Past Grantees and Fellows

The aim of this project is to understand the convergence of digital and neuro-technology in chronic pain amidst the opioid crisis. It compares pain technologies combining neuro-technological interventions with data-driven techniques. The project will analyse their promise as "post-opioid" technologies as well as associated risks.

Through semi-structured interviews, this project aims to understand the facilitators and barriers to the implementation of a food pantry-based Produce Prescription program in New York City from stakeholders’ perspectives. The findings will inform the design and implementation of community-clinic-partnered health interventions.

The proposed work will contribute to the development of a validated outdoor playspace audit tool. Through this initial mixed-methods testing and preliminary validation of a draft tool at 6-8 pilot sites the work will support the final refinement, testing, and publication of this valuable research and design tool.

This study examines the how the community built environment impacts the development and well-being of youth in 2 US cities, and their experience of their community as youth-friendly.  Potential inequities in the provision of environmental resources which support positive youth development will also be examined.

The Civic Playground Project seeks to empower individuals of different backgrounds, languages, and abilities through shared modes of making and collaborative play. The project is anchored in the development and deployment of inclusive frameworks that foster engagement between communities and their built environments.

How do adult residents’ experiences of intergenerational social mobility differ between upgraded and non-upgraded slums in Global South cities? Using semi-structured interviews, I will explore how physical transformations in upgraded and non-upgraded slums in Bogota and Nairobi relate to their residents’ social mobility experiences.

This research project combines agent-based modeling and scenario planning to engage stakeholders in examining reinvestment choices along a spectrum of building reuse to demolition, taking into account larger non-profit and public entities that play an outsized role related to innovation along this spectrum.

In what context is discrimination most severe? Using an experimental approach that updates treatment assignment rules as information is learned, this study will discover the contexts in which human decision-makers make especially discriminatory pairwise choices.

Manville's 2016 fellowship helped him complete two journal articles, and also led to a productive collaboration with ISS fellow Adam Levine, resulting in an additional journal article.

Medical practice variation — individuals with the same characteristics and medical symptoms being prescribed different treatments — is a long-standing and widespread problem. This research proposes a method to discover medical practice variation given historical prescribing data, and evaluates personalized, generative AI-based interventions to reduce such variation in the field.

This research develops approaches and datasets to train LLMs to be logically rational (eg. that do not express self-contradictory beliefs). Our work will enable using LLMs to accurately simulate human responses to information treatments, such as persuasion and propaganda, and thus reduce researchers' and practitioners' reliance on costly belief elicitation from human subjects.