It is well known that memories are not stored as distinct units in the brain, and that one memory can easily trigger another because of shared contextual information.
In a new study published 18 January in the journal Nature Communications, graduate student Halle Dimsdale-Zucker and colleagues investigated the involvement of the hippocampus — a brain region associated with memory — in recalling different experiences.
In the study, 28 participants were asked to watch several videos made with architectural sketching software by the researchers.
Going from one house to another, the participants memorised various different objects, thereby activating both episodic (which video) and spatial (which house) memory.
The second part of the study had the participants answer a number of questions related to the objects they have seen, while being scanned inside an fMRI machine. According to Dimsdale-Zuker, this spontaneously triggered contextual information.
More specifically, an area called CA1 was associated with representing shared contextual information, while another area, called CA23DG, was linked to representing differences in context.
“Our results reveal striking differences in retrieval of contextual information across the hippocampal sub-fields and provide a rare statistical dissociation between CA1 and CA23DG,” wrote the authors in their paper.
While computational specialisation of the hippocampus has already been predicted by mechanistic models, this is one of the first bits of actual empirical evidence.
Another interesting finding was that, contrary to conventional thinking, the hippocampus does not only code primarily for spatial memories, but is also involved in episodic memories, linking both time and space.
“What’s exciting is that it is intuitive that you can remember a unique experience, but the hippocampus is also involved in linking similar experiences,” said Dimsdale-Zucker. “You need both to be able to remember”.
According to Dimsdale-Zucker, virtual reality has been proving itself useful for studying episodic memory within controlled laboratory conditions, with the pragmatic upshot that as our understanding of memory improves, so could medical treatments for neurodegenerative diseases.
The full paper is freely available online.