Research on learning by osmosis?

[marshwiggle]

This is inspired by another thread, and ideas like "teaching across the curriculum".
Does anyone know of serious research indicating the effectiveness (or not) of using examples, etc. from one subject are in another to teach both subjects?

In other words, is there any solid evidence that using data from some subject area in something like math or statistics actually results in significant learning of both the skill area (math or stats) and the example area (basketweaving)?

[mamselle]

Ummm...we do it in the arts all the time.

M.

[marshwiggle]

Ummm...we do it in the arts all the time.

M.

So is there research to show that it is more effective than subjects being siloed? (Or, more commonly, if one subject area provides skills to be used in a different subject area, the "skills" course is a prerequisite for the course in the other subject area.)

[Parasaurolophus]

FWIW, almost twenty years ago, the entire elementary and secondary curriculum in our home province was restructured to promote this kind of hands-on interdisxiplinary learning (definitely not by osmosis, however!). I imagine the move was empirically informed, and that thefe's much more data now. It has a name, which I've forgotten--it'd be fairly easy to find data if I could, however.

I bet Puget has a better idea.

[mamselle]

Project Zero comes to mind:

   http://www.pz.harvard.edu/

M.

[Puget]

FWIW, almost twenty years ago, the entire elementary and secondary curriculum in our home province was restructured to promote this kind of hands-on interdisxiplinary learning (definitely not by osmosis, however!). I imagine the move was empirically informed, and that thefe's much more data now. It has a name, which I've forgotten--it'd be fairly easy to find data if I could, however.

I bet Puget has a better idea.

You called?

First up OP, I don't think "osmosis" means what you think it does. Teaching across the curriculum does not equal learning by osmosis in any way.

Assuming you actually mean to ask if there is a benefit to using examples from one discipline to teach concepts in another-- there is certainly evidence that abstract concepts (e.g., in math, stats, logic) are understood and learned better when concrete, familiar examples are used, because it reduces the working memory demands of the problem by making the pieces easier to represent, among other things.  It is also incontrovertible that people learn best when they are engaged and motivated. So to the extent that the assignments pair math with something the students find more concrete, familiar and interesting, it will help with learning. And of course there may also be benefits to helping them learn to apply quantitative reasoning to whatever the example topic is-- maybe not to poetry, but to many real-world topics.

[Liquidambar]

there is certainly evidence that abstract concepts (e.g., in math, stats, logic) are understood and learned better when concrete, familiar examples are used, because it reduces the working memory demands of the problem by making the pieces easier to represent, among other things.  It is also incontrovertible that people learn best when they are engaged and motivated.

Do you have a source for either of these, especially the former? I was making basically this point to a colleague today, who did not agree with me.  I'm especially interested in the context of in-major courses.  My colleague seems to believe that our majors, who he thinks are the only students that matter, will have the appropriate level of interest regardless of what we use as examples.

[dismalist]

Economics is about life. That's as cross disciplinary as one can get.  Most students and other people hate it.

Trust me.

[Hibush]

As an osmosis nerd, I have to point out that osmosis is the movement of water across a semipermable membrane from a solution of lower concentration to one of higher concentration (or osmolality to be precisely nerdy).

Therefore, if a student were to apply osmosis to knowledge by putting a textbook under their pillow when they sleep, the morning would find them with a shriveled brain and a soggy pillow.

Research on such "learning" would probably not get IRB approval. Maybe that is why so little is published.

[mamselle]

;--}

M.

[Puget]

there is certainly evidence that abstract concepts (e.g., in math, stats, logic) are understood and learned better when concrete, familiar examples are used, because it reduces the working memory demands of the problem by making the pieces easier to represent, among other things.  It is also incontrovertible that people learn best when they are engaged and motivated.

Do you have a source for either of these, especially the former? I was making basically this point to a colleague today, who did not agree with me.  I'm especially interested in the context of in-major courses.  My colleague seems to believe that our majors, who he thinks are the only students that matter, will have the appropriate level of interest regardless of what we use as examples.

The research I know is basic cog psych research, not STL research, though I suspect there is some.

One heavily studied example is the Wason selection task (https://en.wikipedia.org/wiki/Wason_selection_task), in which participants must select which cards to turn over to validate a conditional rule. In it's abstract form (e.g., if there is a vowel on one side there must be an odd number on the other, with 4 cards one each with a vowel, an consonant, an even number and an odd number visible), very few participants get it right, but using a familiar context ("If someone is drinking beer, they must be at least 21"), most participants get it right. Similar context effects have been found for deductive reasoning and algebra problems.

This is going back to stuff I did in grad school, and I haven't really kept up on more recent developments in this area, but here are a few somewhat older citations:

Beller, S., & Spada, H. (2003). The logic of content effects in propositional reasoning: The case of conditional reasoning with a point of view. Thinking & Reasoning, 9, 335–378. doi:10.1080/13546780342000007

Griggs, R. A., & Cox, J. R. (1982). The elusive thematic- materials effect in Wason's selection task. British Journal of Psychology, 73, 407–420. doi:10.1111/j.2044- 8295.1982 .tb01823.x

Koedinger, K. R., Alibali, M. W., & Nathan, M. J. (2008). Trade- offs between grounded and abstract representations: Evidence from algebra problem solving. Cognitive Science, 32, 366–397. doi:10.1080/03640210701863933

Koedinger, K. R., & Nathan, M. J. (2004). The real story behind story problems: Effects of representations on quantitative reasoning. Journal of the Learning Sciences, 13, 129–164. doi:10.1207/s15327809jls1302_1

[marshwiggle]

First up OP, I don't think "osmosis" means what you think it does. Teaching across the curriculum does not equal learning by osmosis in any way.

The situation that prompted my question was from Florida's rejection of math textbooks "due" to CRT, where there was an example of algebra questions being used to find answers to multiple choice questions about Maya Angelou. So it wasn't that that her life or writings produced examples of mathematics; the math was absolutely arbitrarily attached to the factoids about her life. With apologies to biologists and chemists,  "osmosis" was the best term I could come up with for a process which involved no logical connection between the two different subject areas. I'm open to suggestions....

Assuming you actually mean to ask if there is a benefit to using examples from one discipline to teach concepts in another-- there is certainly evidence that abstract concepts (e.g., in math, stats, logic) are understood and learned better when concrete, familiar examples are used, because it reduces the working memory demands of the problem by making the pieces easier to represent, among other things.  It is also incontrovertible that people learn best when they are engaged and motivated. So to the extent that the assignments pair math with something the students find more concrete, familiar and interesting, it will help with learning. And of course there may also be benefits to helping them learn to apply quantitative reasoning to whatever the example topic is-- maybe not to poetry, but to many real-world topics.

Sure; that's pretty rational. Examples that naturally arise in a subject area make the analysis being taught have a clear and useful purpose. In the example above, the math was completely arbitrarily linked to the topic of Maya Angelou, and in fact appears as a pointless hoop for students to jump through to answer a completely unrelated question.

Is there any evidence of such an approach having any pedagogical value?

[mamselle]

Sure, the New Yorker crossworld puzzle.

In the past few months their clues and keywords have been much more global, and as a result, I've learned about many people and places I didn't know about before.

Why not include Angelou in a word problem (or a crossword puzzle, for that matter?) Is there some rule against using well-known poets for setting mathematical tasks to students? if we're as concerned as we all say we are about literacy, encouraging knowledge of a poet would make sense in a student setting.

M.

[marshwiggle]

Sure, the New Yorker crossworld puzzle.

In the past few months their clues and keywords have been much more global, and as a result, I've learned about many people and places I didn't know about before.

Why not include Angelou in a word problem (or a crossword puzzle, for that matter?) Is there some rule against using well-known poets for setting mathematical tasks to students? if we're as concerned as we all say we are about literacy, encouraging knowledge of a poet would make sense in a student setting.

M.

The problem is, for students who don't like math, and think it is pointless, the approach of introducing it as a hoop to be jumped through to find the answer to a question about some other topic reinforces the notion that math has no legitimate purpose, and is just a meaningless obstacle to "real learning".

I've actually heard of teachers in elementary or middle school, who themselves disliked math, telling students that they would never have to do this in real life. I can't imagine a better way to de-motivate students.

(The New Yorker example illustrates the point. If you like doing crosswords, then the learning is a bonus. If you don't, the crossword is an annoying impediment.)

[ciao_yall]

I tried to do my dissertation on the topic of the effectiveness of WAC on remedial college students. I found nothing quantifiable or replicable. Still, everyone thought it was a good idea.

I did one study which showed that remedial students passed WAC classes at the same rate as prepared students. The only difference was that their average grade was a B-minus while the prepared students' average was B-plus. Once remedial students completed their first-year English course, they actually passed and graded the same as prepared students did. I wanted to publish but the PI didn't want to for various internal political reasons.

That said, WAC programs are costly and thus very political. My initial dissertation topic was going to be analyzing the success rates of a state university with a diverse population that had added a WAC program within the last 5 years or so, creating a perfect natural experiment and a large data set.

However, my friend who ran the WAC program got pushed out by the Teaching and Learning Center over a turf war, so my dissertation ended up being about the laws and sausages, as it were, of WAC programs.