clarinetgirl
Posts: 2
Joined: Fri Sep 23, 2005 8:19 pm

### Logic Puzzles

I am doing a project on how musical training is related to logical reasoning, but I don't have any idea of what kind of logic puzzles to use. Could someone please suggest some? FYI, I am in eighth grade, and I am using middle school students for my subjects.

Also, is walking around at lunch and asking the first people I see if they are willing to participate random enough?

EDS
Former Expert
Posts: 67
Joined: Thu Nov 18, 2004 4:23 am

### Re: Logic Puzzles

Hi Clarinetgirl -

Sounds like a neat project.

I'm not sure what sort of puzzles are most likely to yield interesting results. You might consider a battery of several different kinds of puzzles. For example, a spatial reasoning test, a boolean logic puzzle, an ordering (or "Einstein") puzzle, a straightforward math problem, a math word problem.

If you do something like that, try to choose puzzles that won't be ruined if someone has seen similar puzzles before.

There's a nice wikipedia article on logic puzzles that may get you started.
http://en.wikipedia.org/wiki/Logic_puzzle

Standardized tests and IQ tests are another resource to get a feel for the kind of puzzles that are possible. Whether the puzzles on such tests measure what they claim to measure is debatable, but they certainly measure how well someone performs on a particular kind of logic puzzle.

You may have no choice but to pick people at lunch and ask them to participate. If so, try to get people to agree to participate in the test before you tell them any details about it. Say, "would you be willing to spend ten minutes participating in an anonymous, harmless psychological experiment for a science fair project," rather than, "would you like to take a logic puzzle test?"

The danger is that a student's willingness to participate in the test may be correlated with how well they do. For example, you may find that among the people you don't know, only those who really like logic puzzles volunteer, while among the band kids, everyone volunteers whether or not they like logic puzzles because they're your friends. That sort of thing would make data analysis difficult.

Adding a couple survey questions about how familiar your subjects are with different kinds of logic puzzles might be useful in your data analysis.

Finally, one caution: when you're writing up your conclusions, be careful about claiming to have found causal relationships. Say you find that all the music students do better on a particular puzzle. Does that mean that the musical training made them better at the puzzle, or does it mean that people who are good at that kind of puzzle are more likely to choose music class? There's no way to tell from your experiment.

If you had time, you could try to answer that by testing people over the course of several years during which some of them participate in music training and some do not. That's often referred to as a longitudinal study. But, for a science fair project, that sort of thing isn't feasible. All you can say is that you've found a correlation.

Good luck,
Erik
--
Erik Shirokoff
Science Buddies