Recent work has indicated that anthropogenic pollution of floral-scent may have negative impacts on bumblebee foraging behavior. We need quantitative tools to both measure how much pollution of a learned floral-odor bumblebees can tolerate and identify which scent-pollutants are problematic. This study used encoding characteristics of insect olfactory systems to develop a new paradigm for quantifying complex odors. This 'Compounds Without Borders' method builds multidimensional vectors of scents based on physiologically relevant physical characteristics of component odorant-compounds. The angular distance between CWB-vectors then provides a single quantitative variable describing how similar (or dissimilar) two complex odors are. This angular representation of odor similarity is predictive of bumblebees' behavior in an associative odor learning task.