Honeybees can use various kinds of information, including color and pattern, in choosing flowers during foraging. We offered free-flying bees a dimorphic artificial patch of radial and bilateral blue/white flowers in order to examine three hypotheses to explain the noted increase in visitation to the flower type offering a lower caloric reward, i.e., optical resolution, dyslectic interpretation, and cognition related to pattern colors. When bees were offered a color pattern rather than a simple color difference to differentiate flower types, they did not always make choices predicted by theory. Honeybees foraged randomly on both flower morphs when rewards were equal and chose the higher caloric reward more often when rewards were different. However, they visited the less rewarding choice more than 33% of the time. Increasing the size of the flower surface by doubling the dimensions did not decrease visitation to the less rewarding flower type, suggesting that visual acuity is not the limiting factor in flower sizes used. When flower colors that increased contrast (yellow vs, blue) were used in the dimorphic parch, visitation rate to the less rewarding flower type did not decline, nor did this 'error rate' decrease when identical patterns were used with only partial color differences. Adding an orientation reference on each flower decreased the frequency with which the less rewarding flower type was chosen from 36 to 26%, possibly because foragers were induced to switch from a global cue (e.g., patch) to a local cue (e.g.. flower). The rate with which the less rewarding flower type is chosen appears to be a function of honeybee use of cognitive and sensory modalities, rather than limited memory and correlative abilities.