My conclusion has much in common with the article of Gobet, with one big difference though.
In my latest research I found 3 different aspects of solving a position:
- Seeing a series of moves in the mind's eye ("blindfold chess")
- Board visualization
If I can't solve a position while looking at a physical board, and allowing myself to move the pieces on a second analysis board, then I can certainly not solve it blindfolded. Because that only makes it more difficult and not less.
On the other hand, if I, struck by a moment of enlightment, all of a sudden forget to suck at reasoning and get a brainwave how to solve the position, wouldn't it be a pity if I screwed that unlikely precious moment up by poor visualization skills?
So the conclusion can be:
Blindfold chess is not a way to improve at chess because it doesn't teach you how to handle difficult positions. At the same time not being able to play blindfolded can get in the way once your reasoning is able to solve the position. It is the reasoning that decides which moves are candidates and their priority.
About boardvision I have not decided yet. I'm inclined to think that that it is a form of luxury, with which a chessplayer should treat himself, just like he would do with a beautiful wooden board and pieces. If there is a beneficial effect from it to use it as steppingstone as suggested in Tisdall's theory has to be investigated yet.
It is usually advised to solve a position in your head, without moving the pieces.
I advocate to commit the heresy not to do it that way. Since that is actually to play blindfold during the reasoningprocess, which lowers the quality of the problems you can handle.
I advise to make 3 seperate exercises from this in stead of to try it at once:
- Exercise of reasoning
- Exercise of blindfold chess (use low level opponents!)
- Exercise boardvision.