New brain prosthesis re-installs lost memories of patients suffering memory loss

By on
Brain model
Plaster phrenological models of heads, showing different parts of the brain, are seen at an exhibition at the Wellcome Collection in London March 27, 2012. Reuters/Chris Helgren

New brain prosthesis has been developed to help people suffering from memory loss. The device was designed to accurately copy short-term memories to be translated into long-term memory, replacing the ones lost due to brain damage.

The prosthesis has been successfully tested on animals, and is currently being evaluated in humans. The patients have been implanted with a small array of electrodes in the brain, which can re-encode signals, or memories, into a brain region for a long-term storage.

The scientists targeted the hippocampus, or the brain’s memory centre, which the newly created memory travels in a form of a complex electrical signal. However, having damage at any region of the brain, preventing the memory translation, could pose the risk that long-term memory will not be formed.

People with hippocampal damage, like those with Alzheimer's disease, can only recall previous events that were translated into long-term memories before the brain damage occurred. Those patients would probably face difficulty to form new long-term memories, the scientists say.  

But the developers of the prosthesis, Dong Song and Ted Berger, from the University of Southern California Viterbi School of Engineering, have successfully bypassed a damaged hippocampal section to provide the next region the accurately translated memory. The team performs the process when a memory is translated from short-term memory into long-term memory through a collected neural data.

However, the process has been difficult as a method of "reading" a memory by looking at the electrical signal was not yet available. Berger described the work as translating a memory from “Spanish to French without being able to understand either language.”

More than hundreds of trials were conducted with nine human patients, until the team has accurately predicted how they would translate the signals with about 90 percent accuracy. Predicting the neural signals shows that designing a device to support or replace the function of a damaged part of the brain is possible, the developers said.

Contact the writer at or tell us what you think below

Join the Discussion