Scientists have created an artificial intelligence system capable of constructing images of the things that a particular individual is viewing based on the activity happening within their brain.
This technology is referred to as Brain-IT and was developed by researchers at the Weizmann Institute of Science. Functional MRI scans were used to identify the patterns in brain activity that occur when individuals view various types of pictures.
The results demonstrated the technology successfully recreated recognizable images depicting a wide variety of subjects including objects, food, faces, landscapes and more complex visual displays.
Although this type of technology has been described as “mind reading,” Brain-IT does not sit inside someone’s brain where it randomly pulls up thoughts.
Instead, the person must be situated inside an MRI scanner and viewing an image; using the resulting brain activity, Brain-It then works backwards to create an estimated representation of what that person viewed.
There is a significant amount of interest in how quickly the system can adapt to new participants. Older methods typically required many hours of brain scan data from the same participant prior to generating useful reconstruction information.
By contrast, Brain-IT reportedly requires approximately one hour of learning about a new participant. Researchers achieved this by focusing on common patterns that exist throughout all peoples’ brains rather than treating each individual’s brain as completely distinct.
They broke down brain activity into tens-of-thousands of very small components and identified these signals with general characteristics of visual stimuli (such as color and location), as well as higher-level categorizations (e.g., faces, food, sports).
