Does Proofreading Boost Learning?

​​I’ll be honest about my book Learn Better. There were some spelling and grammar mistakes. If you get the hard copy, you can find the gaffes if you spend a few minutes hunting around, and I made sure to fix all the issues for the paperback edition.

​​​That all said, I was surprised about an email that I got from a book reader some weeks ago:


​My English teacher grades rather harshly and the only reason I have a good grade is because of the tests and quizzes about your book. I currently have an 88 in the class, and my teacher is offering extra credit, around 2 points worth that will give me an A if anyone can find the most grammatical errors in your book. I know personally your book is more than grammar and spelling but it would mean the world to me if you can help me get those two points.


​It’s hard to know what the teacher’s intent is with the extra credit offer described in the email.


​But the idea raises an interesting question: Is finding these mistakes a good way of learning the content? Does proofreading lead to deep reading?


​I looped in my colleague Ben Keep for some help, and the answer to this question is, basically, no.


To understand why the answer to this question is no, it would help to consider some learning research, and the problem is that looking for minor mistakes turns attention away from the meaning and toward the form.


​The classic research on this idea is from 1975. Through a series of experiments, two researchers showed how our memory for words depends upon how we process those words. When their research participants paid attention to the meaning of the words (by trying to fit the word into a sentence), they remembered the words better than when they paid attention to the form of the words (like whether the word was capitalized or what the word rhymes with).


This research helped establish a “depth-of-processing” account for learning: meaning is deep, form is shallow. The study complemented decades of earlier memory research on words, sentences, and even faces. All show the same basic principle. For example: thinking about how “economic” a word is (in other words, thinking about meaning) beats copying words or crossing out vowels (paying attention to form).


This “depth-of-processing” account has gotten more complex over the years. How deeply you can process something also has to do with what you already know. A child just learning to read is paying attention to form — she has to because it’s still very unfamiliar to her. An experienced reader has the luxury of reading words automatically and can focus more on meaning. This tradeoff in what we pay attention to is also most acute under conditions of high cognitive load — when our brain is working really hard to understand something.


If you’ve ever answered a reading comprehension question by matching up the words of the question with the words in the passage, then you’ve used shallow processing. Say the question asks about “state actors” and “economic forces.” I might just go to the part of the passage that has the words “state actors” and “economic forces,” and simply repeat whatever the passage says in my answer. This is paying attention to form.


That’s part of the reason why teachers sometimes ask students to “put things in your own words.” The idea is that students will understand the passage, internalize it, and be able to summarize it in their own way. Of course, what often happens is that the student replaces a couple of words with synonyms, moves a couple of clauses around, and calls it a day.


To be fair, some kinds of grammar mistakes are relevant to meaning. Catching something like “‘eats, shoots, and leaves’ should be ‘eats shoots and leaves’” would likely result in deep processing because it’s identifying the intended meaning.


​But lots of grammar mistakes aren’t like this — and certainly not the ones that I make: spelling mistakes, miscapitalized words, using a comma when you should use a semicolon. These are (usually) just about what the convention happens to be. And it (usually) doesn’t take deeply understanding the material to notice these mistakes.


Asking students to edit a chapter (rewrite sentences, change paragraphs around, add relevant material, remove irrelevant material, etc.) would probably be better than asking students to proofread a chapter (list all of the spelling and grammar errors). Editing is about clarifying the meaning; proofreading is about polishing the form.


So, if the purpose of the teacher’s extra credit assignment is to learn more about the content of the book, then it’s a bad idea. Students are more likely to remember that I misspelled “context” on page 53 than the point of chapter 5.


But if the purpose is to get practice at proofreading, then — you guessed it — it’s a fine thing to do. It’s probably not the best text to practice on because it’s all the same book, written in the same tone, and pitched to the same audience. So there’s just not much variation. Also (fingers crossed), there’s not that many mistakes to find.


What are some other things to do that would encourage deep reading? Here’s some ideas from Ben:


  • Identify surprising or non-intuitive ideas in the material. This is a form of elaboration and helps avoid the “I knew it all along” trap.
  • Determine whether a variety of teaching practices follow the recommendations laid out in the book (or not). This approach asks students to apply knowledge to practical settings, which also leads to deeper understanding of the material.
  • Summarize each chapter. Bonus points if students do this a few weeks after having read it, and without looking back at the book (this takes advantage of the mechanism of retrieval). Extra bonus points if students use a different medium to do so, instead of just a paragraph of text (this takes advantage of the value of alternative representations).
  • Create a concept map of the book as a whole. Bonus points, again, if you make this part of retrieval practice.
  • Teach the material to someone else. This requires more work than just summarizing, because students have to think about how to structure the learning experience, avoid misconceptions, determine what the most important points are, etc.
  • Come up with a list of tensions in the work (and try to reconcile them). Nearly all research grapples with contrasting ideas. Is the distinction between form and meaning really tenable? When does one cross over to the other? Debating these questions leads to deeper situational understanding.
​As for the student who wrote the email, well, it seems like he may not have been that interested in deep learning in either case.


​At the end of his note, the student wrote:


Is it possible you have a list of where the grammatical/spelling errors are? If not I just wanted to say thank you for writing this book as it is in my top 5 all-time and I have read a number of books in my short lifetime.


In other words, he reached out to get the answer key, and even without science, I can say: That approach does not help learning.
-Ulrich Boser and Ben Keep

Leave a Comment

Your email address will not be published. Required fields are marked *