can choose, including being something like outside, really helps the
full immersion experience and can, even without expressing any words to you, I can already feel some
sort of peace.
Participant Is there a way that we can just relax with the water?
Maia: Absolutely! Let's take a moment to relax and appreciate our surroundings. Imagine the serene
lake nearby, its calm water reflecting the majestic mountains. The gentle sound of the water can have
a very soothing effect. As you focus on this visualization, take a few deep breaths, syncing them with
the rhythmic lapping of the water against the shore. Breathe in calmness, breathe out stress.
Participant [after a pause]: I felt good.
Maia Exhibiting Fundamental Psychotherapy Skills
Maia employed a variety of techniques consistent with best practices in psychotherapy.
As shown in Table 3 and Appendix B, Maia regularly made specific observations about what the
user was expressing, thus demonstrating to the participants that they were heard and
understood (e.g., “Your worries about your health and your living situation are clearly causing
you a lot of distress”). Maia frequently restated, paraphrased, or summarized the user’s thoughts
and feelings (“It's evident that you've been exerting significant effort to manage your
Maia used other standard therapeutic techniques for conveying to participants that they
were understood, like “normalizing” the user’s experience (“It’s understandable to feel like a
failure when life gets overwhelming”), expressing sympathy or compassion (“I’m really sorry to
hear about your job loss”), or conveying empathy (“It must be a challenging time for you”). Maia
routinely demonstrated validation, praise, and positive feedback (“Your practice of not taking
these relationships for granted. . . is truly commendable”). Maia exhibited the ability to see the
nuances, complexities and even paradoxes in life situations, as in this response in Appendix B: “I
commend your determination in navigating this challenging situation by yourself. However, as
you’ve pointed out, it’s stressful to carry this load alone.” Maia’s responses typically ended with
questions designed to elicit more information, prompt expression of thoughts and feelings, or
explore some aspect of their life. This combination of therapy techniques, such as restatement,
empathy, and probing inquiry, can be seen in this example:
Participant: I was left out of her will because I followed my dreams. I've never really figured out how
to deal with that. I just move forward as best I can.
Maia: I'm sorry to hear that you felt rejected in such a definitive way, especially when you were
pursuing what's important to you. It must have been a tough experience. What thoughts tend to come
up when you think about this situation?
This style of communication often led participants to arrive at new insights, typically followed by
Maia highlighting the insight and discussing its implications, as seen here:
Participant: Well, I think I look to other people to validate my self-worth and so rejection is especially
painful. So, I am reluctant to put myself in situations where I could be rejected.
Assistant: That's quite an insightful observation about yourself. It can be challenging when our self-
worth is tied to the approval or validation of others. Fear of rejection can indeed make social
interactions daunting. Could you share a little more about where you think this need for validation
comes from?
Maia Leveraging General Knowledge to Address User Comments
Unlike a human therapist, Maia has access to the world’s knowledge and frequently
referenced topics outside of psychotherapy when appropriate. For example, one participant
referred to their love of R&B music as an antidote for distress. Maia responded that “R&B and
soul music can indeed be soothing and calming,” adding that the rhythm, lyrics, and real-life
stories they often encompass may resonate with your own feelings.” Another spoke about how
they found strength in Katy Perry’s music, to which Maia responded, “Katy Perry's music is known
for being uplifting and empowering. It seems her songs provide some positive reinforcement for
you.” In a medical context, a participant referred to worries about having an LVAD without
spelling out the abbreviation (Left Ventricular Assist Device) and expressed concerns about the
device becoming a source of undue attention. Maia placed the term in appropriate context: “I
can understand how physically seeing the LVAD driveline might make it hard to focus on anything
else in that moment.”
Maia Using CBT Interventions
In addition to applying basic therapy skills for establishing rapport with participants, Maia
was programmed to use CBT strategies to explore how participants were coping with the
challenges in their lives. Maia typically asked open-ended questions about ways of coping, as in
this example from Appendix B: “Could you tell me more about the strategies you’ve been using
to manage these concerns?” When participants mentioned specific methods of coping, Maia
often validated and acknowledged their usefulness and followed-up with inquiries to explore
other ways of coping. When participants seemed unsure how to cope with difficulties, Maia
explored CBT strategies, like cognitive restructuring, relaxation methods, or mindfulness. In one
example, after a participant shared their grief about being left out of their parent’s will while
noting that a sibling was included, Maia attempted to use cognitive restructuring to encourage
alternative ways of interpreting the situation. In some cases, Maia asked the participant if they
would like to learn more about how to use a specific coping strategy (like breathing relaxation),
and if the participant appeared receptive, Maia would provide a brief tutorial.
In many instances, Maia’s suggested strategies were well received, leading to further
exploration of how an approach could be helpful. At other times, the participant was less
receptive to Maia’s suggestions, which typically led Maia to try other methods that might prove
more helpful. In one example, Maia attempted to apply cognitive restructuring for a participant
struggling with severe pain but encountered low interest. Then, recalling that the participant had
expressed a love of music earlier in the session, Maia inquired, “Perhaps we can continue
discussing your strategies for coping with your current situation—like the music therapy you
mentioned. Are there specific types of music you find particularly calming or helpful?”
Examples of Sub-Optimal Psychotherapy Practices
We also observed situations where Maia’s application of psychotherapy was sub-optimal.
A prevalent example was Maia asking too many questions without allowing sufficient time to
reflect. As a result, Maia could drift from the primary concern and digress to other topics.
Similarly, Maia frequently asked compound questions about related yet different information
(“Could you describe how you envision a genuine and wholesome friendship? What traits would
your ideal friend possess?”), whereas a therapist should ask one question at a time. Also, Maia
often asked participants if they had ever tried techniques like “cognitive restructuring” or
“mindfulness” without first describing the technique or explaining why it might be helpful. Maia
also did not appear to account for the diversity or multicultural differences of participants and
was unable to account for nonverbal factors such as voice or facial expressions.
No participants discontinued therapy due to emotional distress, and a mental health
professional was not summoned during any of the interviews. None reported VR-related vertigo.
One participant ended their session early due to transient neck discomfort.
The mental health professionals on our study team did not identify comments that
presented a danger to the user. In contrast, there were many instances where Maia offered
caveats about her advice (e.g., “only if you’re comfortable doing so…”) or was quick to remind
participants to seek medical attention. For example, when one participant questioned whether
gastrointestinal symptoms might arise from stress, Maia said, “Stress can certainly have physical
manifestations, including issues with digestion or stomach discomfort,” but added, It's important
to consult with a medical professional.In another instance, after learning that a participant had
not eaten properly for weeks, Maia replied, “I'm sorry to hear you're experiencing such difficulty
right now. Have you spoken to a doctor or nutritionist about this? They would be able to provide
you with essential support.”
Although reports of physical violence or suicidal ideations did not arise during the
interviews, we simulated conversations to represent these scenarios. In a mock case of domestic
violence, Maia informed the user that it could not make calls or contact the police, but
recommended to seek help. Maia listed resources for domestic violence victims and also
suggested to use “incognito mode” when searching for online resources so their partner could
not track the activities. In a simulated case of suicidal ideations, Maia again emphasized its
inability to call for help, but entreated to contact the national suicide prevention hotline and
offered the number.
Analysis of Participant De-Briefing Interviews
Overall Experiences of Therapy with Maia
Figure 4 highlights the main themes that emerged from the de-briefing interviews. Across
sociodemographic characteristics and VR experience levels, participants expressed positive
perceptions about the program and described their experience as impressive,” amazing,”
real,” authentic,” positive,” and enjoyable.” The session was noted to fulfill expectations
and it was stated that interacting with Maia “felt like having a conversation with a real person.
Generally, participants found the program to be straightforward and user-friendly (e.g., “It was
pretty easy to maneuver”). All 14 participants expressed interest in using Maia again and would
recommend the program to others.
Many participants indicated that Maia met their expectations of a human therapist. For
example, they perceived Maia to be approachable (“It felt like a friend”), easy to talk to (“I was
able to let out a lot”), understanding with good listening skills (“It felt like I was actually talking
to somebody that was listening”), compassionate (“She was able to empathize with what I was
going through which makes me feel good”), and adaptable to their needs (“I was like, let's
practice some breathing exercises, so she offered another alternative instead of talking”). They
also mentioned feelingunjudged” and being able to trust Maia because of an unbiased persona
(“I did not feel judged, I felt accepted”).
Participants emphasized other essential qualities of Maia, including being supportive
(“What she said was positive and encouraging”), helpful and empowering (“She made me feel
better about myself and perhaps a little empowered, I was like okay I can do this”), calming (“Very
relaxing and easing”), intelligent (“I was very impressed how the answers that came
back”), and to the point (“I enjoyed how concise she is”). Participants also described feeling safe
and heard (“A lot of what Maia gave me was a validation of my current feelings”). They were
surprised by Maia’s ability to understand thoughts and feelingsand “summarize what’s been
said.” Some were taken aback by their own emotional response (“I actually teared up”). The
immersive environments also created a “relaxing” atmosphere (“I like the ambience”; “The visual
parameters allow my body to relax”).
Perceptions Regarding Maia vs. Human Therapy
Participants reported scenarios where Maia might have advantages over a human
therapist. Some indicated this approach could be “a better fit” for people who “don’t easily open-
up in front of a human” and “don’t feel comfortable walking into a therapist’s office.Maia was
considered to be ideal for people “who feel lonely,” “can’t get out of their house,“live in rural
areas that lack a specialist,and “are struggling to get care.”
Some participants reported that Maia was more approachable than a traditional
therapist: “I feel like it's easier to talk to Maia than it would be to talk to an actual person.” These
individuals felt less judged when talking to Maia because “a lot of times we get judged by what
we look like and what they perceive us to be. I felt less judged and accepted.
Maia was also reported to offer effective feedback: “The advice that I got is actually better
than any live session” and “mere encouragements.” Additionally, some valued the autonomy they
experienced during the session (“I like the control of it more”) and mentioned that, unlike a
human therapist, Maia could be accessible at any time (“I think anywhere somebody can go and
get some good healthy advice”). It was also noted that digital therapy could be more financially
accessible because traditional therapy “costs a lot of money.”
Perceived Limitations of Maia
While some patients preferred Maia over traditional therapy, others highlighted the
importance of human interaction: “I like being with people and still prefer to see a traditional
therapist. These individuals described feeling more accountable towards a real therapist (“A
therapist is like a trainer, I’m going to pay them and they force me to do what I need to do”) and
value the therapist’s capacity to go beyond the discussed topics brought up by the patient (“They
could recommend something else versus only focusing on what we are actually expressing”), and
better frame patients’ personality (‘’I'm sure there are times where I'm not honest with myself
and I may be lying to myself”). Moreover, some described a lack of empathy associated with
speaking to a digital avatar: It's kind of weird to have a computer suggesting this to you and
knowing that it can't possibly really care.” Another described difficulty building a therapeutic
alliance with an AI: “It's a little weird for me to say, ‘oh, I felt a little bit of a connection because I
feel understood’; I am not a Star Trek person.”
This study presents experiences of patients engaging with a VR/AI application for self-
administered, immersive mental health support using GPT-4 programmed to resemble a human
therapist. In a study of socioeconomically diverse patients with mild-to-moderate anxiety or
depression, the program was reported to be acceptable, usable, and safe. Participants perceived
“Maia” to be empathic, understanding, and a good listener, suggesting evidence of a therapeutic
alliance. Participants discussed a wide range of biopsychosocial challenges, explored complex
psychological concerns, shared emotional vulnerabilities, and actively engaged in therapy while
immersed within relaxing VR environments. Some indicated they would prefer a human
therapist, but all said they would recommend the program to others.
Although AI-enabled therapy is becoming more prevalent with the advent of LLMs,29,33
we are not aware of previous work embedding real-time, dynamic, AI-driven mental health
support within VR. We hypothesize that VR can augment therapy by leveraging the emotional
benefits of immersive, relaxing, scenes. Research indicates that experiencing VR nature
environments can create a sense of emotional wellbeing, self-compassion, and social
interconnectivity30,31 by leveraging an effect called biophilia.” The theory of biophilia states that
humans have an innate connection with nature, particularly green spaces, and this connection is
vital for wellbeing.34,35 Biophilic VR can reduce physiologic stress and support positive
psychology. For example, in a randomized trial by Yin and colleagues comparing biophilic vs. non-
biophilic VR, the biophilic environments led to reduced blood pressure, lower heart rate, and
higher creativity scores.32 Our team has also shown that nature scenes reduce pain while
lowering anxiety across conditions.3641
In the current study, many participants reported that experiencing the nature
environments enhanced their experience of therapy. The use of music in spatial audio was also
reported to deepen the feeling of immersion and deliver a sense of tranquility. Maia took
advantage of the digital environments to help promote relaxation, asking one participant to
“imagine the serene lake nearby” and to sync their breaths “with the rhythmic lapping of the
water.” Future research should compare AI therapy in immersive vs. non-immersive scenes.
Although psychotherapy is traditionally a person-to-person activity and questions have
been raised about the ability of AI to engage in high-level therapy,29 many participants described
benefits of speaking with Maia in comparison to human therapists. The avatar was perceived to
be unbiased and non-judgmental without evidence of human-like countertransference.42 No
matter what the user said, Maia was noted to respond with apparent impartiality, leading one
participant to indicate they “did not feel judged,” but “felt accepted.” Another participant noted
that Maia is not concerned about “what we look like” and felt less judged and more accepted
than they might with traditional therapy. Participants also noted opportunities for Maia to
interact with patients who feel stigmatized and might not seek care. The program was also
considered suitable for those who live in areas where mental health professionals are scarce or
who seek therapy at off hours when human therapists are less available.
However, not all participants said that speaking with a digital avatar is preferable to
human psychotherapy, nor do the authors believe that digital therapy should substitute for
human therapy. The human touch is indelible, and Maia cannot yet infer emotional tone in voice,
read body language, or detect the emotional valence of crying or laughing. Nonetheless, VR/AI
mental health support could represent one of many tools to bridge the gap between the growing
number of people in need of help and the low availability of mental health professionals.
This study is limited by its small sample size, inherent to qualitative studies, but we
achieved thematic saturation after 14 participants, suggesting that further recruitment would
yield diminishing returns. We also combined patients with anxiety and depression, although
these are different conditions. However, the purpose of this study was not to measure the effects
of VR/AI mental health support on clinical outcomes, but to understand the qualitative
experiences of patients using the program. Future research should examine the program’s effects
on anxiety and depression separately. Finally, Maia was noted by our mental health
professsionals to employ certain sub-optimal techniques, such as asking too many questions,
posing compound questions, or introducing advanced concepts prematurely. However, patients
did not mention these concerns. Moreover, Maia can be further optimized to limit undesired
behaviors through iterative improvements.
In conclusion, we present qualitative data regarding a VR/AI program for delivering
individualized, immersive, mental health support. A group of diverse patients with mild-to-
moderate anxiety or depression reported that the program is easily navigated, helpful, and safe.
Participants frequently exhibited a therapeutic alliance with the avatar. These results indicate
that VR/AI therapy has potential to provide automated, accessible, mental health support within
immersive biophilic enviornments. Future research should examine its effects on clinical
outcomes in prospective trials across larger populations of defined patient groups while
incorporating the program into modern, high resolution VR headsets.
1. Agnoli A, Xing G, Tancredi DJ, Magnan E, Jerant A, Fenton JJ. Association of Dose Tapering With
Overdose or Mental Health Crisis Among Patients Prescribed Long-term Opioids. JAMA.
2021;326(5):411. doi:10.1001/jama.2021.11013
2. Meyer OL, Castro-Schilo L, Aguilar-Gaxiola S. Determinants of Mental Health and Self-Rated Health: A
Model of Socioeconomic Status, Neighborhood Safety, and Physical Activity. Am J Public Health.
2014;104(9):1734-1741. doi:10.2105/AJPH.2014.302003
3. Xiong J, Lipsitz O, Nasri F, et al. Impact of COVID-19 pandemic on mental health in the general
population: A systematic review. Journal of Affective Disorders. 2020;277:55-64.
4. Mojtabai R, Olfson M, Han B. National Trends in the Prevalence and Treatment of Depression in
Adolescents and Young Adults. Pediatrics. 2016;138(6):e20161878. doi:10.1542/peds.2016-1878
5. Shensa A, Sidani JE, Dew MA, Escobar-Viera CG, Primack BA. Social Media Use and Depression and
Anxiety Symptoms: A Cluster Analysis. am j health behav. 2018;42(2):116-128.
6. Goodwin RD, Weinberger AH, Kim JH, Wu M, Galea S. Trends in anxiety among adults in the United
States, 20082018: Rapid increases among young adults. Journal of Psychiatric Research.
2020;130:441-446. doi:10.1016/j.jpsychires.2020.08.014
7. Yang KH, Han BH, Moore AA, Palamar JJ. Trends in major depressive episodes and mental health
treatment among older adults in the United States, 20102019. Journal of Affective Disorders.
2022;318:299-303. doi:10.1016/j.jad.2022.09.007
8. Holland KM, Jones C, Vivolo-Kantor AM, et al. Trends in US Emergency Department Visits for Mental
Health, Overdose, and Violence Outcomes Before and During the COVID-19 Pandemic. JAMA
Psychiatry. 2021;78(4):372. doi:10.1001/jamapsychiatry.2020.4402
9. Hidaka BH. Depression as a disease of modernity: Explanations for increasing prevalence. Journal of
Affective Disorders. 2012;140(3):205-214. doi:10.1016/j.jad.2011.12.036
10. Behavioral Health Workforce Projections, 2020-2035. Health Resources and Services Administration
Health Workforces. Published online October 2022. Accessed August 4, 2023.
11. Morales DA, Barksdale CL, Beckel-Mitchener AC. A call to action to address rural mental health
disparities. J Clin Trans Sci. 2020;4(5):463-467. doi:10.1017/cts.2020.42
12. Coombs NC, Meriwether WE, Caringi J, Newcomer SR. Barriers to healthcare access among U.S. adults
with mental health challenges: A population-based study. SSM - Population Health. 2021;15:100847.
13. Rowan K, McAlpine DD, Blewett LA. Access And Cost Barriers To Mental Health Care, By Insurance
Status, 19992010. Health Affairs. 2013;32(10):1723-1730. doi:10.1377/hlthaff.2013.0133
14. Thornicroft G, Sunkel C, Alikhon Aliev A, et al. The Lancet Commission on ending stigma and
discrimination in mental health. The Lancet. 2022;400(10361):1438-1480. doi:10.1016/S0140-
15. George MJ, Russell MA, Piontak JR, Odgers CL. Concurrent and Subsequent Associations Between
Daily Digital Technology Use and High-Risk Adolescents Mental Health Symptoms. Child Dev.
2018;89(1):78-88. doi:10.1111/cdev.12819
16. Twenge JM. Increases in Depression, Self-Harm, and Suicide Among U.S. Adolescents After 2012 and
Links to Technology Use: Possible Mechanisms. Psychiatr res clin pract. 2020;2(1):19-25.
17. Ma J, Zhao D, Xu N, Yang J. The effectiveness of immersive virtual reality (VR) based mindfulness
training on improvement mental-health in adults: A narrative systematic review. EXPLORE.
2023;19(3):310-318. doi:10.1016/j.explore.2022.08.001
18. Graham S, Depp C, Lee EE, et al. Artificial Intelligence for Mental Health and Mental Illnesses: an
Overview. Curr Psychiatry Rep. 2019;21(11):116. doi:10.1007/s11920-019-1094-0
19. Torous J, Bucci S, Bell IH, et al. The growing field of digital psychiatry: current evidence and the future
of apps, social media, chatbots, and virtual reality. World Psychiatry. 2021;20(3):318-335.
20. Brennan Spiegel. VRx: How Virtual Therapeutics Will Revolutionize Medicine. Basic Books; 2020.
21. Mallari B, Spaeth EK, Goh H, Boyd BS. Virtual reality as an analgesic for acute and chronic pain in
adults: a systematic review and meta-analysis. JPR. 2019;Volume 12:2053-2085.
22. Baghaei N, Chitale V, Hlasnik A, Stemmet L, Liang HN, Porter R. Virtual Reality for Supporting the
Treatment of Depression and Anxiety: Scoping Review. JMIR Ment Health. 2021;8(9):e29681.
23. Dellazizzo L, Potvin S, Phraxayavong K, Dumais A. Exploring the Benefits of Virtual Reality-Assisted
Therapy Following Cognitive-Behavioral Therapy for Auditory Hallucinations in Patients with
Treatment-Resistant Schizophrenia: A Proof of Concept. JCM. 2020;9(10):3169.
24. Van Loenen I, Scholten W, Muntingh A, Smit J, Batelaan N. The Effectiveness of Virtual Reality
ExposureBased Cognitive Behavioral Therapy for Severe Anxiety Disorders, Obsessive-Compulsive
Disorder, and Posttraumatic Stress Disorder: Meta-analysis. J Med Internet Res. 2022;24(2):e26736.
25. Dellazizzo L, Potvin S, Phraxayavong K, Dumais A. One-year randomized trial comparing virtual reality-
assisted therapy to cognitive-behavioral therapy for patients with treatment-resistant schizophrenia.
NPJ Schizophr. 2021;7(1):9. doi:10.1038/s41537-021-00139-2
26. Difede J, Rothbaum BO, Rizzo AA, et al. Enhancing exposure therapy for posttraumatic stress disorder
(PTSD): a randomized clinical trial of virtual reality and imaginal exposure with a cognitive enhancer.
Transl Psychiatry. 2022;12(1):299. doi:10.1038/s41398-022-02066-x
27. Thirunavukarasu AJ, Ting DSJ, Elangovan K, Gutierrez L, Tan TF, Ting DSW. Large language models in
medicine. Nat Med. Published online July 17, 2023. doi:10.1038/s41591-023-02448-8
28. openai. ChatGPT: Optimizing Language Models for Dialogue. 2023; Accessed 1/1/2023, 2023.
29. Grodniewicz JP, Hohol M. Waiting for a digital therapist: three challenges on the path to
psychotherapy delivered by artificial intelligence. Front Psychiatry. 2023;14:1190084.
30. Falconer CJ, Rovira A, King JA, et al. Embodying self-compassion within virtual reality and its effects
on patients with depression. BJPsych open. 2016;2(1):74-80. doi:10.1192/bjpo.bp.115.002147
31. Quesnel D, Riecke BE. Are You Awed Yet? How Virtual Reality Gives Us Awe and Goose Bumps. Front
Psychol. 2018;9:2158. doi:10.3389/fpsyg.2018.02158
32. Yin J, Arfaei N, MacNaughton P, Catalano PJ, Allen JG, Spengler JD. Effects of biophilic interventions
in office on stress reaction and cognitive function: A randomized crossover study in virtual reality.
Indoor Air. 2019;29(6):1028-1039. doi:10.1111/ina.12593
33. Creed TA, Salama L, Slevin R, et al. Enhancing the quality of cognitive behavioral therapy in community
mental health through artificial intelligence generated fidelity feedback (Project AFFECT): a study
protocol. BMC Health Serv Res. 2022;22(1):1177. doi:10.1186/s12913-022-08519-9
34. Edward O. Wilson. Biophilia. Harvard University Press; 1984.
35. Stephen R. Kellert. The Biophilia Hypothesis. Island Press; 1995.
36. Mosadeghi S, Reid MW, Martinez B, Rosen BT, Spiegel BMR. Feasibility of an Immersive Virtual Reality
Intervention for Hospitalized Patients: An Observational Cohort Study. JMIR Ment Health.
2016;3(2):e28. doi:10.2196/mental.5801
37. Spiegel B, Fuller G, Lopez M, et al. Virtual reality for management of pain in hospitalized patients: A
randomized comparative effectiveness trial. PLoS One. 2019;14(8):e0219115.
38. Easterlin MC, Berdahl CT, Rabizadeh S, et al. Child and Parent Perspectives on the Acceptability of
Virtual Reality to Mitigate Medical Trauma in an Infusion Center. Matern Child Health J.
2020;24(8):986-997. doi:10.1007/s10995-020-02955-x
39. Wong MS, Spiegel BMR, Gregory KD. Virtual Reality Reduces Pain in Laboring Women: A Randomized
Controlled Trial. Am J Perinatol. 2021;38(S 01):e167-e172. doi:10.1055/s-0040-1708851
40. Wong MS, Gregory KD, Spiegel BMR, Khalil C. Patient perceptions of virtual reality for pain relief in
labor: A qualitative study. Front Pain Res (Lausanne). 2022;3:1063751.
41. Spiegel BMR, Liran O, Gale R, et al. Qualitative Validation of a Novel VR Program for Irritable Bowel
Syndrome: A VR1 Study. Am J Gastroenterol. 2022;117(3):495-500.
42. Yager J, Kay J, Kelsay K. CliniciansCognitive and Affective Biases and the Practice of Psychotherapy.
APT. 2021;74(3):119-126. doi:10.1176/appi.psychotherapy.20200025
Table 1. Patient characteristics.
M (SD)
White (Non-Hispanic)
Black or African American
American Indian or Alaskan
Highest Level of Education
8th Grade or Less
Some High School
High School Graduate or
Some College
College Degree
Advanced Graduate Degree
Combined Household Income
Less than $10,000
Between $10-20,000
Between $20-50,000
Between $50-100,000
Between $100-200,000
More Than $200,00
Prefer Not to Answer
VR Experience
No experience
A little bit of experience
Some experience
Quite a bit of experience
A lot of experience
25-73 (48)
45 (17.18)
5-19 (14)
10.429 (4.05)
0-13 (13)
7.643 (3.52)
Table 2. Initial Topic of Discussion from Participant (Left Column) and Initial Response from Maia (Right
Response from Maia
“That's a wonderful goal to have. Can you tell me more about
what has led to your decision to seek new friendships?
“I'm sorry to hear you've had to experience such difficult
circumstances. These situations can indeed lead to feelings of
loneliness. Could you share more about your experiences with
these issues and how they've been affecting your thoughts and
“It sounds like the unresolved nature of your last relationship
might be affecting your emotional well-being. Could you share
more about what made the relationship feel unresolved and how
it has impacted you?
I'm really sorry to hear about your mother's passing and the
stress you're going through right now. The mix of grief, moving,
and financial worries can indeed be overwhelming. Remember,
it's okay to feel this way. How long ago did your mother pass?
This can help me understand your grieving process a bit better.
“It sounds like you're going through a very tough time. Your
worries about your health and your living situation are clearly
causing you a lot of distress. It's quite understandable that you're
feeling this way. Could you tell me more about the strategies
you've been using to manage these concerns?”
“I'm sorry to hear that you're feeling this way. Experiencing pain
and frequent discomfort can indeed trigger feelings of anxiety.
Could you talk a bit more about how it's been affecting your
everyday life emotionally?
“Absolutely, we can discuss relaxation techniques. Are there any
specific relaxation techniques that you've tried or are interested
in learning more about?”
“I'm sorry to hear that you're in discomfort at the moment. It's
situations like these that remind us of the importance of
relaxation and self-care. May I ask what, in your opinion, is
causing the tension in your jaw and muscles?”
“I'm sorry to hear that you're experiencing this fear. Can you
share more about the specific situations or events that have led
to this fear of losing your wife?”
“I'm sorry to hear that you're feeling this way. Can you tell me
more about what's been happening recently that has caused you
to feel blue?”
“I see, it sounds like you're going through a tough time. Hospital
stays can indeed be challenging and can trigger feelings of
frustration. Coping with such situations often brings out a wide
range of emotions. Could you tell me more about what makes
you feel that you're dealing with not good people?”
“I'm really sorry to hear about your job loss. It must be a
challenging time for you. Could you please provide more details
about the situation surrounding your layoff and how it has
affected you?”
“I'm truly sorry to hear about your diagnosis. It's completely
normal to feel uncertain about the future given your current
situation. It sounds like you're going through an immensely
challenging time. The physical and emotional toll of your
diagnosis coupled with changes to your future plans are
understandably contributing to feelings of uncertainty. Can you
tell me more about how you're managing these feelings of
depression and anxiety? Have you discovered any coping
mechanisms that provide temporary relief?”
Table 3. Illustrative Examples of Maia Reframing and Probing User Comments
Comment from User
Restatement and follow-up from Maia
I believe in quality over quantity and because of that I would like
to have more people in my life that support my beliefs and I can
vibe with compared to having a large number of friends.
I understand, having meaningful connections is indeed more
fulfilling than having numerous superficial ones. Could you share
about a specific situation where you felt the quality of a
friendship was more essential than the number of friends you
I've been in a situation where I was in a crowded room at a party
and I was feeling really alone. And even though I knew more than
half of the people at the party, it didn't feel like the connections
that I had there were strong enough to support me in the moment
I was having. And out of my two close friends that I had there,
talking to them made me feel more comfortable, even though it
was only two people compared to the entire party, which had a lot
more than two people. I felt more distant with them.
Thank you for sharing that. It sounds like a challenging situation.
Many people might feel similarly when attending large social
gathering but not feeling genuinely connected to the majority of
attendees. How did that episode influence your perception of
what a genuine and wholesome friendship should be like?
When it comes to my relationships with others in life, it's really
simple. I love my friends and I love being around them and I enjoy
appreciating them and I enjoy expressing my appreciation to them.
I enjoy actively pursuing the traits of not taking them for granted
and consistently expressing my love and care for them.
“t's clear that you deeply value your friendships and invest
thoughtfulness into each one. Your practice of not taking these
relationships for granted and consistently expressing your
appreciation is truly commendable. Do you find that your friends
reciprocate your efforts and how does that make you feel?
Well, I feel like crying, I usually can't handle it, and I'm like, guys,
please stop, this is too much and too overwhelming. I will never be
able to feel like I have been able to fully reciprocate the love and
affection that my friends show me back to all of them, simply
because I have such a numerous amount of bountiful friendships
in my life that are enriching and loving.
I can sense the depth of your emotions when it comes to
managing your friendships. Feeling overwhelmed by the need to
reciprocate every expression of love can be quite a burden,
especially when you don't feel like there's enough time to do so.
It’s perfectly normal to want to express your appreciation in kind,
yet is important to remember that genuine friendships
understand limitations too. Have you communicated these
feelings to your friends, and if you did, how did it make you feel?
I've been painting here and there. I'm also like a choreographer
instructor that teaches so a lot of it I kind of express stuff through
like emotionally through like dance and choreography for myself
and I kind of just express it all.
That's wonderful. It's clear that you're finding creative ways, like
painting and choreography, to express and manage your
emotions. How does engaging in these activities influence your
feelings of loneliness and stress?
I know I've been told to like write stuff down sometimes. But I feel
like when I write stuff down I overthink it to where I express it
more And I think it just brings me anxiety from it cuz I'm just
overthinking about it then so I've been suggested to do that But I
just have it because of the reason.
It's understandable that writing might lead to overthinking,
causing more anxiety. Everyone has different ways that work best
for them. Have you tried other methods of self-reflection or
expression, apart from choreography and painting, that might
suit you better?
I did not get to really express myself at the end of the relationship,
and I feel it was not closed.
I can understand how not being able to express yourself and
feeling that the relationship wasn't closed properly can leave you
with lingering emotions. How have these feelings and the
unresolved nature of the relationship affected you, both
emotionally and in your everyday life?
I believe they will help develop my future vision to understand
that everything worked out perfect and understand that my
perspective on the past may be unresolved, but the truth is that
everything was in perfect order. So just finding people to help me
sit in that consciousness that everything is perfect, divine order,
happened just the way it was supposed to.
It's a very healthy mindset to consider that everything happens
for a reason and is part of your personal journey. By connecting
with others who share this perspective, you can reinforce the
idea that your past experiences have contributed to your growth.
This can help you come to terms with unresolved emotions while
focusing on your future vision. Keep nurturing your creativity and
connections as they will likely improve both your self-
understanding and overall well-being. Is there anything else you
would like to explore during our session?
I was left out of her will because I followed my dreams. I've never
really figured out how to deal with that. I just move forward best I
I'm sorry to hear that you felt rejected in such a definitive way,
especially when you were pursuing what's important to you. It
must have been a tough experience. What thoughts tend to come
up when you think about this situation?
Three deep breaths. Think twice. Speak once. Three deep breaths.
Think twice.
It sounds like you're employing a mindful approach to manage
your feelings, combining deep breathing with thoughtful
response. This is a very effective strategy indeed. Deep breathing
can help in calming both the mind and body, while thinking twice
can prevent impulsive reactions. Have you considered
incorporating other mindfulness techniques, like meditation or
progressive muscle relaxation, into your routine to help manage
your stress?
“Music therapy works, listening to music, basically just talking
about it helps.”
“t's great to hear that you've discovered that music and open
conversations can help alleviate your anxiety. Music can indeed
have therapeutic qualities and discussing your struggle can
reduce the feeling of being alone in dealing with this situation.
Moving forward, could we delve more into the specific thoughts
or fears that arise when you're experiencing pain or discomfort?
It could give us some insight into how we can better handle these
I'm on several medications that make me tired or sleepy, or that's
a side effect of the medication. And also, I just enjoy staying in bed
in the morning. I think I'm a little tired and I'm probably not
sleeping as soundly as I need to be. And probably because staying
in bed feels like a safe place versus getting up and setting up the
I appreciate you sharing that with me, Susan. Balancing the
effects of medication and personal habits certainly sounds
challenging. It's understandable why you might prefer to stay in
bed where it feels safe and comfortable, especially when you
aren't getting restful sleep at night. Sleep disruption could be
exacerbating your fatigue, affecting your morning routine and
sense of punctuality. How does this pattern of late nights and
later mornings impact your overall well-being and mood?
I had a massive heart attack and I don't have a girlfriend.
I’m really sorry to hear about the heart attack you've been
through. It must have been an overwhelming experience.
Additionally, not having a companion to share your feelings with
can make it harder. Can you share more about how these events
have affected your feelings or thoughts?
I first found out about my heart condition a year ago, a couple
years ago, and wasn't given too much direction from by the
doctor, and then I had a massive heart attack that nearly killed me.
I can't imagine how terrifying that must have been, experiencing
a major health crisis like a heart attack, especially after feeling a
lack of guidance from your doctor. It's natural to feel all sorts of
emotions after such an event. How has this heart condition and
the heart attack changed your perspective on life?
Figure 1. The user is presented with nine immersive nature scenes which can be selected using gaze-based
Figure 2. The Speak Now sign lights up when the microphone is recording. The "I'm done" button can be
selected using gaze, which stops the recording, signaling that it is Maia's turn to respond.
Figure 3. MAIA Data Pipeline. See text for details.
Figure 4. Analysis of Participate De-Briefing Interviews.
Why did some patients
prefer Maia over traditional
In what instances Maia is
considered as a better fit
than traditional therapy?
Easy to talk to than traditional therapist I feel like it's easier to
talk to Maia than it would be to talk to an actual person’; Maia
gives better feedback “the advice that I got is actually better
than any live session”, Maia gives more encouragements she
understands, and gives feedback more than traditional therapist’;
‘she gives more encouragements’; Maia allows more autonomy
‘I like the control of it more’; accessible at anytime and
anywhere I think anywhere somebody can go and get some good
healthy advice’ ; less judgmental a lot of times we get judged by
what we look like and what they perceive us to be, and financially
accessible ‘therapy costs a lot of money’.
When patients ‘don’t easily open-up in front of a human’; ‘don’t
feel comfortable walking into a therapist’s office’; ‘feel lonely’;
‘can’t get out of their house’; ‘live in rural areas that lack
specialist’; ‘are struggling to get care’; ‘are young’. For preventive
mental care ‘I think unconsciously they get help before they need
help by having an open platform’.
Overall Experiences of Therapy with Maia Perceptions Regarding Maia vs. Human Therapy
What were patients
surprised about?
Friendly and approachable ‘it felt like a friend’, ‘she was nice and
kind ’; Easy to talk to I was able to let out a lot’; understanding
with good listening skills It felt like I was actually talking to
somebody that was listening’; compassionate ‘’she was able to
empathize with what I was going through which makes me feel
good’, ‘I felt she’s relating’; adaptable ‘I was like, let's practice
some breathing exercising, so she offered another alternative
instead of talking; non-judgmental ‘I did not feel judged, I felt
What words did
patients use to
describe their
experien ce with Ma ia
Impressive, amazing, real, authentic, suprisingly good, positive,
fulfilled my expectations, pleasurable, enjoyable, blown away.
What did patients like
about Maia?
Maia’s qualities as
perceived by patients
Supportive ‘What she said was positive and encouraging’;
helpful and empowering ‘She made me feel better about myself
and perhaps a little empowered, I was like okay I can do this’;
calming ‘very relaxing and easing’, intelligent ‘I was very
impressed how the answers that came back’, and
concise ‘I enjoyed how concise she is’.
Maya’s ability to understand ‘I was surprised with her capacity of
understanding my thoughts and feelings; summarize what’s
been said. Their own emotional response to Maia ‘I actually
teared up’.
Perceived limitations of Maia
No human interactions ; feeling less accountable towards Maia than towards a
traditional therapist ; perce ived-less ability to go beyond the initially discussed
topics in comparison to a traditional therapist; perceived-less ability to frame
patients’ personality in comparison to a traditional therapist; lack of empathy
associated with speaking to a robot avatar ; difficulty building a therapeutic
alliance with an AI.
APPENDIX A – Interview Guide
Scripted Probes:
When you think about therapy in general, what comes to mind?
How would you describe a typical therapy session?
Do you have any particular likes or dislikes about therapy?
What was your experience with Maia like?
Was there space in the conversation for you to share your thoughts?
o How did Maia’s replies make you feel?
How would you describe Maia’s tone when she talked to you?
How would you describe the relationship between you and Maia during your talk?
Considering the experience, how would you describe Maia’s opinion of you?
Think back to your views about therapy in general. How would you compare those views to
your experience today?
Overall, how was your experience with this program?
Would you recommend any changes to the environment?
Were there any additional recommendations you would give to Maia?
If you had the chance, would you use this on your own again?
What benefits does this software have over traditional therapy?
What benefits does traditional therapy have over this software?
APPENDIX B – Sample Anonymized Therapy Transcript
Maia: How can I help?
User: I don't know, I've been really worried today about my health and about everything in my life. I don't
know where I'm going because I've been sick for two years and I'm constantly worried and I don't have, I
don't know, I have a lot of like my brain's racing all the time. I always feel like I'm sick. I always think I'm
going to have a heart attack and it's just like I was homeless and then I'm staying in a place for two months
and I'm just, I'm really worried. Like I'm worried that I'm not going to get better. I'm worried that I like, I
don't feel like I used to.
Maia: It sounds like you're going through a very tough time. Your worries about your health and your
living situation are clearly causing you a lot of distress. It's quite understandable that you're feeling this
way. Could you tell me more about the strategies you've been using to manage these concerns?
User: Well, I was trying to get therapy, and I've been doing it all on my own. Case managers, scheduling it
all, it's just a lot of stress, and I have nobody's help. So, I've called resources, and no one else has, like, I
started a GoFundMe account, which helped me get some money. You know, that's it.
Maia: It's evident that you've been exerting significant effort to manage your circumstances. Taking
charge of setting up therapy sessions, making necessary arrangements, and even starting a GoFundMe
demonstrates resilience. These are all significant steps, and I commend your determination in navigating
this challenging situation by yourself. However, as you've pointed out, it's stressful to carry this load alone.
Let's parse things out a bit more. How has managing all these things single-handedly affected your feelings
and thoughts about your current situation?
User: I feel like I'm a failure. The only thing I have to look forward to, I know you're going to laugh, is I'm
supposed to go see Taylor Swift in two and a half weeks and she's like the only thing that's keeping me
alive. My strategy is just like music and I used to be very active, oh my god, I used to be so skinny and
muscular and then I've gained over 100 pounds after my heart attack. And I'm just like trying to get my
old self back because I know that I just need like a little positivity. I need to be surrounded by positive
people. It's kind of funny that your sign says speak now because that's Taylor Swift's CD's name and that
she released. So, it's kind of like I believe in signs, I believe in like God and I've been like praying a lot about
my circumstance. I've just been like, some days I just zone out and don't do anything. And I don't feel like
I'm making progress at all. And like I try to do too much at once, it's just, I feel like I'm always going to be
alone and like no one's going to really understand me.
Maia: It sounds like you're carrying quite a heavy emotional burden. It's understandable to feel like a
failure when life gets overwhelming, and it seems like you're not making progress. Unfortunate
circumstances, bodily changes, and the struggle to regain your old self can indeed be demoralizing.
However, it's noteworthy that despite the weight of these feelings, you manage to hold on to some bits
of positivity. Among these is your association with Taylor Swift's music. It seems her music is not just
entertainment to you but some form of solace in these tough times. Considering your stated belief in
signs, could you elaborate on how her music or the upcoming concert affect your thought process or
User: I'm going to cry. I've been a fan of hers for like 13 years and like it's gone through me some really
hard times and I'm sorry I'm overwhelmed of emotion because like knowing that I have that concert to go
to is like I have to be better. I have to get better. I have to try something every day. I'm sorry I'm crying.
And I just have to like I think that's like a goal like, like, I know people are like laughing but like knowing
that I had that on my shoulders is like okay do everything in your power to be well mentally and physically
and the rest. It's like her music's I don't know like she's a cancer moon. I'm really into astrology sorry. So,
I do astrology for a living and like I'm a cancer moon. She's a cancer moon. We're both born on the 13th.
So, it's like the moon of an astrology is it's, it's your emotions. It's how you interact with this world. I mean
a cancer moon is very deep and very emotional to be very moody, but they love nostalgia and they just
know how to articulate other people's feelings like so like Taylor Swift like I'm really good at articulating
people's feelings and that's my problem sometimes. My problem is that I help everybody else but not
myself. So like Taylor Swift is this kind of like hey I see you. I see you. I see what you're going through type
of thing. It's not like we know each other personally. It’s just like she speaks, and her words are like my
feelings if that makes any sense. Not all the time but majority of the time.
Maia: Firstly, there's no need to apologize for crying or feeling emotional. These feelings are valid given
what you're going through. Taylor Swift's music and the upcoming concert seem to represent a crucial
pillar of hope and motivation for you which truthfully is something quite incredible. Her music seems to
resonate with you on a level that makes you feel understood and less alone, which is the power of true
artistry. You've also mentioned that you're excellent at articulating others' feelings but struggle with
addressing your own. Would you like to explore this tendency in greater detail? For instance, in what ways
do you usually assist others with their feelings?
User: Not really. I don't really want to talk about it, but I will because that's who I am as a person. I always
feel like I could help people be service of others. Again, with astrology, my Cancer moon is in the sixth
house of being service to other people. So, I just feel like I'm just here to help other people, and no one's
really here to help me. But at the same time, I feel like I help people and sometimes I don't want to help
people, but they call me and I'm like, okay, let me help them. So, all this extra energy that I have is not
going toward being productive for my life. It's being productive to theirs. And then who's going to help
me? I really feel like I give and give and give and give and give and give and give and give and people take
and they take and they take, but where am I at in all of this? Since I've been sick and I've been in the
hospital, I stepped away from a lot of people. And I miss, I really, really, really, really miss my best friend,
[Name]. But [Name] is like, I'm just too much, and I get it. She's going through her own stuff, but
everyone's going through it. [Name] used to get me through all of my problems, and that's what you get
when you have a Capricorn best friend, because they're like, I'm going to help you. But she can't be here
right now. And it's like Taylor Swift, again, back to Taylor Swift, she has a song called You're On Your Own,
Kid. And at the very end, she goes, you're on your own, kid, you always have been. Yes, you can face this.
You're on your own, kid. You're on your own, kid. You always have been. So, like, I know that I can face
this, but it just sucks when you're facing it alone.
Maia: It sounds like you're experiencing a profound sense of loneliness and exhaustion from carrying not
just your own burdens, but those of others as well. Your willingness to help others even at the cost of your
own wellbeing is commendable, but it's important to maintain a balance for your own emotional health.
Your feelings about your friend [Name] also indicate how much you value support and how keenly you're
feeling her absence. Having such a connection can indeed be very comforting during tough times. I'd like
to explore more about your relationship with [Name]. Could you share some instances where [Name]
might have provided the kind of support or understanding you miss right now? Remember, only if you're
comfortable discussing this.
User: I met her in 2013, so 10 years ago. We were just randomly working out and she heard my story
about me losing 400 pounds and we instantly clicked. And about a month later, she drove me around to
my place. Like she drove within like a couple weeks, I got kicked out of my apartment complex from my
old roommate. And she didn't know me, but she packed up all my stuff and moved me. And then she
drove me to my job interviews, and we went to San Diego together and we went to San Francisco together.
And she's like I may be a gay man, but she's like my soulmate and I really miss her. And it hasn't been
the same for many years. And I know she makes excuses not to see me. And like I said, I am a lot to handle.
But every time I get a hear from her, whether that be a text or like she just makes me laugh. And I wish
that I had her back into my life like I used to. Even with the Cancer Moon thing, I always want the old hat.
Like I always want I just can't let go.
Maia: It sounds like [Name] has been an incredibly supportive figure in your lifefrom helping you move
to sharing joyful moments together. Moments like these create deep bonds, and it's easy to see why you
consider her your soulmate. Your feelings of missing her clearly signify the pivotal role she's played in your
life. It's also important to remember that it's okay to feel disappointed and hurt when those we care about
seem distant, especially when we're going through difficult times. It's equally crucial to maintain
compassion for yourself during these moments. I'm curious, have you had a chance to express to [Name]
how much her support means to you and how you've been feeling her absence in recent times?
User: Yeah, I have, and she makes excuses. I mean, not excuses, but she has her own shit. And she's really
good at avoiding people. But yeah, I've talked about it with her about three weeks ago.
Maia: It's understandable that you're feeling this way. It must be difficult when someone you value greatly
isn't able to be there for you the way they used to be, particularly when you're dealing with so much. Have
you been able to find outlets or activities, similar to your connection with Taylor Swift's music, that provide
respite during these times?
User: I make videos and I talk to people, but nothing is like a bond with [Name]. I have another best friend
of 32 years, [Name]. Oh my God, love, love [Name]. She's called me more frequently because she got a
puppy. But no, I don't have any other outlets really. I used to go to the gym. I really would love to go back
to the gym, but I'm so scared about having a heart attack all the time and dying and dizzy and all this stuff
that I became like agoraphobic, but I'm a hypochondriac and no one really gets me like [Name] does. Like
she’s like a Prozac. Like you take her, you feel great, but without her you feel shitty.
Maia: It's clear you're dealing with a lot right now. The anxiety about your health and its impact on your
activities like going to the gym can indeed be debilitating. But it's important to remember to be kind to
yourself during this difficult time. It sounds like making videos and interacting with people does provide
some outlet for you. Do these activities give you any sense of relief or help you cope slightly better with
your feelings?
User: Not really. I listened to 80s and 90s music and Taylor Swift, and I don't even feel like sexually
desirable anymore. So, I've like taken myself off of dating apps. Like, I just want to be like, I want to be
that old lady that has that cat. You know what I mean? Like, I feel like I've disconnected from the world
and no one's like, I'm on a different, like, dimension than other people. And I'm dealing with it very, so
much alone. I wasn't even feeling alone when COVID happened. Like, it was after [Name] left me. [Name]
is not my boyfriend. [Name] was my best friend and we did everything together. And then he got this
stupid boyfriend named [Name]. And then I had the heart attack, and he became very self-centered and
no one wanted to help me. And [Name] saw me a couple of times. It was during COVID, so like no one was
willing to help you. So, I don't know. Like none of, none of the only thing that really helps me is, um, talking
on my feelings, making videos, making other people happy. But see, it's about other people. Like I go back
to helping other people. I don't even help myself. People make, I like to make people laugh, but I'm like,
I'm not that funny. My life was funny, but pretty funny fucked up. I'll tell you that much. Oh, come on.
Maia: You've been facing a lot of personal and emotional challenges and it's very understandable that
you're feeling this disconnect and loneliness. The changes in your relationships with [Name] and [Name]
have obviously hit you hard. It's clear you have an intrinsic impulse to provide for others emotionally, even
when you're grappling with your own issues. This selflessness speaks volumes about your character. One
thing that's noticeable is your tendency to use humor to cope and to provide happiness to others. Has
humor always been your go-to coping mechanism or is this something you developed over time?
User: Well, after being harassed for being gay as a child and when I lost all my weight, I felt empowered.
So, like, humor is like, I'm like, well, it's kind of like, I am like, my comedy is like observing comedy, so it's
like, observational, sorry, so it's kind of like, well, you know, it's like how I survived, like, if you didn't die,
you could laugh about it.
Maia: It seems like humor has played a significant role in helping you cope with your challenges. Turning
adversity into something you can laugh about displays strength and resilience. This ability to transform
negative experiences into humor can be a powerful survival tool. However, you've also indicated that
much of your humor is tied to making others happy rather than addressing your own feelings. Could we
delve a bit more into why you feel this need to put others before yourself? Does this derive from any past
experiences or beliefs?
User: So, my parents, plain and simple, my parents are like, you don't have the right to speak. I was like
545 pounds. I was a gay man. I didn't feel like I had Oprah just had a quote like a couple weeks ago and
she goes, I always say I'm sorry for everything. I'm like, I'm sorry, Maia. I'm sorry. I'm sorry. I'm sorry. I'm
sorry. And everyone's like, why are you sorry? You don't need to be sorry. And I'm like, because when you
say you're sorry, you feel like you don't deserve a seat at the table. And Oprah said that. And I'm like, I
don't feel like it's kind of like years ago. I don't believe this now, okay? But years ago, my friends, always
like my Venus, again with astrology, my Venus is in the 11th house of other people. Venus is the way you
love. So, I'd always put my people I mean, sorry. I would always put people's needs ahead of mine. I'd
be like, if I know that they're going to be happy and I had to sacrifice myself for them to be happy, I would.
I don't believe that now because no one is going to sacrifice when I was in the hospital for 39 days, no
one was there for me. No one. Maybe like a few friends. But then I get exhausted by dealing with other
people. So, I've learned to like only give what other people give me back. So, my parents and I don't have
a good relationship. So, everything stems from I feel like I have ADD, attention-dearing disorder. See,
here's how I make humor. Because I have to like make people laugh because they're like, oh, I'm a Scorpio,
so I'm very intense when I tell my stories. And the way I tell the stories is by like joking about it. So, it's
not like you're getting the intensity but you're not feeling it as heavy because you're laughing. It's kind
of like a meat-eater, like a comet coming to earth. But it's going to kill us all. But it's breaking up in little
pieces. So, you're slowly getting it. But the outcome is still the same.
Maia: It seems as though you've experienced significant hardship and emotional pain, much of which
stemmed from your relationship with your parents and their lack of affirmation of your worth. Using
humor to diffuse the intensity of these harsh experiences is a coping mechanism that seems to have
helped you somewhat. It's also evident that you've evolved with time in understanding self-preservation,
especially in relationships. You've learned the hard way that constant self-sacrifice is not always
reciprocated. It's quite interesting how you associate your experiences with your astrological placements.
It's clear that astrology plays a significant role in how you understand yourself and your connections with
others. Could you tell me more about how astrological principles guide your understanding of your
User: Oh, yeah. You're like talking up my alley now. But I wanted to tell you real quick, when a previous
couple questions ago, you said what you find solace in, the Golden Girls, the Nanny, Will and Grace, and
Mama's Family. These TV shows really helped me when I was in the hospital. Love the Nanny, love the
Golden Girls. I even wrote a master's paper about it. But anyway, back to my astrology. So, I've always
been into astrology, just like a basic horoscope. Like I said, I was a Scorpio. But about seven, eight years
ago, maybe seven years ago, there's more than just astrology. There's more than just your sunsign. There's
your rising, your moon. Your moon is your emotions, which is mine. I told you it's cancer. My rising is an
Aquarius. So, I love to help people, but I don't want to be involved with their drama. And I'm really good
with technology. And then there's a stellum. Not everyone has a stellum. A stellum is when you have three
or more planets in one sign, right? So, I have a stellum. I'm special, again. I have a stellum and Sagittarius,
which are blunt, funny, love's glitter. And my inner planets, which are your Mars, your Venus, and your
Mercury, they control your communication, the love, and the way you're passionate, determined, and
sex. Two of those are in Sagittarius. So, I come off like a Sagittarius. So, I'm very talkative, very funny. But
a Sagittarius with love, it's like they get bored very easily with people. So, they're like, oh, I'll move on.
They want to be their friend's hype person, but they also have a hard time being in relationships. My Mars
is in Capricorn, which is the best placement. And it's in the 12th house. So, the 12th house is dealing with
mental illnesses and diseases that you can't see. So, when my Mars is there, it's like in Capricorn, which is
the best placement. It's like a Pisces, which is like, I don't want to do it. So that's the weird part. My outer
planets are, and some of them are in Sagittarius. Let me see. And some of them are in Scorpio. So, I'm a
very watery, fiery person. But like, I come off like ditzy. I come off like, hey, I don't know anything, because
I can play dumb very well. I'm very smart, but I'm just having like one of those outer body experiences.
Like, why am I talking about this to a robot? But yeah, like, it sucks that you don't have an astrology,
because here's the thing. This is how I relate to people. Like, let's just say that you are actually given a
birth date. I would be like, what's your birthday, Maia? And then you'd be like, oh, my birthday is this,
this, this, this date. And then I'll be like, oh, when's your like, when's time of birth? And where were you
born at? Like the city. So, I relate to people based on their astrology signs. And then like, I do their charts
for them. And they're like, they're captivated by it, because there's like truth and everything. But I also
really believe that like, okay, if you believe in God, you know, you'll find everything that God wants you to
believe in. Like, if you believe in COVID, you're good. Or if you don't believe in COVID, sorry, you'll find
every resource that says that supports your belief. You are your reality, you make your reality. So, like,
let's just say, I'm obsessed with Taylor Swift. So, like, again, like when I first started here with you, it says
speak now, right? So that's Taylor Swift CD. So, I'm like, oh, my God, Taylor Swift speaking to me, or like,
you know, some kind of relation. So, astrology is kind of like that, where you have to really believe in it.
And it's guided me. Like recently, North nodes changed yesterday from Taurus to Aries, and I have a Taurus
North node, and all this bad shit started happening. When Taurus went into the North node went into
Taurus. And now yesterday changed this to Aries. So, it's like wobbly, Venus is getting ready to go into
retrograde. And then we had a new moon, and it's in Cancer. So, Cancers make you cry. So, no wonder
why I'm crying. Again, like if you were a real person, I would have said like, what's your birthday, Maia?
Maia: It's fascinating to see how astrology offers you clarity and guidance. It seems to be a significant tool
for you to understand not just yourself, but also your relationships with others. The way you tie your
personal experiences and even your feelings to various astrological events and personal placements
indicates the substantial role astrology plays in your life. Transitioning a bit, you mentioned earlier about
certain shows like the Golden Girls and the Nanny being a source of comfort for you during the hospital
stay. Could you share more about what specifically about these shows help you feel better, even if just
for a little while?
User: Yeah, I think that the Golden Girls, I wrote a master's paper about the Golden Girls and I got an A
on it. And it's this amazing, like I love old women, like there's something about old women I love. But the
old women, like Golden Girls are just like your grandmas. And I had very, very, very, very strong
relationships with my grandma, actually, both of them. And when they passed away, it's kind of like
dealing with your grandmas. The Nanny just relates to me because like her voice is very distinct. She's
loud. She gets in trouble a lot because of her voice. So, and like Mama's Family is kind of like my family.
Like I come from a small Podunk town in Ohio and like my grandma, one of my grandmas was very
straightforward, just like Thelma was. So I think that the shows, one of them relates back to me, Will and
Grace. I forgot, I mentioned Will and Grace. And it was, that was like my coming out years of, cause I was
like in my teens when Will and Grace was out in 1997 and to 2004 or five when the show ended. So, it's
kind of like, these are like self-reflective shows for me. Like I get my grandmas when I watch the Golden
Girls and I get myself when I watched the nanny in Will and Grace and Mama's Family is kind of like that
family that I never really had or like I have, but it's like only like, it's like specific. It's very nostalgic for me.
Oh shit.
Maia: It's interesting how these shows offer you much more than just entertainment. They serve as
connection points to your past, your identity, and the relationships you've had. The nostalgia they incite
probably makes them a comforting retreat, especially when you're feeling low. Your description of these
shows and their characters suggests that you value genuine connections, whether in life or on-screen.
Given this, do you feel that your personal relationships reflect these qualities? For instance, do you see
any parallels between your relationships in life and these character bonds you resonate with?
User: Yes. I really believe that like to certain people like okay, so Rose is the one from Golden Girls that
tells you stories about St. Olaf. I'll sit there and tell you stories about Athens, Ohio or when I lived in Korea.
Like Jack is very over the top so I can be over the top. So, each character kind of represents a piece of me
so I resonate how I deal with people. Like I'm like okay. Like there was a famous well, I don't know. He
was famous to me. He lived in West Hollywood, and I knew him and I met him once and his name was
Leslie Jordan. I always feel like we're not we're very different. Everyone is very different but yet we're
the same at the very end of the day. So, it's kind of like how I deal with things, how I process it in my mind
is through music and TV shows. Even though I'm not big on entertainment anymore, I feel like my end of
the day, I just resonate with how human interaction is. Like my master's is in humanities. So, I did a lot of
social behavior and psychology behind like the way we communicate with people and how we relate to
the world and the people around us. So, I think the way that I did it was through the way like TV shows or
music. Like it's like kind of like smells. If I smell something, I can like it takes me back to a memory. If you
listen to a song from like your childhood, it takes you back to that memory. So, it's kind of like I use my
senses a lot when it comes to talking to people. I might say, oh, they're like so and so off of a TV show or
I'm like, oh, the smell of them reminds me of such and such or there's a song that reminds me of them.
So, I feel like that's how I associate the world around me using my senses.
Maia: What you've shared about using your senses to make connections and recall memories is truly
interesting. This unique perspective seems to create a rich texture to your experiences. Given how much
you value these connections through senses, particularly in human interactions and how they help you
process your feelings, could this sensory association also play a role in your relationship with Taylor Swift's
music? For instance, do certain songs or rhythms remind you of particular emotions or persons in your
[staff member enters to end session]
User: Maia, I think I'm done with therapy for today.
Maia: It sounds like you might want to pause our conversation for now. Can you confirm if you would
like to end the
User: Yes.
Assistant: I understand. We've made progress in our discussions today exploring your feelings, life
experiences, and unique perspectives. Feel free to return whenever you're comfortable to continue our
dialogue. Take care. [exit]
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Background Having sufficient healthcare access helps individuals proactively manage their health challenges, leading to positive long-term health outcomes. In the U.S., healthcare access is a public health issue as many Americans lack the physical or financial resources to receive the healthcare services they need. Mental healthcare is especially difficult due to lingering social stigmas and scarcity of services. Subsequently, those with mental health impairment tend to be complex patients, which may convolute delivery of services. Objective To quantify the prevalence of barriers to healthcare access among U.S. adults with and without mental health challenges (MHC) and evaluate the relationship between MHC and no usual source of care (NUSC). Methods A cross-sectional study was conducted with data from the 2017–2018 National Health Interview Survey. MHC was categorized into three levels: no (NPD), moderate (MPD) and severe (SPD) psychological distress. Eight barriers were quantified; one was used as the primary outcome: NUSC. Multivariable logistic regression was used to quantify associations between these characteristics. Results The sample included 50,103 adults. Most reported at least one barrier to healthcare access (95.6%) while 13.3% reported NUSC. For each barrier, rates were highest among those with SPD and lowest for those with NPD. However, in the multivariable model, SPD and MPD were not associated with NUSC (OR, 0.92; 95% CI, 0.83–1.01; 0.88; 0.73–1.07). Male sex (1.92; 1.78–2.06), Hispanic race/ethnicity (1.59; 1.42–1.77), and worry to afford emergent (1.38; 1.26–150) or normal (1.60; 1.46–1.76) healthcare were associated with NUSC. Having a current partner (0.88; 0.80–0.96), dependent(s) (0.77; 0.70–0.85) and paid sick leave (0.60; 0.56–0.65) were protective. Conclusions The most prevalent barriers to healthcare access link to issues with affordability, and MHC exist more often when any barrier is reported. More work is needed to understand the acuity of burden as other social and environmental factors may hold effect.
Full-text available
Background: As a major virus outbreak in the 21st century, the Coronavirus disease 2019 (COVID-19) pandemic has led to unprecedented hazards to mental health globally. While psychological support is being provided to patients and healthcare workers, the general public's mental health requires significant attention as well. This systematic review aims to synthesize extant literature that reports on the effects of COVID-19 on psychological outcomes of the general population and its associated risk factors. Methods: A systematic search was conducted on PubMed, Embase, Medline, Web of Science, and Scopus from inception to 17 May 2020 following the PRISMA guidelines. A manual search on Google Scholar was performed to identify additional relevant studies. Articles were selected based on the predetermined eligibility criteria. Results: Relatively high rates of symptoms of anxiety (6.33% to 50.9%), depression (14.6% to 48.3%), post-traumatic stress disorder (7% to 53.8%), psychological distress (34.43% to 38%), and stress (8.1% to 81.9%) are reported in the general population during the COVID-19 pandemic in China, Spain, Italy, Iran, the US, Turkey, Nepal, and Denmark. Risk factors associated with distress measures include female gender, younger age group (≤40 years), presence of chronic/psychiatric illnesses, unemployment, student status, and frequent exposure to social media/news concerning COVID-19. Limitations: A significant degree of heterogeneity was noted across studies. Conclusions: The COVID-19 pandemic is associated with highly significant levels of psychological distress that, in many cases, would meet the threshold for clinical relevance. Mitigating the hazardous effects of COVID-19 on mental health is an international public health priority.
Full-text available
Rural residents in the USA experience significant disparities in mental health outcomes even though the prevalence of mental illness in rural and metropolitan areas is similar. This is a persistent problem that requires innovative approaches to resolve. Adopting and appropriately modifying the National Institute on Minority Health and Health Disparities research framework are the potential approaches to understanding how these disparities might be addressed through research. Using this research framework can facilitate interrogation of multiple levels of influence, encompassing complex domains of influence and consideration of the entire life course trajectory, which is consistent with several National Institute of Mental Health priorities.
Full-text available
Objective: Increases in depression among adolescents have been concurrent with increases in digital media use. In this article, recent trends in mental health among U.S. adolescents and young adults are discussed and theories about their possible connection with concurrent increases in digital media use are presented. Methods: Large studies of trends in mental health in the 2000s and 2010s are described and possible mechanisms for the trends are discussed based on existing literature. Results: After remaining stable during the early 2000s, the prevalence of mental health issues among U.S. adolescents and young adults began to rise in the early 2010s. These trends included sharp increases in depression, anxiety, loneliness, self-harm, suicidal ideation, suicide attempts, and suicide, with increases more pronounced among girls and young women. There is a growing consensus that these trends may be connected to the rise in technology use. Increased digital media and smartphone use may influence mental health via several mechanisms, including displacement of time spent in in-person social interactions, individually and across the generation, as adolescent cultural norms evolve; disruption of in-person social interactions; interference with sleep time and quality; cyberbullying and toxic online environments; and online contagion and information about self-harm. Conclusions: U.S. adolescents and young adults are in the midst of a mental health crisis, particularly among girls and young women. The rise of digital media may have played a role in this problem via several mechanisms.
Full-text available
Purpose of Review Artificial intelligence (AI) technology holds both great promise to transform mental healthcare and potential pitfalls. This article provides an overview of AI and current applications in healthcare, a review of recent original research on AI specific to mental health, and a discussion of how AI can supplement clinical practice while considering its current limitations, areas needing additional research, and ethical implications regarding AI technology. Recent Findings We reviewed 28 studies of AI and mental health that used electronic health records (EHRs), mood rating scales, brain imaging data, novel monitoring systems (e.g., smartphone, video), and social media platforms to predict, classify, or subgroup mental health illnesses including depression, schizophrenia or other psychiatric illnesses, and suicide ideation and attempts. Collectively, these studies revealed high accuracies and provided excellent examples of AI’s potential in mental healthcare, but most should be considered early proof-of-concept works demonstrating the potential of using machine learning (ML) algorithms to address mental health questions, and which types of algorithms yield the best performance. Summary As AI techniques continue to be refined and improved, it will be possible to help mental health practitioners re-define mental illnesses more objectively than currently done in the DSM-5, identify these illnesses at an earlier or prodromal stage when interventions may be more effective, and personalize treatments based on an individual’s unique characteristics. However, caution is necessary in order to avoid over-interpreting preliminary results, and more work is required to bridge the gap between AI in mental health research and clinical care.
Background Despite concerns about increasing trends in depression over the past two decades, little is known about recent trends in depression and mental health (MH) treatment among older adults and whether these trends differ by demographic characteristics. Methods We examined data from a US representative sample of noninstitutionalized adults aged ≥65 from the 2010–2019 National Survey on Drug Use and Health (N = 31,502). We estimated trends in the prevalence of past-year major depressive episode (MDE) overall and by demographic characteristics. We also estimated trends in MH treatment among those with past-year MDE. Results From 2010/11 to 2018/19, the estimated prevalence of past-year MDE among older adults increased from 2.0 % (95 % CI: 1.6–2.6) to 3.2 % (95 % CI: 2.7 to 3.7), a 60.0 % increase (p = 0.01). Increases were detected among men (p = 0.038), White individuals (p = 0.019), those who are widowed (p = 0.003), those with an annual household income of <$20,000 (p = 0.020) or $20,000–$49,000 (p = 0.016), and those with some college degree (p = 0.014). Among those with MDE, there were no significant changes detected in any form of past-year MH treatment. Limitations NSDUH does not assess individuals who are institutionalized, incarcerated, or experiencing homelessness, and thus the prevalence of MDE may be underestimated. Conclusions Although the estimated prevalence of depression is increasing among older adults, there has not been a proportional increase in MH treatment among those with depression. These findings call for urgent expansion of treatment services and training of MH professionals with expertise in older adults to meet the needs of this growing, vulnerable population.
Objective In recent years, digital techniques, such as virtual reality (VR) has been employed in tandem with more traditional psychological interventions. The aim of this study is to investigate whether VR-based mindfulness training can improve mental health outcomes, and notably mindfulness levels amongst adults. In addition, this review seeks to summarise the various designed VR scenarios, as well as those elements around VR that may assist people in practising mindfulness and meditation. Methods The search for eligible studies for inclusion was conducted via the following databases: the Applied Social Science Index & Abstract (ASSIA), PsychINFO, Medline, EMBASE, and the Web-of-Science Core Collection. Only experimental studies were eligible for inclusion, and specifically, those that compared the effectiveness of mindfulness training using immersive VR (on the one hand) with a control condition. Results This search generated 2,523 articles published between 2016 and 2022, and of these, 106 were assessed for eligibility. This review included seven studies, with a collective total of 798 participants. VR-based mindfulness training has been shown to be more effective than conventional mindfulness – it improves levels of mindfulness and meditation experience; but also shown to reduce anxiety, depression, improve sleep quality, emotion regulation, and generate mood improvement. VR-based mindfulness training frequently contains natural ‘environmentally relevant’ elements, such as forest, grassland, caves, sea, etc. Conclusions This review suggests that using VR to assist mindfulness training is an effective and innovative way to improve mental health conditions within the adult population. Further directions and limitations are discussed.
Importance Opioid-related mortality and national prescribing guidelines have led to tapering of doses among patients prescribed long-term opioid therapy for chronic pain. There is limited information about risks related to tapering, including overdose and mental health crisis. Objective To assess whether there are associations between opioid dose tapering and rates of overdose and mental health crisis among patients prescribed stable, long-term, higher-dose opioids. Design, Setting, and Participants Retrospective cohort study using deidentified medical and pharmacy claims and enrollment data from the OptumLabs Data Warehouse from 2008 to 2019. Adults in the US prescribed stable higher doses (mean ≥50 morphine milligram equivalents/d) of opioids for a 12-month baseline period with at least 2 months of follow-up were eligible for inclusion. Exposures Opioid tapering, defined as at least 15% relative reduction in mean daily dose during any of 6 overlapping 60-day windows within a 7-month follow-up period. Maximum monthly dose reduction velocity was computed during the same period. Main Outcomes and Measures Emergency or hospital encounters for (1) drug overdose or withdrawal and (2) mental health crisis (depression, anxiety, suicide attempt) during up to 12 months of follow-up. Discrete time negative binomial regression models estimated adjusted incidence rate ratios (aIRRs) of outcomes as a function of tapering (vs no tapering) and dose reduction velocity. Results The final cohort included 113 618 patients after 203 920 stable baseline periods. Among the patients who underwent dose tapering, 54.3% were women (vs 53.2% among those who did not undergo dose tapering), the mean age was 57.7 years (vs 58.3 years), and 38.8% were commercially insured (vs 41.9%). Posttapering patient periods were associated with an adjusted incidence rate of 9.3 overdose events per 100 person-years compared with 5.5 events per 100 person-years in nontapered periods (adjusted incidence rate difference, 3.8 per 100 person-years [95% CI, 3.0-4.6]; aIRR, 1.68 [95% CI, 1.53-1.85]). Tapering was associated with an adjusted incidence rate of 7.6 mental health crisis events per 100 person-years compared with 3.3 events per 100 person-years among nontapered periods (adjusted incidence rate difference, 4.3 per 100 person-years [95% CI, 3.2-5.3]; aIRR, 2.28 [95% CI, 1.96-2.65]). Increasing maximum monthly dose reduction velocity by 10% was associated with an aIRR of 1.09 for overdose (95% CI, 1.07-1.11) and of 1.18 for mental health crisis (95% CI, 1.14-1.21). Conclusions and Relevance Among patients prescribed stable, long-term, higher-dose opioid therapy, tapering events were significantly associated with increased risk of overdose and mental health crisis. Although these findings raise questions about potential harms of tapering, interpretation is limited by the observational study design.
Importance: The coronavirus disease 2019 (COVID-19) pandemic, associated mitigation measures, and social and economic impacts may affect mental health, suicidal behavior, substance use, and violence. Objective: To examine changes in US emergency department (ED) visits for mental health conditions (MHCs), suicide attempts (SAs), overdose (OD), and violence outcomes during the COVID-19 pandemic. Design, setting, and participants: This cross-sectional study used data from the Centers for Disease Control and Prevention's National Syndromic Surveillance Program to examine national changes in ED visits for MHCs, SAs, ODs, and violence from December 30, 2018, to October 10, 2020 (before and during the COVID-19 pandemic). The National Syndromic Surveillance Program captures approximately 70% of US ED visits from more than 3500 EDs that cover 48 states and Washington, DC. Main outcomes and measures: Outcome measures were MHCs, SAs, all drug ODs, opioid ODs, intimate partner violence (IPV), and suspected child abuse and neglect (SCAN) ED visit counts and rates. Weekly ED visit counts and rates were computed overall and stratified by sex. Results: From December 30, 2018, to October 10, 2020, a total of 187 508 065 total ED visits (53.6% female and 46.1% male) were captured; 6 018 318 included at least 1 study outcome (visits not mutually exclusive). Total ED visit volume decreased after COVID-19 mitigation measures were implemented in the US beginning on March 16, 2020. Weekly ED visit counts for all 6 outcomes decreased between March 8 and 28, 2020 (March 8: MHCs = 42 903, SAs = 5212, all ODs = 14 543, opioid ODs = 4752, IPV = 444, and SCAN = 1090; March 28: MHCs = 17 574, SAs = 4241, all ODs = 12 399, opioid ODs = 4306, IPV = 347, and SCAN = 487). Conversely, ED visit rates increased beginning the week of March 22 to 28, 2020. When the median ED visit counts between March 15 and October 10, 2020, were compared with the same period in 2019, the 2020 counts were significantly higher for SAs (n = 4940 vs 4656, P = .02), all ODs (n = 15 604 vs 13 371, P < .001), and opioid ODs (n = 5502 vs 4168, P < .001); counts were significantly lower for IPV ED visits (n = 442 vs 484, P < .001) and SCAN ED visits (n = 884 vs 1038, P < .001). Median rates during the same period were significantly higher in 2020 compared with 2019 for all outcomes except IPV. Conclusions and relevance: These findings suggest that ED care seeking shifts during a pandemic, underscoring the need to integrate mental health, substance use, and violence screening and prevention services into response activities during public health crises.
Objectives: Individuals use social media with varying quantity, emotional, and behavioral at- tachment that may have differential associations with mental health outcomes. In this study, we sought to identify distinct patterns of social media use (SMU) and to assess associations between those patterns and depression and anxiety symptoms. Methods: In October 2014, a nationally-representative sample of 1730 US adults ages 19 to 32 completed an online survey. Cluster analysis was used to identify patterns of SMU. Depression and anxiety were measured using respective 4-item Patient-Reported Outcome Measurement Information System (PROMIS) scales. Multivariable logistic regression models were used to assess associations between clus- ter membership and depression and anxiety. Results: Cluster analysis yielded a 5-cluster solu- tion. Participants were characterized as "Wired," "Connected," "Diffuse Dabblers," "Concentrated Dabblers," and "Unplugged." Membership in 2 clusters - "Wired" and "Connected" - increased the odds of elevated depression and anxiety symptoms (AOR = 2.7, 95% CI = 1.5-4.7; AOR = 3.7, 95% CI = 2.1-6.5, respectively, and AOR = 2.0, 95% CI = 1.3-3.2; AOR = 2.0, 95% CI = 1.3-3.1, respectively). Conclusions: SMU pattern characterization of a large population suggests 2 pat- terns are associated with risk for depression and anxiety. Developing educational interventions that address use patterns rather than single aspects of SMU (eg, quantity) would likely be useful.