The Chromozone / IGCSE Psychology / Topic 1: Memory and forgetting
Murdock's study and your own investigation (lesson 1 of 3)
Share this page with your students: it is all they will see. Students who have joined your class can also open it on their own once the class has moved past this lesson.
Starter (4 marks) 5 min
Answer these three recap questions in your notebook:
- In the multistore model, decay causes forgetting in two stores. Name the two stores. (1.3.3) (1)
- Two students revise the same topic. One reads the textbook chapter three times; the other writes "why" questions about it and answers them. Using the levels of processing model, explain which student is likely to remember more. (1.2.6) (1)
- Confidentiality and privacy are two separate ethical guidelines. Explain how they differ. (7.7.1) (2)
Reveal the starter answers
1. Decay (1)
- The sensory register and short-term memory (both needed) (1)
2. Two ways to revise (1)
- The student who answers "why" questions: thinking about meaning is deep (semantic) processing, which gives a stronger memory than rereading (1)
3. Confidentiality and privacy (2)
- Confidentiality: keeping participants' data anonymous, for example numbers instead of names (1)
- Privacy: not making participants reveal unnecessary personal details (1)
Class test: fifteen words and a hands-up tally 6 min
Everyone takes part. Your teacher reads the words, and the class builds one picture of what it remembered.
- Predict. Write down the three positions in a 15-word list that you think the class will recall best.
- Listen. Your teacher reads 15 words, one every two seconds. Just listen: do not write anything yet.
- Recall. When your teacher says "Write", write down every word you remember, in any order. You have 60 seconds .
- Count. Your teacher reads the list again in order. Put your hand up for each word you wrote down; the number of hands goes into the tally on the board.
| Position | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Hands up |
Then discuss:
- Which positions did the class remember best? Was your prediction right?
- Our class picture comes from one list. Give one reason why it might still not show the true pattern.
Reveal the answers
- (a) Usually the first two or three words and the last three or four, with the fewest hands for the middle words: a U shape. Any honest description of the tally is right.
- (b) With only one list, one easy or striking word can stand out wherever it is; and one class is a small group of similar people.
- Murdock tested 103 people on many lists, so chance effects like these cancel out.
Murdock (1962): aim and procedure
Aim: to investigate the serial position effect in memory: does the position of a word in a list affect how likely it is to be recalled?
Participants: 103 psychology students, male and female.
Procedure:
- Each list had 10 to 40 common words, read aloud one at a time.
- Every word was presented for the same amount of time: one or two seconds, depending on the list.
- Straight after the list, participants had 90 seconds to write down as many words as they could, in any order. This is free recall.
- The test was repeated over several days, with a new list each time.
Task: our demonstration and Murdock's study 3 min
Copy and complete the table. Then answer the question below it.
| Feature | Our demonstration | Murdock (1962) |
|---|---|---|
| Participants | ||
| Words in a list | ||
| Number of lists | ||
| Time for each word | ||
| Time to recall |
Choose two differences that make Murdock's results more trustworthy than ours, and explain each one.
Reveal the answers
- Participants: our class; Murdock had 103 psychology students.
- Words in a list: 15; Murdock 10 to 40.
- Number of lists: one; Murdock many lists, a new one each day over several days.
- Time for each word: two seconds; Murdock one or two seconds.
- Time to recall: 60 seconds; Murdock 90 seconds.
- 103 participants: a few people with unusual memories make little difference to the overall pattern.
- Many lists: a pattern found in every list cannot be caused by a few easy words.
Key words: the serial position effect
- Serial position effect
- How well an item is recalled depends on its position in a list.
- Primacy effect
- Information presented at the beginning is better remembered than information presented at a later stage.
- Recency effect
- Information presented at a recent stage is better remembered than information presented at an earlier stage.
Murdock's results

Recall depended on position, so a graph of the results is U-shaped: the serial position effect.
- Primacy effect: the first three or four words were recalled more often than the middle words.
- Recency effect: the last eight or so words were also recalled more often.
- Words in the middle of the list were recalled least.
- The same U shape appeared for every list length. A longer list just had a longer low middle.
Quick check: read the curve 2 min
Look at the curve in the box above.
- Which part of the curve shows the primacy effect, and which part shows the recency effect?
- Murdock's lists had 10 to 40 words. Why is it more useful to show the percentage of words recalled than the number?
- Give one way the curve is like our class tally, and one way it is different.
Reveal the answers
- (a) Primacy: the high left end (the first few positions). Recency: the high right end (the last positions).
- (b) Percentages let you compare lists and groups of different sizes: 5 out of 10 and 20 out of 40 are both 50%.
- (c) Alike: recall is often higher at one or both ends. Different: our tally comes from one class and one list, so it is bumpy; the curve is smooth because it combines many people and many lists.
Murdock's conclusion: evidence for the multistore model
Murdock concluded that the two ends of the curve come from two different stores.
- The first words had time to be rehearsed, so they were transferred from short-term memory (STM) to long-term memory (LTM). They are recalled from LTM.
- The last words were still in STM when recall began, so they could be written down at once.
So STM and LTM are separate stores. This is the argument for the multistore model (MSM) on the syllabus (1.3.4).
Forgetting in the MSM: why words are lost
Murdock's middle words were heard but not recalled. The MSM explains forgetting in a different way for each store (1.3.3).
- Decay (sensory register and STM): information fades if it is not processed in time. In STM, a word that is not rehearsed is lost within about 18 to 30 seconds.
- Displacement (STM): STM holds only 5 to 9 items, so each new word pushes an earlier one out. In a 15-word list, the middle words are pushed out before they can be rehearsed.
Quick check: decay or displacement? 1 min
Match each sentence to decay or displacement.
- "By the time I had to write, the first words had faded because I had stopped repeating them."
- "Each new word seemed to push the one before it out of my head."
Reveal the answers
- A: decay: a word that is not rehearsed fades from STM.
- B: displacement: STM is full, so new words push old ones out.
Forgetting in the MSM: words stored but not found
A word can also be lost from long-term memory.
- Access problems (LTM): the word is still stored in LTM, but it cannot be retrieved without the right cue.
- It may come back when you see the list again: the list is the cue.
Task: which kind of forgetting? 4 min
For each situation, name the kind of forgetting (decay, displacement or an access problem) and explain it using the MSM.
- Suppose word 8 in our class test was heard by everyone but remembered by very few.
- A researcher repeats Murdock's procedure, but participants count backwards in threes from 100 for 30 seconds before they start writing. Which end of the curve changes, and why?
- A participant cannot recall the first word of the list, but recognises it at once when she sees the list again.
Reveal the answers
- (a) Displacement: later words pushed word 8 out of STM before it could be rehearsed into LTM.
- (b) The recency effect shrinks or disappears; the primacy effect stays. The last words were only in STM: without rehearsal they decay, and the numbers displace them. The first words were already in LTM.
- (c) An access problem: the word was stored in LTM, but she could not retrieve it without a cue. Seeing the list is the cue.
When researchers tried (b), the recency effect did disappear while the primacy effect stayed.
The argument against: depth of processing
- In the MSM, information reaches LTM through rehearsal: the more a word is repeated, the better it should be remembered.
- The levels of processing model (1.2) says memory depends on how deeply a word is processed. Thinking about its meaning (deep, semantic processing) gives a stronger memory than noticing how it looks or sounds (shallow processing).
- Craik and Tulving (1975) found exactly this, with the same time spent on every word.
- The MSM has no way to explain this, because it only counts rehearsal. This is the argument against the model on the syllabus (1.3.5).
Predict, then explain: repeat it or picture it? 4 min
Two groups hear the same 20-word list at the same pace, as in Murdock's study.
- Group R is told: "Repeat each word in your head until the next one arrives."
- Group P is told: "Picture each word and link it to the word before, as a little story."
- Which group does the multistore model predict will recall more middle words? Why?
- Which group does the levels of processing model predict will recall more? Why?
- Suppose group P recalls more middle words. Explain why this is a problem for the multistore model.
Reveal the answers
- (a) Group R, or no difference: the MSM says rehearsal moves words into LTM, and group R rehearses more.
- (b) Group P: picturing and linking each word is deep (semantic) processing, which gives a stronger memory.
- (c) Group P rehearsed less but remembered more, so rehearsal is not the only way into LTM. The MSM cannot explain depth of processing.
Common misconceptions and exam errors
- Naming the effect but not the store. "Murdock found primacy and recency" is only a start. Say which store explains each end and why: primacy from LTM after rehearsal, recency from STM.
- Saying the same point twice. "STM and LTM are separate" and "they are two different stores" earn one mark, not two.
- "Repeating something always makes it stick." Repetition is shallow processing, so it may not help. That is the depth of processing point against the MSM.
Exam-style question (4 marks) 5 min
Dev's teacher says, "Murdock's study of the serial position effect is the best evidence for the multistore model of memory."
Explain why Murdock's (1962) study supports the argument for the multistore model of memory (MSM). (1.3.4) (4)
Reveal the mark scheme
Question (4)
1 mark for each relevant point (× 4)
- The primacy effect: the words at the start of the list were recalled well (1)
- They were recalled from LTM, because they had time to be rehearsed (consolidated) before the middle words arrived (1)
- The recency effect: the words at the end of the list were recalled well (1)
- They were recalled from STM, because they were still there when recall began: no later words had displaced them (1)
- The two effects have different causes, so STM and LTM are separate stores (the MSM) (1)
- The words in the middle were recalled least: they were neither rehearsed into LTM nor still in STM (1)
Max 4. Do not credit a point that only repeats an earlier one (for example, "STM and LTM are different" said twice).
Stretch: long lists and short lists (3 marks)
Murdock used lists from 10 to 40 words long. In every list, the last eight or so words were recalled well. Use the multistore model to explain why the low middle part of the curve was much longer for a 40-word list than for a 10-word list. (1.3.2, 1.3.4) (3)
Reveal the mark scheme
1 mark for each relevant point (× 3)
- STM has a limited capacity (about 5 to 9 items), so only about the last eight words can still be there at recall, whatever the length of the list (1)
- Only the first few words get enough rehearsal to reach LTM, before STM fills up (1)
- In a 40-word list about 28 words fall between these two parts, but in a 10-word list there are almost none, so the dip is much longer (1)
Plenary: learning objectives check 1 min
Rate yourself red, amber or green for each objective.
- Describe the aim, procedure, results and conclusion of Murdock's (1962) study. (1.3.4)
- Explain how Murdock's results support the multistore model. (1.3.2, 1.3.4)
- Explain why the multistore model cannot explain depth of processing. (1.3.5)
- Red: I cannot do this yet.
- Amber: I can do some of it.
- Green: I can do all of it.
If you have time (10 marks)
Exam-style questions for when you finish early: about 10 minutes.
1. A student reads a list of 10 common words aloud, one every two seconds, to 25 classmates. Straight afterwards the classmates write down as many words as they can in any order. The table shows how many of the 25 classmates recalled the word at each position in the list.
| Position | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|
| Classmates | 20 | 16 | 12 | 9 | 7 | 6 | 8 | 12 | 17 | 21 |
- Calculate the percentage of classmates who recalled the word at position 10. (1)
- Calculate the mean number of classmates who recalled the words at positions 4 to 7. (1)
- Identify the effect shown by the high recall of the words at positions 1 and 2. (1)
- Outline why the words at the end of the list were recalled from short-term memory (STM). (1)
2. Mia reads a list of 30 words to a group. She says that the word at position 14 was recalled by very few of them.
- Explain, using the multistore model, why the word at position 14 was unlikely to be recalled. (2)
- Mia later shows the group the list. Several people then say they knew the word at position 14. Identify the type of forgetting this suggests. (1)
3. Ravi revises by reading his notes over and over. Kai writes down what each idea means and links it to his own life. Both spend the same time.
- Suggest which student will remember more, and explain why. Use the levels of processing model. (2)
- Identify the level of processing Ravi is using when he reads his notes over and over. (1)
Reveal the mark scheme
Question 1(a) (1)
1 mark for calculation
- 84% (1)
(21 ÷ 25) × 100 = 84
Question 1(b) (1)
1 mark for calculation
- 7.5 (classmates) (1)
(9 + 7 + 6 + 8) ÷ 4 = 30 ÷ 4 = 7.5
Question 1(c) (1)
1 mark for identification
- Primacy effect (1)
Do not credit recency effect.
Question 1(d) (1)
1 mark for outline
- The last words were still in STM when recall began (so they could be written down at once) / had not yet been displaced or lost from STM (1)
Max 1.
Question 2(a) (2)
1 mark for each relevant point (× 2)
- STM holds only 5 to 9 items (a small capacity) (1)
- Each new word pushed earlier words out (displacement) before the word at position 14 could be rehearsed and transferred to LTM (1)
Question 2(b) (1)
1 mark for identification
- An access problem (the word was in LTM but could not be retrieved without a cue) (1)
Do not credit decay or displacement.
Question 3(a) (2)
1 mark for suggestion
1 mark for explanation
- Kai will remember more (suggestion) (1)
- He processes the meaning of the ideas (deep or semantic processing), which gives a stronger memory than the shallow repetition used by Ravi (explanation) (1)
Kai with no reason: Max 1.
Do not credit "repetition always makes it stick".
Question 3(b) (1)
1 mark for identification
- Structural (shallow) processing (1)
Do not credit semantic.