CoChat

Investigating the Accessibility of Voice Assistants With Impaired Users: Mixed Methods Study

Open access

Published version of a preprint

The DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index; 7 of the 36 sources its claims cite were checked for support.

Abstract

BACKGROUND: Voice assistants allow users to control appliances and functions of a smart home by simply uttering a few words. Such systems hold the potential to significantly help users with motor and cognitive disabilities who currently depend on their caregiver even for basic needs (eg, opening a door). The research on voice assistants is mainly dedicated to able-bodied users, and studies evaluating the accessibility of such systems are still sparse and fail to account for the participants' actual motor, linguistic, and cognitive abilities. OBJECTIVE: The aim of this work is to investigate whether cognitive and/or linguistic functions could predict user performance in operating an off-the-shelf voice assistant (Google Home). METHODS: A group of users with disabilities (n=16) was invited to a living laboratory and asked to interact with the system. Besides collecting data on their performance and experience with the system, their cognitive and linguistic skills were assessed using standardized inventories. The identification of predictors (cognitive and/or linguistic) capable of accounting for an efficient interaction with the voice assistant was investigated by performing multiple linear regression models. The best model was identified by adopting a selection strategy based on the Akaike information criterion (AIC). RESULTS: For users with disabilities, the effectiveness of interacting with a voice assistant is predicted by the Mini-Mental State Examination (MMSE) and the Robertson Dysarthria Profile (specifically, the ability to repeat sentences), as the best model shows (AIC=130.11). CONCLUSIONS: Users with motor, linguistic, and cognitive impairments can effectively interact with voice assistants, given specific levels of residual cognitive and linguistic skills. More specifically, our paper advances practical indicators to predict the level of accessibility of speech-based interactive systems. Finally, accessibility design guidelines are introduced based on the performance results observed in users with disabilities.

Overview

CoChat Summary

TL;DR

In a mixed-methods study with 16 users with motor, linguistic, and cognitive disabilities, performance with Google Home was about 58.5% accurate on average and was best predicted by MMSE and sentence repetition ability. The paper argues that commercial voice assistants can be usable for some disabled users if cognitive and linguistic abilities are sufficient, and it proposes design implications for accessibility.

Key findings

  • Participant accuracy in interacting with the voice assistant was 58.5% (SD 18.6%) on average.
  • The most frequent errors were phrasing errors (75/182, 41.2%) and timing errors (74/182, 40.7%); less frequent were comprehension errors (19/182, 10.4%) and pronunciation errors (14/182, 7.7%).
  • In the final regression model, MMSE and repetition of sentences significantly predicted accuracy: MMSE (β=3.70, t13=3.26, P=.006) and repetition of sentences (β=22.06, t13=4.16, P=.001); the model had AIC=130.11 and R2=.77.
  • Participants’ overall evaluation of the system was highly positive, with a mean score of 9.4 (SD 1.2), and all participants said they would like a voice assistant in their own home.

Methods

  • Design: Two-phase mixed-methods study in a living laboratory: phase 1 involved group sessions where participants performed voice-assistant tasks; phase 2 involved neuropsychological and linguistic assessment. Sessions were video recorded and later coded.
  • Sample/data: 16 participants with ascertained motor impairments and related language difficulties who needed daily caregiver assistance; mean age 38.3 years and mean education 11.8 years.
  • Analysis: Video-coded task attempts were summarized into a performance index, then related to cognitive/linguistic measures using multiple linear regression with AIC-based model selection.

Limitations

  • The paper states that the sample size was limited to 16 participants, and calls for larger and more heterogeneous samples.
  • The paper states that the study examined a group situation in a living-lab setting; it says future work should test individual use to better examine person-device interaction.
  • Evident from the text: the study used a single commercial voice assistant (Google Home) in a controlled setting, so generalization to other systems and real-world home environments is limited (evident from the setting described).

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.2196/18431

Cites

33 of 33 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Representation of the experimental setting.
Figure 1 Representation of the experimental setting.CC BY · PubMed Central

Verification

Claims the paper makes

4 Valid · 6 Not valid · 12 Not checked · 55 Own results and interpretations

47 more claims · 2 with their cited sources checked

Create a free account to see all 77 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 12
  • It may state two separate claims as one. 17

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Research works printed but not linked: 5

  1. Duffy, J, 2013, Motor speech disorders e-book: substrates, differential diagnosis, and management
  2. Ballati, F, 2018, Intell Environ, 1
  3. Fussi, F, 1999, Profilo di valutazione della disartria: adattamento italiano del test di Robertson, raccolta di dati normativi e linee di trattamento
  4. Boersma, P, 2001, Glot Int, 5, 341
  5. Myers, R, 1990, Classical and Modern Regression With Applications, 2nd Edition

Cited by

See all 107

Metadata

Authors
Masina, F., Orso, V., Pluchino, P., Dainese, G., Volpato, S., Nelini, C., Mapelli, D., Spagnolli, A., & Gamberini, L.
Affiliations
University of Padua (IT)
Corresponding author
Gamberini, Luciano
Venue
Journal of Medical Internet Research
Published
2020-09-25
Volume
22
Issue
9
Pages
e18431
ISSN
1438-8871 (ISSN-L)
OpenAlex
W3045900875
Cited by
105
References
32
Citations by year
2026: 14 · 2025: 18 · 2024: 26 · 2023: 15 · 2022: 19 · 2021: 13
FWCI
20.90
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • Journal of Medical Internet Research (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Padua Research Archive (University of Padova) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
European Regional Development Fund (FESR 2014-2020, 2014-2020’, POR FESR 2014-2020)
Topics
  • Health Sciences › Health Professions › Occupational Therapy › Assistive Technology in Communication and Mobility
  • Social Sciences › Human Factors and Ergonomics › Digital Accessibility for Disabilities
  • Physical Sciences › Computer Science › Human-Computer Interaction › Usability and User Interface Design
Keywords
voice assistants, accessibility, disabled users, motor impairment, cognitive impairment, speech disorders, Google Home, home automation, speech-based interaction, Mini-Mental State Examination, living labs
MeSH terms
Adult, Female, Persons with Disabilities, Humans, Language, Male, Middle Aged, Voice, Young Adult, Cognitive Dysfunction

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.