Idiosyncratic learning performance in flies generalizes across modalities

M. A. Smith, Kyle S. Honegger, Glenn Turner, Benjamin de BivortbioRxiv 2021 · DOI

Abstract

Individuals vary in their innate behaviors, even when they share the same genome and have been reared in identical environments. While this individuality is well-documented in innate behaviors, its extent in plastic behaviors like learning remains less characterized. We investigated this using the fruit fly Drosophila melanogaster, a model organism for studying the mechanistic basis of memory. We found that isogenic flies exhibit clear individuality in their learning responses. Crucially, flies that performed well in one associative task tended to perform well in others, even when the odorants or the reinforcement modalities (e.g., electric shock vs. bitter taste) were changed. These results suggest that individual differences in learning performance are a prominent feature of isogenic animals and that these differences generalize across stimulus modalities, likely originating from central brain processes.

Introduction

Biologists have long recognized that even genetically identical (isogenic) animals raised in the same laboratory environment show "personality"—idiosyncratic biases in innate behaviors such as locomotor patterns or light responses. However, it was previously unclear if this individuality extends to plastic behaviors like learning. This study addresses two primary questions: Do identical flies vary in their learning performance, and if so, is this variation a general learning trait or merely task-specific noise?

Methods

To move beyond the "bulk" testing typical in Drosophila research, the researchers developed a custom high-throughput behavioral instrument designed for individualized tracking.

Schematic of the odor conditioning experimental apparatus, featuring 15 linear tunnels for individualized tracking.

Experimental Design

  • Individualized Tracking: The apparatus utilizes 15 linear tunnels, allowing the precise movement and learning curves of individual flies to be tracked simultaneously.
  • Classical Conditioning: A standard aversive conditioning protocol was used, where an odor (Conditioned Stimulus, CS) is paired with a negative reinforcement (Unconditioned Stimulus, US), such as an 80V electric shock.
  • Cross-Modality Testing: To determine if learning ability generalizes, the same flies were tested across different odors (e.g., octanol vs. methylcyclohexanol) and different reinforcements.
  • Optogenetics: In a novel approach, physical electric shocks were replaced with the optogenetic activation of bitter-taste neurons (Gr66a). This allowed researchers to test if a fly’s learning ability was tied to its physical sensitivity to shock or to a more central associative capability.

Key Insights and Results

The "General Learning" Factor

The most striking finding is that learning performance correlates across tasks. A fly that excels at "Forward Conditioning" (pairing Odor A with a shock) is significantly more likely to excel at "Reversal Learning" (pairing Odor B with a shock). These individual differences were found to be stable over at least 24 hours.

Comprehensive summary of experimental data, including the training protocol, individual kymographs, and the correlation of learning responses across different conditioning trials.

Beyond Sensory Periphery

Because performance correlated even when the reinforcement changed from electric shock to optogenetic bitter-taste stimulation ($r = 0.45, p = 0.01$), the researchers concluded that the source of individuality does not reside in the sensory organs (the "nose" or the "skin"). Instead, it likely resides in higher-order brain centers like the Mushroom Body, where associations are formed. Furthermore, the researchers ruled out general activity levels (distance traveled) as a factor, confirming that the variation was specific to learning performance rather than "laziness."

Key Results Summary

  • Correlation across odors: Learning responses for different odor pairs showed significant positive correlations (ranging from $r = 0.31$ to $r = 0.59$).
  • Correlation across reinforcements: There was a significant correlation between a fly’s ability to learn using electric shocks and its ability to learn using optogenetic bitter-taste stimulation.
  • Developmental Noise: Since the flies are genetically identical and reared in controlled environments, this individuality likely arises from "developmental noise"—stochastic variations in how brain circuits wire up during growth.

Discussion and Future Work

This study establishes that individuality in learning is a robust trait in Drosophila. While the paper points toward the Mushroom Body as the likely site of this individuality, the specific neurons or synaptic variations responsible have yet to be identified. Future work could explore whether these "smart" flies also excel in appetitive tasks (e.g., seeking sugar rewards) or in other sensory domains like visual or mechanosensory learning. Establishing these results in flies opens the door to studying the genetic and neural circuit basis of individual differences in learning in a highly tractable model organism.

Summary updated

Abstract

Abstract Individuals vary in their innate behaviors, even when they have the same genome and have been reared in the same environment. The extent of individuality in plastic behaviors, like learning, is less well characterized. Also unknown is the extent to which intragenotypic differences in learning generalize: if an individual performs well in one assay, will it perform well in other assays? We investigated this using the fruit fly Drosophila melanogaster , an organism long-used to study the mechanistic basis of learning and memory. We found that isogenic flies, reared in identical lab conditions, and subject to classical conditioning that associated odorants with electric shock, exhibit clear individuality in their learning responses. Flies that performed well when an odor was paired with shock tended to perform well when other odors were paired with shock, or when the original odor was paired with bitter taste. Thus, individuality in learning performance appears to be prominent in isogenic animals reared identically, and individual differences in learning performance generalize across stimulus modalities. Establishing these results in flies opens up the possibility of studying the genetic and neural circuit basis of individual differences in learning in a highly suitable model organism.
BibTeX
@misc{smith2021idiosyncratic,
  author = {M. A. Smith and Kyle S. Honegger and Glenn Turner and Benjamin de Bivort},
  title = {{Idiosyncratic learning performance in flies generalizes across modalities}},
  year = {2021},
  doi = {10.1101/2021.01.23.427920},
  howpublished = {bioRxiv},
}
Figure