Can the human brain multi-task?
The loss of efficiency and effectiveness due to context switching among various tasks is one of the underpinning tenets of why a Kanban approach helps organizations to reduce the lead times of their work. But is it really true that the human brain cannot multi-task?
The Internet is full of pop “experts” who proclaim that the brain cannot multi-task, as if they really had any knowledge of the subject. It would be interesting to have a better grasp of the reality of how the brain can work, rather than depend on the muddle promulgated by such folklore. I am not a neurologist and do not pretend to have the answers to these questions, but I do have sufficient powers of ratiocination to know when others, even scientists, are somehow missing the point.
It’s a higher brain function issue
Whether or not it is useful to try to understand the brain via the analogy of a binary electronic data processor is moot. But let us take for granted that the brain has various types of storage, does perform input and output operations and also does some type of processing in between those inputs and outputs. Now, since many of us can both walk and chew gum at the same time, it is obvious that some form of multi-tasking is eminently possible. So the issue is restricted to some of the higher brain functions.
Rapid context switching vs. true multi-tasking
Researchers tell us that the brain is good at switching contexts very rapidly, giving the impression of a certain degree of multi-tasking. Some use the example of the short order cook. As I have done this sort of work professionally, I have some first hand experience that helps me assess the situation. And I would agree that a professional cook may be successful due to the ability to rapidly switch contexts, without data loss or confusion, rather than due to true multi-tasking. But the experiments that support this view concern issues of perception, or how the brain handles certain types of sense input. So these experiments are really only relevant to the I/O functions, not to the processing that might be occurring in between the input and the output.
If music be the food of multi-tasking, play on
Can we identify any cases of higher functions that are examples of true multi-tasking? Again, I think the answer is “yes”. The principal example that comes to mind is in the performance of music. Instrumentalists in an orchestra must simultaneously produce their own music, while hearing many other instruments and ensuring that all are coordinated. Once again, I speak from many years of direct experience. A singing guitar player must simultaneously control a hand that strums, a different hand and fingers that position themselves along the neck of the guitar and also a voice, often having a rhythm and a melody different from that of the guitar. A pianist, too, coordinates two hands and ten fingers, each playing different notes, different rhythms and reading two different notation systems. The poly-rhythmic requirements can become quite sophisticated, with one hand playing triplets, or even quintuplets, against the rhythm of the other hand in 4/4 time. Think of the earlier music of Stravinsky, whose Sacré du printemps is filled with staccato cords timed to defy any notions of predictable timing. Charles Ives takes this principle to an extreme. In short, musicians not only multi-task; they must simultaneously coordinate complex physical movements with acoustic properties and esthetics.
An acquired skill
Most interesting about multi-tasking in the production of music is that it is generally an acquired skill, even if the art may come separately. Debutants, who might readily produce a beautiful one-part song, tend to be no good at all at multi-part music. As the man said, to get to Carnegie Hall takes practice.
What does this mean for Kanban and for work in general? In my view, it means that we should not take for granted that context switching among tasks will necessarily cause major increases in lead times. I would agree that there is abundant proof that the typical team doing knowledge work performs much better when work in progress is limited. But could this be because we have not yet found out how to acquire the skills needed to multi-task knowledge work efficiently and effectively? To use another analogy, one might create a sports team from a group of natural athletes. But they will be hopeless as team unless they have learned the specific skills of the sport and how to work together, all requiring extensive training, coaching and practice. The challenge, then, is to learn how to multi-task effectively, given that we do multi-task in certain contexts and can produce beautiful results.
Limited WIP will remain valuable
I do not know if we will readily find out how to multi-task more efficiently. Until that time, limiting work in progress is a proven strategy for delivering results more frequently and more quickly. And even we reached that nirvana where single-tasking and multi-tasking the same batch of work would take the same amount of time, there are still numerous benefits to limiting batch size and work in progress. Kanban has a long future ahead of it.