Back

How we draw and recognize things that don't exist

A-Izzeddin, E. J.; Schmidt, F.; Houborg, C.; Tiedemann, H.; Fleming, R. W.

2026-04-23 neuroscience
10.64898/2026.04.22.720107 bioRxiv
Show abstract

How do we make sense of something weve never seen before? Classifying objects into superordinate classes like animal is a key step in interpreting novel experiences, but is challenging because radically different items (e.g., octopus, rabbit) must somehow be grouped together. In general, no single feature is shared by all members. We reasoned that to classify or imagine novel items from outside the distribution of previous experiences, observers parse objects into meaningful component features that they can mentally recombine ( compositionality). To test this, we asked participants to draw familiar and novel members of nine superordinate object classes. We then asked other participants to classify the drawings, and mark and label their defining parts. We find that human classification performance is well predicted by a Bayesian classifier that optimally combines the part labels, suggesting humans can create and classify out-of-distribution experiences through a compositional generative representation of object features.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

1
Proceedings of the National Academy of Sciences
2130 papers in training set
Top 3%
14.3%
2
Nature Communications
4913 papers in training set
Top 20%
9.8%
3
Nature Human Behaviour
85 papers in training set
Top 0.2%
8.9%
4
eLife
5422 papers in training set
Top 12%
6.6%
5
Neuron
282 papers in training set
Top 3%
6.2%
6
PLOS Computational Biology
1633 papers in training set
Top 6%
6.2%
50% of probability mass above
7
Science
429 papers in training set
Top 5%
6.2%
8
Current Biology
596 papers in training set
Top 4%
4.7%
9
Nature
575 papers in training set
Top 6%
4.7%
10
Nature Neuroscience
216 papers in training set
Top 2%
3.9%
11
The Journal of Neuroscience
928 papers in training set
Top 4%
3.6%
12
Communications Psychology
20 papers in training set
Top 0.1%
2.7%
13
Cognition
44 papers in training set
Top 0.2%
2.0%
14
eneuro
389 papers in training set
Top 4%
2.0%
15
Science Advances
1098 papers in training set
Top 13%
2.0%
16
Philosophical Transactions of the Royal Society B
51 papers in training set
Top 2%
2.0%
17
Scientific Reports
3102 papers in training set
Top 60%
1.7%
18
Psychological Science
14 papers in training set
Top 0.1%
1.6%
19
Cell
370 papers in training set
Top 13%
1.4%
20
iScience
1063 papers in training set
Top 27%
0.9%
21
Cell Reports
1338 papers in training set
Top 34%
0.7%
22
PLOS ONE
4510 papers in training set
Top 69%
0.7%
23
PLOS Biology
408 papers in training set
Top 22%
0.7%