Manage episode 492071577 series 2966529
How are memories represented in the brain and how can we model this?
Professor Anna Schapiro from University of Pennsylvania as we discuss her work on computational modelling of sleep and memory. In this episode, we discuss several kinds of Neural Network models as well as diving into the history of Jay McClleland's Complementary Learning Systems theory and Anna's recent work on combining neural network modeling with EEG.
Find out more about Anna Schapiro and her research group here.
Links to articles discussed in this episode:
- Why there are Complementary Learning Systems in the hippocampus and neocortex
- [Replace with correct Norman 2005 paper]
- Self-recovery of memory via generative replay
- A unifying account of reply as context-driven memory reactivation
Host: Penny Lewis
Editor: Sophie Smith
Check out our NaPS website to find out more about our team, our research and events.
This recording is the property of the Sleep Science Podcast and not for resale.
Chapters
1. S3E5 - Anna Schapiro - Can we model the way REM and NREM replay consolidate memories? (00:00:00)
2. What are computational models? (00:01:20)
3. Neural Network Models (00:04:18)
4. Complementary Learning Systems (00:08:40)
5. NREM & REM Models (00:21:10)
6. Lab based experiments (00:32:00)
24 episodes