[EricK]

The threads on pattern recognition and squeezes have got me thinking...

Here's a common scenario. I read a book of declarer problems and struggle on one of them. When I look at the answer I think to myself "Yes, that makes sense". Later (even months later), if I reread the book I will be able to solve the problem, but if I come across a different setting for the same technique in another book I will struggle again.

In other words, I am recognising the exact problem but not recognizing the pattern.

Although I don't have the Bridgemaster programs, I expect the same thing would happen. I would, having seen the answers for any I couldn't do the first time, be able to solve them all, but my technique would not have improved.

What would be nice is a piece of software which was able to construct new examples of the techniques in question, so instead of getting exactly the same problems you got variations of them. Obviously, to be useful, the variations would have to be more than simply swapping suits or low cards.

Does anybody else agree that this would be a good thing?

Could those of you know about programming say whether this is a realisable concept?

Eric