Banded sunflower moth, Cochylis hospes Walsingham, is one of the most destructive seed-feeding insect pests of sunflowers, causing significant economic yield losses in the northern Great Plains. In an attempt to understand host-plant resistance mechanisms for this pest, we field-tested, over several years, the effects of seven sunflower accessions, rated as resistant to C. hospes in previous screening trials, and a susceptible control (Par 1673-2), on the ovipositional preference and larval performance of C. hospes and its larval parasitoids. Of the resistant accessions, PI 494859 was the most preferred for oviposition, receiving a significantly greater number of eggs per head than did the susceptible Par 1673-2 in 2 of 3 yr. However, the numbers of larvae, and consequently the rate of seed infestation, found in PI 494859 heads were significantly lower than those in Par 1673-2 heads over all 3 yr. Female moths laid relatively few eggs on accessions PI 170385, 291403, and 251902, compared with on Par 1673-2, resulting in lower numbers of larvae per head and lower percentages of seed damaged. No association was observed between the concentrations of two diterpenoid alcohols or two diterpenoid acids in sunflower bracts and the numbers of eggs laid on the heads of the accessions. The number of banded sunflower moth larvae and the proportion of seeds damaged were positively correlated with kaurenoic acid concentrations and negatively correlated with kauranol concentrations. A positive association between resistance to larval feeding and parasitism was found in years 2006 and 2008, with resistant accessions having significantly greater proportions of parasitized larvae than did the susceptible Par 1673-2.