NLP Research Literature Overview

A Summary of Research Reviews

A large amount of research on NLP was done in the early 1980s on a construct called the “Preferred Representational System” (PRS).

It has been surmised that proponents of NLP have, in the past (Sharpley, 1987) had “little to support them and much to answer to in the research literature.” Commenting on the research which had already been done by that time, Sharpley (1987) commented to researchers that “there [was] little use to the field of counseling research in further replications of previous studies,” until researchers had taken the time to perform a “careful meta-analysis of the large amount of data already gathered.”

Here it is hoped that such an undertaking will be able to promote, as Sharpley suggested, “future research that can contribute new data … via methodological advances [and a] consideration of different aspects of NLP.”

The majority of research into NLP concepts was primarily concerned with the concept of a “Preferred Representational System,” or PRS. It is important to understand the issues surrounding this concept before examining the research.

A large number of these studies only have limited value as evidence, however, due to significant confounding methodological and theoretical issues.

One common misunderstanding of the field of NLP in the psychological research is the important idea that NLP is an epistemology and a methodology, not a single theory or model.

“NLP is an epistemology and a methodology, not a single theory or model.”

Once these issues are understood, it will be possible to engage in useful and interesting research into the field of NLP, through adhering to a number of factors which are important in designing good research into NLP.

Sharpley’s (1984) Review: “NLP Is Unsupported in The Literature”

The first review of the experimental literature in NLP by Sharpley (1984) suggested little supportive evidence and a large amount of data opposing the validity of the concept of a PRS.

The review cited 15 studies which were concerned with determining:
  • The presence of a PRS,
  • Adequate methods to discover a PRS,
  • The effects of matching PRS with verbal predicates

However, the review did not consider numerous methodological dimensions which are important in examining and evaluating NLP research.

Einspruch & Forman’s (1985) Critique of Methodological Problems

Einspruch & Forman (1985), in a later review of 39 empirical studies (including all of the 15 initial studies) promoted the idea that NLP research is, in theory, testable and verifiable, but that past research was fraught with methodological confounds such as:

  • A lack of understanding important psychological patterns such as the effect of context;
  • An unfamiliarity with NLP as an approach to therapy;
  • Lack of familiarity with the “meta-model”;
  • Failure to consider the role of stimulus-response associations;
  • Inadequate interviewer training in rapport-building;
  • Logical mistakes

As a result of these criticisms, Einspruch & Forman (1985) classified all 39 studies as unreliable and concluded that it was “not possible at this time to determine the validity of either NLP concepts or whether NLP-based therapeutic procedures are effective for achieving therapeutic outcomes,” and that “only when well-designed empirical investigations are carried out may we be assured of NLP’s validity as a model of therapy.”

Bearing in mind New Code principles, specifically the significance of framing and context, at least two of these can be considered useful guiding principles in evaluating past research and planning future research.

1. Lack of understanding patterns of behaviour and control for confounding contextual factors surrounding the application of NLP models

Einspruch & Forman (1985) recognised that people behave in a way that is dependent upon the environment – the context – within which they find themselves. 15 studies were criticised for not taking this into account. An additional pattern not incorporated into studies was that of individual calibration; people are systematic with their own behaviour but it is potentially erroneous to assume that behaviour is systematic across any given population of people, and thus, procedures and interventions should be applied at the individual level and calibrated to each person’s characteristics.

2. Failure to consider the role of stimulus-response associations

Einspruch & Forman (1985) also understood that the basic concept of association was a powerful influencer of behaviour, and that people were likely to respond based on the associations they had learned, if the experimental settings did not control for this possibility.

These, and other examples of methodological errors in past research, can be summed up by using the following dimensions, which are core concepts to NLP as a discipline, and well recognised in psychological literature:

  • Associations influence processing
  • Features of a context (contextual markers) function as influential associations
  • Language creates task demands , serving as a contextual marker which provides a frame, thereby influencing responses

It is important to understand these concepts in order to conduct useful research.

Einspruch & Forman (1985) made other important and often misunderstood observations:

  • Representational systems are an important part of NLP, but are only one model within NLP
  • Researchers should be adequately trained in NLP so that the procedures and interventions generated can be used within the presuppositions contained in the model

Sharpley’s (1987) Reply

Although Einspruch & Forman did develop some useful methodological distinctions, Sharpley (1987) objected to the resultant dismissal of valuable research and produced a doubly-binding suggestion that either:

  • The past experimental results a lack of conclusive effects, or that
  • The procedures examined were not able to be adequately assessed.

Sharpley (1987) considered that not all of the criticisms Einspruch & Forman (1985) raised were reasonable, and resulted in a somewhat unnecessary dismissal of research.

For example, the criticism that experimenters had a “lack of familiarity with the meta-model” because they measure nominalisations such as anxiety, ease, empathy and hostility (used to discount the validity of 3 studies), can be upheld only insofar as the particular psychometric inventories used can be questioned as reliable.

Einspruch & Forman rejected the validity of 7 studies for an “unfamiliarity with NLP as an approach to therapy,” based on the criticism that asking questions about a client’s past displayed a misunderstanding of the nature of NLP as a generative as opposed to an archaeological approach. This appears to be a rather extreme view and is not necessarily true, in that theoretically, basing experiments around questions in a client’s past is not enough in itself to invalidate the testing of NLP.

Sharpley (1987) critisizes their exclusion of 12 studies because “researchers were not adequately trained in NLP” by commenting that “it is difficult to accept that none of the studies were performed by persons with enough of an understanding to perform the various procedures that were evaluated.” Unfortunately this is difficult to confirm or deny, given that up until now, there has been no officially recognised standard for NLP practitioners.

Sharpley (1987) cited a further 7 studies which showed no evidence in support of the concept of a PRS. However, these studies also contained various methodological issues. Elich, Thompson, & Miller (1985) and Graunke & Roberts (1985), for example, are among those studies discussed as examples of flawed methodological design.

Sharpley’s Conclusions

Some of Sharpley’s (1987) conclusions (which mirror Elich, Thompson, & Miller’s, 1985), provide similar conclusions to the current model of representational systems:

(a) PRS may change over time,

(b) It is not certain that PRS exists,

(c) PRS may merely reflect current language style,

(d) PRS may be heavily influenced by language.

While these conclusions are similar to our current position on representational systems – which mirror John Grinder’s – the methodology used to arrive at them leaves a great deal to be desired and points directly to factors to consider in any analysis of current research, and during the design phase of future research projects.

NLP as a Field and a Discipline

Unfortunately Sharpley and others have made a common mistake, revealing a failure to recognise that NLP is a methodology for creating outcome-oriented models, in considering that the PRS constitutes one of the basic tenets of NLP.

PRS is one aspect of a single model of NLP, and as such it is a mistake to assume that lack of evidence for one model within the field of NLP in any way compromises the validity or utility of subsequent NLP research, or on the field of NLP as practised currently.

(Note: If you would like to learn more about the NLP as a field you can, claim your free copy of , our e-book ‘Neuro-Linguistic Programming; An Overview of the Field’, by Chris Collingwood, NLP Trainer. For a limited time only, here).

Sharpley concludes by putting NLP “in the same category as psychoanalysis, that is, with principles not easily demonstrated in laboratory settings but, nevertheless, strongly supported by clinicians in the field.”

Our proposal is that NLP can and should be researched, quantitatively and qualitatively.

Requirements For Future Research

If adequate research is to take place, concise, highly specified and empirically verifiable descriptions of the models of NLP need to be in place.

In order for this to happen there must be an agreement upon the definitions within the NLP community.

The Graduate Certificate NLP is the first step towards establishing an accepted standard, and along with Grinder & Bostic-St.Clair, Inspiritive and its collaborators are moving towards developing agreed-upon vocabulary and explicit definitions of the patterns and models of NLP.

If you would like to take part in this project, please contact Inspiritive. This is a vital endeavour if the field of NLP is to reclaim its roots within the Cognitive Sciences and consolidate its establishment as a valid field of endeavour, study and research.


Bostic St. Clair, C., & Grinder, J. (2001). Whispering In The Wind. Scotts Valley, California 950666: J & C Enterprises.

Einspruch, E. L., & Forman, B. D. (1985). Observations Concerning Research Literature on Neuro-Linguistic Programming. Journal of Counseling Psychology, 32(4), 589-596.

Elich, M., Thompson, R. W., & Miller, L. (1985). Mental imagery as revealed by eye movements and spoken predicates: A test of neurolinguistic programming. Journal of Counseling Psychology, 32(4), 622-625.

Graunke, B., & Roberts, T. (1985). Neurolinguistic programming: The impact of imagery tasks on sensory predicate usage. Journal of Counseling Psychology, 32(4), 525-530.

Sharpley, C. F. (1984). Predicate matching in NLP: A review of research on the preferred representational system. Journal of Counseling Psychology, 31(2), 238-248.

Sharpley, C. F. (1987). Research findings on neurolinguistic programming: non-supportive data or an untestable theory? Journal of Counselling Psychology, 34, 103-107.

About the Author

Richard Thompson, BSc. (Cognitive Science), is a Graduate of Exeter University, and is a freelance writer and web consultant. He holds the Graduate Certificate in NLP and enjoys receiving responses to his work.

Article content copyright 2006. Richard Thompson. All rights reserved.

All other material copyright 2006. INSPIRITIVE Pty Ltd. All rights reserved

Related articles

Learn more about NLP by reading our Ultimate Compendium of NLP

If you found this article useful please share.

Subscribe Now!

Stay Up-to-Date with Our Latest Courses and Special Offers

Stay in the loop with new course releases and opportunities by completing our form. Never miss out!

What would you like to be updated on?