Disclosure Matters: How Self-Disclosure Statements in Song Signing Videos Shape d/Deaf Audiences' Acceptance of Culturally Sensitive Content

This CHI 2026 paper investigates whether self-disclosure statements (SDS) — short first-person texts in which a song signer describes their identity, motivations, training, and relationship to the Deaf community — can improve how d/Deaf audiences receive song signing videos on platforms like YouTube. Song signing (the performance of song lyrics through sign language, typically ASL) has grown rapidly online, largely produced by non-Deaf amateurs, and the Deaf community has raised persistent concerns about misrepresentation, incorrect signs being replicated by learners, and cultural appropriation for profit or clout. The authors, grounded in Cross-Cultural Adaptation Theory and Intergroup Contact Theory, frame SDS as a communicative design mechanism for positionality and trust-building. The work consists of two studies. Study 1 is a mixed-methods investigation with 11 hearing and hard-of-hearing song signers interviewed via Zoom plus 50 d/Deaf and hearing viewers surveyed about what they want song signers to disclose — producing a catalogue of 8 SDS elements with rationales, trade-offs, and drawbacks. Study 2 is a within-subjects controlled experiment with 24 d/Deaf participants (12 Deaf, 12 Hard of Hearing) who rated four song signing videos both without and with guided SDS (600–1,050 words each), using 5-point scales, open-ended comments, and post-study semi-structured interviews.

Across all 24 d/Deaf participants, mean perception ratings rose from 3.11 (video only) to 3.58 (video + SDS), a statistically significant improvement (Mann–Whitney U=2053, p=.027, r=-.20). The effect was driven almost entirely by Deaf participants, whose ratings rose from 2.50 to 3.08 (p=.042, Kendall's tau r=.73 — a strong positive correlation between SDS presence and rating). Hard-of-hearing participants showed a smaller, non-significant change (3.78 to 4.33, p=.184). The three SDS elements participants rated most important were song signing style (MD=5), sign language used (MD=5), and relationship to the Deaf community (MD=4.5) — confirming that Deaf viewers care most about cultural-linguistic precision and community positioning. The full 8-element catalogue includes: signing style, relationship to the Deaf community, sign language used, background, limitations, Deaf education/certification, experience, and motivation. Caveats surfaced: some limitation statements read as "self-deprecating" and lowered viewer confidence; oversharing personal detail exposed performers to harassment (one signer reported having an SDS weaponised against her); and some participants felt SDS risked distracting from the performance itself.

For accessibility practitioners working on social-video platforms, Deaf-community content, or cross-cultural creator tools, this paper gives concrete, evidence-backed guidance on what to disclose and how. The 8-element SDS framework is directly usable as a template or authoring scaffold for creators working outside a community they represent. For platform designers, the finding that Deaf viewers respond strongly to positionality cues — while Hard-of-Hearing viewers do not — argues against one-size-fits-all disclosure UI and in favour of audience-aware presentation. The authors' recommendations on placement (non-intrusive overlays, text annotations, and structured narratives that avoid overshadowing the performance) translate to captioning, content-moderation, and creator-tool design more broadly. Key limitations: the sample is small (N=24 for Study 2), Study 1 deliberately excluded Deaf song signers (only examined non-Deaf disclosers), and video stimuli were limited to four pop/hip-hop tracks with varied caption styles. The work extends beyond song signing to any HCI context where cultural outsiders produce content — AI-generated sign language, ally-led disability storytelling, or cross-cultural translation platforms all face the same disclosure design problem.