Category Archives: Working Memory

Making Grammar Memorable

I have been thinking a little too much about the differences between the students who get it and those who do not. I was reminded of this basic distinction when I was reading through another one of Jonah Lehrer’s posts on the nature of expertise and the research he cited by a few folks at various universities.

One of the behaviors that research pointed to as essential to learning well was the ability to tolerate boring and repetitive practice without distractions. Lehrer cited to work of K. Anders Ericsson at the University of Florida and Angela Duckworth at the University of Pennsylvania who have both done a great deal of research on memory development.

Ericsson has worked with music performers to find out why some excel while others languish in mediocrity, and Duckworth led research that focused on students in the National Spelling Bee to find out how some students nailed those impossible words while others made it there but received nothing more than a congratulation and a pat on the back.

Ericsson found that the best musicians tended not to be those born to perform but those who spent hours and hours in their rooms practicing by themselves. This reminded me of a story told by Paco de Lucia’s father about how the young Paco was not able to go to school, so he ended up staying in his room hour after hour practicing his guitar and listening to records of flamenco guitarists. For musicians, this ability to practice scales to the point of being able to go through them without thinking is part of internalizing complicated movements into something the brains considers to be a single motion.

Duckworth found essentially the same discipline in the top spellers at the National Spelling Bee. They were were not the ones who asked other people to quiz them, and, contrary to what many English teachers would like to believe, they were not the ones who read a great deal. They were the ones who were willing to put up with the drudgery of repetitive practice. They tended to practice even after they had already spelled the word correctly once or twice, and made sure they had, essentially, nailed it.

For anyone who has been following memory research, these findings are nothing new. They are quite consistent with what has been showing up in studies for decades. Research on chess players showed that the ones who were willing to practice hour after hour developed better memories for various strategies and the conditions under which they can be used. The research on chess players showed that practice and not intelligence was the key to being able to learn and employ the various strategies during matches.

In spite of all the mounting research on the utility of repetitive practice, I have not found many books or teachers who put this approach into practice. It seems grammar is already such a boring topic that making parts of it repetitive would push it over the edge and make teaching it a violation of the Geneva Convention.

The issue is determining which parts of grammar instruction should be drilled. According to Ericsson’s and Duckworth’s studies, only those materials that can be reduced to small segments of randomly accessed memory should be learned through repetitive drill. Doing this would allow students to focus on the structural elements without sweating over whether a certain word were a preposition or an article.


Writing and Grammatical Fluency

I struggled a great deal with writing in my early college years and for some time even into graduate school. I would never have written a blog, for example, as a freshman in college because I simply could not write well. I would have loved the idea of having my writing be public, but I know I would have sat and stared at the screen.

I can vividly recall sitting in the office of New University at UCI trying to write up a quick story on some event that happened the day before, but I struggled for an hour to put together about a dozen sentences on a demonstration I had been a part of. It was not a matter of not knowing what had transpired but of not having the phrases and clauses to put the piece together.

Writing well was not so much a matter of having the words well thought out and sounding somewhat intelligent, but was more a matter of just feeling comfortable with the words, phrases and clauses that I needed to compose. The use of grammatical structures was simply something I was not able to undertake with any confidence.

Watching someone write fluently was much like watching someone speak Finnish or Chinese. It just did not make sense to me and seemed so foreign at so many levels.

Now I find it relatively easy to write. I cannot say I write well since I have reread enough of my work to know it is far too often filled with non-sensical phrases and clauses, but writing today is nothing like writing at UCI so many years ago. So what happened over the course of a few years?

I do not know for certain, but I know it was not a matter of learning some concept or following someone else’s directions. I did undertake a good deal of writing while in graduate school in the 1990’s, but I also started writing at message boards a good deal during the 2000s. My penchant for combativeness and argumentation had me posting frequently and at length.

My sense is that I became familiar with using not just certain words but with using certain phrase and clause structures in my writing. Fluency in speech seems to be some thing that is easy enough to recognize, but fluency is writing seems to be something few people discuss. My sense is that it is developed in much the same way that verbal fluency is developed in second language students. Basically, practice breeds internalization of certain grammatical structures.

Teaching grammar cannot help a student develop a written fluency all that much since fluency comes from extensive practice and not a few exercises. In fact, instruction might even increase levels of anxiety over writing since the student can become too aware of the potential for making what others might see as an error.

Misconceptions and Refutations in Grammar Textbooks

Reading Research Quarterly has an excellent study from a group of researchers based at the University of Cyprus that examines the differences between expository and refutation writing for teaching college students about scientific concepts.

The study is fairly involved and has a number of nuances that are beyond the scope of this entry, but basically the researchers take two groups of students and have them read one of two different pieces about how energy works in physics. The researchers picked this topic because energy is something that students often have trouble understanding or come up with misconceptions to explain situations around them that seem to use energy. Many students, for example, see energy as akin to a natural force or believe it has mass that can be seen in a microscope or measured with a scale.

The first group was given an expository text written in much the way textbooks are written while the second group was given a refutation text written in much the same manner as the first but with refutations of common misconceptions about energy at certain key places. The refutation texts were slightly longer but they did not provide extra information, but instead simply debunked common misconceptions about energy.

The abilities of students to understand the core concepts regarding energy were essentially the same in both groups, but the students reading the refutation texts tended to be better at making inferences about energy and its application to situations. This difference was especially pronounced among students who were assessed as having very little knowledge about energy or as having knowledge that was made up primarily of misconceptions. The students who started the study with lower levels of knowledge about energy and physics in general were prone to making poor inferences and were easily confused by situations that required application of the knowledge to a concrete situation. Could you, for example, see the energy within sugar with a microscope?

To me, this was interesting.

Students of grammar come to class with a great number of misconceptions about how the language works and what learning grammar is supposed to involve. Misconceptions seem to be more common and more deeply rooted than actually understandings. There is, of course, the comma rule that is not a rule (“Put a comma wherever there is a pause”); then there is the clear and concise definition of a sentence that makes no sense (“A sentence is a complete thought”); and, there are many many others that students have picked up from well meaning family members and poorly educated English teachers. All these misconceptions add up to a great deal of interference when the students are trying to learn this messy language.

I see the research as having application to grammar instruction not only because of the number of misconceptions students bring with them but also because it showed how useful refutation writing is for developing the ability to make inferences. The reality is that knowing grammar is not all that impressive if it cannot be applied directly to writing more effectively.

Oddly enough refutations of common misconceptions rarely make it into the basic grammar books. There are plenty of books out there that deal with how silly many of the “rules” that English teachers have come up with over the years are, but the books are written by linguists for other linguists and never make it into the textbooks students use.

The teachers tend to use textbooks that answer certain questions and resolve the issues that students bring to class without taking the time to explain how arbitrary many of these “rules” actually are. There are plenty of books by linguists that blow holes right through these “rules” but there are no books that I am aware of that balance the need to teach students how to use the language but make them aware that these rules are really just useful conventions.

Knowledge of these misconceptions has been around for quite some time. Mina Shaughnessy, David Bartholomae, Joseph Williams and many others have written about the nature of “error” in student writing decades ago, but the students do not have their error patterns directly addressed in the textbooks themselves. A more direct approach to the misconceptions students have would make teaching them new concepts all that easier. No major rewrite of the grammar textbooks would be needed. Just a revision, a revision that would incorporate refutations of the many misconceptions students already have on the first day of class.

Some Not So Surprising Things About Reading and Recall

The New York Times has an article summarizing the research of Jeffrey Karpike of the Memory and Cognition Lab at Purdue University. The study Karpike conducted focuses on the effectiveness of different approaches to learning a reading passage on a scientific topic and was published in Science, a well-known scientific journal.

Karpike has put together four different studying scenarios together and measured their effectiveness soon after the study sessions and then a week after the study sessions.

  1. Have the students read the material in five minutes and then move on with their lives.
  2. Have the students read the material four different times for five minutes at a time.
  3. Have the students read the material and then draw diagrams of the material with the reading in front of them.
  4. Have the students read the material and then “test” themselves on it by writing for 10 minutes without the reading in front of them.

The assessments used by Karpike and his team seem to be well constructed since they ask questions that require the students to recall facts from the articles but also to make inferences from the materials. This type of assessment is common practice in reading research because it checks whether the student understands the materials but also whether he or she can apply it to a situation not covered in the reading.

Predictably, the first scenario where the student reads for five minutes and goes about his or her day does not work all that well in the assessment taken right after the study session nor in the assessment taken a week later. The next three scenarios all work fairly well both right after the study session and a week later, but the New York Times and Karpike go out of their way to point out that the fourth scenario works best, especially after a week’s time.

Here is the diagram the New York Times reproduced from the Science article to make this point:

Everything Karpike and his team are doing is well within the practices of the research I have read before. He has also conducted and published many studies quite similar to the one that the New York Times is treating as if it were ground breaking stuff. It is not, and that is a good thing.

I do, of course, have few problems with this article and the implications of it. (This is a blog, after all, so I must have a problem with somebody, right?) The New York Times’ headline writer is misleading by stating that students should quite studying and take a test. The “test” Karpike has the students use is not a question and answer test, but a free write where the student just rehashes as much as he or she can for ten minutes. It is just writing. There is nothing innovative in this approach, and that is a good thing.

Another problem is the one that all teachers confront when they read what appears to be relevant research. How can this be used with the lessons I teach to the students I have? Apparently, we should revise our lesson plans pronto:

“It really bumps it up a level of importance by contrasting it with concept mapping, which many educators think of as sort of the gold standard,” said Daniel Willingham, a psychology professor at the University of Virginia. Although “it’s not totally obvious that this is shovel-ready — put it in the classroom and it’s good to go — for educators this ought to be a big deal.”

I guess I am not sure why research that Karpike has been conducting and publishing regularly for the past ten years or so should suddenly be a “big deal,” especially since it deals with methods most teachers have been using for quite some time. The reality is that Karpike has conducted some solid research over the course of a number of years and deserves to get the attention that comes with being published in Science. He makes no grandiose claims for his work, and that is a good thing.

When I look through the list and try to see how my students might make sense of them, things get different. The first approach of reading for five minutes and then moving on is clearly not going to work as well when the only goal is reading comprehension. The problem is that students use this method most often not because they mistakenly think it works well, but because it is easy and fast. This approach is going to be quite enticing to both lazy and overworked students who, depending on the work ethic, either want to get back to their video games or finish studying for three other courses. Depending on the context, the first approach might be the best one.

Even if writing for ten minutes works better than reading repeatedly or doodling a concept map, it might not be the best approach all the time for all students. For some students, the task of writing bring in too many other issues and anxieties and might interfere with learning the material. And concept mapping is not only useful for learning how to present material to others but is also a good deal easier and more fun for most students. When you have been cramming for a number of classes, a change in approach, especially if it is less taxing on your working memory, might be a better way to go.

Writing, as Karpike points out in the article and other research he has done, is more difficult to do that re-reading or coming up with a concept map. This means it is probably far more taxing on working memory and less likely to be something a student can do over and over again when there are multiple assignments to be completed in a limited amount of time.

In the end, the New York Times article reads much like the articles in car magazines that compare the relative worths of different models of sports cars. They put the cars through idealized testing situations and then quantify the results with clearly defined winners and losers. The problem is that there is usually more to the situation than the numbers can pick up. In this case, writing a summary after reading an assignment is the high horsepower option, but it seems to burn more gas than the other models.

[At this point I was going to connect this article to grammar instruction, but I would rather post a picture of a really cool Alfa Romeo 159 from 1951 that produced 425 hp from a 1.5 liter engine. Yes, folks, that year and those numbers are correct:


The only problem is that it got 1 mpg. And that is not such a good thing.]