Looking Beyond the Screen to Study the Technology Use of Older People Experiencing Cognitive Concerns
Ruipu Hu, Eun Kyoung Choe, Amanda Lazar · 2026 · Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI '26) · doi:10.1145/3772318.3790526
Summary
This CHI 2026 paper is a week-long qualitative study of how ten older adults with cognitive concerns — nine with Subjective Cognitive Decline (SCD) and one with Mild Cognitive Impairment (MCI), ages 50s–80s, six Mandarin-speaking — use videoconferencing in their everyday lives. The authors argue that HCI research on older adults has focused too narrowly on the software interface and has overlooked the wider material environment that makes videoconferencing possible. Combining two in-person interviews (one demographic/clinical, one reflective), in-person observation during an actual videoconference, and a novel modified diary study in which participants documented objects onto Polaroid-and-yarn corkboards, they assembled 281 object mentions across 15 hours of diary audio, 14 hours of observation, and ~14 hours of interviews. Data was analysed via reflexive thematic analysis through the lens of Hutchins' distributed cognition theory. The paper's research questions are RQ1: what kinds of objects participants interact with in the context of a videoconference, and RQ2: how those objects support videoconferencing for people with cognitive concerns. Contributions are threefold: (1) a seven-category typology of objects involved in videoconferencing — hardware, software/web platforms, paper-based objects, personal items, built environment, and 'other' (pets, people, weather); (2) three mechanisms by which these objects support cognition — holding information, distributing effort across time, and distributing effort across space via grabbable/glanceable/switchable affordances; and (3) design implications that re-centre the material environment as a legitimate site for accessibility intervention rather than treating the on-screen interface as the whole problem.
Key findings
Participants drew on a far wider object ecology than interface-centric studies capture. Paper-based artefacts (handwritten to-do lists, printed attendee rosters, wall calendars, AI class handouts) and built-environment elements (rooms chosen for lighting and quietness, doors closed to block construction noise, curtains drawn to manage glare, landlines used as backup clocks) were as consequential as hardware and software for successful participation. Weather, pets, and family members also appeared as objects shaping videoconferencing (e.g., a cat removed from the room to reduce distraction). Three cognitive-support mechanisms recurred. (1) Objects hold information: Phoenix kept a printed attendee roster at arm's reach to overcome name-recall difficulties while teaching language classes; Echo placed a physical calendar beside her videoconferencing spot to track dates; Rowan relied on a landline's built-in clock when the videoconferencing software hid her e-calendar. (2) Objects distribute effort across time: Kai wrote videoconference notes early in the day when her cognition was sharpest and returned to them later; Skyler annotated unfamiliar concepts during the call to process them afterwards; Taylor maintained an 'AI bag' prepared the night before. (3) Objects distribute effort across space via three affordances — grabbable (items within arm's reach, like glasses, phones, exercise gear), glanceable (sticky notes on monitor edges, wall calendars in peripheral view), and switchable (windows, lights, TVs, hearing aids toggled to regulate sensory load). The authors also surface a methodological contribution: deficit-framed codes ('difficulties recalling information') were revised toward strength-based codes that foregrounded participants' successful workarounds.
Relevance
For accessibility researchers and AT designers working with older adults or cognitive-accessibility populations, this paper is a strong argument that 'one-touch' or 'simplified' videoconferencing interfaces miss the bulk of the actual work. The typology and three affordance categories (grabbable, glanceable, switchable) are directly usable heuristics for auditing videoconferencing tools and wider digital systems: does the tool let critical reference materials stay glanceable, or does it force fullscreen takeover that hides the user's external memory? Does it accommodate paper-based workflows alongside the screen? Does it tolerate 'distribution across time' by letting users prepare notes before and revisit them after? The discussion also introduces the distributed-cognition lens as an alternative to deficit-framed dementia/MCI research, which aligns with the social model of disability by treating the environment — not the individual body — as the primary unit of analysis. Limitations worth noting: the sample is ten people in a single US region, social partners declined to participate so dyadic coordination is absent, affordances were inferred rather than tested multi-user, and accessibility for blind users is explicitly not addressed (the 'glanceable' category assumes vision). Despite these boundaries, this is a rare empirical reframing that practitioners can cite when arguing for wider material and social considerations in cognitive-accessibility design.
Tags: older adults · aging · cognitive accessibility · cognitive concerns · subjective cognitive decline · mild cognitive impairment · videoconferencing · distributed cognition · diary study · reflexive thematic analysis · aging in place · design for aging