Improve CAs
- The way we and others have simulated CAs they're pretty
much binary switches, active or not.
- Memories don't work this way.
- They are activated and decay over time (seconds).
- This means each memory has a time and activity course.
- Different memories can have different behaviours.
- How do CAs fit in with working memory limitations?
- Can we link them more firmly to biological behaviour?
- Memories are also formed and related to each other in
complex ways.
- Can we develop a full CA based memory system?
