Why we have brain farts, and what scientists are doing to stop them

I was absent-mindedly shoveling cereal into my mouth when the brainfart struck: my hand decided to reroute the incoming spoon's flight trajectory into my cheek. As I sat there with milk dripping down my chin, my immediate reaction was to blame my hand. But then I realized that my hand had just been following orders. If anyone was to blame here, it was my brain. Turns out that neuroscientists agree with me.
Brain farts, the momentary lapses in attention that strike when you least expect them, may actually be rooted in abnormal patterns of brain activity. Neuroscientists call them "maladaptive brain states." We spoke to researchers in an emerging field of neuroscience that examines these brain states to learn about the neurological basis of brain farts, their potential evolutionary origins, and how they might one day be a thing of the past.
For a long time, the mistakes caused by brain farts during monotonous tasks were chalked up to momentary, unavoidable fluctuations in brain activity. Consequently, much of the research since the early nineties surrounding human error and brain activity has been focused on how our brains react to mistakes in order to facilitate processes like correction and learning. In contrast, very little attention has been paid to what goes on in our brains in the moments leading up to a mistake.
But a handful of recent studies have demonstrated that what we refer to colloquially as "brain farts" may actually be rooted in a number maladaptive brain states. These unusual neural patterns happen when you're doing monotonous or repetitive activities — and they can develop as many as thirty seconds before a mistake occurs. It's this aspect of the maladaptive brain state that has led some scientists to conclude that we may one day be capable of predicting, and preventing, the elusive brain fart.

How Neuroscientists Catch Brains in the Act (of Farting)
To understand why our brains sometimes fail to properly execute the most straightforward of tasks, neuroscientists need to look at what's going on in our brains in the seconds leading up to a mistake. To accomplish this, they use imaging techniques like functional magnetic resonance imaging (fMRI), electroencephalography (EEG), and magnetoencephalography (MEG). Redmond O'Connell from Trinity College Institute of Neuroscience in Dublin explained how EEG and MEG allow neuroscientists to peek inside our brains:

MEG and EEG are both based on the fact that active neurons generate a tiny electric field. In the case of EEG's, this electrical activity is picked up on the scalp by placing electrodes there. The activity of individual neurons propagates through the skull onto the scalp, and the activity of large groups of neurons will sum together to produce the jagged line you see on an EEG trace.
The electrical field produced by active neurons also generates a magnetic field, and this can be measured with MEG. Both measures have the advantage of providing a milisecond-by-milisecond index of [what's going on] in the brain.

Dr. O'Connell went on to explain that the primary advantage of MEG over EEG is that magnetic waves undergo less distortion as they pass through living tissue, which can make it easier to identify what part of the brain the magnetic signal is coming from. The increased spacial precision of MEG comes at a cost, however – it requires significantly more expensive (not to mention larger) machinery (pictured on the left) compared to the recording nets used in EEG (below).

Since different areas of the brain interact with one another to control different actions, being able to identify activity in a specific brain region or regions can be very important. Fortunately, this is something fMRI does very well — even better than MEG. Ali Mazaheri from the UC Davis Center for Mind and Brain explains:

[fMRI] measures the hemodynamic response (change in blood flow) in different brain areas…the basic idea being that the more active the neurons are in a region of the cortex, the more fuel they need to consume. It has a relatively slow time resolution, which means that for an event to reach the cortex and a response to be obtained takes about 5 seconds. In contrast, EEG/MEG have a resolution in the range of milliseconds.
Having said that, fMRI has an extremely nice spatial resolution (i.e. you can identify, within millimeters, where activity is occurring in the brain), and it's also nicer for studying network brain activity (i.e. how different regions of the brain are interacting).

The Neurological Basis of Brain Farts
Back in 2007, Dr. Tom Eichele, a researcher at the University of Bergen in Norway, used fMRI to spy on the brains of volunteers as they performed a monotonous task. This task was designed to mimic the repetitive behavior that tends to give rise to brain farts. Dr. Eichele and his colleagues made two surprising, and seemingly counterintuitive, observations.
Image via PNAS
As much as 30 seconds before test subjects made a mistake, blood flow began to decrease in the regions of the brain associated with maintaining focus and task effort (labeled blue in the figure on the left). At the same time, activity began to increase in regions of the brain that are typically only active during periods of wakeful rest – regions that are usually kept deactivated during goal-oriented activity (labeled red in the figure on the left).
Interestingly, the abnormal brain activity vanished as soon as a test subject made a mistake; the same way you recognize a mistake resulting from a brain fart at almost the precise moment that it's happening, an error committed by a test subject reset the brain activity in the abnormally behaving regions back to their expected levels.
So what's causing the brain to wig out? As far as the researchers could tell, the test subjects' brains weren't getting tired, and they certainly weren't falling asleep.
"Autopilot would be a better metaphor," explains Dr. Stefan Debener, who collaborated with Dr. Eichele on the study. "We can assume that the tendency to economize task performance leads to an inappropriate reduction of effort, thus causing errors," he told BBC news.
Translation? Eichele and his colleagues don't think your brain gets tired, they think it gets lazy. When your brain goes on cruise control it's like it's taking a break, and in doing so it saves on energy. Pretty brilliant, really, until your brain gets a little too comfy and forgets to do its job in the first place.
Dr. Eichele's theory that brain farts occur when our brains switch to autopilot is one that has been supported by studies utilizing EEG and MEG, as well. Dr. O'Connell and his colleagues were the first researchers to use EEG to detect increases in the brain's emission of alpha waves - a brain wave associated with wakeful relaxation. By monitoring the behavior of these alpha waves, O'Connell and his colleagues were able to predict performance errors made by volunteers engaged in a mundane task scenario up to 20 seconds before they occurred.
Dr. Mazaheri, in turn, used MEG to monitor the brain activity of volunteers engaged in a response task similar to the one used by Dr. Eichele. Mazaheri found that in the seconds leading up to performance errors, the visual and sensorimotor regions of the brain exhibit a significant increase in strength of alpha and mu rhythms, respectively. (Mu waves are an alpha-like variant found over the motor cortex.)
Why Keeping Brain Farts in Check is a Good Idea – and why it Might Not Be
The predictive power of the maladaptive brain activities observed by Drs. Eichele, O'Connell, and Mazaheri could be used in the future to develop a number of practical applications. Imagine, for example, a feedback device that could warn air traffic controllers when their attention waned too close a dangerously low threshold, or help keep you alert on the highway. With such a device in every human's pocket, brain farts could one day become a thing of the past.
There are, of course, limitations to consider. The enormous size of the imaging devices used to detect these brain patterns would pose a hurdle in the development of such a device, at least for a few years. As of today, brain imaging with an EEG net would be the most feasible option for a portable brain-monitoring system, considering that the smallest MEG and fMRI machines are comparable in size to a car (on the left is an fMRI machine). And while we'll probably look back on the massive size of these imaging devices in 15 years and laugh, the cost of a device that could incorporate the imaging power of all three techniques would likely remain well outside the vast majority of the population's budget.
Then there's the question of whether we'd even want such a device alerting us every time we switched into an idle state of mind.
"Sometimes you just want the brain to be on auto-pilot/automatic mode," Dr. Mazaheri told io9. "A simple example is a gymnast...imagine if he or she had to actively think about every single aspect of the routine. Things could get messy."
Something else to consider is that this brain mechanism has survived thousands upon thousands of years of natural selection – could the conditions of human existence really have changed so drastically in the last couple millennia that these so called "maladaptive" brain activity changes no longer confer any benefit? Dr. O'Connell welcomed the opportunity to speculate:

It is always hard to prove whether something like this confers an evolutionary advantage, but one hunch we have is that a natural circuit breaker might have been a good thing for our ancestors. If you think about it, really monotonous monitoring tasks are a relatively recent thing – sitting still in school for hours listening to a teacher drone on; driving a train and monitoring the signals for hours at a time; working quality control on a factor assembly line; working security at the airport – the modern age, with the gradual take over of technology in our lives, has placed a lot more emphasis on our ability to monitor for long periods of time.
For our early ancestors, who lived in a far less safe environment, staying focused on one task for a very long period of time might have been too dangerous. For example, if you were to spend ages looking through bushes for berries without ever drawing your attention away, you might miss somebody creeping up behind you intending to club you over the head! So a brain mechanism that kicks in every 20-30 seconds to draw you briefly away from the current task and to encourage you to check your surroundings could be a good thing. These ideas however, are purely speculative, as there is no way of knowing. On the other hand, lapsing attention could simply be a flaw that was never smoothed out because, again, monitoring for extended periods of time wasn't a major requirement for our ancestors.

And isn't there something to be said for self-referential thought, mind-wandering, and creativity (all of which have been tied to brain activity associated with the "maladaptive" brain states characterized in the studies referenced earlier)? Absolutely. Still, it would be nice to have breakfast without shoveling cereal into my cheek.