Efficient signalling requires coordination of signal form and receiver design. To maintain signal function, parallel changes in signaller and receiver traits are required. Genetic correlation and co-evolution among signal and response traits have been proposed to preserve signal function (i.e. coordination) during the evolution of mate recognition systems. Empirical studies have provided support for both mechanisms; however, there is debate regarding the interpretation of some of these studies. Tests for a genetic correlation typically hybridize divergent signalling systems and look at hybrid signal form and receiver design, or impose artificial selection on signal form and look for an indirect response to selection in receiver design. Some of the hybridization studies did not achieve reassortment of genes from the parental types, whereas some of the artificial selection studies incorporated random mating in their designs. As a result of these limitations, the hybridization studies cannot discriminate between genetic correlation and co-evolution with primarily additive genetic effects underlying signal and response traits. Similarly, the artificial selection experiments cannot discriminate between genetic correlation because of linkage disequilibrium and co-evolution. This study examined the mating preferences of male almond moths, Cadra cautella, before and after female moths were artificially selected (using a design incorporating assortative mating) for novel pheromone blend ratios. Our results demonstrate the absence of a genetic correlation between signal and response traits in the almond moth.