Human-Like Cognitive Models as a Basis for Cognitive Systems

Cognitive models using oscillator systems along with a suite of semantic knowledge, episodic memory, and other modules offer potential benefit to human operated control systems. Not only do the cognitive models offer safety checks for the human operator, but they are also capable of explaining how they came to their decision. Dr. Chris Forsythe gave a brief high-level overview of the cognitive modelling program active at Sandia National Laboratories. Most of his presentation focused on systems capable of second checking a live human operator. The main system focus as presented was on an AWAC operator’s interface. The system as presented currently relies on a pre-constructed semantic database, and a cognitive model that is built by observing individual operators. It appears that episodic memory is available in some form, however this was not discussed at length in the lecture. The real ”selling point” of this system is the potential interface this system offers to users. Since the system is ”trained up” with it’s operator, it is a unique solution built expressly for the operator. This is in stark contrast to the ”one size fits all” approaches typical AI/expert systems take. Because of this individual knowledge the system possesses it is able to effectively ”scale” it’s feedback to the operator according to their level of understanding of the problem. This is not a new idea, however not enough HCI (human computer interaction) experts are devoting time and money to the