4 ms·
There's actually a simple explanation for this: meditation trains brain functions that operate at a sufficiently general level to see cross-domain impacts, whil
by westoncb 10y ago
There's actually a simple explanation for this: meditation trains brain functions that operate at a sufficiently general level to see cross-domain impacts, while brain training games don't (or at least the range of domains is much smaller).
Let me give you an example of a popular form of meditation:
You sit up straight in a comfortable place, not doing anything special, just listening to what's going on in your environment and tuning into sensations in your body (this happens automatically as long as you haven't 'assigned' yourself something else to do; once you get into this state, you might recognize it as familiar from when you were younger, or from a time when you were feeling particularly like "everything's okay" and you are just at home). Before starting this, you understand that you are going to watch/listen/feel your breath—and if you've had some practice, you don't try to do anything about it: you just wait and feel whatever comes up as time passes. Eventually it's like the queue of other things clears and you start tuning into your breath more and more until the sensations become very rich and relaxing.
Thoughts will continue emerging for a good long while during this, and meditation teaches a certain response to noticing their arrival: you acknowledge them, as if you'd heard a small sound in the room, turned around and saw what it was, and then returned to what you were doing (often times meditators are encouraged to use a quick label for whatever mental thing comes up, e.g. 'thinking' or 'fear')—then, assuming you haven't got too wrapped up in the thought (either by following it, or by trying somehow to making it go away), the sensations of your breath return to the forefront.
Two big things come from this: you become more relaxed afterward, which aids task performance in most domains directly; and, you learn all these pitfalls and techniques for maintaining focus on what you choose to focus on. One big one, for instance, is that a lot of people approach focus by seeing distractions as problems to be solved (whether they are thoughts, pains, environmental sounds—whatever), but eventually you see that what works is maintaining an equanimous attitude toward distractions, and simply returning your attention to what you're doing, while understanding you might get pulled away again shortly (and it's not enough to just 'know' this; you have to practice it).
Lastly, meditators are typically encouraged to practice being in this 'aware but not trying to do anything' state while doing other things than sitting, which has got to help with generalizing. (Not to mention the desired mental state is just domain agnostic to begin with. When you're doing a task you either believe you know how to do it and go at it confidently, with little meta-thought; or you're observing/critiquing how you're doing it and trying to change routines you've stored for doing it. In the first state you're way more tuned into sense data which is relevant for actually executing the task in the present moment. Meditation helps you learn to enter the first state intentionally.)
- tonystubblebine 10y agoThat's an interesting hypothesis that brain games are too specific.