The car driving is considered a very complex activity, consisting of different tasks and subtasks. For such a reason, in particular situations the cognitive demand to the driver could be very high inducing a strong mental workload, and consequently a performance decreasing and an error probability increasing. To this regard, it has been demonstrated that human error is the main cause of the 57 % of road accidents and a contributing factor in over 90 % of them (Treat et al, 1979). Therefore, it becomes crucial to prevent drivers' performance decreasing in order to reduce the probability of error commission.
The car industry has focused its recent research activity on developing algorithms able to forecast aberrant driver's mental states, such as mental overload, fatigue, drowsiness, inattention, on the basis of changes in driver's performance and behaviour (head movements, steering control, frequency of actions on the steering wheel, etc.). However, in the last decades the neuroscientific and neuroergonomic research widely demonstrated how the neurophysiological measures are able to provide a human mental state evaluation more objective, more sensitive, and almost instantaneous, if compared with behavioural and performance measures (Scerbo et al., 2001). Also, thanks to the advances in the field of Brain-Computer Interface (BCI), the neurophysiological measures, and brain signals in particular, have been demonstrated to be the best candidate to trigger the operative systems (the so-called "Adaptive Automation" (AA) solutions), i.e. an automated interface that is able to adapt and reallocate its activities on the basis of the user's mental state (Aricò et al., 2016).
The DriveME project aims to validate an EEG-based mental workload index in real driving settings, in order to provide a tool for: (i) developing AA solutions triggered by neurometrics to support the drivers; (ii) investigating the effect of road infrastructure and car equipment on the driver's workload.
The concept of a ¿closed-loop system¿, i.e. the mitigation of an operator's level of mental workload through a closed-loop system driven by the operator's own EEG, was theorized during the past decade (Prinzel et al., 2000). The major limits in applying EEG measures in operational and everyday life contexts were technological, since the neurophysiological monitoring devices could be perceived invasive and uncomfortable. In 1996, Byrne an Parasuraman (Byrne and Parasuraman, 1996) assessed that the advantage of applying neurophysiological measures in triggering Adaptive Automation was very clear, but the ¿effective application of psychophysiology in the regulatory role may require years of effort and considerable maturation in technology¿. Nowadays, twenty years later, such ¿effective application¿ could become a reality thanks to the progresses in Brain-Computer Interfaces (BCI) and in sensors research: more and more devices for neurophysiological measures (e.g. EEG, ECG, GSR, etc.) are available on the market, with limited costs, limited invasiveness and fancy designs. However, despite the scientific evidences on the possibility of using neurophysiological measures (i.e. by using passive BCI) to trigger AA solutions, only few examples have been proposed in this regard, the most of them in the aviation domain and, in general, in laboratory settings.
The DriveME project aims to fill these gaps, investigating the use of the EEG-based Mental Workload (MW) index, already validated in aviation domain (Aricò et al., 2016), during real car driving situations. With respect to the state of the art, the innovation of the DriveME project could be summarized in the following key points:
- the realization of a system able to calculate an online MW index on the basis of the EEG activity of the driver, providing an output, i.e. the EEG-based MW index, potentially usable to trigger pBCI-based systems;
- application to the car driving domain of a concept developed in aviation;
- validation in real driving situations.
In addition, the DriveME project would provide a useful tool for other domains of applied research related to the human mental workload. Since the prototype will be a user-friendly system, based on simple and intuitive graphical interfaces, that will allows the operator (without any specific technical skills about signal processing) to obtain a reliable index of MW, it would be employed in all those areas of research on Ergonomics and Human Factors, where it is interesting to evaluate how, for instance, road infrastructure or car equipment or traffic flows affect the human mental workload while driving, but neuroscientists are not available.