I used to multitask a lot. My girlfriend in college would enter my room and see me gaming on one monitor, watching NFL Redzone on another, and completing school assignments on my laptop. I’ve since consciously and aggressively reduced the number of tasks I work on at any given time, especially as those tasks require deeper and more intense focus. This change was a response to a gradual realization that my brain is “single-threaded” in nature and any meaningful work requires minimizing costly context switching.

I began really noticing it Junior year of college once I got to higher level maths and programming courses. When completing assignments I often found myself turning off everything including any music playing and working in complete silence, sometimes even with my phone locked away in a drawer if necessary. I was being challenged in a way that required genuine focus without distraction. It’s not that I physically couldn’t have a football game on in the background it’s just if I did I wouldn’t have noticed a single thing happening in it and if I did then it meant I had lost my flow state.

It still wasn’t very often when I had assignments that challenged me such as a that so I didn’t take much conscious action but then I graduated and joined Microsoft and now Census and it’s all different. The scale and complexity of the systems I am working on and the pace at which things change is incredibly rapid compared to school. I’m being pushed so much harder every day than I have the past decade and it’s a welcome change. Especially at Census the agency I’m granted and the impact I can have leaves me with a wide open field of extremely creative work and it can only really be done effectively when focusing.

But what does it mean to be single-threaded? It means that at any given time the processor (your brain) can only make progress on one active task. To work on two tasks at once it needs to switch back and forth. Every time it switches the entire context of what task its working on needs to change. This doesn’t matter if the cost for switching contexts is low and for the past decade most of my “work” in school has had a low cost. When I would watch football, play games, and do schoolwork it wasn’t that I was actually doing all three at once it was simply that the cost of switching my attention rapidly from a game to a tv screen to a worksheet was negligible. In fact if you find that you can switch between activities rapidly without burden it might be a sign that those activities aren’t worth doing. Or at least they might not be very fulfilling.

I have tried coming up with a good heuristic for what makes a task costly to context switch in and out of but I have been unsuccessful. My best approximation is that the cost to switch contexts increases as the complexity and novelty of the task increases. It has to be sufficiently complex because if it wasn’t you wouldn’t need to devote your full attention to it but it also needs to be novel. Programming can be extremely complex but because I’m very familiar with it– it’s less novel– it’s easier for me to switch into a programming context. My brain is familiar with the state of mind required for programming so it’s easier to switch from a natural resting state into it. In fact because I program so much the natural resting state of my mind has gotten a lot closer to that of when I’m programming. The times when programming has a high cost to context switch is when I’m analyzing a new system or a new part of the codebase or debugging something and I need to understand all the interactions happening. In other words, when I’m doing something novel. Similarly when writing blog posts such as these it’s far more unfamiliar and therefore far more costly for me to switch contexts. It can take hours for me to switch into the right mindset for writing– it quite literally feels like my neurons are firing in a different direction– so when I finally get into that flow state if I lose it I’m back to square one. That’s a massive cost and it’s entirely because writing is unexpectedly complex and it’s also not something I do very often.

There’s no real takeaway here, just a development in how I approach work. The more important a task is the more I try to focus only on it. Solve one problem at a time. Get comfortable not jumping around to different tasks even when you hit a dopamine plateau. Get comfortable not looking at your phone in lulls. Learn how to not even think about whether you should look at your phone (that’s the hard part). As I’ve grown more conscious of the single threaded nature of my brain I’ve realized the need to curate what tasks are running on it more carefully because it’s the only way to achieve meaningful deep work that I find fulfilling.