For more than 50 years now, doctors have known that the hippocampus is important for memory in humans. Patients with hippocampal damage presented with the striking inability to form new memories. These cases ultimately led to the general understanding that the hippocampus serves as sort of a central hub that talks to other brain areas to store and retrieve memories — much like a computer processor storing and accessing information in a hard drive. While these processes are exactly understood in a computer, the parallel memory mechanisms that exist within the human brain are unknown.
Studying these mechanisms is difficult, but neurosurgical patients with implanted electrodes onto the brain surface provide an opportunity for high quality recording of brain activity. Researchers can therefore gain a more detailed view of what the brain is doing while patients perform different cognitive tasks. As we were interested in measuring brain activity associated with memory, we leveraged leveraged this protocol to measure the electrical signals from the brains of patients as they played memory games.
Patients memorized pairs of words such as “Orange” and “Navy” and later were presented with one word and asked to recall its pair. The game therefore allowed for us to choose the specific memory that we wanted the patient to remember. All the while, we recorded the electrical signals directly from the brain surface in order to discern any memory relevant activity.
As such, we were in a privileged position to see what was happening in the human medial temporal lobe (including the hippocampus) and other brain regions during the formation and retrieval of memory. We found that when the patients correctly recalled memories, specific brain waves known as ripple oscillations, emerged at the same time between these brain areas. Conversely, the ripples did not occur when patients could not recall the memory or in parts of the brain that are not related to verbal memory such as the motor cortex. Taken together, the fact that these ripples occurred preferentially during successful memory recall indicates that they are likely a mechanism for communicating memory information across the brain.
To support this idea, we additionally found that when these electrical ripples occur during memory recall, they elicit an “echo” that resembles a period in the past when the memory was first formed. For example, when you think of your high school graduation, this ripple mechanism would reinstate the brain waves associated with that event to help you remember what you were wearing, the smell of the graduation gown, and even the elevated feelings of excitement.
Likewise, ripples would be happening when remembering items on a grocery list, the directions to a friend’s house, or any task that requires memory retrieval. Generally speaking, our findings suggest a way for the hippocampus to ping information throughout the brain so that we can use past information to guide us through our lives. In other words, these electrical ripples underlie that “light bulb” feeling when you are searching around for your lost keys and all of a sudden remember that you left them in your coat pocket.
Thinking clinically, these findings are likely important in the pathophysiology of Alzheimer’s disease and other dementias. Any advance in this field is particularly important, since over 200 clinical trials of new therapeutics for neurodegenerative disorders have failed in the past decade. If ripples indeed underlie memory retrieval in humans as our data suggests, then this mechanism would likely be disrupted in these disorders. Indeed, recent work has already indicated that ripples are affected in mouse models of Alzheimer’s disease, lending weight to this claim.
Therefore the identification of a mechanism for memory retrieval in humans not only illuminates how the brain can function as a computer in saving and retrieving memories, but it is also a major step towards narrowing the scope of current therapies for dementia. Simply put, this work has identified a narrow target, instead of the currently wide sea of possibilities, that future researchers can try to protect with drugs or neural stimulation in order to address the most debilitating symptoms of neurodegenerative diseases.
These findings are described in the article entitled Coupled ripple oscillations between the medial temporal lobe and neocortex retrieve human memory, recently published in the journal Science.