Thursday, June 18, 2020

Secret Science Club Remote Lecture: MRI and Memories

Tonight, my friends at the Secret Science Club are sponsoring a lecture by neurologist Dr Daphna Shohamy of Columbia University and the Dana Alliance for Brain Initiatives. I had a problem booting my laptop (nice timing, Dell Inspiron!), so I popped into the conference a half hour into Dr Shohamy's lecture. This lecture was the annual Dana Foundation lecture and originally been scheduled for March.

When I was added to the chat, Dr Shohamy was discussing the role of the hippocampus in the formation of memories. Much of what we know about the role of the hippocampus came from the study of Henry Molaison (known in the scientific literature as Patient H.M.), who suffered from epilepsy resulting from a traumatic brain injury as a youth. At the age of 27, his seizures were localized to the medial temporal lobe, specifically the hippocampus. In order to cure his epilepsy, H.M.'s hippocampus was surgically removed. This led to H.M. having difficulty in forming new memories. His older memories were intact, but he had problems forming memories of newly encountered people, scenes, and moments. This phenomenon is known as anterograde amnesia. H.M. was able to learn new tasks, such as drawing a shape observed in a mirror, but did not remember the learning process even as he got better at the task itself. Neuropsychologists Suzanne Corkin and Brenda Milner studied H.M.'s memory and cognition after his operation, and was instrumental in discovering the role of the hippocampus in memory formation.

The hippocampus is named due to its perceived resemblance to a seahorse. Imagining the hippocampus using the brainbow staining technique, displays a complex, and gorgeous, network of neurons. To give the audience some idea of the complexity of neural connections, Dr Shohamy displayed Felleman and Van Essen's 1991 diagram of the visual cortex of the brain. The 'HC' at the top represents the hippocampus, with 'ER' being the entorhinal cortex:




The hippocampus has a 'privileged position' for receiving sensory input. It is high on the brain 'hierarchy', funneling sensory input to other regions of the brain for higher cognition. Study of the brain typically involves the use of functional Magnetic Resonance Imaging, which measures blood flow, allowing metabolic activity to be tracked. The prefrontal codex is active in memory coding. Dopamine, which Dr Shohamy jokingly described as 'the celebrity of neurotransmitters', plays a role in reward learning (as well as playing a role in addiction). The corpus striatum also plays a role in action and movement.

Dr Shohamy then shifted to the topic of the role of reward in memory. She wistfully noted 'we used to be able to travel', then recounted a trip to Paris when she was twenty years old- she would often leave her map behind and explore the city. She got to the point where she could locate a favorite cafe, even without making a conscious effort to memorize the location. How can we prioritize important information if we don't know at the time that it is important?

Dr Shohamy's former student Kendall Braun developed a computer 'maze' game which involved a reward half of the time. Participants would follow a series of images until they found, in fifty percent of the image boards, the image of a gold coin. Fifteen minutes after playing the game, participants would remember images from 'no reward' mazes almost as well as they remembered images from 'reward' mazes. Twenty-four hours later, they would remember images from 'reward' mazes much better than they remembered those from 'no reward' mazes. After twenty-four hours, and a consolidation process which takes place during sleep, there was a retroactive improvement of memories.

Memories are shaped by their relevance to future events (which affect you in the future), the prioritization of memories occurs retroactively, and depends on a consolidation process involving dopamine, and a reactivation of patterns after sleep.

It's easier to access memories which lead to better decision making. Links between memories can lead to 'false' memories, such as a memory of a conversation in a restaurant being conflated with a different conversation with the same person in a different location. Dr Shohamy presented us with a hypothetical example of integrating memories to result in a 'false' memory. Noting that her colleague Dr Nim Tottenham delivered the SSC Dana lecture last year, she posed a situation in which remembering having a cocktail in the lecture setting could be inspired by drinking a cocktail while watching her Zoom lecture, which could set off a cascade of other memories of the lecture.

In one experiment, the factors were associations between two events, the possibility of a reward, and a resultant decision, a basic Sensory Preconditioning situation. Create an association of a neutral image with a reward, and that image will become preferred to another image even in situations in which a reward is absent. In the case of faces, real or imagined, specific areas of the visual cortex, the fusiform face area, are active.

Memories are integrated into a network of associations, a 'model of the world'. This integration depends on the reactivation of associated memories. Access spread across memories shapes decision making. Dr Shohamy posed the question, should these retroactive associations be considered false memories, or flexible memories?

The hippocampus also plays a role in imagining the future, and damage to the hippocampus can result in trouble articulating the details of future plans. Neurologist Akram Bakkour formulated an experiment based on Buridan's Paradox, a seemingly easy decision making process made difficult by making the choice between two equally desired items. Dr Shohamy used the example of her two favorite candies, a choice between a packet of M&Ms or a Kit-Kat bar- put simply she would want them equally. In healthy people, the hippocampus supports deliberation even in 'simple decisions'. While anteroamnesiacs would remember long-desired objects, they would be slow in making decisions regarding preferences, but would be unhampered in choosing between two sets of facts, such as a question regarding whether objects were yellow or blue. We need memories in complicated times to help us make proper decisions.

Dr Shohamy concluded the lecture with a set of bullet points... The hippocampus allows the formation of rich, vivid, long-term records of the past. Memory is shaped by priorities, meaning, and intention. The brain trades off accuracy for flexibility so we can use the past to adaptively prepare for the future. The role of memory has implications in society, economically, culturally, and in terms of policy.

The lecture was followed by a Q&A session, but your Bastard was unable to get a question in (I was still catching up, and was also somewhat occupied by work-related tasks). Neurology is a young, but rapidly evolving field, so any shortcomings in the use of fMRI in brain studies should be ironed out as the field matures. Regarding eyewitness accounts, Elizabeth Loftus did the groundbreaking research on the unreliability of eyewitness accounts- things are misremembered, mistaken connections are made. Regarding the retention of memory in seniors, cardiovascular activity is crucial to the health of the hippocampus, and is beneficial to memory. Regarding ADHD, more research is needed, but persons with ADHD have no noticeable memory deficit, though occasional coding issues may arise. That being said, curiosity helps with memory, and persons with ADHD tend to be more curious as well as more impulsive. Regarding stress, and its affect on memory, the hippocampus is packed with trauma/stress receptors and corticosteroids, stress hormones, can damage the hippocampus... she indicated that SSC alumnus and rockstar Dr. Joseph LeDoux was conducting research in this field.

Once again, the Secret Science Club served up a fun, informative lecture, though the boozy, beautiful Bell House and its lovely staff are sorely missed. Kudos to Dr Shohamy, the Dana Foundation, and Margaret and Dorian. Virtual high fives all around! I don't know when the next Cyber Secret Science Club lecture will take place, but put yourselves on the mailing list so you can get notifications. In the audience of this lecture, their were audience members of Chicago, LA, and Australia... so join in and soak in that SCIENCE!!!

Here's a recent video of Dr Shohamy lecturing on conscious memories:





If you dig that, get your name on the list.

3 comments:

Anonymous said...

It's a "no brainer" for me; M&Ms!

Big Bad Bald Bastard said...

I dunno, M&Ms are so hard to peel, and when you finally get the rinds off, there’s hardly any chocolate left.

Anonymous said...

Dang I've been doing it wrong, and that was the only aspect of your post about which I thought I could offer an intelligent comment! Just kidding man, always an interesting read.