Song Mi Lee

Information Behavior Researcher

Practicing Information Literacy on Social Media

Ongoing dissertation research

How can we re-imagine social media design to support practicing information literacy? My doctoral prelim exam and dissertation research examine the psychosocial influences of social media design on practicing information literacy—not just that of everyday users but of social media content creators, the powerhouse of today’s information ecosystem. I propose design interventions to nudge purposeful and reflective social media use and discuss implications for information literacy campaigns and policies around social media.

Prelim exams passed with distinction. A milestone “pass with distinction” is awarded when the literature review, research proposal, and oral defense components are all exemplary, representing the top 10% of students in this endeavor in any given year.
Prelim committee: Cliff Lampe, Andrea Thomer, Irene Pasquetto

Highlights

  • Information literacy (IL) campaigns so far have largely relied on instructions and individuals’ voluntary self-control. The big next step should demand systemic changes—to make social media a more information literacy friendly environment.
  • Practicing information literacy (IL) under three key design features of the current commercial social media stream (i.e., feeds):
    • Algorithmic content curation
    • Normative cues (identity cues + social reactions)
    • Attention economy
  • I connect the literature in social media studies, behavioral intervention, library and information science, human-computer interaction, and computer-mediated communication.
  • I invite social media content creators and library/information professionals to envision social media design that facilitates informed and agentic information interactions.

Methods

Literature review, semi-structured interviews, mock-up designs, surveys, focus groups

expected Contributions

  • Seek ways to reduce the burden/cognitive load of stakeholders in the information disorder
    • The platform’s burden of content moderation, the information consumers’ burden of fact-checking, the information producers’ burden of algorithmic labor
  • Provide social media design recommendations with regard to content arrangement, interface cues, and behavioral nudges
  • Develop deeper insights into how people understand the interplay between user agency and system agency

Understanding Perpetrators and Public Perceptions of Online Harms

I’m a research assistant/project manager/lead author of the NSF-funded research project: CHS: Medium: Drawing from Theories of Justice to Respond to Online Harassment. With the team, I design and conduct survey studies that look into how justice theories and perceptions help understand online harms. I also shed more light on the psychological mechanisms behind online harm perpetration (as opposed to victimization) to imagine more preemptive remediation, as an alternative to after-the-fact content moderation.

Faculty members: Sarita Schoenebeck, Cliff Lampe, JJ Prescott

oral & written Communication

some issues we tackle

  • How do we design a less-stigmatizing survey instrument to access people who engage in online harassment?
  • We know online harassment is “serious”…can we quantitatively measure the perceived severity by the general public? What factors contribute to the perceived severity, and how do they differ by different demographics/experiences?
  • ‘Online harassment’ is a vague, controversial umbrella term. What are some alternatives?

Contributions

  • Explicate the multi-dimensionality of online aggression and conflict behaviors that are often bundled under the term ‘online harassment’
  • Recommend design interventions that account for people’s psychological characteristics, following the safety by design principles
  • Provide large-scale, general public survey data (US representative sampling) and survey design on online harassment perpetration and perceptions

Negative Interface Cues Against Online Harassment

Can Dislikes and Flags be used against online harassment—even when the harassment is justified and supported through Likes as an act of moral outrage? Online retributive harassment, or harassment targeted at social wrongdoers (e.g., criminals, racists), is a prevalent, often seemingly justifiable, and thus hard-to-moderate type of harassment that’s intertwined with online shaming and cancel culture. We experiment if negative interface cues (i.e., Dislikes and flags) can signal a social norm against retributive harassment. The negative cues are predicted to overpower positive cues (e.g., Likes) based on the theory of negativity effect—but we reveal a more complicated picture.

Happy to mentor and collaborate with a master’s student, Shriya Hardikar (U of Michigan)

ORAL & Written communication

  • Conference presentation at CSCW 2022, Changes in Platform Designs Session
  • Song Mi Lee, Andrea K. Thomer, and Cliff Lampe. 2022. The Use of Negative Interface Cues to Change Perceptions of Online Retributive Harassment. Proc. ACM Hum.-Comput. Interact. 6, CSCW2, Article 334 (November 2022), 23 pages. https://doi.org/10.1145/3555226
  • Song Mi Lee. Nov. 12, 2021. Can Dislikes and Flags Change Perceptions of Online Retributive Harassment? Presented to Twitter Advisory Board.

Some assumptions we challenge

  • People assume that the Dislike button will be abused FOR harassment… but can we experiment if it can be used AGAINST harassment?
  • Online bystanders are often assumed to be helpful and empathetic towards the victim of the harassment. However, they might as well be antagonistic and escalate the situation depending on the social context and their own dispositions.

COntributions

  • Identify, test, and compare theoretical frameworks of interface cues around social norms and harassment
  • Suggest interface design for bystander intervention that accounts for people’s psychosocial tendencies

Sense-Making of COVID-19 Deaths

WORK in progress

Speaking ill of the dead, particularly those who were average citizens without high social profiles is a widespread social taboo—but we are seeing numerous counterexamples around COVID-19 deaths. I’m investigating how people make sense of COVID-19 deaths, especially the deaths of anti-vaxxers in frustration and anger, which involves online harassment and shaming. Why are people engaging in online shaming of the dead (and sometimes the bereaved)? Rather than dismissing it as an immoral, malevolent activity, I bring in a different theoretical perspective.

Happy to mentor and collaborate with an undergrad student, Angela Voit (U of Michigan), and a master’s student, Manvitha Madem (U of Michigan)

Usability Testing: Mental Health Intervention App for Senior Citizens

I provide UX consultation to the University of Michigan School of Social Work research team to develop Empower@Home, an online Cognitive Behavioral Therapy (iCBT) tool designed for older adults. I primarily design usability tests and competitive analyses with a team of UI/UX designers.

Product Lead: Xiaoling Xiang
Keywords: UX of older adults