Microphone Security Project Overview
The Microphone Security Project was a research project out of the University of Washington’s User Empowerment Lab that explored users’ perceptions of audio data collection on smart devices, such as smartphones and smart speakers, to inform interface design recommendations for developers. This project was led by Dr. Alexis Hiniker and Ph.C. Julia Dubar.
-
I worked as an Undergraduate Research Assistant from March 2019 to September 2021, where my key roles were:
Qualitatively analyzing over 30 generative interviews and focus groups
Creating and facilitating co-design sessions
Synthesizing results into a manuscript that was published in the Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
Giving an invited talk at the ACM International Joint Conference on Pervasive and Ubiquitous Computing
-
I worked directly with Dr. Alexis Hiniker, Ph.C. Julia Dunbar, and Ashley Boone on this work.
-
My key accomplishments on this project:
Being second author on a paper titled “Is Someone Listening? Audio-Related Privacy Perceptions and Design Recommendations from Guardians, Pragmatists, and Cynics“ that was published in the Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
Giving an invited talk on the project at the ACM International Joint Conference on Pervasive and Ubiquitous Computing
-
The key methodologies and skills I used during this project were:
Generative User Interviews & Focus Groups
Generative Co-Design
Inductive-Deductive Qualitative Analysis
Scientific Writing
Presenting Findings
Figma Prototyping
The Details
We conducted a two-part study to identify user-centered, privacy-conscious designs for devices that record audio data. We first conducted interviews and focus groups to explore users’ perspectives on audio recording, using this formative work to create prompts for subsequent design sessions.
Interviews & Focus Groups
We conducted 32 semi-structured interviews and focus groups that probed the participants’ perceptions of audio data recording, concerns about privacy, risk-management strategies, usage habits, and other related topics.
Our goal was to assemble a participant pool with a wide range of privacy and security needs, thus we purposely recruited people who are blind or low-vision, immigrants with varying legal statuses, and members of the general public.
We conducted an iterative inductive-deductive qualitative analysis of the interviews and focus groups. Our initial code categories categorized participants as concerned or unconcered. Following best practices, we then consulted the literature to examine our emergent themes against existing constructs and ultimately found that Schomakers et al framework was a useful lense for our analysis moving forward.
Co-Design Sessions
From our interview and focus group analysis, We developed three design session protocols for an in-person, general-population session, an in-person session for blind or low-vision participants, and a remote session. We created eight “scenario cards” and eight “parameter cards” as prompts for the design sessions.
Scenario cards described concrete user scenarios, inspired by the concerns participants expressed during interviews and focus groups, focusing on misuses of smartphones and cellphones because these were frequently raised by participants. Parameter cards were provided to help participants scope their ideas and included topics such as push notifications, icons in the status bar, and text-based disclaimer.
We then repeated our analysis process, categorizing our participants based on their responses during a pre-task discussion. Over the course of several weeks, we identified design themes within each group and created wireframe interpretations of participants’ ideas.
Accomplishments
UbiComp Invited Talk
I reported our findings on behalf of our research team at the ACM International Joint Conference on Pervasive and Ubiquitous Computing. My presentation can be watched above.
IMWUT Full Paper
We published this work a paper titled “Is Someone Listening? Audio-Related Privacy Perceptions and Design Recommendations from Guardians, Pragmatists, and Cynics.” that was published in the Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies.