Background: Rear-end collisions represent about 30% of all car crashes and generate a significant economic cost for society. Driver inattention has been identified as the most important contributing factor in rear-end collisions. One possible countermeasure is the use of systems that warn drivers of potential collisions. Objective: The main objective of this study is to assess the impact of such a warning system on the processing of a relevant visual cue while taking into account the attentional state of the participants. Method: For this, we designed a laboratory experiment during which we recorded behavioral data and brain activity (event related potential, ERP) following the detection of a visual target preceded or not by an auditory alert. Participants had to perform this visual task either alone (simple task) or with a concurrent problem-solving task (dual task). Results: Behavioral and electrophysiological data contribute to revealing (1) that there is a behavioral gain induced by the alert and (2) that this gain is at least linked with a time-saving aspect at both the sensory and cognitive stages of neural information processing. Nevertheless, this impact depends on the attentional states of the participant. Conclusion: In conclusion, we observed an impact of the warning system at both behavioral and brain levels. In addition, our results suggest that this impact has to be tested according to the attentional state of the driver. The main innovation in this work has been the use of ERP measures enabling a better understanding of how this kind of system acts on the information processing.