Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 38 of the 38 references it cites were found in a registry or an index; 5 of the 45 sources its claims cite were checked for support.

Eye-tracking reveals agency in assisted autistic communication

Open access
Jaswal, Vikram K., Wayne, Allison, Golino, Hudson · 2020 · Scientific Reports

Abstract

About one-third of autistic people have limited ability to use speech. Some have learned to communicate by pointing to letters of the alphabet. But this method is controversial because it requires the assistance of another person—someone who holds a letterboard in front of users and so could theoretically cue them to point to particular letters. Indeed, some scientists have dismissed the possibility that any nonspeaking autistic person who communicates with assistance could be conveying their own thoughts. In the study reported here, we used head-mounted eye-tracking to investigate communicative agency in a sample of nine nonspeaking autistic letterboard users. We measured the speed and accuracy with which they looked at and pointed to letters as they responded to novel questions. Participants pointed to about one letter per second, rarely made spelling errors, and visually fixated most letters about half a second before pointing to them. Additionally, their response times reflected planning and production processes characteristic of fluent spelling in non-autistic typists. These findings render a cueing account of participants’ performance unlikely: The speed, accuracy, timing, and visual fixation patterns suggest that participants pointed to letters they selected themselves, not letters they were directed to by the assistant. The blanket dismissal of assisted autistic communication is therefore unwarranted.

Overview

CoChat Summary

TL;DR

Using head-mounted eye-tracking during a familiar letterboard spelling task, the authors found that nine experienced nonspeaking autistic users looked at and pointed to letters quickly, accurately, and in patterns consistent with self-generated spelling rather than assistant cueing. The study argues that these in situ communication behaviors cast doubt on the idea that assisted autistic communication is necessarily driven by subtle cues from an assistant.

Key findings

  • Across the nine participants, most words were spelled correctly: mean 83% correct per participant, ranging from 71–92%; letter accuracy was mean 94% correct, ranging from 88–97%.
  • Participants spelled fluently: within correctly spelled words, the median inter-point interval was 952 ms; within each participant’s longest string of consecutive correct letters in a single response, it was 1054 ms.
  • Participants showed planning-like timing effects: they were slower to point to the first letter of a new word than to letters within words (b = 0.75, SE = 0.07, t(7.96) = 10.76, p < 0.0001), and faster to point to letters in more frequent English letter pairs than less frequent ones (β = −0.18, SE = 0.02, t(8.66) = 9.25, p < 0.0001).
  • Visual behavior was consistent with goal-directed spelling: participants fixated, on average, 71% of correct letters before pointing to them (range 32–99%), and these fixations preceded the point by a median of 476 ms.

Methods

  • Design: in-situ observational study using head-mounted eye-tracking while participants answered questions about a text read aloud by a familiar assistant; responses were composed by independently pointing to letters on a vertically held letterboard.
  • Sample: 9 nonspeaking autistic young adults, average age 20.04 years (range 14.67–26.25), with at least 2.25 years of letterboard experience; all had limited speech and no usable verbal open-ended responses.
  • Analysis: frame-by-frame video coding of pointing and gaze, calculation of word/letter accuracy and inter-point intervals, and linear mixed-effects models for timing effects; anticipatory fixations were evaluated with participant-specific simulations and 1,000 repeated randomization runs per participant.

Limitations

  • The authors state that this was a unique sample of experienced letterboard users and that they are not claiming all nonspeaking autistic people can learn to convey their thoughts using a letterboard.
  • They also state they are not claiming that someone who spells slowly or whose eyes cannot be tracked is incapable of conveying their own thoughts, and that additional research is needed to develop other methods.
  • Evident from the text: the study was conducted with a small sample (9 participants) at one center with one assistant, so generalizability is limited.

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

Integrity signals

Registry

Crossref · 10.1038/s41598-020-64553-9

Cites

38 of 38 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Quantitative characterization of in situ letterboard use. Participants wore eye-tracking glasses that provided a video record of their field of view and their right eye’s movements. Shown here are frames from Participant 2’s processed video (produced using Yarbus software, version 2.5.0, www.positivescience.com) as he pointed to the first two letters in the word “like” in the response available in Supplementary Video 1. The red circle shows where the eye-tracker estimated he was looking; the circle was not visible during his session (see Methods). The left frame shows the end of his point to the letter “L”; the position of the red circle indicates he was also looking at “L” at that time. The middle frame shows that 374 ms later, he began fixating the letter “I” as his finger was moving toward it. The right frame shows that 510 ms after that, the participant began pointing to the letter “I” (and he continued to fixate the letter “I”). The time between the end of the point to the letter “L” and the beginning of the point to the letter “I” (the inter-point interval) was 884 ms. Ellipses represent frames not shown in the figure.
Figure 1 Quantitative characterization of in situ letterboard use. Participants wore eye-tracking glasses that provided a video record of their field of view and their right eye’s movements. Shown here are frames from Participant 2’s processed video (produced using Yarbus software, version 2.5.0, www.positivescience.com) as he pointed to the first two letters in the word “like” in the response available in Supplementary Video 1. The red circle shows where the eye-tracker estimated he was looking; the circle was not visible during his session (see Methods). The left frame shows the end of his point to the letter “L”; the position of the red circle indicates he was also looking at “L” at that time. The middle frame shows that 374 ms later, he began fixating the letter “I” as his finger was moving toward it. The right frame shows that 510 ms after that, the participant began pointing to the letter “I” (and he continued to fixate the letter “I”). The time between the end of the point to the letter “L” and the beginning of the point to the letter “I” (the inter-point interval) was 884 ms. Ellipses represent frames not shown in the figure.CC BY · PubMed Central
Figure 2 Inter-point interval (IPI). (a) IPI between letters within correctly spelled words. (6 observations of 1976 total were slower than 4250 ms and are not shown.) (b) IPI between letters in each participant’s longest sequence of points to correct letters within a response. (1 observation of 252 total was slower than 4250 ms and is not shown.) Each dot represents an individual datapoint. Yellow lines show the medians, and red lines show the means.
Figure 2 Inter-point interval (IPI). (a) IPI between letters within correctly spelled words. (6 observations of 1976 total were slower than 4250 ms and are not shown.) (b) IPI between letters in each participant’s longest sequence of points to correct letters within a response. (1 observation of 252 total was slower than 4250 ms and is not shown.) Each dot represents an individual datapoint. Yellow lines show the medians, and red lines show the means.CC BY · PubMed Central
Figure 3 Inter-point interval (IPI) patterns. (a) Time between end of a point to correct lettern-1 and point to correct lettern as a function of whether the transition from lettern-1 to lettern crossed a word boundary (light blue) or was within a word (dark blue). Each dot represents an individual datapoint. Yellow lines show the medians, and red lines show the means. (7 of 350 between-word observations and 6 of 2417 within-word observations were slower than 4250 ms and are not shown.) (b) Individual model fits for each participant predicting IPI between consecutive points to correct letters within words, using as predictors the distance between those letters on the letterboard and how frequently they occur consecutively in English. Bigram frequency is shown as a median split but was a continuous variable in the linear mixed-effects analysis reported in the text. Shading represents 95% confidence intervals.
Figure 3 Inter-point interval (IPI) patterns. (a) Time between end of a point to correct lettern-1 and point to correct lettern as a function of whether the transition from lettern-1 to lettern crossed a word boundary (light blue) or was within a word (dark blue). Each dot represents an individual datapoint. Yellow lines show the medians, and red lines show the means. (7 of 350 between-word observations and 6 of 2417 within-word observations were slower than 4250 ms and are not shown.) (b) Individual model fits for each participant predicting IPI between consecutive points to correct letters within words, using as predictors the distance between those letters on the letterboard and how frequently they occur consecutively in English. Bigram frequency is shown as a median split but was a continuous variable in the linear mixed-effects analysis reported in the text. Shading represents 95% confidence intervals.CC BY · PubMed Central
Figure 4 Anticipatory fixations of correct letters. (a) For fixated letters, time between the beginning of the first fixation of lettern and the point to lettern. Minimum fixation length was 99 ms. (24 observations out of 2192 total were slower than 2500 ms and are not shown.) (b) Observed percentage of points to correct letters preceded by a fixation of that letter (red lines) vs. simulated percentage of points to correct letters preceded by a fixation of that letter if fixations had been random (blue boxplots representing 1000 simulations). (c) Time between the end of the point to lettern-1 and the beginning of the first fixation of lettern. (30 observations out of 2192 total were slower than 2500 ms and are not shown.) For violin plots, each dot represents an individual datapoint. Dots at or below 0 represent occasions where a participant’s fixation of lettern began before or during the point to lettern-1 and continued after the point to lettern-1 had ended. Yellow lines show the medians, and red lines show the means.
Figure 4 Anticipatory fixations of correct letters. (a) For fixated letters, time between the beginning of the first fixation of lettern and the point to lettern. Minimum fixation length was 99 ms. (24 observations out of 2192 total were slower than 2500 ms and are not shown.) (b) Observed percentage of points to correct letters preceded by a fixation of that letter (red lines) vs. simulated percentage of points to correct letters preceded by a fixation of that letter if fixations had been random (blue boxplots representing 1000 simulations). (c) Time between the end of the point to lettern-1 and the beginning of the first fixation of lettern. (30 observations out of 2192 total were slower than 2500 ms and are not shown.) For violin plots, each dot represents an individual datapoint. Dots at or below 0 represent occasions where a participant’s fixation of lettern began before or during the point to lettern-1 and continued after the point to lettern-1 had ended. Yellow lines show the medians, and red lines show the means.CC BY · PubMed Central
Figure 5 Timing of visual fixations and points in Participant 2’s response shown in Supplementary Video 1. Question: Can you think of something you have to wait for? Answer: “That is hard. (reset) I feel like world is waiting on me (done) (reset) not the other way around (done)”. Purple shading shows the time between the end of the point to lettern-1 (or when letterboard was placed) and the first fixation of lettern; yellow shading shows the time between the beginning of the first fixation of lettern and the beginning of the point to lettern. For letters without a shaded column, coders did not code an anticipatory fixation. The lower red line shows when the participant’s pointing finger made contact with lettern; the upper red line shows when his finger stopped being in contact with lettern (i.e., the end of the point to lettern). The red shading shows how long the participant’s finger remained in contact with lettern.
Figure 5 Timing of visual fixations and points in Participant 2’s response shown in Supplementary Video 1. Question: Can you think of something you have to wait for? Answer: “That is hard. (reset) I feel like world is waiting on me (done) (reset) not the other way around (done)”. Purple shading shows the time between the end of the point to lettern-1 (or when letterboard was placed) and the first fixation of lettern; yellow shading shows the time between the beginning of the first fixation of lettern and the beginning of the point to lettern. For letters without a shaded column, coders did not code an anticipatory fixation. The lower red line shows when the participant’s pointing finger made contact with lettern; the upper red line shows when his finger stopped being in contact with lettern (i.e., the end of the point to lettern). The red shading shows how long the participant’s finger remained in contact with lettern.CC BY · PubMed Central

Verification

Claims the paper makes

4 Valid · 0 Not valid · 27 Not checked · 53 Own results and interpretations

41 more claims

Create a free account to see all 84 claims and to verify the 40 references still open.

See all claims — free account
  • Probably not a substantial claim. 29
  • It may state two separate claims as one. 17
  • The quote only partly states this claim. 1

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?

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Consistentt(7.96) = 10.76reported p < .0001 · recomputed p < .001

    “t (7.96) = 10.76, p < .0001”

  • Consistentt(8.66) = 9.25reported p < .0001 · recomputed p < .001

    “t (8.66) = 9.25, p < .0001”

  • Consistentt(8.69) = 4.36reported p = .002 · recomputed p = .002

    “t (8.69) = 4.36, p = .0020”

  • Consistentt(8.71) = 5.13reported p = .0007 · recomputed p < .001

    “t (8.71) = 5.13, p = .0007”

4 of 4

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Self-check results appear here once an independent check has confirmed them.

Citations

Cites

Show all 38 references

Research works printed but not linked: 6

  1. Peña, E. V. (Ed.). Leaders Around Me: Autobiographies of Autistics who Type, Point, and Spell to Communicate (E. V. Peña, 2019).
  2. Savarese, D. J. Passive plants. The Iowa Review 47, 34–37 (2017).
  3. Rogoff, B. The Cultural Nature of Human Development (Oxford Univ. Press, 2003).
  4. Mukhopadhyay, S. Understanding Autism Through Rapid Prompting Method (Outskirts Press, 2008).
  5. Land, M. F. & Tatler, B. W. Looking and Acting: Vision and Eye Movements in Natural Behaviour (Oxford Univ. Press, 2013).
  6. Constantino, J. N. & Gruber, C. P. Social Responsiveness Scale, 2 Edition (SRS-2) (Western Psychological Services, 2012).

Cited by

See all 68

Metadata

Authors
Jaswal, V. K., Wayne, A., & Golino, H.
Affiliations
University of Virginia (US)
Corresponding author
Jaswal, Vikram K.
Venue
Scientific Reports
Published
2020-05-12
Volume
10
Issue
1
Pages
7882-7882
ISSN
2045-2322 (ISSN-L)
OpenAlex
W3024768142
Cited by
68
References
38
Citations by year
2026: 3 · 2025: 25 · 2024: 9 · 2023: 16 · 2022: 6 · 2021: 5 · 2020: 3
FWCI
13.34
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • OSF Preprints (OSF Preprints) (submittedVersion)
Funders
University of Virginia
Topics
  • Health Sciences › Health Professions › Occupational Therapy › Assistive Technology in Communication and Mobility
  • Physical Sciences › Computer Science › Human-Computer Interaction › Gaze Tracking and Assistive Technology
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Autism Spectrum Disorder Research
Keywords
autism spectrum disorder, communicative agency, pointing behavior, visual fixation, reaction time (psychology), spelling accuracy, head-mounted eye tracking
MeSH terms
Adolescent, Adult, Autistic Disorder, Communication, Eye Movements, Female, Fixation, Ocular, Humans, Language, Male, Surveys and Questionnaires, Reaction Time, Speech, Young Adult

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 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-11 (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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Eye-tracking reveals agency in assisted autistic communication. https://app.cochat.ai/papers/10.1038/s41598-020-64553-9
The paper
Jaswal, V. K., Wayne, A., & Golino, H. (2020). Eye-tracking reveals agency in assisted autistic communication. Scientific Reports. https://doi.org/10.1038/s41598-020-64553-9

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.