Scania : Tracking brainwaves for better safety

With increasing automation, the interaction between drivers and vehicles is becoming more and more important. One area of development is the vehicle's warning sound - which sound attracts the most attention and has the desired effect?

In a previous research project, Scania R&D's Driver Vehicle Interaction team has examined how experienced drivers perceive a range of sounds, including warnings relating to obstacles, load shifting and bad weather conditions.

'We still don't know why drivers prefer some sounds over others, though,' says Stas Krupenia, Senior Cognitive Engineer. 'It may simply be because they thought that they sounded better.'

Recording brain activity

To investigate the relationship between subjective experience and actual brain activity, Scania R&D has entered into a partnership with the Max Planck Institute for Biological Cybernetics in Tübingen, Germany, a world-leading research centre. A recent research project recorded the brain activity of 16 drivers. Each driver twice drove the same stretch of road in the simulator for 25 minutes. Meanwhile, twelve different alert sounds were played a total of 20 times. The sounds came in different ranges, while other, irrelevant sounds were also played. The drivers pressed a button to confirm when they had heard each warning sound.

'It was extremely tiring, much more so than normal driving,' noted Jeffrey Joslin, one of the test subject drivers.

Each driver was equipped with an electroencephalography (EEG) cap from the German company Brain Products, which saw each test subject have 64 electrodes attached to their scalp. The experiments were led by Christiane Glatz, a doctoral student in cognitive neuroscience at the Max Planck Institute.

Selecting warning signals

A similar technology is commonly used for diagnosing brain diseases such as epilepsy, seizures, inflammation of the brain and some types of dementia. A disease can be identified based on the waves that are recorded in various parts of the brain. But the technology can also provide insight into the suitability of different warning sounds in eliciting the appropriate in-vehicle response.

'A warning sound prepares the driver for taking or avoiding an action,' says Glatz. 'A good warning sound should be understood immediately and without ambiguity. There should be no need for deep contemplation. Verbal commands can be understood clearly, but they might require more mental resources to process than auditory cues that we are familiar with. For example, we immediately recognise an ambulance siren and its changing pitch as a indication of its moving direction.'

Cultural factors eliminated

The research aims to measure brain activity while eliminating cultural factors that influence our perception of warning sounds and the workload involved in comprehending them.

'If a horn is sounded on German or Swedish roads we probably pay attention. But an Italian driver perhaps wouldn't care at all,' says Glatz.

The huge amount of data generated by the tests will now be analysed to determine whether the test subjects' own experiences are consistent with brain activity.

'This will give us a solid basis for establishing good simulation methods that increase our capacity to perform virtual tests,' explains Krupenia.