Building Narratives and Probing Concepts: Preparing Materials for Co-Design with Autistic Livestreamers
Terrance Mok, Tyson Hartley, Anthony Tang, Adam McCrimmon, Lora Oehlberg · 2025 · ACM Transactions on Accessible Computing · doi:10.1145/3706575
Summary
This paper presents scenario-based design (SBD) materials developed to facilitate future co-design sessions with autistic livestreamers on Twitch. Building on prior ethnographic research involving semi-structured interviews with 10 autistic adult streamers, the authors transform qualitative findings into concrete design artifacts: narrative scenarios depicting challenges faced by autistic streamers paired with speculative technology concepts that could address those challenges. The work uses identity-first language ("autistic person" rather than "person with autism"), reflecting the preferences of participants and a growing trend in autism research. The research addresses a gap between ethnographic insights and actionable design. While prior work documented how autistic users experience livestreaming, this paper bridges that understanding to materials that can prompt co-creative dialogue. The authors identified five intersecting themes from their interviews: Access to People and Social Experiences (livestreaming provides community and relationships otherwise difficult to access), Sharing Autistic Identity (opportunities for representation and "unmasking"), Achieving Personal Growth (improved social skills that transfer offline), Control of Interaction and Space (streamers set rules, pace, and moderation), and Accessibility Challenges of Being an Autistic Streamer (negative viewers, social exhaustion, sensory issues). Livestreaming emerged as a uniquely valuable social space for autistic adults. Unlike face-to-face interactions that demand neurotypical communication norms, streaming offers asymmetric communication where streamers control incoming social information, can take time to process and respond, and build communities around shared interests. Participants reported that streaming helped them practice social skills, express their authentic selves, and form meaningful relationships—benefits often difficult to access through conventional social channels.
Key findings
The paper presents six detailed scenarios, each pairing a narrative with a "strawman" technology concept designed to provoke discussion rather than prescribe solutions: 1. **Light Sensitivity** (Remy's narrative): Addresses sensory challenges with bright streaming lights. Proposes "Light Up AI"—an AI tool that creates a well-lit digital image of the streamer from low-light camera input, allowing comfortable streaming without harsh lighting. 2. **Moderating in a Meltdown** (Emma's narrative): Depicts a streamer overwhelmed by chat toxicity during an autistic meltdown who makes hasty banning decisions. Proposes "!buffered ban"—a tool that temporarily bans users but allows review when the streamer is calmer. 3. **What Do These Mean?** (Sarah's narrative): Addresses difficulty interpreting Twitch emotes with ambiguous meanings. Proposes "Emotes Explained"—hovering over emotes reveals definitions plus contextual usage notes indicating tone (genuine vs. sarcastic). 4. **Do What We Say** (Joshua's narrative): Depicts challenges maintaining chat norms without constant streamer intervention. Proposes "AI Community Member"—an AI chat participant that models desired behavior, responds to others, and gently corrects norm violations. 5. **Talking to the Void** (Jordan's narrative): Addresses the challenge of streaming to minimal audience engagement. Proposes "Stream Connect"—a matchmaking tool connecting streamers with similar interests for collaborative broadcasts. 6. **Livestream Multi-Tasking** (Kim's narrative): Depicts missing important viewer messages while focused on gameplay. Proposes "Emotion Aware"—a system that analyzes chat for emotional content and highlights/reads aloud significant messages. Each scenario includes discussion questions and consideration prompts designed for use in co-design sessions.
Relevance
This work provides a methodological contribution for accessibility researchers: a detailed, replicable process for transforming ethnographic findings into co-design materials for neurodiverse populations. The four-step approach—identifying stories from raw data, developing questions, building narratives, and preparing technology concepts—offers a template for practitioners working with any underrepresented user group. The technology concepts themselves, while intentionally provocative rather than prescriptive, identify concrete design opportunities for streaming platforms. Features like buffered moderation decisions, contextual emote explanations, and emotion-aware chat highlighting could benefit all streamers while being particularly valuable for autistic users who may struggle with social ambiguity, sensory overload, or decision-making under stress. For accessibility practitioners, the research reinforces that autistic users are not a monolithic group requiring "fixing"—participants valued livestreaming precisely because it let them be authentically autistic rather than masking. Design interventions should support autistic preferences and communication styles, not normalize them toward neurotypical expectations. The emphasis on user control (over environment, pace, rules) aligns with broader accessibility principles: providing options and agency rather than one-size-fits-all solutions. The materials are publicly available at GitHub for other researchers to adapt for their own co-design work.
Tags: autism · livestreaming · Twitch · co-design · scenario-based design · neurodiversity · participatory design · social media accessibility