Queueing network-adaptive control of thought rational (QN-ACTR): an integrated cognitive architecture for modelling complex cognitive and multi-task performance

How to computationally model human performance in complex cognitive and multi-task scenarios has become an important yet challenging question for human performance modelling and simulation. This paper reports the work that develops an integrated cognitive architecture for this purpose. The resulting architecture – queueing network-adaptive control of thought rational (QN-ACTR) – is an integration of the QN mathematical architecture and the ACT-R symbolic architecture. This integration allows QN-ACTR to overcome the limitations in each method and model a wider range of tasks. Implemented as a computerised simulation programme, QN-ACTR has been verified in the simulation of 20 typical tasks from the ACT-R literature. The benefits of the integration have been demonstrated in the simulation of 29 transcription typing phenomena, showing its capability in modelling complex cognitive and multi-task scenarios that have not been modelled by either QN or ACT-R.