Associative memory in social robots, using LIDA's PAM module?

[Jaro Kuiken]

Hello CCRG!

I am currently a 3rd year Creative Technology student at the University of Twente and I'm doing my bachelor thesis right now. My thesis is for the RISE research center, and is about improving the social interaction abilities of a social robot by giving it more empathic abilities. The way we want to give a social robot more empathy, is by implementing an associative memory into the robot. This way, when a robot that could be situated in many situations (like elderly care, a receptionist, etc.) could recognize people that they see more often, and interact with them on a more personal level by remembering specific things about a person. I have been scouring the net for all kinds of associative memory models, and i have found many, but almost none that are nearing the stage of implementation. Your PAM module however, you have already implemented in a framework! I have read quite some of your papers about the LIDA model and the PAM module itself and i have also gone through the first couple of exercises in the tutorial. I have been wondering if the PAM module that you have created, could somehow be used or experimented with for our goal. And thus I was wondering if you could maybe give some of your thoughts on if you think the PAM module could help me in reaching my goal, or if it is too unrelated. Please let me know! And if you do think it could help me, any ideas on what I could try with your PAM module?

Best regards,

Jaro Kuiken