My PhD research focused on contextual memory: how implicit context helps us retrieve and use information more appropriately. Over the course of our lives we learn thousands or millions of facts, most of which are irrelevant to our current situation. Even when we can clearly specify the information we need, there may be many facts that fit the criteria, and it would be most efficient to consider the most relevant facts first.
Fortunately, the situations we are in often contain enough information to appropriately rank the facts we know in the most appropriate order – if we collect information about the situation in a context, and structure memory retrieval to take advantage of that context. My work on context sensitive asynchronous memory explores this idea as applied to information retrieval and case-based planning.
This can include not just memory for facts, but memory for emotion: annotating experiences with the emotions they cause can be used to learn emotional responses that are appropriate for different situations. My work on emotional long term memory explores this idea as applied to robotic control and game artificial intelligence.
