Mind reading - the science behind the superpower

Will it be possible to read minds in the future? Sophie Dixon examines the current research and future possibilities of mind reading

Mind reading, the phenomena that has for so long been considered a fantasy, is becoming a more realistic possibility. While mind reading devices for casual communication are still a long way off, the ability to translate a person’s brain activity into written text, a process known as neurotelepathy, has already been achieved. The ‘brain-to-text’ system developed by German and U.S. researchers uses electrocorticography to record the neural activity in specific locations in the brain. Research participants were instructed to read sample text aloud while having this neural activity recorded; this data then could be accurately decoded into the text that had been read. The ability to translate brain activity into text is a pretty mind-blowing concept, and today this may have life-changing consequences for patients suffering from conditions such as locked-in syndrome. However, in the future we may consider these to be the first few steps towards conquering the popular sci-fi fantasy of mind reading.

Meanwhile, Professor Marvin Chun at Yale University has succeeded in extracting images from the brains of participants looking at pictures while inside fMRI scanners. Volunteers looked at a picture of a face whilst inside the brain imaging machine, and researchers were able to create digital reconstructions of the face by decoding the electrical activity patterns of the brain. Neurotelepathy is possible because of our knowledge of the different specialised regions of the brain. Whilst one area may be active when we view faces, another will be active when we view a car, and in general these specialised regions are the same across the population. Of course there are some individual differences, and this could mean an image triggers slightly different neurons in one person’s brain compared to another’s, perhaps because of different experiences and emotions associated with the image. Scientists, however, are able to combat this by creating a database big enough to account for these differences, and allowing a kind of neural signature to be developed for different stimuli.

The results of this research have already pointed towards some potential future applications. For example, the reconstructed images have been found to be more accurate than other facial reconstruction technologies used in the investigation of crime, such as PhotoFit. As the technology improves and our understanding of the human brain advances, it is a very real possibility that such methods may one day aid the criminal justice system, in ways we never imagined would be possible.

From Issue 10