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

. 2020 Sep 2;20(9):2. doi: 10.1167/jov.20.9.2

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

Copyright 2020 The Authors

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

PMC Copyright notice

Figure 3. — Relationships among fixation map consistency, fixation count, and scene memorability. (a) The Edinburgh results. Scene memorability (“recog accuracy”) was obtained from G1. Fixation map (“fixmap”) consistency and fixation counts were obtained from G2. Raw values were plotted in the scatterplots. The filled square and triangles indicate the scenes presented in Figure 1. The solid line represents a significant correlation, ⍴(130) = 0.23; 95% CI, 0.09–0.36; p = 0.007. The dashed line represents a nonsignificant correlation, ⍴(130) = 0.11; 95% CI, –0.03 to 0.25; p = 0.234, The gray shades represent the 95% confidence bands. (b) Relationship between fixation map consistency and fixation counts in the Edinburgh dataset. The dashed line represents a nonsignificant correlation, ⍴(130) = 0.08; 95% CI, –0.07 to 0.22; p = 0.365. (c) Explained variance of the linear regression models for predicting scene memorability in the Edinburgh dataset. Models E_Fcnt and E_FMC used z-scored fixation counts and fixation map consistency as the predictor, respectively. Model E_Both used both z-scored variables as the predictor. (d) The FIGRIM results. Scene memorability was obtained from AMT participants. Fixation map consistency and fixation counts were obtained from the lab participants. The solid lines represent significant correlations. For recognition accuracy and fixation map consistency, ⍴(628) = 0.21; 95% CI, 0.15–0.27, and for recognition accuracy and fixation count, ⍴(628) = 0.18; 95% CI, 0.12–0.25 (both p < 0.001). (e) Relationship between fixation map consistency and fixation counts in the FIGRIM dataset. The solid line represents a significant correlation, ⍴(628) = –0.13; 95% CI, –0.2 to –0.07; p < 0.001. (f) Explained variance of the linear regression models for predicting scene memorability in FIGRIM dataset. Models F_Fcnt and F_FMC used z-scored fixation counts and fixation map consistency as the predictor, respectively. Model F_Both used both z-scored variables as the predictor.