Machine learning enables largest study to date of aesthetic preferences and fish ecology

What is the relationship between people’s perception of beauty and the conservation needs of animals? According to a machine learning study by Nicolas Mouquet at the University of Montpellier, France, and his colleagues, published on June 7ein the open access journal PLOS Biology, the reef fish that people find the most beautiful tend to be the lowest priority for conservation support.

The researchers asked 13,000 members of the public to rate the aesthetic appeal of 481 photographs of ray-finned reef fish in an online survey and used that data to train a convolutional neural network. They then used the trained neural network to generate predictions for an additional 4,400 photographs featuring 2,417 of the most commonly encountered reef fish species.

By combining audience ratings with neural network predictions, they found that bright, colorful fish species with rounder bodies tended to be considered the most beautiful. However, the species ranked as the most attractive tended to be less distinctive in terms of ecological characteristics and evolutionary history. In addition, species listed on the IUCN Red List as ‘Threatened’ or whose conservation status has not yet been assessed had on average lower aesthetic value than species categorized as ‘Least Concern’. Unattractive species were also of greater commercial interest, while aesthetic value was not correlated with a species’ importance to subsistence fishing.

Our innate preferences for shape and color are likely a consequence of how the human brain processes color and pattern, say the authors, but mismatches between aesthetic value, ecological function and vulnerability to extinction may mean that the species most in need of public support are least likely to receive it. The ecological and evolutionary specificity of unattractive fish makes them important to the functioning of the entire reef, and their loss could have a disproportionate impact on these highly biodiverse ecosystems.

Mouquet adds: “Our study provides, for the first time, the aesthetic value of 2,417 species of reef fish. We found that the least beautiful fish are the most ecological and evolutionary distinct species and those recognized as endangered. Our study highlights likely large mismatches between potential public support for conservation and the species that need it most.”

Source of the story:

Material provided by OLP. Note: Content may be edited for style and length.

Sherry J. Basler