Perceiving animacy in ‘identical’ images
- Johns Hopkins University, United States
Figures
Figure 1
Visual anagrams for the study of perceived animacy.
(Top row) This dog and boot differ in animacy, but they also differ in many lower-level features. (Bottom row) This dog and boot are visual anagrams: They differ in animacy while preserving nearly all lower-level features because they are the very same image rotated 90°.
Figure 2
Animacy itself guides visual search and visual working memory.
(A) In a visual working memory task, changes that altered animacy (e.g. rabbit→boot) were more detectable than changes that did not (e.g. rabbit→dog), even though animacy was task-irrelevant. (B) In a visual search task (modeled after Long et al., 2017), participants were faster to find targets in mixed-animacy arrays than in uniform-animacy arrays. (C) This search effect also arose in a present/absent search task. Error bars represent standard errors of the differences across conditions. Post-exclusion sample sizes for each graph: Experiment 1, n = 28; Experiment 2, n = 28; Experiment 3, n = 29; Experiment 4, n = 30; Experiment 5, n = 30; Experiment 6, n = 29. Readers may view our tasks at https://perceptionresearch.org/anagrams_animacy.
Figure 3
Perceived animacy goes beyond orientation and semantics.
(A) To control for orientation (perhaps the only feature left uncontrolled by visual anagrams), we replicated the present/absent search task with silhouetted versions of the anagrams (within-subjects). While anagrams again showed a search advantage, silhouettes did not, suggesting that orientation differences cannot account for our results. (B) To control for linguistic labeling and/or effects of semantics, we replicated the present/absent search task once again with linguistic labels for the anagrams (within-subjects). We observed a significant search advantage for the anagrams, but not for the words, meaning that linguistic labeling cannot account for our results. Error bars represent standard errors of the differences across conditions. Post-exclusion sample sizes for each graph: Experiment 7, n = 145; Experiment 8, n = 30.
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Perceiving animacy in ‘identical’ images
eLife 15:RP111179.
https://doi.org/10.7554/eLife.111179.3
