If you want to see some screenshots and videos (including audio!) of the VR Memory Palace, check out this awesome article from Columbia! Importantly,
each room featured music from all over the world, including tracks by neuroscientists you may know: @jmasis.bsky.social and @jayneuro.bsky.social!
Read the full paper for all the juicy details, but if you want a yummy bite-sized appetizer, I highly recommend Wilma Bainbridge's excellent write-up: www.nature.com/articles/s41...
But, for those that are curious about how we 𝗺𝗶𝗴𝗵𝘁 𝗱𝗲𝘀𝗶𝗴𝗻 𝘀𝗽𝗮𝗰𝗲𝘀 𝗳𝗼𝗿 𝗯𝗲𝘁𝘁𝗲𝗿 𝗺𝗲𝗺𝗼𝗿𝘆:
We did find that in the default mode network regions, locations that were 𝘀𝗺𝗮𝗹𝗹, had a 𝘃𝗶𝗲𝘄 𝘁𝗼 𝘁𝗵𝗲 𝗼𝘂𝘁𝘀𝗶𝗱𝗲, and had 𝗺𝗮𝗻𝘆 𝗰𝗼𝗿𝗻𝗲𝗿𝘀 tended to be more reliable across people.
Importantly, this wasn’t just about some places being inherently more reliable than others.
Turns out, 𝗲𝗮𝗰𝗵 𝗽𝗲𝗿𝘀𝗼𝗻 𝗵𝗮𝘀 𝗮 𝘂𝗻𝗶𝗾𝘂𝗲 𝘄𝗮𝘆 𝗼𝗳 𝗿𝗲𝗽𝗿𝗲𝘀𝗲𝗻𝘁𝗶𝗻𝗴 𝘁𝗵𝗲 𝘀𝗮𝗺𝗲 𝗹𝗼𝗰𝗮𝘁𝗶𝗼𝗻 𝗶𝗻 𝘁𝗵𝗲𝗶𝗿 𝗯𝗿𝗮𝗶𝗻 and this was also important for predicting object reinstatement.
For memories to be properly remembered, the brain needs to: 1) retrieve specific contexts without confusing them with similar ones, and 2) use those contexts as stable scaffolds for new experiences.
𝗔 𝘀𝘁𝗿𝗼𝗻𝗴 𝗲𝘅𝗶𝘀𝘁𝗶𝗻𝗴 𝗺𝗲𝗺𝗼𝗿𝘆 𝗼𝗳 𝘁𝗵𝗲 ‘𝘄𝗵𝗲𝗿𝗲’ 𝘄𝗶𝗹𝗹 𝗵𝗲𝗹𝗽 𝘆𝗼𝘂 𝗿𝗲𝗺𝗲𝗺𝗯𝗲𝗿 𝗯𝗲𝘁𝘁𝗲𝗿 𝘁𝗵𝗲 ‘𝘄𝗵𝗮𝘁’!
This means that before your next coffee shop visit, I could take a snapshot of your brain’s existing coffeeshop representation, audit it, and then estimate how well you’ll remember your next experience in it!
So, why would your knowledge of a space help you to form new memories?
For memories to stick, it’s not just about the episode itself, 𝗯𝘂𝘁 𝗮𝗯𝗼𝘂𝘁 *𝘄𝗵𝗲𝗿𝗲* 𝘁𝗵𝗲 𝗲𝗽𝗶𝘀𝗼𝗱𝗲 𝘁𝗮𝗸𝗲𝘀 𝗽𝗹𝗮𝗰𝗲.
We found that spatial contexts (the ‘where’ 🏡) with reliable neural representations predict better reinstatement of the objects (the ‘episode’ ☕️ ) experienced there.
What if we could tell you how well you’ll remember your next visit to your local coffee shop? ☕️
In our new Nature Human Behaviour paper, we show that the 𝗾𝘂𝗮𝗹𝗶𝘁𝘆 𝗼𝗳 𝗮 𝘀𝗽𝗮𝘁𝗶𝗮𝗹 𝗿𝗲𝗽𝗿𝗲𝘀𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻 can be measured with neuroimaging – and 𝘁𝗵𝗮𝘁 𝘀𝗰𝗼𝗿𝗲 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝘀 𝗵𝗼𝘄 𝘄𝗲𝗹𝗹 𝗻𝗲𝘄 𝗲𝘅𝗽𝗲𝗿𝗶𝗲𝗻𝗰𝗲𝘀 𝘄𝗶𝗹𝗹 𝘀𝘁𝗶𝗰𝗸.
CHARLIE, we have to go to candy mountain! [burn this after reading] Shooting my shot in the Brooker multiverse — I sent emails to Superkaylo, TVGoHome, and Horny Estelle. All bounced harder than my optimism. Please ask CAA to forward my message before reality resets. HALP!
super cool work from @jay_neuro on how music shapes our memories of the movies we watch: folks watched a jim carrey movie & the musical themes reactivated memories from the movie! ||| sounds like i should hire a composer to score my life so i can improve my memory 🎹🎵🎶
#CNS2025 was such a blast! It was a delight to work & present our symposium on how VR can be leveraged in neuro/psych research! Make sure to check out the cool research being done by my fellow speakers @tammyish.bsky.social & @manusmad.bsky.social & Birgit Peña Häufler! Excited for #CNS2026 !
For anyone at #CNS2025 - check out @xrmasiso.bsky.social's talk tomorrow afternoon, showing that we can use fMRI to predict which (VR) locations will be good anchors for creating *future* memories!
www.cogneurosociety.org/talk/?id=5579
#neuroskyence #psychscisky
Check out our preprint to learn about our methodology and additional finds! LINK: www.biorxiv.org/content/10.1... (11/11)
This means we were able to 𝐝𝐢𝐚𝐠𝐧𝐨𝐬𝐞 whether a room will serve as an effective memory scaffold 𝐁𝐄𝐅𝐎𝐑𝐄 objects are placed in the room! (10/11)
In accordance with our predictions, we found that we were able 𝐭𝐨 𝐮𝐬𝐞 𝐞𝐚𝐜𝐡 𝐢𝐧𝐝𝐢𝐯𝐢𝐝𝐮𝐚𝐥 𝐩𝐚𝐫𝐭𝐢𝐜𝐢𝐩𝐚𝐧𝐭’𝐬 𝐫𝐨𝐨𝐦 𝐫𝐞𝐥𝐢𝐚𝐛𝐢𝐥𝐢𝐭𝐲 (measured before they saw the newly placed objects) 𝐭𝐨 𝐩𝐫𝐞𝐝𝐢𝐜𝐭 𝐭𝐡𝐞 𝐝𝐞𝐠𝐫𝐞𝐞 𝐨𝐟 𝐨𝐛𝐣𝐞𝐜𝐭 𝐫𝐞𝐢𝐧𝐬𝐭𝐚𝐭𝐞𝐦𝐞𝐧𝐭 𝐝𝐮𝐫𝐢𝐧𝐠 𝐯𝐞𝐫𝐛𝐚𝐥 𝐫𝐞𝐜𝐚𝐥𝐥! (9/11)
…then, we placed an object randomly in each of the rooms and had participants go back into VR to learn the room-object associations. Finally, we scanned their brains again while they verbally recalled the objects and the rooms. (8/11)
Treating each room video as a ‘localizer’, we quantified the reliability of each room representation by comparing the neural pattern similarity of a room to itself vs other rooms… (7/11)
To test this, we built a 23-room virtual reality (VR) environment that participants explored with a head-mounted display. Afterwards, we scanned their brains while they watched two sets of videos of the rooms they visited. (6/11)
…we would predict that when a new experience occurs (such as encountering a yummy banana), he would be better at encoding his experience and consequently, better at retrieving it later. (5/11)
So, we hypothesized that a 𝐡𝐢𝐠𝐡 𝐪𝐮𝐚𝐥𝐢𝐭𝐲 (𝐫𝐞𝐥𝐢𝐚𝐛𝐥𝐞) 𝐫𝐨𝐨𝐦 𝐫𝐞𝐩𝐫𝐞𝐬𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧 (one that is stable and distinct) 𝐰𝐨𝐮𝐥𝐝 𝐥𝐞𝐚𝐝 𝐭𝐨 𝐢𝐦𝐩𝐫𝐨𝐯𝐞𝐝 𝐦𝐞𝐦𝐨𝐫𝐲 𝐟𝐨𝐫 𝐨𝐛𝐣𝐞𝐜𝐭𝐬 𝐬𝐮𝐛𝐬𝐞𝐪𝐮𝐞𝐧𝐭𝐥𝐲 𝐩𝐥𝐚𝐜𝐞𝐝 𝐢𝐧 𝐭𝐡𝐞 𝐫𝐨𝐨𝐦. For example, if the typing minion has a reliable room representation… (4/11)
However, it’s important that the 𝐫𝐞𝐩𝐫𝐞𝐬𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐚 𝐥𝐨𝐜𝐚𝐭𝐢𝐨𝐧 𝐢𝐬 𝐫𝐞𝐥𝐢𝐚𝐛𝐥𝐞! In other words, it 𝐧𝐞𝐞𝐝𝐬 𝐭𝐨 𝐛𝐞 𝐬𝐭𝐚𝐛𝐥𝐞 (so that we can retrieve the details consistently), 𝐛𝐮𝐭 𝐚𝐥𝐬𝐨 𝐝𝐢𝐬𝐭𝐢𝐧𝐜𝐭 (so that the memories across different locations don’t interfere!) (3/11)
You know when you spontaneously remember old memories when you enter your childhood home? Our lived experiences are often tied to the location in which they occur! 𝐎𝐮𝐫 𝐛𝐫𝐚𝐢𝐧 𝐜𝐚𝐧 𝐮𝐬𝐞 𝐭𝐡𝐞 𝐫𝐞𝐩𝐫𝐞𝐬𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐚 𝐥𝐨𝐜𝐚𝐭𝐢𝐨𝐧 𝐭𝐨 𝐡𝐞𝐥𝐩 𝐠𝐮𝐢𝐝𝐞 𝐨𝐮𝐫 𝐫𝐞𝐜𝐨𝐥𝐥𝐞𝐜𝐭𝐢𝐨𝐧 𝐭𝐨𝐰𝐚𝐫𝐝𝐬 𝐭𝐡𝐞 𝐦𝐞𝐦𝐨𝐫𝐢𝐞𝐬 𝐭𝐡𝐚𝐭 𝐨𝐜𝐜𝐮𝐫𝐫𝐞𝐝 𝐭𝐡𝐞𝐫𝐞. (2/11)
🔔𝐍𝐄𝐖 𝐏𝐑𝐄𝐏𝐑𝐈𝐍𝐓 𝐀𝐋𝐄𝐑𝐓🔔 Beyond excited to present our new work showcasing 𝐡𝐨𝐰 𝐰𝐞 𝐜𝐚𝐧 𝐩𝐫𝐞𝐝𝐢𝐜𝐭 𝐡𝐨𝐰 𝐰𝐞𝐥𝐥 𝐲𝐨𝐮 𝐰𝐢𝐥𝐥 𝐫𝐞𝐢𝐧𝐬𝐭𝐚𝐭𝐞 𝐚 𝐧𝐞𝐰 𝐦𝐞𝐦𝐨𝐫𝐲 𝐛𝐞𝐟𝐨𝐫𝐞 𝐭𝐡𝐚𝐭 𝐦𝐞𝐦𝐨𝐫𝐲 𝐢𝐬 𝐟𝐨𝐫𝐦𝐞𝐝! Wait what? Exciting collab w/ @ptoncompmemlab.bsky.social & @chrisbaldassano.bsky.social Link: www.biorxiv.org/content/10.1... (1/11)
beautiful.
I'm also starting to migrate! Thanks for putting this together, Avi!
