Memories – fact or fiction?
Everyday slips in memory can have consequences that range from mundane to potentially life-threatening. Much of neuroscience research has aimed to understand memory through very pure, simple, stripped-down tests – such as remembering lists of unrelated works or pictures of objects. However, as Chris Bird, professor of Cognitive Neuroscience at the University of Sussex, explains, these measures don’t capture the fact that our real-world experiences unfold over time, in specific locations, and involve cause-and-effect. The EVENTS(opens in new window) project, which was supported by the European Research Council(opens in new window), adopted a different approach. The team used stories or videos to simulate everyday experiences to test memory, and functional MRI(opens in new window) (fMRI) to investigate how memory processes are underpinned by different brain networks. “This provides insights into how normal memory processes work and why and when slips in memory might occur, as well as why memory impairment is a defining feature of brain diseases such as Alzheimer’s disease,” he says.
How information processed in independent neural modules is combined within a mental ‘event model’
EVENTS considered whether people with severe memory problems were still able to understand and follow a narrative storyline ‘in-the-moment’, even if they were unable to recall it later. Around 400 people aged 18-40, recruited from the university, and the surrounding area, were scanned by fMRI, and approximately 50 adults with memory problems, aged 65-85, were tested, along with a similar number of healthy age-matched, ‘control’ participants. A further 1 000 online volunteers also participated in experiments.
Remembering what happened? Or what we think should have happened?
The team have improved our understanding of why and when our memory might fail and point to ways that we can stop memory failures from happening. They have also shown that individuals with memory problems have difficulty with not only recalling events in the hours and days after they occurred but also comprehending how events unfold in the present. Other mechanisms behind the processing of information by the brain were revealed. Using stop-motion videos of people performing familiar actions, such as doing the laundry, which participants watched in the scanner, the work addressed how the hippocampus signals surprise. “This resolved a theoretical debate about the role of the hippocampus in ‘prediction error’ processing,” adds Bird, who co-authored a paper(opens in new window) setting out the findings in the ‘Proceedings of the National Academy of Sciences’. Another paper, ‘False memories for ending of events’(opens in new window), presented findings showing that if people watch events that end abruptly before a natural ‘scene change’, then, when people recall those events from memory, they very frequently ‘add in’ the ending, as if they had watched the scene play out to its conclusion. They are unaware that they are falsely recalling things that they never saw happen. In several publications, the researchers address the question: How much do we remember of what happened in a situation versus what we think should happen in a situation? “We showed that the video-specific ‘fingerprints’ of fMRI activity are largely based on the reliable and predictable information that we expect from a situation, rather than the less predictable information that might make the event unique. For example, if we recall a scene from the TV show ‘Friends’, it’s our knowledge of the main characters and their context that forms the main scaffold of our memory – the specific events of a particular scene are just the icing on the cake.” As Bird explains: “We are teasing apart the brain systems that process knowledge about the world and provide our expectations of what should happen in a given situation.”