Automated quantification of real-time food ingestion and foraging in Drosophila The EXPRESSO device automates the conventional CAFE assay and allows us to measure real-time feeding behavior for extended periods for dozens of flies in parallel. The Expresso sensor bank consists of five independent feeding channels. Each feeding channel has its glass feeding capillary which sits along a linear array optical sensor. A microcontroller samples the data from the optical sensor and detects the liquid food level in the capillary. The first version of the Expresso system has allowed us to capture the microstructure of food ingestion in flies. We plan to add new features to the Expresso system which will enhance its capacity to capture different steps in feeding behavior.
Foraging is a fundamental survival behavior, requiring animals to make complex decisions under uncertainty—balancing the need to find food with the risks of predation or energy loss. In natural environments, mice navigate dynamic, unpredictable settings, using sensory cues and memory to locate food while avoiding danger. To study the cognitive and neural mechanisms underlying such behaviors, we developed a task that mimics key features of naturalistic foraging. Our paradigm, Amaze, is a large maze with 32 endpoints representing potential food sources, each requiring a sequence of binary decisions to reach. This setup enables detailed behavioral analysis and computational modeling of foraging strategies. By tracking decision paths and outcomes, we can apply learning models to understand how mice estimate reward probabilities, adjust strategies, and integrate past experiences. This task provides a robust platform for investigating the neural and algorithmic basis of foraging behavior and adaptive decision-making.