Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2021 Nov 8;24(12):1772–1779. doi: 10.1038/s41593-021-00947-w

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2021, corrected publication 2023

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

PMC Copyright notice

Fig. 2 — a, Single-participant examples of successful gaze decoding for three viewing behaviors. b, Predicted error (PE) correlates with the Euclidean error (EE) between real and predicted gaze positions. This allows for filtering of the test set after decoding based on estimated reliability. Single-participant data with a regression line are plotted. Participants were split into 80% most reliable (low PE, blue) and 20% least reliable participants (high PE, orange). Scores are normalized for visualization. c, Group results. Top, gaze decoding expressed as the Pearson correlation between true and decoded gaze trajectory for the five key datasets featuring fixations, 3× smooth pursuit and visual search. Participants are color coded according to PE. Whisker box plots for low-PE participants (central line, median; box, 25th and 75th percentile; whiskers, all data points not considered outliers; outliers, data points outside 1.5× interquartile range) and single-participant data (blue and orange dots) for all are plotted. Bottom, time-collapsed group average histograms of decoded positions relative to the true positions (0, 0) in visual degrees. The color depicts decoding probability (black represents high). d, Test error as a function of how many participants were used for model training. e, Gaze decoding from the eyeballs and early visual cortex for time-shifted gaze labels. f, Subimaging temporal resolution. The model performance (explained variance normalized for each participant) depending on how many subimaging samples were decoded is plotted. The data in d–f show group-level mean ± s.e.m. (black lines) as well as single-participant data (blue dots) of the results obtained for the visual search dataset 5. Source data are provided.

Source data