This study examines experimentally processing response time (1) (Jiang 2012) of two types of written corrective feedback (WCF) in the treatment of errors of preposition a in the values of date, manner, and its use as a direct object marker. The researched types of WCF are Direct, and Indirect Written Metalinguistic Feedback (DWMF and IWMF, respectively) whose rt are analyzed according to its effectiveness and subjects' variables of proficiency (Levels A2 and B1), and mother tongue (German, French, and English). Main findings include greater rt generated by IWMF, manner, and date. The latter also presents a cognitive load increase in level B1. Additionally, rt also appears to be affected by mother tongue, with the lowest cognitive load in English L1, and the greatest in German and French. This study contributes with evidence that indicates the complexity of WCF processing, particularly for the treatment of preposition a related errors.