The Wordometer eye-tracking app estimates the number of words a person reads with an average error rate of 6.5 percent¶
Claim · All claims
Evidence · 1 study · designr2·q2
Currency · newest study 2013 · not yet checked against its sources · not reviewed by a person · overdue for a recheck since 2016: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? A support vector regression algorithm using detected line breaks and eye-gaze features estimates words read with 6.5 percent average error for 8 users over 10 documents. → Kai Kunze 2013
Evidence¶
Kai Kunze 2013¶
Kai Kunze, Masakazu Iwamura, Koichi Kise, Seiichi Uchida, and Shinichiro Omachi. (2013). Activity Recognition for the Mind: Toward a Cognitive “Quantified Self.” Computer, IEEE Computer Society, October 2013. http://kaikunze.de/papers/2013Kunze-1.pdf
q2 · i? · design · r2
Prototype evaluation of the Wordometer app on the SMI iView X eye tracker: eye movements of users reading English documents were recorded, a line-break detector looked for long saccades against the reading direction, and support vector regression estimated word counts. The article reports an "average error rate of only 6.5 percent for 8 users over 10 documents"; no effect size is printed.
"Our word-count estimation method achieves an average error rate of only 6.5 percent for 8 users over 10 documents."
Discussion¶
Related Claims¶
- Mobile eye tracking distinguishes five document types with 99 percent user-dependent and 74 percent user-independent recognition from about one minute of gaze data
- Eye tracking can identify which words a second-language reader had trouble understanding
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
wordometer-word-count-6-5-percent-error |
title |
The Wordometer eye-tracking app estimates the number of words a person reads with an average error rate of 6.5 percent |
description |
The Wordometer eye-tracking app estimates the number of words a person reads with an average error rate of 6.5 percent |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2013 |
recheck_by |
2016 |
generated |
by process:wiki-ingest on 2026-10-16 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — kai-kunze-2013 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.