Testing the equivalency of human “predators” and deep neural networks in the detection of cryptic moths

Researchers have shown growing interest in using deep neural networks (DNNs) to efficiently test the effects of perceptual processes on the evolution of color patterns and morphologies. Whether this is a valid approach remains unclear, as it is unknown whether the relative detectability of ecologically relevant stimuli to DNNs actually matches that of biological neural networks. To test this, we compare image classification performance by humans and six DNNs (AlexNet, VGG-16, VGG-19, ResNet-18, SqueezeNet, and GoogLeNet) trained to detect artificial moths on tree trunks. Moths varied in their degree of crypsis, conferred by different sizes and spatial configurations of transparent wing elements. Like humans, four of six DNN architectures found moths with larger transparent elements harder to detect. However, humans and only one DNN architecture (GoogLeNet) found moths with transparent elements touching one side of the moth’s outline harder to detect than moths with untouched outlines. W..., , , # Testing the equivalency of human “predators” and deep neural networks in the detection of cryptic moths

[https://doi.org/10.5061/dryad.w0vt4b92k](https://doi.org/10.5061/dryad.w0vt4b92k)

## Description of the data and file structure

This data was collected for the paper entitled \"Testing the equivalency of human “predators” and deep neural networks in the detection of cryptic moths.\"

### Files and variables

#### File: Data\_\_\_Code.zip

**Description:** This archive contains:

1\) Folder entitled \"224x224 images,\" which contains images of moths for DNNs requiring images of this resolution. Each subfolder contains images of a given moth morph. Abbreviations should be interpreted thusly: O=opaque morph, LW=morph with large windows, SW=morph with small windows, BE=morph with large windows touching bottom edges of wings, B3E=morph with large windows touching all three wing edges. Image names are the default names assigned by the camera that took the images.

2\) Folder entitled \"227...