Thinking About You Thinking About Me

How Recursive Thinking Exercises Actually Work

People throw around the phrase thinking about you thinking about me when they want to describe any kind of back-and-forth mental loop. It is a simple enough concept once you strip away the mystique. One person projects a thought at another person, then imagines that person projecting a thought back. That is all there is to it. The confusion comes when people try to treat it like something supernatural or deeply predictive. Before you attempt anything, you need to understand what you are actually manipulating. This is not mind reading. It is a structural exercise in perspective shifting. You hold your own mental state, then overlay a second layer that represents what you believe the other person is thinking, and sometimes a third layer on top of that. The layers stack. The trick is keeping them from collapsing into confusion. I spent years working with performers and coaches who used this recursively. The method itself is clean but fragile. A lot of people skip the warm-up and go straight to complex scenarios, which is why most attempts feel clunky. You build it from single steps. Hold a simple scenario in your head, notice where the layers blur, adjust your framing, then add another layer only when the previous one holds steady.

Here is the part most tutorials leave out. The moment you introduce emotion into the loop, the accuracy drops. Think about a neutral situation first. Picture someone deciding whether to send a text message back. Map your thought, then map their likely thought. Once that feels natural, introduce mild stakes. A follow-up question after an awkward conversation. The emotional weight changes the projection, and your loop will start drifting unless you anchor each layer to concrete cues instead of guesses. I ran into a problem once where a client kept producing the same false pattern no matter how hard she tried to break it. She was projecting her own anxiety onto the other person's supposed thoughts, which made every layer collapse into the same emotional tone. The workaround was almost stupid in its simplicity. We wrote down each layer on separate cards, physically separated on the table, and forced her to read the next card only after she had stated the previous one aloud. It slowed her down enough that the contamination between layers became visible instead of hidden. The deeper you push this, the more you will notice that people over-index on their own biases. If you assume someone is angry, you will interpret their silence as confirmation of anger. The loop reinforces itself. This happens constantly. Beginners miss it because the exercise feels subjectively convincing even when it is internally consistent but externally wrong.

There is also a practical limit to how many layers most people can manage. Two layers works for casual situations. Three layers is the ceiling for trained practitioners under normal conditions. Beyond that, the exercise stops being useful and starts becoming a form of rumination. You are no longer mapping perspectives. You are just looping your own thoughts with different labels attached. If you are using this for creative writing, character development, or interpersonal coaching, the structure holds up well. If you are using it to claim insight into someone else's actual mind, you are doing it wrong. The exercise maps your model of their thinking, not their thinking itself. Those two things are not interchangeable. The closest thing to a standard reference is work from folk psychology theory on mentalizing and perspective taking, though that academic framework does not always translate cleanly into practical application. For practitioners, the technique often shows up in improv training, theatrical rehearsal, and certain coaching models that focus on conversational alignment. None of those sources are perfect, and none of them spell it out using the exact phrase. The concept exists across a few domains but is rarely consolidated into a single manual.

What helps most is keeping a record of your projections after the fact. Write down what you thought the other person was considering, then check it against what they actually said or did. The gap between the two is where you learn. Most people skip that step because it is uncomfortable. They would rather stay in the loop than break it open to see what they got wrong. You can practice this with everyday interactions. A barista. A coworker. Someone you have a routine with. Low stakes make the learning faster because the cost of being wrong is minimal. The exercise becomes less useful once you apply it to people who do not share your context or communication style. Accuracy drops sharply in those cases, and the method gives no built-in correction mechanism for cross-cultural or cross-context mismatch. The real bottleneck is time. A single clean three-layer pass takes about five to ten minutes when you are doing it deliberately. Most people rush through it in under two minutes and get sloppy results. Slowing down is the difference between a useful exercise and a waste of time.