Research

I am Maria Wimber, a Professor of Cognitive neuroscience here in the School of Psychology & Neuroscience. The scientific work in my group is centered around the question how remembering works. We study how the human brain reconstructs memories of past events, and how our memories adaptively change each time we recall them.

Memories are reconstructions of the past, not veridical copies. One important line of research is concerned with the question how the memory recall process unfolds in time, starting with a reminder and ending with a more or less full-blown recollection. We use behavioural and multivariate ("decoding") approaches of neural signals (EEG, MEG, iEEG, fMRI) to track the neural footprints of memories, and to directly observe what features of a memory are reconstructed step by step in the human brain. We are also interested in how brain oscillations, and in particular the hippocampal theta rhythm, help to orchestrate the reactivation of memories.

On a slower time scale, we also study how memories qualitatively change over time and with repeated remembering. What aspects of an event are most likely to be remembered, and what aspects are most rapidly forgotten over time? Retrieving a memory has been shown to have two sides. On the positive side, memories become more stable and permanent each time we reactivate them. On the other hand, remembering can also induce forgetting of related memories. This forgetting is in fact a highly adaptive capacity of human memory: our brains appear to function on a "use it or lose it" basis, retaining the information that is frequently reactivated, and discarding irrelevant or competing information.

We study these adaptive memory processes using a range of neuroimaging and electrophysiological techniques, including functional MRI, EEG, MEG, and various combinations of those techniques. We also record signals directly from the human hippocampus, with the help of epilepsy patients who undergo presurgical monitoring and volunteer to take part in our research. In terms of analytic approaches, the group has a strong focus on multivariate pattern techniques (RSA, MVPA) that allow us to isolate the neural fingerprints of individual memory representations in neural activation patterns, and to observe how these mnemonic patterns dynamically change over time.