The ability to generate and recall memory is a behavior that is evolutionarily conserved across the animal kingdom from humans to jellyfish. Memory not only allows previous experiences to inform future decision making, but it also amasses information essential to life, such as memory of quality food sources, shelter, and predator-related associations. Associative memory forms a relationship between two or more distinct and initially unrelated stimuli and can be defined by its temporal characteristics, such as shortand long-term duration, as well as the memory being appetitive or aversive, generating approach or avoidance behavior, respectively. Since its introduction as a memory model in the 1970s, the fruit fly, Drosophila melanogaster, has emerged as a powerful tool for the investigation of memory-related processes. While a variety of memory paradigms have been used extensively in Drosophila, such as appetitive and aversive olfactory memory, the use of appetitive visual memory remains infrequent. A previous study introduced a visual shortterm memory (STM) paradigm that could be used for the study of both appetitive and aversive visual memory in Drosophila. However, this protocol required 50+ flies per condition, with three conditions per experiment, and 15 or more replications were frequently used to assess memory. As a result, this paradigm requires substantial numbers of flies, time, and is impractical for large genetic screens. Here, building upon this previous work, we describe an optimized appetite visual STM paradigm in freely moving Drosophila. Using recently published data on sexual dimorphism, innate color preferences, and borrowing practices from related appetitive assays, we have established an approach that minimizes confounding factors, such as sexually dimorphic starvation survival and sucrose preference, as well as pre-training color preference variation between groups. In doing so, we present an appetitive visual STM paradigm requiring substantially fewer replicates and numbers of flies to produce significant learning.