Anna Stróż May 15, 2020 - Interior sensing

Development of Neuroergonomics: applications of EEG in Automotive

The rapidly growing development of neuroscience created new opportunities and expectations for this scientific domain. Although the interest in human cognition and perception is constantly present since Ancient Greek philosophy – e.g. Heraclitus of Ephesus and his epistemological insights [1] – today we possess various techniques to finally deep dive into our cognitive processes and in a step-by-step manner, investigate the relationships between our brains and our behaviour. What is also extremely interesting is that the field of neuroscience is present not only in clinical applications – neuroscientific concepts are applied in other domains, such as: marketing, UX, and ergonomics.

There is a plethora of neuroimaging techniques, which give researchers the ability to “take a look into” our skulls – you may have seen following abbreviations: MRI, CT, PET, EEG – believe me, there are many more of them and each decade brings us another technique. Out of all of them, EEG (i.e. electroencephalography) is widely used by researchers around the globe as this technique is inexpensive and provides great temporal resolution. As with all of the abovementioned techniques, it has its pros and cons, however, we will not focus on that now. 

EEG records the electrical activity of cerebral cortex, which – depending on internal and external circumstances – may manifest various oscillations occurring with different frequencies and amplitudes. What may be interesting is to observe what is the brain’s “response” to a certain stimulus or what happens in our brain if someone asks us to focus on e.g. reading a book. Maybe we can look at the brain activity of an aircraft pilot? A Mathematician? Or an automobile driver? 

This is how we got to the main point of the article. In the past decade, automotive researchers and engineers presented their technological advances or early-stage research in acquisition of other bioelectrical signals – such as electrocardiography (ECG) [2] or galvanic skin response (GSR) [3], which are the responses changing due to the influence of the autonomic nervous system. While we feel stress or a strong emotion, our pulse or sweating intensity may rise or fall.  As you correctly think, these indicators at least theoretically may be utilised in order to detect a behaviour different than normal, such as a stressful event. What about EEG then? How can the fact that we read brain electrical activity be useful for automotive research?   

EEG may be a potentially interesting measure for mental fatigue, drowsiness or while one deals with a difficult task [4, 5]. It may be interesting to see what the variance of drivers’ behaviour is [4] and what is truly fantastic is that through the simulation environment you can create road tasks of various complexity and observe the behaviour while being sure that everyone is safe – e.g. the task of keeping the lane with simulated crosswind in the experiment of Karthaus and colleagues [4]. In the study of Karthaus et al. [4] authors describe the usage of EEG in the assessment of driver performance variability. The simulated experiment forced the participants to try to keep the lane. It is noteworthy that based on the acquired EEG and driving behaviour data, authors could confirm previous observations of various strategies of road behaviour [4]. Even if most of the studies related to driver behaviour are conducted in simulated environments, there are as well some designated test tracks (such as AstaZero in Sweden [6]) which may help immerse the driver more intensively without losing the aspect of safety. 

Yet another study, performed by Cao et al. [5] was an investigation of what happens inside the brain while one focuses his/her attention on a given driving manoeuvre. It is indeed very interesting how our driving performance and brain responses may change if we, for example, switch our attention to our smartphone or engage in a deep conversation.  

Yan et al. [7] provided an interesting study in which they attempted to classify driving states using EEG, driving behaviour data and by applying machine learning algorithms to do so. What is promising right here is that there is an amazing (even if now only theoretical) potential to integrate real-time EEG processing to recognise dangerous behaviour on the road and… to force the car to intervene with the driver [7]! Seems like a complicated thing but in the Bitbrain article you can read about Nissan’s brain-to-vehicle technology presented at CES 2018 [8].  

Now you know that bioelectrical signals, even EEG, have the potential to push the development of human-oriented automobile design further. What is more, together with the rapid development of artificial intelligence and machine learning approaches, there will be a growing number of R&D projects utilising bioelectrical signals from the body sensors, the data from the automobile and from the environment to infer what the proper behaviour should be or simply to assess what is going on.     


[1] Tatarkiewicz, W. (1981). Historia filozofii. Tom pierwszy. Filozofia starożytna i średniowieczna. Państwowe Wydawnictwo Naukowe, Warszawa, p. 32. 

[2] (Access: 06.04.2020) 

[3] (Access: 06.04.2020) 

[4] Karthaus, M., Wascher, E., & Getzmann, S. (2018). Proactive vs. reactive car driving: EEG evidence for different driving strategies of older drivers. PloS one, 13(1). (Distributed under Creative Commons Attribution License).  

[5] Cao, Z., Chuang, C. H., King, J. K., & Lin, C. T. (2019). Multi-channel EEG recordings during a sustained-attention driving task. Scientific data, 6(1), 1-8.  (Distributed under Creative Commons Attribution 4.0 International License,  

[6] (Access: 06.04.2020) 

[7] Yan, F., Liu, M., Ding, C., Wang, Y., & Yan, L. (2019). Driving style recognition based on electroencephalography data from a simulated driving experiment. Frontiers in psychology, 10, 1254. (Distributed under Creative Commons Attribution License (CC BY),  

[8] (Access: 06.04.2020) 

[9] Ma, Y., Chen, B., Li, R., Wang, C., Wang, J., She, Q., … & Zhang, Y. (2019). Driving fatigue detection from EEG using a modified PCANet method. Computational intelligence and neuroscience, 2019. (Distributed under Creative Commons Attribution License)

Share post: