Identity Domains Reveal How Life Experiences and the Microbiome Shape Individuality and Uncover the Presence of Social Memory in Drosophila
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Personality is comprised of enduring traits that shape behavior across contexts and over time. Yet outside humans, research often equates individuality with personality, focusing on how one animal differs from another in each behavior. This overlooks the latent space that spans an animal’s behavioral repertoire, how it is shaped by innate tendencies and prior experiences, and how it differs from transient factors. Here, we bridge this gap by integrating a data-driven framework with unbiased trait interpretation via large language models, revealing how social, sexual, and physiological conditions systematically alter the expression of personality-like traits in Drosophila melanogaster . This approach yielded four identities: space-use (exploration), time-use (social investment), avoidance, and aggression. Large language models produced strong consensus on the interpretation of some of these axes and divergence on others. These dimensions proved stable within individuals yet shifted predictably following mating, isolation, and microbiome manipulation, often in sex dependent ways. Notably, flies behaved differently toward familiar and unfamiliar peers, revealing a capacity for social memory. These findings bridge temperament and personality, showing how life history reshapes baseline tendencies into individualized profiles within constrained dimensions. They establish Drosophila as a tractable system for uncovering the biological logic of personality and advancing cross-species principles for individuality research.
Significance Statement
Personality is often treated as a hallmark of complex brains, yet we show that fruit flies express stable behavioral identities with systematic, experience driven plasticity. Using a mathematical framework, we resolve four stable traits and map how sex, social experience, and the gut microbiome shift their expression. We further demonstrate social recognition in flies: they treat familiar cage mates differently from newcomers over repeated encounters, evidencing social memory, a striking result given their simple nervous system. Finally, we introduce large language models to provide objective, reproducible interpretations of trait dimensions, reducing human bias. These advances establish a general, quantitative approach to personality and social behavior in a tractable genetic model.
