In partially trained DNN+HMM models simulating children, eye movement consistency (lower entropy) predicts recognition accuracy but eye movement pattern does not¶
Claim · All claims
Evidence · 1 study (2 entries) · associationalr2·q2
Currency · newest study 2022 · not yet checked against its sources · not reviewed by a person
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? In partially trained (epoch 100) models, lower HMM entropy was negatively correlated with recognition accuracy. → Hsiao 2022
q2 i? In partially trained models, the eyes- vs nose-focused patterns did not differ in validation accuracy and AB scale was not correlated with accuracy. → Hsiao 2022 (2)
Evidence¶
Hsiao 2022¶
Hsiao, J. H., An, J., Hui, V. K. S., Zheng, Y. & Chan, A. B. (2022). Understanding the role of eye movement consistency in face recognition and autism through integrating deep neural networks and hidden Markov models. npj Science of Learning 7:28. https://doi.org/10.1038/s41539-022-00139-6
q2 · i? · associational · r2
Simulation result: among 80 partially trained DNN+HMM models, entropy (a consistency measure) correlated negatively with accuracy; the article prints "p < 0.001 R2 = 0.205" but no standardised effect size.
"The entropy of partially trained models was negatively correlated with accuracy, p < 0.001 R2 = 0.205"
Hsiao 2022 (2)¶
Hsiao, J. H., An, J., Hui, V. K. S., Zheng, Y. & Chan, A. B. (2022). Understanding the role of eye movement consistency in face recognition and autism through integrating deep neural networks and hidden Markov models. npj Science of Learning 7:28. https://doi.org/10.1038/s41539-022-00139-6
q2 · i? · associational · r2
Simulation: at epoch 100, eyes- and nose-focused representative patterns showed no accuracy difference (d = 0.16, MA = 0.464 vs MB = 0.457); equivalence was not tested. The article also reports "the AB scale was not correlated with accuracy".
"the patterns A and B of partially trained models did not differ in accuracy on the validation data, t(78) = 0.72, p = 0.475, d = 0.16, MA = 0.464, MB = 0.457"
Discussion¶
Related Claims¶
- In typically developing children aged 6–11, face recognition performance is predicted by eye movement consistency but not by eye movement pattern — related
- Children's eye movement consistency is associated with selective attention and executive function, whereas eye movement pattern relates to working memory — related
- A less eyes-focused pattern and lower eye movement consistency jointly predict poorer recognition beyond age and cognitive abilities — related
- A more eyes-focused scanning pattern and higher eye movement consistency are each correlated with better face recognition performance — related
- In adults, face recognition performance is predicted by eye movement pattern, with eye movement consistency adding no additional variance — related
- In well-trained DNN+HMM face-recognition models, an eyes-focused eye movement pattern yields higher recognition accuracy than a nose-focused pattern — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
partially-trained-model-consistency-predicts-accuracy |
title |
In partially trained DNN+HMM models simulating children, eye movement consistency (lower entropy) predicts recognition accuracy but eye movement pattern does not |
description |
In partially trained DNN+HMM models simulating children, eye movement consistency (lower entropy) predicts recognition accuracy but eye movement pattern does not |
status |
review |
evidence_strength |
weak |
newest_evidence |
2022 |
generated |
by process:wiki-ingest on 2026-10-26 |
trust |
unverified — no one has confirmed this page's content |
sources |
2 — hsiao-2022, hsiao-2022-2 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.