The ability to concentrate our attention is essential for carrying out everyday tasks. Our brain has built-in mechanisms to stop our attention from fragmenting and our minds from wandering, and while they may be highly useful for, say, work, or driving a car, they can deceive us
Building up on the work of Mack and Rock at the end of the 1990s, there is now a strong body of evidence that too much focus can make our brains “blind” to very obvious events or stimuli in our environment. Simons and Chabris from Harvard University, brilliantly demonstrated this phenomenon, called “inattentional blindness”.
In their experiment, they asked volunteers to watch a video of a basketball training practice and count the number of passes made by players in white t-shirts. More than half of the participants did not see the gorilla that appeared centre screen and beat its chest midway through, so absorbed were they in the task.
These attentional blinkers are not limited to vision: our auditory system can, astoundingly, be affected to the point of becoming completely deaf. This “inattentional deafness” has been highlighted in studies where, for instance, people listening to Strauss’ famous _Also Sprach Zarathustra _failed to notice an unexpected electric guitar solo smack in the middle of the symphonic poem.
Research on our brains’ sudden deafness and the associated cerebral imaging studies have provided vital scientific knowledge for understanding human performance in practical situations, like piloting a plane. The history of aviation is littered with accidents in which human operators in the cockpit did not notice auditory warnings, and persisted in their mistakes.
How could professional pilots ignore such critical information? To answer this question, we must adopt a scientific approach, beyond traditional ergonomic approaches, based on both subjective and objective observations of human behaviour. Recent progress in cognitive neuroscience has indeed revolutionised our understanding of the brain mechanisms underlying our perceptual, cognitive and motor behaviour.
A key factor has been the development of neuroimaging techniques such as fMRI and wearable systems like EEG and fNIRS. But, neuroscience is generally confined to highly controlled laboratory studies, often too removed from the situations we face in our everyday lives.
In the last few years, a new discipline has emerged, combining the tools and techniques of neuroscience with the field approach of ergonomics. This discipline, called neuroergonomics, is defined by its founder, Prof. Raja Parasuraman, as the study of the brain at work. This is the approach we adopted in order to understand the mechanisms underlying pilots’ deafness to alarms, under the auspices of a chair funded by AXA Research Fund at the ISAE-SUPAERO aerospace engineering school, in Toulouse, France.
Our initial study was conducted in collaboration with Daniel Callan from NICT at Osaka University, an associate researcher with our laboratory. Pilots were placed in an fMRI and asked to fly an acrobatic plane in a simulator, projected into the scanner through the use of mirrors. Participants were asked to report auditory alarms, set off at regular intervals during the flight scenario.
The results showed that around 35% of alarms went unheard. Even more interestingly, analyses revealed that certain areas of the prefrontal cortex, the “executive arm” of the brain, were activated when flight conditions became critical, and effectively switched off the auditory cortex, rendering the pilots incapable of processing and responding to the alarms. At the same time, there was increased activity in certain visual areas, associated with processing movement.
It is as if the brain decided that visual information was primordial and smothered the processing of auditory signals. This would likely explain why when driving, for example, we can no longer hear the voices of passengers or the radio during emergency breaking. The brain reconfigures itself to avoid danger and activate the most relevant sense.
While fMRI is a vital tool for identifying which areas of the brain are responsible for this deafness, its temporal resolution is not sufficient to allow us to measure exactly when the phenomenon arises. In order to find this out, we conducted a second experiment with my colleagues Raphaëlle Roy and Sébastien Scannella, using an EEG, a useful technique for studying the brain in action.
‘Visual’ pilots vs. ‘auditory’ pilots
Participants were placed in our simulator, mounted on hydraulic jacks, and confronted with a scenario where smoke was filling the cockpit, meaning they had to make an emergency landing in difficult conditions. Prior to the experiment, the volunteer pilots had to complete a test to determine whether they were more visual or auditory.
According to our results, more than 50% of the alarms were ignored, and “visual” pilots were more likely to fail to respond to alarms than “auditory” pilots. Moreover, analyses of the neurophysiological signals revealed that the mechanism behind deafness to alarms was automatic and intervened very early on, at 100ms post-stimuli – that is, long before the sound reaches consciousness (300ms).
We also developed algorithms to determine whether the pilots were able to hear the alarms, by looking at their neurophysiological responses. In 70% of cases, our algorithm managed to detect when the pilot’s brain was no longer capable of processing the auditory alarms.
A final experiment was conducted in real flight conditions, using light planes from ISAE-SUPAREO, with Prof. Callan. Student pilots wearing EEG headsets were asked to take an instructional flight and manage a number of unexpected situations, while responding to auditory alarms. The use of advanced mathematical tools for signal processing enabled us to further understand the phenomenon of deafness to alarms. When pilots missed the alarms, their auditory cortex was out of phase with the rest of the brain and the environment.
These results add to our initial fMRI study by showing that, in these situations, the auditory cortex likely no longer communicates with the brain. Beyond deepening our understanding of the mechanisms underlying attention, this work also opens up the interesting possibility of integrating sensors into pilots’ helmets in order to monitor their attentional states in real time. Ultimately, we should be able to adapt the cockpit and alarms so as to make them more effective for pilots under stress.
Created in 2007, Axa Research Fund supports more than 500 projects worldwide led by researchers of 51 nationalities. To learn more about Frédéric Dehais’ work, visit the[Axa Research Fund] website (https://www.axa-research.org/fr/projets/frederic-dehais).
Translated from the French by Alice Heathwood for Fast for Word