We all forget things – but are the things that we do remember as accurate as we’d like to believe?
Mind gamers: How good do you reckon your memory is? We might forget things from time to time, but the stuff we do remember is pretty accurate, right? The trouble is, our memory isn’t as infallible as we might want to believe, and you can test this for yourself using the simple experiment below.
Try it yourself
Take a look at the three lists above, and read each word for about a second.
All done? Great.
Now we’re going to do a simple recognition test – below is another list of words for you to look at. Without looking back, note down which of them appeared in the three lists you just scanned. No cheating!
Top – Chair – Sleep – Seat – Slow – Strong – Yawn – Mountain – Sweet
If you said that top, seat and yawn were in the lists, you’re spot on. Likewise, if you think that slow, sweet and strong didn’t appear anywhere, you’re also right. What about chair, mountain and sleep though? They sound like they should have been in the lists, but they never made an appearance. Some of you may have spotted this, but a lot of people tend to say, with a fair amount of certainty, that the words were present.
False recognition and the DRM paradigm
This experiment comes from a classic 1995 study by Henry L. Roediger and Kathleen McDermott at Rice University in Texas. Based on earlier work by James Deese (hence the name Deese-Roediger-McDermott, or DRM, paradigm), participants heard a series of word lists, which they then had to recall from memory. After a brief conversation with the researcher, the participants were then given a new list of words. Critically, this new list contained some words that were associated with every single item on each of the initial lists – for example, while sleep doesn’t appear on list 3 above, it’s related to each word that does appear (bed, rest, tired, and so on).
Next, the participants had to say how confident they were that the words in the new list had appeared previously. Roediger and McDermott’s results showed that people claimed to recognise the associated words (like sleep) about as often as words that were actually presented on the list – around 85% of the time. In other words, people were claiming to remember things, fairly confidently, that hadn’t happened.
There are a number of reasons why this effect occurs. One suggestion put forward by Roediger and McDermott relates to something known as associative processes; because all of the words in a given list are related to each other, they are more likely to activate related words in our memory. Word stem completion tasks help to highlight this point. If I say the word ‘beer’, and ask you to fill in the blank here:
_ I N E
You might be more likely to say W, even though D, F, L and M would all be perfectly acceptable. That’s because beer and wine are related concepts, and saying one makes the other easier to recall in memory.
Of course, other factors come into play with the DRM paradigm. It might also happen because you’re thinking of the word sleep when you read the related list, which leads you mistakenly think that you actually did read the word later on. Regardless though, studies of false recognition make one thing clear: our memories aren’t always as accurate as we’d like to believe.
guardian.co.uk © Guardian News & Media Limited 2010