Why Do We Forget

228
Chapter 6
FIGURE 6 .1 0
Ebbinghaus’s Curve of Forgetting
doing
2
100
90
Mean retention
List thirty words, selected at
random from a dictionary, and
by
spend a few minutes memorizing them. After an hour has passed, write
down as many words as you can remember, but don’t look at the original list
again. Test yourself again eight hours later,
a day later, and two days later. Now look
at the original list and see how well you
did on each recall test. Ebbinghaus found
that most forgetting occurs during the
first nine hours after learning, and especially during the first hour. If this was not
the case for you, why do you think your results were different?
learn
Memory
80
70
60
50
40
30
20
10
1 hour
20
minutes
9 hours
24 hours
2 days
6 days 31 days
Retention interval
aloud a list of nonsense syllables, such as POF, XEM, and QAL, at a constant pace, and
then tried to recall the syllables.
Ebbinghaus devised the method of savings to measure how much he forgot over
time. This method compares the number of repetitions (or trials) it takes to learn a
list of items and the number of trials needed to relearn that same list later. Any difference in the number of learning trials represents the savings from one learning to
the next. If it took Ebbinghaus ten trials to learn a list and ten more trials to relearn
it, there would be no savings. Forgetting would have been complete. If it took him
ten trials to learn the list and only five trials to relearn it, there would be a savings of
50 percent.
Ebbinghaus’s research produced two lasting discoveries. One is the shape of the forgetting curve shown in Figure 6.10. Even when psychologists have substituted words,
sentences, and stories for nonsense syllables, the forgetting curve shows the same strong
initial drop in memory, followed by a more moderate decrease over time (Slamecka &
McElree, 1983; Wixted, 2004). Of course, we remember sensible stories better than nonsense syllables, but the shape of the curve is the same no matter what type of material
is involved (Davis & Moore, 1935). Even the forgetting of events from daily life tends
to follow Ebbinghaus’s forgetting curve (Thomson, 1982).
Ebbinghaus also discovered just how long-lasting “savings” in long-term memory
can be. Psychologists now know from the method of savings that information about
everything from algebra to bike riding is often retained for decades (Matlin, 1998). So,
although you may forget something you have learned if you do not use the information, it is very easy to relearn the material if the need arises, indicating that the forgetting was not complete (Hall & Bahrick, 1998).
Why Do We Forget?
IT’S ALL COMING BACK TO ME
This grandfather hasn’t fed an infant for
decades, but his memory of how to do it
is not entirely gone. He showed some
“savings”; it took him less time to relearn
the skill than it took him to learn it
initially.
We have seen how forgetting occurs, but why does it happen? In principle, one of two
processes can be responsible (Best, 1999). One process is decay, the gradual disappearance of the information from memory. Decay occurs in memory in much the same
way as the inscription on a ring or bracelet wears away and fades over time. Forgetting
might also occur because of interference. Through interference, either the storage or
the retrieval of information is impaired by the presence of other information. Interference might occur because one piece of information actually displaces other information, pushing it out of memory. It might also occur because one piece of information makes storing or recalling other information more difficult.
In the case of short-term memory, if an item is not rehearsed or thought about,
memory of it decreases consistently over the course of about eighteen seconds. So decay
appears to play the main role in forgetting information in short-term memory. But
interference through displacement can also be operating. Like a desktop, short-term
229
Forgetting
FIGURE
6.11
Procedures for Studying Interference
To remember the difference between the two types of interference, keep in mind that the prefixes—pro and retro—indicate directions in time.
Pro means “forward,” and retro means “backward.” In proactive interference, previously learned material “comes forward” to interfere with
new learning; retroactive interference occurs when new information “goes back” to interfere with the recall of past learning.
PROACTIVE INTERFERENCE EXPERIMENT
Group
Time 1
Experimental
Learn list A
Control
Time 2
Time 3
Learn list B
Recall list B
Learn list B
Recall list B
Result
The experimental group
will suffer from proactive
interference, and the control
group will be able to recall
more material from list B.
RETROACTIVE INTERFERENCE EXPERIMENT
Group
Time 1
Experimental
Learn list A
Control
Learn list A
method of savings A method for
measuring forgetting.
decay The gradual disappearance of
information from memory.
interference The process through
which storage or retrieval of information is impaired by the presence of
other information.
retroactive interference A cause of
forgetting whereby new information
placed in memory interferes with the
ability to recall information already in
memory.
proactive interference A cause of forgetting whereby previously learned information interferes with the ability to
remember new information.
Time 2
Learn list B
Time 3
Recall list A
Recall list A
Result
The experimental group
will suffer from retroactive
interference, and the control
group will be able to recall
more material from list A.
memory can hold only so much. Once it is full, adding additional items tends to make
others “fall off ” and become unavailable (Haberlandt, 1999). Displacement is one reason why the phone number you just looked up is likely to drop out of short-term memory if you read another number immediately afterward. Rehearsal prevents displacement by continually reentering the same information into short-term memory.
The cause of forgetting from long-term memory appears to be more directly tied to
interference. Long-term memory can be affected by retroactive interference,
in which learning new information interferes with our recall of older information
(Wixted, 2005). Proactive interference can also occur, in which old information
interferes with learning or remembering new information. Retroactive interference
would help explain why studying French vocabulary this term might make it more difficult to remember the Spanish words you learned last term. And because of proactive
interference, the French words you are learning now might make it harder to learn
German next term. Figure 6.11 outlines the types of experiments used to study the
influence of each form of interference in long-term memory.
Does interference push information out of memory, or does it merely make it harder
to retrieve the information? To find out, Endel Tulving and Joseph Psotka (1971) presented people with lists of words that represented a particular category. For example,
there was a “buildings” list (e.g., hut, cottage, cabin, hotel) and a geographical features
list (e.g., cliff, river, hill, volcano). Some people learned a list and then recalled as many
of its words as possible. Other groups learned one list and then learned up to five additional lists before trying to recall the first one.
The results were dramatic. As the number of additional lists increased, the number
of words that people could recall from the original list decreased. This finding reflected
strong retroactive interference; the new lists were interfering with recall of the first one.
Then the researchers gave a second test, but this time they provided a retrieval cue by
telling the category of the words (such as “types of buildings”) to be recalled. Now the
number of additional lists had almost no effect on the number of words recalled from
the original list, as Figure 6.12 shows. These results indicate that the words from the