Memory for Word Lists
You will hear six short lists of words, read at a steady pace, and then take a memory test on what you heard. It takes about five minutes. Listen carefully and try to remember as many words as you can; you will not need to write anything down.
Please do this before reading about it. The explanation appears below when you finish.
The lists
Your results
Every list
Your answers stay in this page only. Nothing is saved; reloading or closing the page erases them.
What the demo shows
Roediger, H. L., III, & McDermott, K. B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(4), 803–814. https://doi.org/10.1037/0278-7393.21.4.803
Deese, J. (1959). On the prediction of occurrence of particular verbal intrusions in immediate recall. Journal of Experimental Psychology, 58(1), 17–22. https://doi.org/10.1037/h0046671
Each list was made of the words people most often give when asked to free-associate to one word, its theme word: bed, rest, awake, tired, dream … for sleep. In most lists the theme word itself was never read. The question is whether you “heard” it anyway.
Roediger and McDermott, building on Deese (1959), found that students wrote down the unread theme word on 40% of lists when recalling them, and on 55% with longer lists. On a later recognition test the unread theme words were called old about as often as words that had actually been read (84% against 86% in Experiment 1), and students were sure they had heard them 58% of the time. In Experiment 2 most of these false recognitions came with a sense of remembering the word being spoken, not just a feeling that it was familiar. They noted that “wily graduate students who knew we were trying to induce false memories” showed the effect quite strongly too.
The usual explanation has two parts. Hearing many close associates activates the theme word, often enough that it comes to mind while you listen. At test, a word that came to mind that easily, and that fits the list so well, is hard to tell apart from a word you heard: memory for the source of the thought fails. The theme words of lists you never heard are the comparison. They are just as common and just as “themey,” but without the list they are rarely called old.
Two of your six lists did include their theme word. That way the rule “reject every theme word” does not work: to get those items right you have to actually remember whether you heard the word. Compare how often you said you heard the theme words that were read with how often you said you heard the ones that were not.
How this demo differs from Roediger and McDermott’s experiments
- The lists come from their Appendix (15 words each, in their order): the six lists of Experiment 1 (chair, mountain, needle, rough, sleep, sweet) and two more (window, doctor). Each run uses all eight: four are heard without the theme word, two are heard with it, and two are not heard at all. Which lists go where is random.
- Including the theme word in some lists is not part of their design. It was added so that students who know about the effect cannot simply reject every theme word. In those lists the theme word is the 8th word, and the list’s last word is left out to keep 15 words.
- They used 16 lists in Experiment 2 and 6 in Experiment 1, read aloud by the experimenter or from tape at about one word per 1.5 s. Here 6 lists are spoken by a computer voice at the same rate, or shown on screen.
- Their students recalled lists before the recognition test (all lists in Experiment 1, half in Experiment 2). Recall raised later false recognition, so this demo skips it and compares your rates with their lists that were followed by arithmetic instead of recall: 65% “old” for studied words, 72% for unread theme words, 11% for words from unheard lists and 16% for theme words of unheard lists.
- The test uses the 4-point confidence scale from their Experiment 1 instead of the remember–know judgments from Experiment 2. It tests 24 words: the theme word and two list words (the 1st and 10th) from each of the eight lists, in random order, with each theme word placed after at least one word from its list.