Hello Bagpuss. Have a look here. You'll find plenty of stuff on both sides of the fence. As thread creep seems inevitable here it's probably worth saying that when Bandler and Grinder started their work they studied individuals who they felt were - in their terms - effective communicators. They realised that they could break down the components of their performance and teach them to others. They found that if the parts were broken down into small segments thay could be described and quantified and patterns seen for each individual.

The basis of training using NLP, therefore, is to identify the patterns found in excellent performers and transfer these patterns to individuals wishing to emulate their performance.

I wish when I'd been at school my teachers had known more about learning style preference, for instance. Kolb, Honey and Mumford are not part of the NLP movement (for want of a better word) but their work in this respect comes from more or less the same direction - although more scientifically based and, as I understand it is being used more and more in mainstream education.

For more evidence of mumbo-jumbo about, in this case, multiple intelligence - again coming from the same general direction - being used successfully, have a look at this, about an elementary school in Indianapolis.

Sorry to go on - I just knew I shouldn't engage on this!!