Neuro Linguistic Programming

Development in the 1970s

A 2012 systematic review by Sturt and colleagues says NLP emerged from scholarly work in the late 1970s and moved into commercial settings during the 1980s, and it lists Bandler and Grinder texts such as Frogs into Princes (1979) among the foundational works. Wiseman and colleagues trace the eye-movement hypothesis to a 1975 work by Bandler and Grinder.

The sources opened for this page do not describe which therapists or settings Bandler and Grinder drew on when building the framework, and no book by Bandler or Grinder was opened. Dated claims about the early years therefore rest on the secondary accounts above.

Spread through therapy and training

Michael Heap, a psychologist who reviewed the early claims, writes that in the early 1980s NLP was promoted among psychotherapists, particularly those practicing hypnosis, as a revolutionary breakthrough in delivering fast and effective therapy. He notes that its central doctrine of representational systems was absent from cognitive psychology textbooks, and that the experimental studies he located were mostly postgraduate dissertation projects.

By the 2010s NLP had reached health care. Sturt and colleagues report that interest had been captured among doctors, healthcare professionals and managers, and that NLP training in the United Kingdom is informally regulated by the Association for NLP and, internationally, by the International NLP Trainers Association.

Names and definition

The name is written as neuro-linguistic programming, neurolinguistic programming or NLP, and the sources reviewed use these forms interchangeably. Sturt and colleagues define NLP as a communication framework that uses techniques to understand and facilitate change in thinking and behavior. That is a definition given in a review, not a single agreed standard, and no source reviewed describes a regulated professional title.

Preferred representational systems

Sturt and colleagues identify the preferred representational system as the central hypothesis: the idea that people favor particular sensory modes, such as visual, auditory or kinesthetic, and that communication becomes more effective when matched to that preference. Heap describes the doctrine as holding that a person's dominant mode is signalled by behavioral indices, notably verbal expression and eye movements.

Passmore and Rowson list eye-movement patterns, preferred representational matching and a technique called the Fast Phobia Fix as the three NLP practices they examine. These are terms of the NLP tradition, not measured constructs established in cognitive science.

Professional bodies and levels

The Association for NLP publishes a unified code of ethics, its own code of ethical conduct and a trainers code, and it accredits trainers against minimum course content. It describes itself as an independent professional association and not as a training provider. Sturt and colleagues add that international training runs through three diploma levels based on hours of study.

Techniques named in the literature

Sturt and colleagues list rapport development, modeling, well-formed outcomes and anchoring among the core techniques, and they describe anchoring as based on conditioning. Modeling means studying and copying the patterns of people regarded as effective. The sources reviewed do not give procedures for reframing, so it is not described further here.

Format and duration in studies

In the ten studies reviewed by Sturt and colleagues, NLP interventions were mostly delivered over 4 to 20 sessions, while three were single sessions. One acrophobia study identified in a separate systematic review used a 15-minute session in which the participant imagined a real experience of heights, rather than using computer software.

Eye accessing cues

As described by Wiseman and colleagues, the eye-movement hypothesis holds that when right-handed people look up and to their right they are likely visualizing a constructed, imagined event, and that looking up and to their left suggests recalling a memory. Some practitioners apply this to reading how a person is thinking or whether a statement is true, which is how the hypothesis has been tested.

Health-related applications tested

The experimental studies identified by Sturt and colleagues targeted anxiety disorders, weight maintenance, morning sickness, substance misuse and claustrophobia during MRI scanning. A separate systematic review of acrophobia treatments included an NLP technique alongside desensitization, in vivo exposure and virtual reality exposure.

Coaching and workplace use

Passmore and Rowson reviewed NLP specifically in coaching, where it forms part of the training of some coaches. Sturt and colleagues report from Freedom of Information data that over 700 National Health Service staff received NLP training across three years, predominantly in administrative and managerial roles, which indicates use for organizational as well as clinical purposes.

Reading deception from eye movements

Wiseman and colleagues describe a claim, circulated in NLP training and popular materials, that eye direction can reveal whether a person is lying. They tested this claim directly. Their results appear under research and evidence below.

Systematic review of health outcomes

Sturt and colleagues searched MEDLINE, PsycINFO, CINAHL and other sources and screened 1,459 titles. Ten experimental studies were included: five randomized trials, with sample sizes from 22 to 106, and five pre-post studies. Four trials found no significant difference between groups and one favored NLP on quality of life. Three trials and all five pre-post studies reported improvement within groups, and risk of bias was high or uncertain throughout.

The authors conclude that there is little evidence that NLP improves health-related outcomes, and they state that this reflects the limited quantity and quality of research rather than robust evidence of no effect.

A phobia technique in a wider review

Arroll and colleagues pooled 16 studies of acrophobia treatments from 1973 to 2016. One study of a brief NLP technique compared it with a mindfulness exercise and gave a number needed to treat of 6.0 (95 percent confidence interval 2.8 to 35.5). The review notes that it was often unclear whether the studies were biased, that reporting quality was poor, and that the evidence cannot show which therapy is more effective. Short-term effects were seen for several therapies and long-term effects were not established.

Direct tests of the eye-movement claim

Wiseman and colleagues ran three studies. In the first, 32 right-handed participants were filmed lying and telling the truth, and eye direction did not differ between the two. In the second, 21 people instructed in the NLP hypothesis and 29 controls judged the recordings, and the two groups performed alike. In the third, 52 televised appeals by people seeking missing relatives, 26 later confirmed as deceptive, also showed no significant difference in eye movements. The authors conclude the hypothesis lacks empirical support.

A 2023 eye-tracking study of 31 right-handed adults who answered 100 questions found more eye movements for complex questions, but the gaze directions did not match those predicted by NLP. The authors describe it as a preliminary study; 40 percent of data were excluded for quality reasons and 74 percent of participants were women. An earlier one-page 1980 report titled as a test of the eye-movement hypothesis exists, but only its citation was accessible, so no result is reported here.

Critical reviews of the wider literature

Witkowski reviewed three and a half decades of NLP research drawn from a research database, and framed the question in the title as state of the art or pseudoscientific decoration. Passmore and Rowson, writing for coaching psychologists, conclude that unique NLP practices are poorly supported by research evidence. Heap reports having found many experimental studies of representational systems, mostly dissertations, and his account is critical. Full texts of these reviews were not all accessible, so their specific counts and findings are not reproduced here.

Care boundary

The sources reviewed describe NLP in health outcome trials and coaching, and none supports using it in place of established medical or mental health care. Anyone with persistent anxiety, a phobia that limits daily life, substance misuse or pregnancy-related symptoms should not delay or replace necessary conventional care in favor of NLP. Sturt and colleagues do not give data on adverse events, so the absence of reported harms should not be read as evidence of safety.

Cost and resource use

Sturt and colleagues report an NHS expenditure of 802,468 pounds on NLP-related activity over three years, and conclude that evidence is insufficient to support allocating NHS resources to NLP outside research. This is a cost and opportunity concern raised by the reviewers, not a direct harm to individual clients.

Reliance on unsupported claims

Wiseman and colleagues state that it would be irresponsible for practitioners to keep encouraging people to make important decisions based on the eye-movement lie-detection claim, since their studies found it did not distinguish liars from truth-tellers. This is an indirect harm of relying on a claim, as stated by the authors, and not a report of injury.

Training and oversight

Sturt and colleagues describe NLP training in the United Kingdom as informally regulated through the Association for NLP. That association states it maintains a code of ethics, accredits trainers and operates a complaints policy, and that it is an independent association. No statutory regulation or training standard for NLP practitioners was found in the sources reviewed, and requirements may vary by jurisdiction and by organization.