Embedded Files
Feel project taps into our emotional state and its association with individual mood and psycho-socio cognitive functions. We study this complex network by either inducing affective states and/or examining affective disorders by employing digital or VR behavioral, physiological, and eye-tracking methods. Despite potential advances in emotion, cognition, and motivation research, we lack understanding of underlying cognitive mechanisms that might lead to affective disorders and, in turn, lack in realizing effective diagnostics and treatments with successful interventions. The project envisions better explanations for affective disorder and its relationship with motivation, emotion, attention, working memory, and executive control. Further plans to aid diagnostics, treatment/ monitoring clinical health care by providing more objective assessments and monitoring methods using digital and extended reality (e.g., VR) technologies and physiological monitoring devices, like smart watches.
We focus on questions like:
Do people diagnosed with depression and vulnerable to depression differ in their cognitive functions, especially mental representation, attention, executive control, affective state, motivation and associated psychomotor responses (such as head-tracking, gait analyses, etc.), and psychosocial functions?
Does depression influence 'general' or 'global' attentional processing or specific sub-domains of the attention network?
Can characterizing general cognitive processes help identify the cognitive-biological risk factors for depression in college-going young adults?
Can tuning general-cognitive networks alongside affective network will help modulate cognitive biases and may help in developing better risk detection for depression and or delay the onset or better intervention programs for treatments and monitoring of depression?
Can digital interfaces and VR be used as alternative and effective tools to aid in detecting risks for depression, monitoring depression proneness and major depression disorder (MDD)?
Below are details of current ongoing projects' focus and measures:
Feel Project Focus: The image depicts the research focus that aims to investigate the relationship between emotion, mood, predisposition like depression, social factors like interpersonal conflicts, and cognition.
Feel Project Focus: The image depicts the Feel Project focus that envisions a better understanding of emotion and mood and their association with other cognitive processing and psychosocial strategies. It also aims to offer better objective diagnostic, monitoring and intervention method using digital and VR technologies.
Feel Project Measures: We use multiple measures to examine the qualitative and quantitative measures. Current ongoing studies, Left top to bottom: Behavioural including RTs, VR headtracking and eye-tracking; Gait Analyses; Speech and Text analyses, and Interview/Survey data. The right side depicts the future plan in progress.
PUBLICATIONS/PRESENTATIONS
What Do Head Scans Reveal About Depression? Insights from VR 360-degree Psychomotor Assessment.
In Proceedings of Cognitive Science Society, 2025
San Francisco, July 30-August 2
https://cognitivesciencesociety.org/cogsci-2025/
Priyanka Srivastava, Rohan Lahane, Vivek R. and Prudhvi Pulapa
Abstract: Psychomotor changes, while crucial indicators of depression, remain underrepresented in clinical observations. We examined the relationship between depression and psychomotor behavior by analyzing head-tracking data related to yaw movements made during exploration of 360° emotional videos, alongside valence and arousal ratings on 9-point Likert scale. Symptoms of depression were recorded using the Patient Health Questionnaire (PHQ-9). While subjective ratings for valence and arousal showed no differences across depression groups, the head-tracking data revealed novel results. Individuals with moderate to severe depression exhibited significantly lower scanning speed and standard deviation in yaw movement compared to minimal to mild depression. Although preliminary, these results underscore the importance of psychomotor measures in diagnosis, risk assessment, and monitoring in psychiatric care, alongside subjective evaluations.
Novel Findings:
Despite no significant group differences in subjective ratings for valence and arousal, head-tracking data revealed distinct psychomotor patterns in those with higher depression severity.
Individuals with moderate to severe depression exhibited significantly reduced scanning speed and standard deviation compared to individuals with minimal to mild depression.
VR head-tracking data can be considered as potential psychomotor marker of emotional responsiveness and a realiable indicator of vulnerability to depression.
These findings are particularly important in the light of rising cases of moderate to severe depression and anxiety among youths.
Figure above, with head-tracking data, depicts an example of head-movement data visualization in a 360° virtual environment by four participants, representing varying severity of depression and anxiety. The visualization demonstrates consistency in change in head- movement in individuals with severe depression and anxiety, suggesting that head-tracking data in VR videos exploration could be considered a reliable psychomotor assessment for depression. (Refer to figure 4 in the manuscript)
Figure: Raincloud plot depicting comparison between scanning speed while exploring 360° affective virtual environments across depression groups. SD = severe depression, MD = mild depression, and ND = no depression.
The raincloud plot above, depicts ANCOVA results indicating a significant effect of depression symptoms on scanning speed, after controlling for state and trait anxiety (GAD-7 scores and STAI-T scores, respectively) (p <.001, η2 = .295) (Refer to Figure 3 in the mansucript).
Further, the post hoc comparisons revealed that individuals with severe symptoms of depression demonstrated significantly slowest scanning speed (M = 3.88, SD = 0.88) compared to those with mild depression (M = 4.94, SD =1.07) and no-depression (M = 5.29, SD = 0.88), (p = .028, d = -1.22; and p = .021, d = 1.64, respectively) (Figure 3). However, no significant interaction effects were observed between depression and state and trait anxiety for scanning behavior. Depression scores did not significantly influence the yaw and standard deviation of yaw movement.
Current Trends in Virtual Reality-Based Mental Health Research and Applications - Symposium at APS's ICPS, 2023
Association for Psychological Science, International Convention of Psychological Science, (APS's ICPS) 2023
Brussels, Belgium, March 9-11
Symposium, Priyanka Srivastava (Chair and Speaker),
other speakers: Tammie Hydee (Envision Healthcare, US); Iva Georgieva (Assistant Professor at the Bulgarian Academy of Sciences, Bulgaria); Tanusree Mustafi (Founder of Begin Mental Health Care and Rehabilitation, and Mental Health Consultant at IIITH, India) and Allen Olson-Utrecho (Studio Bahia, Design Lead and Founder, US)
Virtual reality (VR) like Oculus Rift and HTCvive are changing the landscape of clinical research and practices, especially mental health and well-being. The upsurge in VR-based mental health research aligns with the growing demands for research and evidence-based mental health practices. Currently, mental health screening, diagnosis, and treatment are time-consuming and psychologically exhausting for both clinical professionals and patients. The diagnosis of any psychological disorder, including as common as depression, mainly relies on clinical examination, subjective evaluation, and self-reported symptoms. There are no globally accepted or approved biological, ecological, social, and psychological markers as part of diagnostic criteria (DSM-V, ICD-10) for any psychiatric disorders. The lack of standardized health markers for mental illnesses becomes all the more concerning for low-and-middle-income countries like India, where the doctor: patient ratio is approximately 1:3 lakhs for mental health cases. Despite advancements in medical sciences characterizing, mental health conditions are still challenging. The inability to create a phenomenological, psychosocial, and ecologically valid environment (e.g., exposure therapy) appears to be a critical bottleneck in addressing the gaps in understanding mental health conditions. Clinical research using VR technologies seems promising, as it could offer phenomenological, psychosocial, and ecological interventions alongside behavioral and biological measures in clinical and non-clinical settings. The proliferation of virtual reality technologies, like HTCvive or FOVE, wearable technologies, ease of accessibility, unobtrusive nature, flexibility to design ecologically valid environments with controlled settings, use of quantitative and qualitative measures, along with high display and interaction fidelity make VR systems a potential alternative aid for mental health diagnostics, monitoring, and treatment purposes. However, it is worth noting that VR is not devoid of challenges that may arise due to the highly immersive virtual environments and non-patient-centric designs, despite umpteen possibilities to create more ecologically valid stimuli with better-controlled settings. The symposium broadly focuses on virtual reality-based mental health research. We aim to address the diversity of mental health concerns ranging from VR-based screening to behavioral psychotherapy, alongside diverse population profiles, and psychosocial and phenomenological challenges. The symposium brings together researchers and practitioners from different continents: Asia, the US, and Europe, to share the current trends and emerging practices in VR-based mental health and well-being research. The session will help us examine the limitations of the existing clinical practices and will discuss the need to shift the focus from a paper-pencil and clinical interview-based examination to more objective, cognitive, biological, socio-ecological, and phenomenological perspectives of clinical observations. Further, the session will focus on virtual reality clinical research trends in mental health screening and psychosocial and phenomenological therapy research. We will end with a panel discussion to address the VR-based mental health research challenges to realize the future avenues of such research while recognizing a paradigm shift from traditional clinical practices to more advanced, futuristic clinical research and applications.
Self-Reported Depression Is Associated With Aberration in Emotional Reactivity and Emotional Concept Coding
Frontiers in Psychology, Emotion Science, 24th June 2022https://doi.org/10.3389/fpsyg.2022.814234ABSTRACT: Cognitive impairment, alterations in mood, emotion dysregulation are just a few of the consequences of depression. Despite depression being reported as the most common mental disorder worldwide, examining depression or risks to depression is still challenging. Emotional reactivity has been observed to predict the risk of depression, but the results have been mixed for negative emotional reactivity (NER). To better understand the emotional response conflict, we asked our participants to describe their feeling in meaningful sentences alongside reporting their reactions to the emotionally evocative words. We presented a word on the screen and asked participants to perform two tasks, rate their feeling after reading the word using the self-assessment manikin (SAM) scale, and describe their feeling using the property generation task. The emotional content was analysed using a novel machine-learning algorithm approach. We performed these two tasks in blocks and randomized their order across participants. Beck Depression Inventory (BDI) was used to categorize participants into self-reported non-depressed (ND) and depressed (D) groups. Compared to the ND, the D group reported reduced positive emotional reactivity when presented with extremely pleasant words regardless of their arousal levels. However, no significant difference was observed between the D and ND groups for negative emotional reactivity. In contrast, we observed increased sadness and inclination towards negative context from descriptive content by the D compared to the ND group. The positive content analyses showed a mixed result. The contrasting results between the emotional reactivity and emotional content analyses demand further examination between cohorts of self-reported depressive symptoms, no-symptoms, and MDD patients to better examine the risks to depression.
What networks of attention are affected by depression? A meta-analysis of studies that used the attention network test
Journal of Affective Disorders Reports, April, 2022https://doi.org/10.1016/j.jadr.2021.100302ABSTRACT: Depression is an all-too-common mood disorder that has been linked in various ways to the cognitive concept of attention. The Attention Network Test, based on the influential taxonomy of attention originally proposed by Posner, can be used to measure the efficacy of three components or networks of attention: Alerting, Orienting and Executive Control. Our focus here is to discover how depression might affect these networks. Studies related to attention and depression were selected for this meta-analytic review using the online Attention Network Test (ANT) Database (Arora, Lawrence & Klein, 2020) resource. Based on the inclusion and exclusion criteria, 5 publications were analyzed with a Bayesian hierarchical model to compare participants with and without a diagnosis of depression. The posterior distribution of our model revealed no credible differences in the Alerting and Orienting networks but showed credible non-zero values for Executive Control. This suggests that Executive Control is less efficient in individuals suffering from depression. As literature using the ANT with depressed participants is limited, the methodological variability between studies in this meta-analysis should be considered when interpreting these results. This meta-analysis review is the first quantitative review that aimed to address the ANT-depression literature. A consistent finding of no alerting and no orienting deficit but a deficit in the executive control suggests that some, but not all, components of attention are impaired in depression.
Affective Experience Correlates with Head Movement in VR
(accepted at 2021 IEEE 7th International Conference on Virtual Reality, China)10.1109/ICVR51878.2021.9483847
(a) VR exploration settings. (b) & (c) 3D plots of head-movement data while exploring VR 360°videosby a normal healthy participant. (b) 360°exploration with pleasant experience for "Tahiti Surf" video (index 70). (c) 360°exploration with unpleasant experience for "Abandoned City" video (index 3).
HIGHLIGHTS:
Request to use Head Tracking and self-report Data: Please write to -priyanka.srivastava@iiit.ac.in with the following information: Applicant Name, Academic Email id, Lab and University, Advisor Name, and Advisor email id.
- Virtual Reality (VR) could be used as an analysis of psychomotor response to affective experience.
- The current study shows a novel assessment of spatio-temporal scope of attention while exploring affective 3600 videos using a head-mounted display.
- Significant positive correlation between the standard deviation of yaw (spread of head-orientation around the mean) and valence (affective experience), and significant positive correlation between scanning speed (rate of change of angular movement) and valence, extend the support for ‘theory of positive emotion’.
- Widespread use of VR technology in affective research demands a standardized VR dynamic affective database.
Request to use Head Tracking and self-report Data: Please write to -priyanka.srivastava@iiit.ac.in with the following information: Applicant Name, Academic Email id, Lab and University, Advisor Name, and Advisor email id.
Virtual and Augmented Reality mental health research and practices
Invited book chapter, 2022, in Articiial Intellienence Taylor and Francis Group
Abstract: Virtual and augmented reality (VR and AR) technologies like Oculus Rift and Microsoft HoloLens are changing the landscape of medical health research and practices, especially mental health and well-being. VR and AR are emerging as alternative aids for treatments and diagnostics of psychological disorders. There has been criticism for traditional neuropsychological and psycho-therapeutic assessments for scoping the clinical based reviews to everyday complex functional scenarios. Currently, diagnosis of any psychological disorder, including depression and anxiety, mainly relies on clinical interviews and subjective reporting of the symptoms. Despite cognitive decline and psycho-motor disturbance, there are no globally accepted or approved cognitive markers comprising the diagnostic criteria (DSM-V and ICD–10) for any psychological disorders. The clinical research with digital technologies, especially VR and AR, may bring more quantitative and objective assessment using cognitive measures, besides the traditional clinical examination and reporting. Envisioning the future of mental health clinical practices, this chapter describes the state-of-the-art virtual and augmented reality clinical research and discusses the limitations and future directions of VR/AR mental health research.
Pilot data evaluating the effect of propensity to depression on 360-degree affective experience and 360-degree exploration behavior
Indian VR Affective Database with Self-Report and EDA Measure
published at 25th ACM Symposium on Virtual Reality Software Technology, 2019, Sydney
Abstract: The current work assesses the physiological and psychological responses to the 360° emotional videos, selected from Stanford virtual reality (VR) affective database, and presented using VR head-mounted display (HMD). Participants were asked to report valence and arousal level after watching each video. The electro-dermal activity (EDA) was recorded while watching the videos. The current pilot study shows no significant difference in skin-conductance response (SCR) between the high and low arousal experience. Similar trends were observed during high and low valence. The self-report pilot data on valence and arousal shows no statistical significant difference between Stanford VR affective responses and the corresponding Indian population psychological responses. Despite positive result of no-significant difference in self-report across cultures, we are limited to generalize the result because of small sample size.
Request to use EDA and self-report Data: Please write to -priyanka.srivastava@iiit.ac.in with the following information: Applicant Name, Academic Email id, Lab, University, Advisor Name, and Advisor email id.
Indian VR Affective Database - Self report and EDA.pdf
Virtual Reality Affective Database with Indian Population
Presented at the ACCS 2018, India
Abstract: Virtual reality (VR) is blurring the gaps between direct and indirect access to the perceived reality. Feeling of being present and the ability to interact with the environment is changing the way human behavior can be studied under control experimental settings. One of the critical applications of VR is to understand threat/ fear perception and related action, and further develop a rehabilitation program to prepare patients to encounter such situations more effectively. With increased demand for dynamic use of emotional stimuli, it becomes indispensable to develop a VR affective database. Currently, we have Stanford public database available for VR affective research, which certainly lacks the responses from Indian population that might influence the VR affective interaction. Therefore, validation of Stanford immersive VR affective database becomes essential for future research related to the Indian population. Our pilot data showed encouraging result, indicating that the Stanford database can be used for Indian population as well.
Google Sites
Report abuse