Spatial Intelligence and Computer Science

If you're weary of the level of the hackneyed "learning styles, yea or nay" blog debate, you may find Mark Guzdial's latest post interesting:

One of the really interesting findings I learned this week was that high spatial ability is a predictor for a major in mathematics and computer science at least in a study of mathematically precocious youth.  I believed that after only a little reflection.  The terms "mapping" and "models" are originally about spatial ideas. The whole idea of arrays and object references (two of the hardest early ideas) are spatial--we talk about "dimensionality" of arrays and matrices, and "directionality" of links.  Everything we do with data structures is a matter of "navigation" from a current position.  We as teachers rely on spatial language and students' ability to "picture" spatial references.

One of the big questions in SILC is how much of spatial ability is malleable and how can it be improved.  They cited a study by Sheryl Sorby who taught civil engineering at Michigan Tech. She gave her students a measure of spatial ability on the first day of class, then encouraged students who had low scores to take a one credit hour course to improve their spatial ability.  She found that students succeeded much better after taking that course.