The current dissertation presents an ERP-investigation of metrical stress processing in French. Indeed, while metrical stress is well known to play an invaluable role in speech comprehension, its functions in French speech processing are unclear. French is a language traditionally described as having no accent. This dissertation questions the traditional view and aligns to two metrical models on French accentuation, which propose stress to be encoded in cognitive templates underlying the abstract representation of the word. In our interdisciplinary investigation of metrical stress processing in French, we take a functional, yet metrically rooted, approach. We use the method of Event-Related Potentials (ERP), which provides us with a highly sensitive and temporally precise measure allowing us to determine whether there is metrical stress in French, and to what extent metrical stress aids the listener in speech comprehension. We will show that metrical stress facilitates processing throughout French comprehension and argue for metrical stress, as well as the domain of the word, to be given a more prominent place in the descriptions of French prosody.