Savlucci & Taatgen (2010) argue that as we automate tasks (through expertise for example) we become better at multitasking because we rely less on declarative memory to plan our actions, thereby, our "procedural resource" is not bottlenecked by retrieving knowledge of task instructions (e.g. over time we do no longer need to think about how to change gears). Nonetheless, automated tasks can still be involved in a multitasking situation (or be a multitasking situations themselves).
I understand that making divides between two tasks (e.g. primary and secondary) is pragmatically useful when we are assessing multitasking at a broad level of abstraction; however, to arbitrarily cut out all automated tasks from the multitasking continuum altogether (until a non-automated task is activated) seems like a strong theoretical claim.
Let me know if I am missing something here. Cheers.
Thank you for your well thought out response. Your explanation certainly made my understanding more clear, and given the introductory student context, I understand the difficulties faced in actually clearly defining what constitutes multitasking 'scientifically'.
The problem you raise regarding a definition or criterion for a multitasking situation is an issue I have been thinking about as well. My research pertains to the study of interruptions (sequential task switching) in complex workplace tasks. The literature as it currently stands makes no clear distinction regarding what exactly an interruption is and how an interruption differs from multitasking. The distinction is a simplistic one to make from a layman's point of view, however, scientifically speaking it is much more difficult.
Idealistically, a good rule of thumb is that an interruption is a situation in which two tasks are performed sequentially (sequential switching). However, as you mentioned a lot of research paradigms would consider this a unique form of multitasking. I have frequently found myself wondering at what level does multitasking actually become sequential switching. Trafton & Monk, made the distinction by stating the defining feature of an interruption paradigm is the presence of a primary task which must be returned to. Nonetheless, the question still remains how disengaged one must be from the primary task to consider it an interruption. Is it a matter of all modalities (visual, auditory, tactile, cognitive) being disengaged from a primary task to be considered an interruption, or just some of them?
As you can tell, I seem to have more questions than definitive answers at this stage of my research. Cheers for your reply. If I have more thoughts on this, I will send them through if you are happy to discuss.