Switching contexts has a cost
Assuming that Multitasking does not increase productivity (in people), it is possible that this can be justified by a cost function when switching contexts. This is not unlike what happens with multithreading in Python (and probably other programming languages).
Switching contexts while we are thinking means jumping from one train of thought to another. While writing, for example, this would be the difference between finding the words and copyediting what we have written. They require different types of attention (202010161703) and probably access to different neural paths on our brains.
If switching from one context to another has a cost which is dependent on the distance between tasks (i.e. receiving a notification on the phone while replying an e-mail), and we assume that there may be some idle time while performing tasks, then we can postulate the following:
- Switching between tasks that require a certain level of attention will be lower than performing the tasks consecutively (i.e. Multitasking does not increase productivity).
- Performing long (hard to define what long means in this context) tasks that require attention will have many idle blocks of time
- Switching tasks in the idle moments may be a better way of using the time overall
Therefore, when planning to work, it could be useful to have more than one task at hand, focus on one and when we feel we are stuck or not being productive, we can switch to a new context. Ideally these two tasks should be correlated in such a way that the activation cost is not too large, but also that the type of attention required is different. For example, we could write several paragraphs for 20 minutes, stop and focus on editing, or on outlining.
Bear in mind that this approach does not imply taking breaks, but simply changing tasks in a way that we can optimize the use of the idle time.