A judge and jury can never truly get into the mind of a suspected criminal. So it’s distinctly hard to tell if a crime is committed either out of recklessness, carelessness, passion, or cold-hearted intent.

But, neuroscientists now believe that it could be possible to “mind read” suspects using biomedical imaging and machine-learning data analysis. Fortunately, it’s not quite as 1984 as it sounds.

A team of researchers has used functional magnetic resonance imaging (fMRI) to analyze the brains of people in the hopes of finding out whether or not they committed a crime "knowingly" or out of "recklessness". Their findings were published in the journal Proceedings of the National Academy of Sciences.

It started with a series of mock crimes to test-run the idea. They used the fMRI scanners – which pick up on changes of blood flow in the brain – on 40 people while they took part in a computerized task that offered them rewards to carry suitcases across a border. In one scenario, they knew the case contained illegal drugs. In another, they were not sure whether the suitcase they were assigned contained contraband, but they were aware of the risk.

They found that those knowingly breaking the law had significant different brain activity than those who were acting recklessly. Those knowingly committing the crime had far greater levels of activity in multiple specific brain regions. Those being “merely” reckless had fewer bursts of activity, mainly limited to just the occipital cortex. A machine-learning algorithm then kept a watch out for these patterns and gave a “fairly accurate” prediction of what mental state they committed the crime under, based just on their brain imaging data.

It’s early days for the research and right now serves more as a “proof of concept”. Most obviously, of course, it’s unlikely you’re going commit a crime while you're plugged into an fMRI scanner. That’s even before you consider the mammoth legal and ethical questions of using such a technique.

The authors also point out “It would be absurd to suggest, in light of our results, that the task of assessing the mental state of a defendant could or should, even in principle, be reduced to the classification of brain data.”

Nevertheless, they say “this is not to suggest that our results have no legal significance.” They argue that this research could be developed to provide juries with more insight into the intent of the suspect, helping them make a more informed decision.

“Whether a reckless drug courier should be punished any less than a knowing one will of course always remain a normative question," the study concludes "However, that question may be informed by the comfort that our legally relevant mental state categories have a psychological foundation.”