One of the most costly and intensive mental operations anyone performs daily is task switching. It’s as though a brain just isn’t geared for clearing its RAM to make space for a new process. Instead, it gets stuck trying to compute the new task using some ancient page file mechanism, while maintaining the previous task in the RAM.
There are a few scenarios that can play out when switching.
- I can switch perfectly (100%), get into the new mindspace, and start work. Clean RAM, ready to plow.
- I can fail completely to switch (0%) and be rendered incapable of completing my new task, because I can’t get the required data loaded into memory.
- I can land somewhere in between.
Scenario 3 is most dangerous, because it can feel like it’s a clean, 100% switch. This is one of the brain’s many shortcomings – it doesn’t have great (or any, really) integration tests, so you really can’t tell when it’s bugging out. Incomplete switching – when it feels like complete switching – leads to subpar work output, poor reasoning, bad conversations, and a bunch of other nasty issues.
Recently, I moved from a sparring conversation with someone who runs a CPM-bid channel, to someone who runs a CPC-bid channel. Failing to fully switch mindspace caused me to deliver a series of dangerous recommendations, based on a CPM mindset. I had to take time later to go and repair the potential damage.
The above situation is extremely mild, compared to what can happen in professions where more is at stake, and speed is higher.
Ensuring total switching
The best way I’ve found to ensure a total switch is to wipe the RAM before attempting the switch.
For humans, that means stopping work, going to get a drink, doing some pull-ups, eating a piece of fruit, and looking at dank memes for 2 minutes.
Alternatively, getting in a flight or fight situation also wipes the RAM, but may prove stressful if made into a routine occurrence at the workplace.
If you find yourself in a situation where you’ve been interrupted, and can’t take enough time to complete a wipe of your RAM, you can either invite your interrupter to do the activities, or enter the conversation assuming that you know nothing, and asking lots of questions. Excellent way to enter a conversation anyway, most of the time.
Integration testing the switch
Even when the RAM is attempted wiped, the operation doesn’t always succeed. So how do you know? The brain doesn’t just let you know of its own accord, that’s for-damn-sure.
Doing a bit of manual integration testing is valuable, but super unreliable. Brains don’t like to admit when they’re subtly undermining you – it makes the whole thing way less subtle.
The best integration test therefore, happens via a 3rd party tool. You can choose between tools such as Friends, Family or Colleagues. Effectively, your only option is to vet the output before you commit it to production.
Showing your work output to someone else, explaining that you’re concerned about partial switching, and asking them to validate the base principles of the work.
If you’re in a conversation, it’s easier – you can just preface the talk with a comment about the switching issue, and let your conversational partner pay special attention to whether you’re aligned to the correct mind-space.
General Principles/Transferability
In general, the approaches used to debug task switching are safe to bake into your normal routines. If you’re suffering from task switching errors, you’ll put in safeguards and improve. If you’re somehow NOT suffering from task switching errors, you’ll just improve the way you have a conversation, and open yourself up to more feedback. Win.