The Internal Dialogue That Drags You Down Is a Language — and It Can Be Edited

In 1988, when I was thirteen, the system stamped a label on me that said bad student. My Primary School Leaving Examination score was a few points away from being pushed into the normal stream, and I barely squeezed into the express one. I finished near the bottom of my class. My parents tried tuition, nagging, monitoring — none of it moved the needle.

But the label was not the worst part. The worst part was what I started to say to myself about it, on repeat, inside my own head. I am not academic. I am one of the slow ones. This is not for me. Nobody had to say those things to me after a while, because I had taken the job over from them. My internal dialogue agreed with the report card, and the report card kept agreeing with my internal dialogue, and the two of them spun together until they felt like a fact about who I was.

Then, in June that year, my parents sent me to a preview of a course on something called accelerated learning. The man on the stage did not teach me a trick for remembering dates. He changed what I believed was possible for a person like me. I spent the holidays learning the methodology. In July, at our next monthly test, I came first in my class, and my average jumped by more than twenty percent. A friend asked me if I had cheated. I had not cheated the test. I had found a way out of the sentence I had been repeating, and the moment that sentence cracked, the results changed.

That is the subject of this article: the internal dialogue — the running commentary you play on yourself — and why it is more than a metaphor. It is a language, and every language can be edited. Once you see it that way, a defeated state stops being a mystery and becomes a document with revisions available.

There is a temptation in the parts of the self-help world that talk about this material to treat the internal dialogue as a spiritual idea — something you feel but cannot measure. The most useful thing that has happened in the last two years is that researchers started to try to measure it, and what they found lines up with what practitioners had been watching in rooms for decades.

In a study published in 2024 in the Technium Social Sciences Journal, and then again in the Philippine Social Science Journal in 2025, researchers Charito Ybias, Mark Ortibano, and D.V. Madrigal ran a small single-case intervention with three unmedicated participants aged twenty to thirty who had been diagnosed with depression and carried histories of trauma. Over four weekly sessions, an internationally certified Master Practitioner of NLP delivered an integrated set of techniques — the Meta Model, reframing, anchoring, Timeline Therapy, and hypnotherapy — as a supplement to standard care, not as a replacement for it. The researchers tracked symptoms with the Hopkins Symptom Checklist and the Automatic Thoughts Questionnaire.

Here is the part I want you to sit with, because it is the part that matters most and the part nobody talks about. The intervention produced significant reductions in depressive symptoms, and the strongest improvements clustered in three specific areas: negative self-concept, negative expectations, and low self-esteem.

Read those three words again. Negative self-concept. Negative expectations. Low self-esteem. Those are not vague moods. Those are statements. They are sentences about yourself that you are running on repeat. And the study found that when you work on the internal language directly, those are precisely the dimensions that move.

I want to be honest with you about the limits, because this material deserves honesty more than it deserves cheerleading. This is a small, self-selected, single-case study with three participants and no control arm. It is not a proof that NLP cures depression, and I am not claiming it is. What it is, is a real, published piece of evidence that points in the same direction the practitioners have been pointing for years: the internal dialogue is a teachable, editable component of a person's state, and it shows up in the data.

You have heard the verdict, probably in the same words every time: there is insufficient evidence for NLP as a health intervention. The critique has been around since the discipline was named. But if you actually read the skeptics' own review, the conclusion is stranger and far more useful than the slogan version.

In the spring of 2019, the International Coaching Psychology Review devoted an entire special issue to NLP coaching. In the editorial and in the papers that followed — including the contributions by Jonathan Passmore and Tatiana Rowson, and by Lisa de Rijk and her colleagues — the finding was not that NLP has been proven not to work. The finding was that there is little evidence that NLP interventions improve health-related outcomes, and that this is attributable to the limited quantity and quality of NLP research rather than to robust evidence of no effect. That distinction is the whole ballgame.

In other words, the honest state of the field is not “nlp is fake.” It is “we have not studied it well enough to know.” Those are loaded sentences pointing in opposite directions. One closes the door. The other leaves the door open for the evidence to grow — which is exactly what the 2024 and 2025 integrated findings, small as they are, are starting to do.

There is a practical lesson buried in the skeptics' own review, and it is one I have built my whole way of teaching on. The critique that carries real weight is about NLP sold as a standalone treatment — a cure you substitute for everything else. The 2019 issue itself concluded that there is little evidence for coaching based solely on NLP tools. That word — solely — is doing more work than anyone gives it credit for. When you stop selling a cure and start using NLP as a layer inside a broader system — sleep, exercise, real therapy where it is indicated, a business built on actual operations — you stop inviting the all-or-nothing fight and you start letting the evidence accumulate where it can genuinely land.

That reframe is not a marketing dodge. It is better science. Integration is how weak but real signals survive contact with hostile reviewers.

Let me put the mechanism in plain English, because this is where the whole thing becomes usable.

Every human being runs a continuous, mostly automatic verbal commentary. Psychologists have measured it. The clinical tradition that studied our self-talk directly — beginning with Aaron Beck's cognitive therapy in the 1960s, which identified the automatic thoughts that accompany depression — found that this commentary is not a side effect of mood. It is a driver of it. Beck's core move was the same one NLP's Meta Model performs on language: you take a statement like “I always fail” and you ask what evidence exists, what the boundaries are, and what the words actually refer to. When you interrogate the sentence, it tends to collapse under its own overgeneralization.

Here is the loop, and once you see it you cannot unsee it.

A negative self-concept is a sentence: I am not the kind of person who can do this. That sentence produces a predictable behavior: avoidance. You do not apply for the role, you do not make the call, you do not open the accounts. Avoidance produces a predictable outcome: no result. And no result is read back as evidence that confirms the original sentence. See? I was right.

Negative self-concept over generalizes (one failure becomes a permanent identity), deletes the exceptions (you forget the twenty wins and remember the one miss), and distorts the feedback (neutral events get filed as confirmations). The loop feeds itself. That is why a defeated state feels so airtight from the inside — every piece of contrary evidence is filtered out before it can arrive.

Now the good news, which is the exact thing the 1988 classroom taught me and the integrated study quietly confirmed. Because these three components — self-concept, expectations, self-esteem — are sentences, they are accessible to language-based methods. A technique that edits the sentence edits the loop's fuel. You do not need to argue with your whole mood. You need to re-author one sentence, produce one new result from it, and let that result supply the evidence the old loop refused to admit. That is how a thirteen-year-old went from bad student to first in class in a month: not by trying harder at being smart, but by changing the operating instructions and letting the results rebuild the identity.