When you apply, we will ask you to rank your top three interests from the research projects listed below. To frame the social relevance of the research, each project is aligned with at least one Sustainable Development Goal (SDG) defined by the United Nations.
Your project choices are important because they will determine which mentors will evaluate your application. We therefore encourage applicants to explore each mentor’s website to learn more about the individual research activities of each lab.
Facilitating Transgender Identity Development with Virtual Reality
Individuals that identify as transgender often face discrimination, harassment, or physical violence, making it uncomfortable or potentially unsafe to express a non-conforming gender identity, even with family or close friends. The stressors can have dire implications, leading to significantly increased risks for mental illness, substance abuse disorders, and suicide. This project aims to investigate how virtual reality can used by transgender and gender-questioning individuals to explore alternative body representations and gender expressions. REU students will collaborate with researchers in human-computer interaction and family social science to explore novel technologies that aim to improve wellbeing of transgender adolescents and adults.
Reducing Inequitable Barriers for Engaging with Virtual Reality Technologies
Half or more of all people who have ever used VR technology have at some point experienced cybersickness – feelings of nausea, disorientation and eye strain – and cybersickness is becoming a major obstacle to the wider deployment of VR for socially beneficial purposes in areas such as education, psychotherapy, job training, implicit/unconscious bias reduction training, cultural heritage, manufacturing, design, and more. As cybersickness disproportionately affects women, developing strategies to predict and prevent cybersickness onset or mitigate cybersickness severity is especially important to ensure equal opportunity of access to these emerging technologies. The REU student will work with a team of students and faculty from Computer Science and Kinesiology to help design, develop, and evaluate novel cybersickness mitigation techniques.
Technologies for Connecting Older Adults and Elementary School Children
Intergenerational mentorship provides children with one-on-one support that can improve academic performance and increase resilience, while also helping older adults remain active and connected. Computing allows to expand access to intergenerational mentorship and make the experience more fun for both the mentors and the students. Working closely with a local non-profit for older adults and Twin Cities public school teachers, we will develop and build new prototypes to support reading together, helping with homework, and playing games across distance. As an example, we may build a tablet-based augmented reality application that allows older adults and children to do a virtual scavenger hunt together while reading through a book.
Computational Support for Recovery from Addiction
Addiction and alcoholism are one of the greatest threats to people’s health and well-being. Early recovery is a particularly sensitive time, with as many as three fourths of people experiencing relapse. Computing provides an opportunity to connect people with the support they need to get and stay in recovery. Working closely with members of the recovery community, we will conduct and analyze qualitative formative work to understand people's needs and values. Based on insights gained from this work, students will contribute to the development and deployment of interactive technologies for computational support for recovery. As an example, we may work on a mobile application to help connect people who are new to recovery with more experienced members of the recovery community.
Design of Augmented Reality (AR) Technology-mediated Disclosure
People who live with stigmatized identities or conditions (e.g., members of the LGBTQA+ community, people living with chronic conditions such as HIV) experience high levels of stigma from society. Public or societal stigma has a highly negative impact on stigmatized individuals’ physical and psychological health. Self-disclosure to others about their stigmatized conditions has proven to be a key important step prior to activating social support exchange from others as a means to both cope better with stigma as well as to reduce the prevalence of such stigma in society. Thus, working with participants from stigmatized groups, we would explore their feedback on an augmented reality (AR) technology prototype that could assist in disclosure in collocated scenarios. For example, the use of AR eyeglasses that could allow people to disclose non-verbally via sending or showing visual cues to others in a collocated space.
Computational Analysis of Alternative Treatments for Opioid Addiction in Online Communities
There are online communities that support those struggling with opioid use disorder (colloquially called opioid addiction) and facilitate recovery outcomes. In addition to promoting well-ground treatment strategies for recovery, some online communities also promote clinically unverified treatments for recovery. These include untested drugs and substances, questionably legal drugs, and over-the-counter medications. Little research exists on which alternative treatments people use, whether these treatments are effective for recovery, or the side effects of their use. This project will use computational techniques from natural language processing, applied machine learning, and data mining to study these alternative therapies for opioid adduction. This may include automatic identification of what substances people are using; studying changes in the alternative therapies over time; or the connection of substance use to other mental illnesses or symptoms, like depression. Students will also learn about what it takes to develop a data pipeline for projects like this, including data gathering, annotation, and cleaning; model training and engineering; and model testing and evaluation on real-world examples; and ethical outcomes of these systems.
Equitable Algorithms: How Intelligent Systems Can Serve the Needs of Community Members
Organizations have been deploying intelligent algorithms to support their decision making processes for several decades, but awareness of this trend has exploded recently. Algorithms have been used to support judges in making parole decisions, social workers in evaluating whether a child should be removed from their family home, and employers in evaluating resumes of job applicants. In many cases, fundamental inequities have been identified, leading some researchers and advocates to argue that algorithms should not be used at all for these sorts of sensitive decisions. We are fortunate to study these issues in the context of Wikipedia, the world's largest encyclopedia and collaborative knowledge production project. Algorithms play a prominent role in Wikipedia, for example, identifying edits that are likely to be damaging and recommending articles for people to work on. However, Wikipedia also is known to suffer from biases, particularly against contributions by and about women. Algorithms have the potential to make such problems worse (e.g., by reinforcing existing biases) or better (e.g., by directing people toward editing underrepresented topics or blocking harmful behaviors). We have been studying how algorithms operate in Wikipedia and coming up with methods for designing more effective and equitable algorithms. We will define a specific REU project in this space that meets your particular interests and abilities.
Exploring Virtual Reality to Mitigate Implicit Bias
Implicit biases are unconscious stereotypes that can affect our interpretation of factual information and influence the decisions we make in ways that we are not overtly aware of, leading to negative consequences in a very wide range of important areas including: public safety; hiring, promotion and salary decisions; and the provision of timely and appropriate medical treatment. Virtual reality technology has tremendous potential as a medium for implicit bias reduction training, yet there are many open questions in how best to deploy VR for maximum benefit. The REU student will work with a small team including a Ph.D. student and a social psychologist to help design and pilot potential VR interventions.
Applications of Virtual Reality in Addressing Neuropsychiatric Disorders
This project focuses on efforts to develop and test the use of immersive virtual reality technology to assist in the early detection of childhood-onset neuropsychiatric disorders and to support the implementation and quantitative assessment of therapeutic interventions. The summer intern will work closely with a team that includes faculty from the Medical School (Psychiatry) as well as experts in computer vision, visualization, and virtual reality. The principle targeted disorder for detection and quantification is Tourette syndrome, which is characterized by observable, involuntary repetitive motor movements and vocalizations. We will also be considering potential VR-supported interventions for obsessive-compulsive disorder, attention-deficit hyperactivity disorder, and autism; many of these neuropsychiatric conditions are co-occurring. Research efforts will focus on computational methods and hardware for human behavior analysis and on the development of immersive virtual environments that may be useful in diagnosis and treatment.
Applications of Virtual Reality in Forestry
In this project, the summer intern will work in an interdisciplinary team with faculty from computer science and forestry to explore the use of VR for forestry-related aims, potentially including: research into the potential benefits of virtual "forest-bathing"; using VR to support a better understanding of the potential impacts of climate change on forest resources; using VR to solicit user feedback on alternative strategies for supporting forest recovery from disease/damage.
Reducing Dependency on Large Datasets for Visual Underwater Human-Robot Collaboration
This project will focus on reducing the dependency on large datasets for vision-based autonomous underwater robotics, particularly when collaborating with human partners. Underwater robots capable of following divers, understanding their hand gestures, or their actions and intent are essential for collaborative task execution underwater, but developing such algorithms using machine learning requires a significant amount of imagery. Collecting such datasets can be extremely costly and dangerous to humans, robots, and the environment, and annotating these is an even more laborious process. This project will look into creating visual learning-based methods that do not require large-scale datasets for underwater human-robot collaborative missions.