If I can talk about our experience with IQ test, where was the need seek pieces of shapes, important in prediction was sex. Girls were better than boys statistically significant (p < 0,01) in vizual perception. Children were 12, 13 and 14 years old.

Gender...excellent point. Didn't even think about gender. Thanks.

Yes, it sounds like a joke :-) And for us (men) it is not proven, but in children it works. If you need, here is a link: http://www.educationalrev.us.edu.pl/vol/tner_1_2012.pdf

Search term: visual perception

I think is motivation, comprehension of the task, self regulation, memory, attention, comprehension of the goal...

I prefer to share some papers:

http://psp.sagepub.com/content/31/12/1739.full.pdf+html

http://rstb.royalsocietypublishing.org/content/367/1594/1338.full.pdf+html

https://www.lehigh.edu/~sml3/publs/2012_FDG_PCGws.pdf

http://www.dcs.gla.ac.uk/~mounia/Papers/Chi2012.pdf

http://books.google.com/books?hl=en&lr=&id=xNKy-jZEw6YC&oi=fnd&pg=PT387&dq=predict+performance+puzzle+completion+task&ots=1lCdOCSJ13&sig=Oe0_I32eSjO9ROVeNvAmMyHDWgI#v=onepage&q&f=false

http://www.aaai.org/ocs/index.php/AAAI/AAAI12/paper/viewFile/5094/5383

Performance on a task is really sensitive to experience with that task. Is the puzzle similar to anything people might have done before?

It may also be useful to look up research on individual differences on visual search tasks (which your puzzle sounds similar to).

No, puzzle will be pretty new, unless by some chance they saw it before, but you can't avoid that.

I want to stay away from dispositions, random assignment will take care of that. I just want to make sure I don't run into some acquirable skill that the subjects will all have as a function of where the sample is being drawn from.

Even if you have a new disposition/task, you can't avoid previous knowledge from other tasks... how will your run your puzzle? paper?, screen? will people use mouse or tap?, because most of research use classic paper or mouse, but just few use tap and that can add different skills

I think the question differes a little bit from its explanation. In case of the abilities envolved in visual puzzle completion the right hemisphere is more envolved than the left hemisphere. Since the former is more visual, global, deals with colors and spacial locations and seeks harmonization. While the latter is more dealing with words, aural and verbal, logical, sequential and deal with reason. yet in comparing two pictures in case you will look at the main picture only once I think you test the photographical memory, but in case you keep loking and comparing I think it is a matter of both perception and problem solving.

I think it is patience and discipline that is needed too.

Yes, perceptual ability would ultimately be measured. However, other factors that could effect measurement include the subject’s: speed of processing, attentional skills (ability to focus on the task-at-hand), hyperactivity/impulsivity, short-term visual memory skills, eyesight, ability to discriminate color, fine-motor skills (depending on the nature of the test – tactile manipulation of puzzle pieces vs. pointing to responses.), neurological considerations (TBI, low birth weight, sustained high fever, oxygen deprivation, etc.), autistic tendencies, and general cognitive ability.

As for the gender of the subject, it has been my experience that males tend to perform better on non-verbal, visual-spatial oriented tasks, than females. Conversely, females typically perform better on verbally oriented tests of reasoning/cognition.

As such, you may wish to screen test subjects for these conditions prior to administration of your puzzle task in order to account for such variables in your findings.

Sarmila Banerjee(scholar V.B .Santineketan) . Ithink it is Attention and Patience.IQ is also an important factor to be measured.

1.Thinking,memory,concentration ability.

The abilities children apply in this context is apptitude and creativity.Several researches have pointed out that people who have high IQ also have high creativity. Such people can often display value skills that are amazing. However this does not mean that those with high IQ will necessarily manage a task puzzle .Often they might not. Meaning they have a special cognitive ability . This is embeded in the way they perceive the task and the reasoning process applied in problem solving.

As has been mentioned: visual search, but definitely also visuo-spatial construction. This ability is specifically impaired after certain lesions in the parietal lobe. I reccommend looking into this term especially. E.g. look at block construction tests. I would also suppose attention - no tests exists that does not involve attention (lezak says so under "attention" in her main book: psychological testing). I would also suppose IQ, and perhaps sex, where men might be superior, but this is just my hunch.

You guys are great. Thanks for all the suggestions.

There has been a little shift in the task though. Technically it is still a puzzle. But we are leaning to using Sudoku puzzle. We need a complex, independent, cognitive task. So sudoku seems to offer all that. With all our suggestions currently I have the following to consider:

perception

problem solving

attention

short-term visual memory

G

gender

People may have experience with sudoku, but we are restricting the time to complete and are using a puzzle of higher level of difficulty, so they should still have trouble completing within the time allowed.

I should mention that the task completion is of interest based on the condition we experimentally create. Given the random assignment all these things should be taken care of but I would like to collect some measures prior to the task to have a more complete data set. More importantly I would also like to account for something that could be more prevalent in our accessible population, which is in a tech school, that would be different from the general population and thus would have to be considered in terms of external validity.

Thanks again.

You are welcome!,I wish you sucess!

Adults, range from 25-40

Attention,reasoning,view problem as a whole

How about a 3-back task? A lot of Sudoku solving entails remembering what you have already noticed elsewhere in the puzzle.

Yes I think it would fit into 3-back task model. I actually found a study that examined cognitive processes on the Sudoku puzzle. Very interesting stuff. They were saying that difficult Sudoku puzzles do not really differ in the number of empty cells but their solution depends on much more complex strategies, rather than just elimination. Estimating the simultaneous probability of several numbers in a given cell is required. I wonder if it actually would constitute more than 3-back?

Undoubtedly, Sudoku taps much more than a 3-back would. However, if you take it to a 4-back you risk having a good portion of your students at chance levels. I would definitely include a paper-and-pencil stroop, Trails A & B, digit span backward, and maybe a Tower of Hanoi task. I'm not a real believer in cognitive predicter tasks that are so complex you don't know what they are testing.

Yep, thanks. The task is not used here as a predictor. It's actually just something for them to do, to determine how specific type of directions influence a specific type of task behavior. But it has to be something complex cognitively, so Sudoku seems to fit well.

Reasoning ability, intelligence,memory

Visual working memory, spatial reasoning processes, such as mental rotation, and general reasoning processes such induction and analogical reasoning. The relative contribution of each would vary with task complexity, experience, and developmental level.

problem solving abilities,function of memory,reasoning abilities and S-Factor of intelligence.