Jolyon T. Troscianko

DigiFish creates realistic virtual fish animations and reconstructs visual fields from tracking data to study visually guided behavior.

This study uses genetic algorithms to explore how lighting and habitat geometry influence optimal camouflage in prey, showing their combined effects on patterning and color.

This study examines how different artificial light characteristics affect moth flight behavior, finding that light intensity and structure influence flight-to-light responses and tortuosity.

This study shows that recent breeding experience improves egg ejection behavior in barn swallows, suggesting awareness of own egg-laying is key, not a template image.

BehaveAI detects and classifies objects and behaviors from motion using a color-from-motion encoding strategy, enabling efficient and accurate analysis with minimal annotation effort.

This study compares human and machine learning accuracy in identifying parasitic eggs using UV-visible photographs, finding machine learning models significantly outperform humans.

This study investigates clownfish color patterns and finds that specialists and generalists have distinct ecological and adaptive functions linked to their anemone hosts.

The paper shows that light microhabitats in forests drive sensory convergence and divergence in mimetic butterflies' visual systems through mutualistic interactions.

This study examines how temperature and turbidity affect prey detectability by predators, finding that warmer water increases prey conspicuousness, counteracting the negative effects of turbidity.

The paper quantifies how lighting and 3D environment alter scene and object appearance, finding that blue-yellow contrast and pattern directionality are lighting-sensitive, while green-red information remains stable.

This paper introduces a low-cost, open-source hyperspectral imaging system for measuring light environments across various applications.

The paper presents a model of colour appearance based on efficient coding of natural images, showing it predicts human perception and primate retinal responses across 35 phenomena.