Many hosts of obligate brood‐parasitic birds use variation in the coloration and pattern of eggshells to identify and reject foreign eggs from their nests. However, egg‐rejection behavior of several hosts is not tightly predicted by the modeled output of overall avian‐perceivable chromatic differences between foreign and host eggs. This demands a re‐investigation into the proximate sensory cues and perceptual processes used for egg recognition. One plausible mechanism is that rejection by hosts depends on the relative stimulation of one (or a combination) of their photoreceptors when comparing their own and parasitic eggs (i.e., self‐referencing). To test this, we placed dark‐blue egg models (with a known ~ 50% rejection rate) in nests of American Robins (Turdus migratorius), which are occasional hosts of Brown‐headed Cowbirds (Molothrus ater). We found that relative metrics of individually repeatable size or shape and overall avian‐perceivable differences in color between American Robin eggs and model eggs did not explain the probability of an experimental egg being rejected. However, despite a limited R2 value, hosts were significantly more likely to reject a model egg when the estimated contrast between their own egg and the model egg was greater for the blue (SWS) photoreceptor. These results are consistent with female American Robins using a simple physiological algorithm that would confer widespread benefits by reducing the likelihood of rejecting their own eggs, while still eliminating parasitic eggs because of their dissimilar novel coloration.