Abstract
Biogenic amines are fundamental for physiological homeostasis and behavioral control in both vertebrates and invertebrates. Monoamine neurotransmitters released in target brain regions conjointly regulate adaptive learning and plasticity. However, our understanding of these multi-analyte mechanisms remains nascent, in part due to limitations in measurement technology. Here, during associative conditioning in honey bees, we concurrently tracked sub-second fluctuations in octopamine, tyramine, dopamine, and serotonin in the antennal lobe, where plasticity influences odorant representations. By repeatedly pairing an odorant with subsequent sucrose delivery, we observed individual differences in the conditioned response to odor, which occurred after a variable number of pairings (learners) or not at all (nonlearners). The distinction between learners and non-learners was reflected in neurotransmitter responses across experimental conditions. Remarkably, the speed of learning – the number of pairings prior to a proboscis extension reflex – could be predicted from monoamine opponent signaling (octopamine–tyramine), from both the first presentation of the odorant alone, prior to any pairing with sucrose, and from the first conditioned response to the odorant, coming after a number of sucrose pairings. These results suggest monoamine signaling phenotypes may relate directly to the now widely-reported socially-relevant genetic differences in honey bee learning.
Introduction
Individual differences in learning are universal, yet their biological origins remain elusive. In the honey bee (Apis mellifera), some individuals are "fast learners," while others are "slow" or even "non-learners." While it is established that biogenic amines like dopamine and serotonin regulate behavior, real-time tracking of their dynamics has historically been limited by measurement bottlenecks. Specifically, the invertebrate equivalents of the adrenergic system—octopamine (OA) and tyramine (TYM)—are known to influence plasticity, but their concurrent dynamics during the learning process have remained largely unobserved. This study aims to bridge the gap between neurochemical "states" and behavioral "traits" by tracking sub-second fluctuations of multiple monoamines simultaneously.
Methods
The researchers employed a multi-analyte approach combining advanced electrochemical sensing with machine learning.
ML-Augmented Voltammetry
Fast-Scan Cyclic Voltammetry (FSCV) with carbon fiber microelectrodes was used in the bee's antennal lobe. To overcome the overlapping electrochemical signatures of different neurotransmitters, an ensemble of Deep Convolutional Neural Networks (CNNs) was trained on in vitro data. This allowed for the disentanglement of signals from four distinct molecules: Dopamine (DA), Serotonin (5HT), Tyramine (TYM), and Octopamine (OA).
Behavioral Conditioning
The Proboscis Extension Response (PER) paradigm was utilized, where an odor (conditioned stimulus, CS) is paired with a sucrose reward (unconditioned stimulus, US). Bees were categorized into "learners" (those exhibiting PER after a variable number of pairings) and "non-learners."
Computational Analysis
Beyond concentration tracking, the study calculated "opponent signals" (e.g., OA minus TYM) and used Singular Value Decomposition (SVD) to identify latent patterns in the neurochemical time series correlating with learning speed.
Key Results
The "Pre-emptive" Predictor
A primary finding is that the OA–TYM signal during the very first exposure to an odor—before any reward is offered—predicts the number of trials a bee will require to learn the association. This suggests that "learning readiness" is a measurable neurochemical state. The onset time of the maximum OA–TYM cluster during pre-conditioning was highly correlated with the subsequent learning rate (rho = 0.85, p = 0.002).
Opponency and Learning Trajectories
The balance between OA and TYM, rather than individual levels, proved critical. In learners, the OA–TYM signal starts positive and flips to negative following the first successful conditioned response (the "PER trial"). This "sign-flip" may act as a de-sensitization signal, indicating the association has been successfully updated. In contrast, non-learners' neurochemical responses were statistically indistinguishable from zero.
Monoamine Trends
While DA and 5HT levels decreased linearly as conditioning progressed, they did not serve as primary predictors for individual learning speed in the same way the OA–TYM pair did. The rate of change (slope) of the OA–TYM signal during the first successful conditioned response also predicted the overall learning rate (rho = -0.78, p = 0.008).
Discussion
These results establish that monoamine signaling phenotypes relate directly to individual learning rates. The ability to predict learning speed from initial odor exposure suggests a pre-existing neurochemical state that governs plasticity.
Limitations and Future Directions
The study's sample size (N=18) warrants expansion in future research to confirm these signaling phenotypes across broader populations. Furthermore, while the antennal lobe was the focus here, exploring higher-order integration centers like the mushroom bodies could reveal how these dynamics are transformed. Finally, the correlation between these phenotypes and known genetic differences in honey bee social roles (e.g., scouts vs. recruits) presents a promising avenue for socio-genomic research.
Summary updated