Back

Art's Hidden Topology: A window into human perception

Dmitriuk, E.; Bajno, B.; Dreszer, J.; Bałaj, B.; Ratajczak, E.; Hajnowski, M.; Janik, R.; Kus, M.; Kadir, S. N.; Rogala, J.

2024-10-18 neuroscience
10.1101/2024.10.16.618741 bioRxiv
Show abstract

Generations of researchers have sought a link between features of an artistic image and the audiences experience. However, a direct link between the properties of an image and the responses evoked has still not been established. Given the importance of shape to human perception and artistic creation, it can be assumed that one of the most important aspects of an artistic image is the use of different visual structures. We show that a method from the field of computational topology, persistent homology, can be used to analyse properties of image structures and composition at multiple scales. In order to determine the reliability of this method as a tool for analysing visual artworks, we analysed two different sets of abstract paintings that revealed significant discrepancies in the eye tracking and electroencephalography (EEG) activity of viewers. Our research showed that our newly developed method using persistent homology, not only clearly distinguished between two sets of images, which was not possible with common statistical image properties, but also allowed us to map topological features onto gaze fixation heat maps.

Published in PLOS Computational Biology (predicted rank #1) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

PLOS Computational Biology · published here
● 1863 papers in training set
Top 0.6%
29.0%
Similar papers in this journal
2
Scientific Reports
○ 3612 papers in training set
Top 4%
9.8%
Similar papers in this journal
3
Journal of Vision
● 110 papers in training set
Top 0.2%
8.0%
Similar papers in this journal
4
PLOS ONE
● 5266 papers in training set
Top 26%
5.6%
Similar papers in this journal
50% of probability mass above
5
Royal Society Open Science
● 214 papers in training set
Top 1.0%
4.1%
Similar papers in this journal
6
Vision Research
○ 29 papers in training set
Top 0.1%
3.3%
7
eLife
● 5828 papers in training set
Top 34%
3.3%
8
iScience
○ 1154 papers in training set
Top 10%
2.5%
9
Nature Communications
○ 5641 papers in training set
Top 41%
2.2%
10
Neuroscience of Consciousness
◐ 16 papers in training set
Top 0.1%
2.2%
11
Computers in Biology and Medicine
○ 128 papers in training set
Top 3%
1.5%
12
eneuro
● 439 papers in training set
Top 5%
1.4%
13
Frontiers in Computational Neuroscience
○ 60 papers in training set
Top 0.9%
1.1%
14
Network Neuroscience
● 126 papers in training set
Top 1%
1.1%
15
NeuroImage
○ 903 papers in training set
Top 5%
1.1%
16
Journal of The Royal Society Interface
● 235 papers in training set
Top 3%
1.1%
17
IEEE Access
● 35 papers in training set
Top 1%
0.9%
18
Communications Biology
○ 993 papers in training set
Top 29%
0.9%
19
Entropy
○ 21 papers in training set
Top 0.3%
0.9%
20
Neural Networks
○ 35 papers in training set
Top 0.6%
0.9%
21
Advanced Science
○ 286 papers in training set
Top 10%
0.6%
22
Peer Community Journal
● 281 papers in training set
Top 6%
0.6%
23
Cognitive Neurodynamics
○ 18 papers in training set
Top 0.4%
0.6%
24
Neural Computation
● 39 papers in training set
Top 0.8%
0.6%
25
Proceedings of the National Academy of Sciences
● 2444 papers in training set
Top 44%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.