For the purpose of deriving meaningful numbers for scoring I make it a rule never to handle the pieces until I try them to see if they fit. This is partly to practice the skill of memorizing the shapes and color patterns regardless of their relative orientations. For bifocal reasons (I cannot clearly see the details across a table) I occasionally reorient the whole table, puzzle and all. As the assembled puzzle grows I transfer pieces to make room without reorienting them. This I don't regard as cheating. When I try a piece and find my expectations were wrong I score myself an error and I count and record the number of pieces assembled at that point. Because of this system of scoring I never assemble any sub-groups. When I complete a puzzle I have a record of the numbers of pieces assembled versus the number of errors. This ratio is a measure, of sorts, of my skill and the difficulty of the puzzle. Buried in this figure are the other unknown variables relating to how much I happen to know about pictures in general and this particular picture - and the nature of jigsaw puzzles and how they happen to be cut.
My contact with working and making jigsaw puzzles dates from about 1911. My work in illustrating as well as architecture and painting has always involved visual activity. Working puzzles seemed, as a kind of rationalization, a way of practicing both short-term and long-term visual memory. While I am drawing or painting from nature the act of remembering what I see, from the time when I observe until I commit the memory to paper or canvas, is an exercise of the short-term memory. In the matter of working 'from memory' I am exercising my long-term memory. In the puzzle analogue the corresponding memories could be called very-short-term and longish-short-term memory. Quite naturally I am most interested in the visual acuity and visual memory. Therefore, as an analogue, it is visually slanted but, as I see it, the essential memory character is not impaired as to its generality. Visual ideas may be just as original and peculiar as any other kind.
One of the benefits of these rules is that they make the puzzles last longer. Most of the commercially available puzzles contain directions for working them which include the suggestion that the worker consult the picture (usually reproduced on the box) and then to segregate the pieces into color groups, edge groups or similar categories. These rules are designed to reduce the frustrations of the uninitiated. I prefer to work in just the opposite way. I arrange to have my wife purchase the puzzle, open the box and turn the pieces out onto the table top. I avoid seeing the puzzle altogether until I have finally assembled all the pieces. Since I am in no hurry to finish the puzzle I do not attempt to score the time I work. For one thing, because of the incidental cogitation it is practically impossible to assess the time actually devoted to working the puzzle.