What has been tested most
- Representational systems and eye-accessing cues: the foundational claims, tested across decades — the empirical record is poor. Our journal piece on eye-accessing cues details this
- Preferred representational modality as a stable trait: little supporting evidence
- Specific techniques (swish, anchoring as standalone interventions): thin direct research; laboratory analogues of associated procedures show mixed results
What adjacent research supports
Pieces of NLP's practical territory are supported by neighbouring literatures: mental practice and imagery have real effects in sports-science research; expectancy and framing effects are robust in social psychology; solution-focused and motivational-interviewing approaches — cousins of NLP's outcome orientation — have decent clinical-trial records; rapport and mimicry research supports parts of the pacing story. None of this "proves NLP"; it means some components sit on respectable science while the original theoretical core does not.
What has barely been studied
- NLP practitioner training as a whole package — almost no controlled research exists
- Coaching-with-NLP outcomes in field settings
- The modelling methodology itself — NLP's core process has essentially no research literature
IONI's reading of this record
Hold the claims to the evidence: drop the eye-cue mind-reading, drop "scientifically proven" marketing, keep the components that work in practice and are plausible in mechanism. That position costs marketing punch and buys credibility. The journal's is-NLP-scientific piece sets out the full reasoning.