In general terms automaticity could be defined as the subconscious condition wherein 'we perform a complex series of tasks very quickly and efficiently, without having to think about the various components and subcomponents of action involved' (DeKeyser 2001: 125). For language learning, Segalowitz (2003) characterised automaticity as a more efficient, more accurate, and more stable performance. As such, automaticity is often associated with systematicity and a merely instructional approach. However, task-based learning seems not incompatible with automaticity either, since it incorporates activities that respect 'transfer-appropriate processing and other positive features of communicative practices' (Segalowitz 2003: 402) and thus allows students to creatively apply previously acquired knowledge in new communicative contexts. In order to test this assumption, an experiment was conducted at Antwerp University with a group of intermediate-level students of Spanish. Two groups were evaluated: an experimental group and a control group. The control group attended a traditional communicative course, whereas the experimental group's course had a task-based component built into it. The results of the experiment indicate that the experimental group outperformed the control group for automatization (as defined by a number of criteria).