Toward a Computational Investigation of Bee Visual Perception

This publication establishes a scientific and computational foundation for two connected, ongoing lines of investigation: one concerned with bee visual perception, and another concerned with the effects of pesticides on visually guided behaviour. The studies remain in development. They are presented here as a framework of questions, evidence, and proposed methods, not as completed simulations or claims to reproduce a bee's subjective experience. This paper develops a transparent computational methodology informed by honeybee ultraviolet, blue, and green photoreception, compound-eye organisation, motion sensitivity and optic flow, visual learning and navigation, and pesticide-related impairment of visually guided behaviour. It distinguishes RGB imagery from multispectral evidence, optical models from neural and behavioural processes, and biologically informed transformations from generative interpretation. The proposed modular Python framework records spectral inputs, spatial sampling, temporal processing, disruption parameters, and generative reconstruction as separate stages. Together, these stages establish an inspectable route from biological evidence to computational experiment. The paper includes seven figures documenting historical microscope preparations from the Andrea Abbatangelo Microscope Slide Archive.