This study investigated the relative effectiveness of two instructional interventions (implicit and explicit feedback) as a function of implicit sequence-learning ability. Second language (L2) attainment was measured by means of a self-paced reading task, which shows online sensitivity to language errors. Implicit sequence-learning ability was measured by means of the serial reaction time task from the High-Level Language Aptitude Battery. Two Spanish target structures were investigated (gender agreement and differential object marking). The results showed that neither explicit nor implicit feedback had an effect on L2 learners' grammatical sensitivity to the target structures and, therefore, that feedback was not effective in changing L2 learning outcomes, as measured through an online processing measure. The results also showed a significant relationship between implicit sequence-learning ability and sensitivity to feminine gender agreement on the posttest, after controlling for pretest scores, but only in the implicit feedback group.