Two important differences between LTMs and libraries

  1. Reconstruction: A book in the library is stored as a complete unit-- every word has already been written. All the pages of the book exist whether you check it out or not, and every time you check it out, you are checking out the same book. If and when a book is revised, we know the book has been revised: No librarian would mistake a 2nd edition for a 1st edition.

    An event stored in memory, on the other hand, can be revised without you knowing it--and is often not stored in its entirety. Instead, of an entire event being stored, often only a few bits and pieces of the original event make their way into memory.

    So, why does it seem like you remembered the whole thing when you only stored an incomplete outline of the event? Because when you recall an event, you  use what you know about how the world works (i.e., your semantic memory) to turn that bare bones outline into a vivid memory that seems complete.  (See 17-second animation)

    Four amazing facts about reconstruction: the process of using your knowledge of how the world works to take some bits of stored information to create a "complete" memory.

    1. When you reconstruct a memory, you don't realize that you are "writing" much of that memory: Instead, you think you are just copying what was already written in memory.
    2. You are not very good at knowing which bits of your recall of an event are
    3. What you copied from what was originally written in memory and which bits you just made up. Your not knowing which part you made up makes makes sense given you didn't realize that you made up any of it. However, it creates a problem: As research on eyewitness testimony shows, there is little relationship between how sure you are about a memory and the memory's accuracy (As you will discover when you and your partner both claim to be 100% sure of correctly remembering a detail of an event, but you both remember it differently. Warning: When your partner says, "I know I'm right because I'm sure," pointing out that science shows that their being sure doesn't mean they're right may not help you win the argument.).
    4. You may continue to rewrite your memory every time you recall it, so memories may get revised without you knowing it. One implication:  That 5-foot jump shot you made to win the game may, after telling the story numerous times, be remembered as  a 15-foot jumper.  Or, as basketball great Connie Hawkins said, "The older I get, the better I used to be."
    5. Memories are not copies of reality but instead are re-creations, so memories aren't as accurate as we would like to think (Examples of your memory's inaccuracy). 

    Experimental evidence for reconstruction
    The bad news about reconstruction: Because memory, rather than being like a video recorder, relies on reconstruction, memories that we are confident about can be wrong. Examples:

    The good news about reconstruction: By noting what you can reconstruct and memorizing only what you can't, it can seem like you have remembered everything without memorizing much. If visuals help you, mouse over this text.

    So, oddly enough, the key to seeming like you have memorized more is to memorize less! For example, imagine that a friend tries to memorize every word of a 30-page chapter whereas you boil down that chapter to less than a page of notes from which you can reconstruct the chapter. You will memorize less than a page of notes but appear to know much more than your friend. But can you really summarize an entire chapter in one page of notes? Yes. In fact, one highly paid memory expert advises his clients to end the semester with one page of notes for each course!
    How do you boil down a long chapter to less than one page of notes?
    1. Preview the chapter so you are ready to ask
      1. Which information in the chapter can you reconstruct from what you already know?
      2. What are the main points from which you can reconstruct the rest of the chapter?
    2. Keep in mind that you should be able to reconstruct most of the chapter, so either do not highlight anything--take notes instead--or highlight only the most important information on the page and go back later to try to reconstruct the rest of the page from the few words you highlighted. (Students who use highlighters to paint their books are obviously not being selective and thus are not taking advantage of reconstruction.)
    3. Realize that condensing a chapter down to less than one page of notes may not happen in one step. You may have to rewrite your summary several times, shortening it by (1) eliminating material in the previous summary that is either redundant (e.g., two examples of the same thing) or material that you could reconstruct and (2) replacing relatively narrow principles with broader principles.


  2. Need for overlearning: In a well-organized  library, books don't get misplaced. So, if you retrieve a certain book once, you can always retrieve it (as long as the book is in the library). Your memory, on the other hand, is not so well-organized. As a result, even if you have retrieved a certain piece of information from your LTM before, you may not be able to retrieve that material later. The solution is to overlearn: Study after you already know it. Effective overlearning is practicing retrieval even after you have recalled the information perfectly (Since most people overestimate how much they know, you can't rely on thinking that you could recall the information:  The only way to know that you can recall the information is to actually recall it).

Why do you need to overlearn? You need to overlearn--which should be called "Super Reviewing"-- to defeat the forgetting curve. (Perhaps it was the forgetting curve and the power of overlearning that caused Quintilian to write, "Nothing is so much strengthened by practice, or weakened by neglect, as memory.") Put another way, since retrieval is the big problem in LTM, you need to practice retrieval. For students, this usually means taking practice tests and quizzing each other.  In a sense, practicing retrieval is like mowing the memory's retrieval path. Because of the need for overlearning, you should retrieve newly learned information at least 3 times after learning it (e.g., a day after learning it, a week after learning it, and a month after learning it).

 How can you use overlearning to do better in school? One way is to repeatedly write out answers to essay questions based on course material. If you tire of writing out answers, you could say your answers aloud to a critical friend or to a recording device. Alternatively, instead of writing entire essays, you could outline answers to essay questions or develop rubrics for grading answers to those essay questions (if you have a study partner, you could use your rubric to grade your partner's essays). If sample essay questions aren't available from your text, your professor, or online, you could create your own essay questions or you may be able to get them from the Psychology Problem Solver series.  

Another way to practice overlearning is to repeat parts of lectures to your parents, a stuffed animal, or a recording device. Tutoring also provides a chance to overlearn. Note that because everything you need to overlearn is already in  your head, you can practice overlearning almost any time. For example, you can recite answers to questions or recite the main points of a lecture while walking between classes or while waiting in line.

To 7 Memory Myths

Back to how LTM is like a library

Back to Memory Menu

Back to Lecture Menu

Copyright 2020-2023 Mark L. Mitchell