NLP has loyal fans and plenty of testimonials. People say it changed their confidence, their career or their relationships. So why do psychologists remain unconvinced? The answer lies in what has been tested.
What the reviews found
A 2010 review (Witkowski). It looked at 33 studies testing NLP’s own claims. About 18 per cent supported them, about 55 per cent did not, and the remaining 27 per cent were uncertain.
A 2012 systematic review (Sturt and colleagues). This looked at NLP and health outcomes and found only ten experimental studies: five randomised controlled trials and five before-and-after studies. Four of the five trials found no significant difference between groups, and one favoured NLP. The conditions covered were anxiety, weight maintenance, morning sickness, substance misuse and claustrophobia during MRI scans. Risk of bias was high or unclear throughout. The authors concluded that there is little evidence that NLP improves health outcomes.
Earlier reviews. Studies in the 1980s tested NLP’s central theory, that people have a preferred “representational system” revealed by their word choice. Reviews such as Sharpley’s found that the research did not support it.
“Little evidence” is not “proven not to work”
The 2012 authors were careful on this point. Their conclusion reflects how little good research there is, not strong proof that every NLP technique is useless. A technique can be untested and still helpful, or tested and found helpful for reasons that have nothing to do with NLP’s theory.
Why NLP can still feel like it works
Several things can make a method feel effective without its theory being correct:
- Expectation. Believing something will help is itself helpful. This is the placebo effect, covered in hypnosis series part 33.
- Good communication. An NLP course often teaches listening, goal-setting and rapport. Those are useful whatever they are called.
- Regression to the mean. People seek help when things are at their worst. They often improve afterwards whatever they do.
- Borrowed ideas. Much of NLP repackages methods that already have an evidence base, such as imagery and cognitive reframing.
- Selective memory. The success stories are told. The shrugs are not.
How to read NLP claims
- Ask which technique, for which problem, tested how.
- Prefer randomised trials over testimonials.
- Be wary of claims that NLP can “cure” phobias, trauma or illness.
- Treat “NLP” and “evidence-based” as separate questions.
The bottom line
The research does not support NLP’s big claims, and it has been tested far less than the people selling it suggest. If you find a technique useful, you have lost nothing by using it. Just don’t swap it for treatment with a real evidence base when your health is at stake.
Next in the series: NLP and Hypnosis: How They’re Linked (and How They Differ)
This article is for general information and is not medical advice.