The Feedback Loop Your Brain Runs Every Second (And Why You Never Noticed)
Your brain runs a feedback loop between 200 and 400 times per second. Every micro-adjustment your eyes make to track this sentence. Every shift in your posture as you read. Every correction you make without thinking about making it. All of it is a single repeating architecture called the TOTE.
Test. Operate. Test. Exit.
That is the cycle. Four phases. And I spent the first decade of my NLP practice not realising that every single technique I was using — anchoring, reframing, the Swish Pattern, perceptual positions — was a specialised instance of this one loop.
I learned NLP the way most people do: technique by technique. Anchoring on Tuesday. Reframing on Thursday. Submodalities the following week. Each one presented as a standalone tool. And for years I taught them the same way — as independent protocols, not as expressions of a single underlying architecture.
The shift came when I started modelling strategies — not just techniques, but the complete decision-making sequences that top performers run without conscious awareness. I realised that every effective strategy maps to the same structure. And every broken strategy breaks at a predictable point in that structure.
That structure is the TOTE model. It was originally described by psychologists George Miller, Eugene Galanter, and Karl Pribram in their 1960 book Plans and the Structure of Behavior. Bandler and Grinder adopted it as the organising architecture for NLP in the 1970s because it explained something no other model could: how human behaviour corrects itself in real time without conscious thought.
This article covers:
- What the TOTE model actually is (and why the textbook version misses the most important part)
- The three places your strategies break — and how to find yours in under five minutes
- How the SAF feedback-first leadership story demonstrates TOTE at the organisational level
- Why every NLP technique is a specialised TOTE (and what that means for your practice)
- A five-step protocol for diagnosing and repairing broken strategies
The Four Phases That Run Everything You Do
Test: How Do You Know When to Start?
Every strategy begins with a test. Not an action. A comparison. Your brain compares the current state against a stored reference — what Miller, Galanter, and Pribram called the "incongruity" between what is and what should be.
A simple example: you feel thirsty. Your brain holds a reference for "hydrated." It tests the current state against that reference. Thirsty = incongruity. That incongruity triggers the next phase.
The critical insight for leaders: most broken strategies have a broken test phase. The reference point is missing, or vague, or borrowed from someone else's standards. A founder who says "I want to grow the business" but cannot describe the sensory evidence for "grown" will loop indefinitely in the operate phase — trying one thing, then another, never reaching a test that can pass.
Operate: The Intervention Phase
The operate phase is the action. This is where most NLP training focuses because it is visible. You set an anchor. You reframe a belief. You shift a submodality. The operate phase is the intervention.
But here is what the textbook leaves out: the operate phase itself contains micro-TOTEs. Every sub-action within an intervention is itself a feedback loop. When you test whether the anchor has fired, you are running a TOTE inside a TOTE. Nested loops. The architecture is fractal.
Test Again: Did It Work?
This is the phase most people skip. Not on purpose — they skip it because they never defined the test criteria in phase one. If you do not know what "done" looks like, you cannot test for it. So you either:
- Over-operate (keep fixing something that is already fixed)
- Under-operate (stop before the change is stable)
- Or mistake activity for progress (operate feels productive, so you keep doing it)
The second test phase is a reality check. It compares the post-operation state against the original reference. If they match, you exit. If they do not, you loop back to operate with a different approach.
Exit: The Decision
Exit is not passive. It is an active decision to stop the loop. People who cannot exit a strategy — who keep refining, tweaking, and optimising past the point of diminishing returns — have a broken exit gate.
Bandler's expanded TOTE model adds something the original Miller-Galanter-Pribram version did not: explicit decision gates. The simple TOTE assumes you exit when the test passes. The expanded TOTE acknowledges that sometimes the test passes and you still do not exit — because the reference moved, or because you are addicted to the operate phase itself.
A Reddit thread on learning NLP identified that most learners accumulate concepts without calibration — they loop in the operate phase indefinitely, never testing whether their technique produces the intended effect because they never defined the test criteria.
The Three Strategy Failure Patterns Every Founder Runs (And How to Fix Each One)
Failure 1: Missing Test Criteria — You Cannot Hit a Target You Did Not Define
I see this in nearly every founder I work with. They describe a goal with energy and conviction. Then I ask: "How will you know when you have achieved it?"
The answer is usually a feeling. They will feel successful. They will feel satisfied. They will feel like they made it.
A feeling is not a test criterion. Feelings fluctuate with blood sugar, sleep quality, and whether a client smiled at you this morning. A reliable test criterion must be sensory-specific — something you can see, hear, or feel that is independent of your current state.
The fix: Define your test criterion before you start operating. Write it as if a video camera has to verify it. "Revenue of $2M trailing twelve months" is a test criterion. "Feeling abundant" is not.
Failure 2: Over-Operation — Mistaking Motion for Progress
I watched a founder rebuild his website four times in six months. Each rebuild was justified: better design, better messaging, better funnel. Meanwhile, his core offer had not changed. His market had not moved. He was operating because operating felt like progress, not because the test required it.
The second test phase protects against this. But only if you build it into the strategy from the start. Schedule a decision point at a specific date, not a specific feeling. "On August 1, I will test whether version three of the site produced a 15% conversion improvement. If yes, exit. If no, switch to a different approach (not version four)."
The fix: Set a clock-based decision gate, not an outcome-based one. The decision gate prevents endless iteration by making the stop condition independent of how you feel about the current version.
Failure 3: Broken Exit — Winning and Not Stopping
The most dangerous failure pattern is the one that looks like success. You hit the target. The test passes. And you keep going. More features. More optimisation. More complexity. The strategy that produced the win is now the strategy that will undo it.
This is where the expanded TOTE's explicit decision point becomes critical. The decision point is not "did the test pass?" — it is "now that the test passed, do I exit, adjust, or override?"
The fix: Pre-commit to the exit condition before you start. Write down: "When I hit $2M revenue, I will stop scaling this product line and start developing the next one." Then when you hit it, you execute the pre-commitment. No debate. No "just one more quarter."
The SAF Feedback Loop: How I Learned That Leadership Is a TOTE, Not a Transaction
In 2019, after taking over a major training exercise as Head of Branch in the Singapore Armed Forces, I sat down with my team for the post-exercise interview. Standard military procedure. Senior officer reviews performance. Subordinates receive feedback. Done.
Except I did something that, at the time, struck my team as odd. I started the session by sharing what I saw — their strengths that played out, the strengths they had misapplied, and the strengths they did not realise they should have used. Then I asked: "What feedback do you have for me?"
An unprecedented question from a senior officer. Leaders give feedback. They do not solicit it. But I was running a TOTE without knowing I was running one.
Test: As the new Head of Branch, I was aware of a blind spot. I had not developed myself in leadership conversations — feedback delivery, post-exercise interviews, the hard conversations. The incongruity between where I was and where I needed to be triggered the operate phase.
Operate: I shifted attention from how I felt to what my team needed to perform better. I maintained a mental journal — anytime a conversation touched on someone's need, I logged it and followed up. The interview format itself became an operate: share observations first (strengths, misapplied strengths, unseen potential), then ask for their feedback.
Test again: I watched their responses. The soldiers engaged. They gave genuine feedback. The loop was producing the intended effect — better communication, stronger trust, higher performance in the next exercise.
Exit: The loop became a permanent operating system. Not a one-time intervention. An embedded protocol.
Here is what I missed at the time: I was intuitively using the TOTE architecture without knowing its name. I had never studied Miller, Galanter, and Pribram. I had never mapped my leadership behaviour to a feedback loop. I just did what the situation required, and the situation rewarded it.
When I later learned the TOTE model in NLP training, the recognition was immediate. Not because the model was new — but because it named something I was already doing. That is the power of a true systematic model: it does not teach you new behaviour. It makes visible the architecture you are already running, so you can run it with intention instead of instinct.