The 52-year-old was able to react to pictures of human faces showing emotions such as anger, happiness or fear, the researchers said on Sunday. 

The man, identified only as Patient X, has suffered two strokes which damaged the brain areas that process visual signals, leaving him blind. 

However, his eyes and optic nerves are intact, and brain scans showed that he appeared to somehow use a part of the brain not usually used for sight to process visual signals linked to some emotions. 

When researchers from the University of Wales showed the man images of shapes such as circles and squares, he could only guess what they were, and had a similar lack of success determining the gender of emotionless male and female faces. 

Detecting emotions

But when presented with angry or happy human faces, his accuracy improved to 59%, significantly better than what would be expected by random chance, with similar results for distinguishing between sad and happy or fearful and happy faces. 

"What Patient X has assisted us in establishing is that this area undoubtedly processes visual facial signals connected with all types of emotional facial expressions"

Dr Alan Pegna,
head of team,
University of Wales, UK

He was unable, however, to tell apart images of animals which appeared either threatening or non-threatening. 

Brain scans showed that when the man looked at faces expressing emotion, it activated a part of his brain called the right amygdala, which is known to respond to non-verbal emotional signs.

The findings, published in the journal Nature Neuroscience, suggested the man was able to process information gathered by his eyes in a different part of the brain from the visual centre. 

"This discovery is interesting for behavioural scientists as the right amygdala has been associated with subliminal processing of emotional stimuli in clinically healthy individuals," said Dr Alan Pegna, who led the study. 

"What Patient X has assisted us in establishing is that this area undoubtedly processes visual facial signals connected with all types of emotional facial expressions."