Xinlan Emily Hu

Xinlan Emily Hu

Auto-generated profile, based on publicly available information

Xinlan Emily Hu

Operations, Information, and Decisions Department, The Wharton School, University of Pennsylvania, Philadelphia, Pennsylvania 19104; and Institute for Data, Systems, and Society, Massachusetts Institute of Technology, Cambridge, Massachusetts 02142

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 February 2025 in Proceedings of the National Academy of Sciences
Verified
Journal Article
Open access
NLNaijia Liu
Naijia Liu
XHXinlan Emily Hu
Xinlan Emily Hu
YSYasemin Savas
Yasemin Savas
MBMatthew A. Baum
Matthew A. Baum
ABAdam J. Berinsky
Adam J. Berinsky
BSBrandon M. Stewart
Brandon M. Stewart
An enormous body of literature argues that recommendation algorithms drive political polarization by creating “filter bubbles” and “rabbit holes.” Using four experiments with nearly 9,000 participants, we show that manipulating algorithmic recommendations to create these conditions has limited effects on opinions. Our experiments employ a custom-built video platform with a naturalistic, YouTube-like interface presenting real YouTube videos and recommendations. We experimentally manipulate YouTube’s actual recommendation algorithm to simulate filter bubbles and rabbit holes by presenting ideologically balanced and slanted choices. Our design allows us to intervene in a feedback loop that has confounded the study of algorithmic polarization—the complex interplay between supply of recommendations and user demand for content—to examine downstream effects on policy attitudes. We use over 130,000 experimentally manipulated recommendations and 31,000 platform interactions to estimate how recommendation algorithms alter users’ media consumption decisions and, indirectly, their political attitudes. Our results cast doubt on widely circulating theories of algorithmic polarization by showing that even heavy-handed (although short-term) perturbations of real-world recommendations have limited causal effects on policy attitudes. Given our inability to detect consistent evidence for algorithmic effects, we argue the burden of proof for claims about algorithm-induced polarization has shifted. Our methodology, which captures and modifies the output of real-world recommendation algorithms, offers a path forward for future investigations of black-box artificial intelligence systems. Our findings reveal practical limits to effect sizes that are feasibly detectable in academic experiments.

1 November 2026 in Management Science
Verified
Journal Article
Closed access
XHXinlan Emily Hu
Xinlan Emily Hu
MWMark Whiting
Mark Whiting
LGLinnea Gandhi
Linnea Gandhi
DWDuncan J. Watts
Duncan J. Watts
AAAbdullah Almaatouq
Abdullah Almaatouq
Research on teams spans many contexts, but integrating knowledge from heterogeneous sources is challenging because studies typically examine different tasks that cannot be directly compared. Most investigations involve teams working on just one or a handful of tasks, and researchers lack principled ways to quantify how similar or different these tasks are from one another. We address this challenge by introducing the “Task Space,” a multidimensional space in which tasks—and the distances between them—can be represented formally, and use it to create a “Task Map” of 102 crowd-annotated tasks from the published experimental literature. We then demonstrate the Task Space’s utility by performing an integrative experiment that addresses a fundamental question in team research: when do interacting groups outperform individuals? Our experiment samples 20 diverse tasks from the Task Map at three complexity levels and recruits 1,231 participants to work either individually or in groups of three or six (180 experimental conditions). We find striking heterogeneity in group advantage, with groups performing anywhere from three times worse to 60% better than the best individual working alone, depending on the task context. Critically, the Task Space makes this heterogeneity predictable: it significantly outperforms traditional typologies in predicting group advantage on unseen tasks. Our models also reveal theoretically meaningful interactions between task features; for example, group advantage on creative tasks depends on whether the answers are objectively verifiable. We conclude by arguing that the Task Space enables researchers to integrate findings across different experiments, thereby building cumulative knowledge about team performance. This paper was accepted by Sameer Srivastava, organizations. Funding: The authors thank the Alfred P. Sloan Foundation [Grant #202-13924] and the MIT Wade Fund for their generous support of this research. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.03544 .

28 February 2026 in PsyArXiv
Verified
Preprint
Open access
XHXinlan Emily Hu
Xinlan Emily Hu
MWMark Whiting
Mark Whiting
LGLinnea Gandhi
Linnea Gandhi
DWDuncan J Watts
Duncan J Watts
AAAbdullah Almaatouq
Abdullah Almaatouq
Research on teams spans many contexts, but integrating knowledge from heterogeneous sources is challenging because studies typically examine different tasks that cannot be directly compared. Most investigations involve teams working on just one or a handful of tasks, and researchers lack principled ways to quantify how similar or different these tasks are from one another. We address this challenge by introducing the “Task Space,” a multidimensional space in which tasks—and the distances between them—can be represented formally, and use it to create a “Task Map” of 102 crowd-annotated tasks from the published experimental literature. We then demonstrate the Task Space’s utility by performing an integrative experiment that addresses a fundamental question in team research: when do interacting groups outperform individuals? Our experiment samples 20 diverse tasks from the Task Map at three complexity levels and recruits 1,231 participants to work either individually or in groups of three or six (180 experimental conditions). We find striking heterogeneity in group advantage, with groups performing anywhere from three times worse to 60% better than the best individual working alone, depending on the task context. Critically, the Task Space makes this heterogeneity predictable: it significantly outperforms traditional typologies in predicting group advantage on unseen tasks. Our models also reveal theoretically meaningful interactions between task features; for example, group advantage on creative tasks depends on whether the answers are objectively verifiable. We conclude by arguing that the Task Space enables researchers to integrate findings across different experiments, thereby building cumulative knowledge about team performance.

8 September 2026 in PsyArXiv
Verified
Preprint
Open access
XHXinlan Emily Hu
Xinlan Emily Hu
Conversation is a rich window to our social world. Through it, humans coordinate their activities, resolve interpersonal disputes, and build lasting social bonds. However, the full potential of conversation data often goes unrealized, because analyzing a conversation requires making contingent and costly decisions about which constructs to study, how to measure them, and at what unit of analysis to do so. At the start of a project, however, an analyst may not know which constructs merit this investment. To lower the cost of a first-pass analysis, I introduce an omnibus tool for quantifying conversation: the Team Communication Toolkit (team_comm_tools on the Python Package Index; abbreviated TCT). The TCT is an open-source Python package that extracts 166 features at three units of analysis (the utterance, the speaker, and the conversation) from any corpus of text-based conversations. Requiring just four input columns and fewer than 10 lines of code, the TCT replaces what would otherwise be a patchwork of processing pipelines with a single streamlined interface. In a case study of 803 online public goods games, I use the TCT to explore which attributes of the talk preceding a decision predict how much a group contributes. I show that conversation is a meaningful signal of group outcomes only during a critical early-game period. 27 measures survive false-discovery-rate correction and replicate on held-out games. Together, they suggest a high-level interpretation that early-game conversations provide an honest signal of engagement, which is influential because social impressions are still nascent.