classic study: the influence of acoustic and semantic similarity on ltm for word sequences (Baddeley, 1966b)

Question 1

what does acoustic mean?

Answer 1

• sound

Question 2

what does acoustic similarity mean?

Answer 2

• words that sound alike

Question 3

what does semantic mean?

Answer 3

• meaning

Question 4

what does semantic similarity mean?

Answer 4

• words that have similar meaning

Question 5

what did Baddeley want to test?

Answer 5

• whether stm and ltm were different or whether they existed on a continuum
• from prev research, he knew that stm for word sequences is worse for acoustically similar sounds than semantically similar words
• so he said that stm is acoustically encoded - by sound

Question 6

aim of study

Answer 6

• to find out if ltm encodes acoustically or semantically
• Baddeley thought if ppts struggle to recall word order it suggests ltm is confused by similarity which means that this is how ltm tends to encode (by meaning rather than by sound)

Question 7

what was the type of exp used?

Answer 7

• a laboratory exp was designed to test order recall of acoustically (sound alike) and semantically (similar meaning) word lists

Question 8

what did the sample consist of?

Answer 8

• a mix of men and women volunteers
• from Applied Psychology Research Unit Subject Panel, Cambridge
• each group consisted of approximately 20 ppts

Question 9

what was the iv?

Answer 9

• acoustic similarity/semantic similarity

Question 10

what is the dv?

Answer 10

• score of recall test of 10 words
• words must be recalled in correct order

Question 11

what were the ppts assigned to?

Answer 11

• they were assigned to one of the four conditions as an independent groups design

Question 12

how many word lists were used?

Answer 12

• four lists of 10 words were used (one list for each condition)

Question 13

list A:

Answer 13

• 10 acoustically (sound) similar words
• man, cab, can, cap, mad, max, may, cat, map

Question 14

list B:

Answer 14

• 10 acoustically (sound) dissimilar words
• pit, few, cow, pen, sup, bar, day, hot, rig, bun

Question 15

list C:

Answer 15

• 10 semantically (meaning) similar words
• great, large, big, huge, broad, long, tall, fat, wide, high

Question 16

list D:

Answer 16

• 10 semantically (meaning) dissimilar words
• good, huge, hot, safe, thin, deep, strong, foul, old, late

Question 17

what were the lists?

Answer 17

• lists B and D acted as control groups for list A and C
• also Baddeley was careful to ensure that all the words were one syllable
• each ppt only gets one list

Question 18

what were the lists presented by?

Answer 18

• each list of 10 words was presented by slide projector
• words were presented in the correct order one at a time for three seconds

Question 19

what happened after the presentation?

Answer 19

• the ppts were given an ’interference task’ where six eight-digit (numbers) sequences were read out (eg 64239173) at one per second
• ppts had eight seconds to write down the numbers
• interference task ensured the ppts couldn’t keep the word in STM by rehearsing

Question 20

what happened after the interference task?

Answer 20

• ppts were allowed 1 min to write down the 10 words in order they were presented
• repeated 4 times (same list 4 times) - independent measures

Question 21

what happened after the four learning trials?

Answer 21

• ppts were given 15 min interference task which involved copying 8 digit number sequences at own pace
• after this, ppts were given surprise retest on word list order

Question 22

what were the results of the study?

Answer 22

• recall of acoustically similar sounding words (list A) and acoustically dissimilar sounding words (list B) were very similar
  
  • shows that acoustic encoding didn’t affect ltm recall
• however, ppts found semantically similar words (list c) was much worse than semantically dissimilar words (list d)
  
  • recalled significantly fewer similar words in retest

Question 23

what were the conclusions of the study?

Answer 23

• results showed ltm learning was affected by meaning of words - evidence that ltm uses semantic coding (coding by meaning)
• baddeley showed that coding in ltm is different from coding in stm (stm = acoustic coding, ltm = semantic coding)

Question 24

which statistical test did Baddeley use to test whether his results were significant or not?

Answer 24

• Baddeley used the Mann-Whitney U test because:
  
  • Baddeley was looking for a difference
  • the data collected was at least interval level – being the number of words recalled in the
    right order from 10
  • the experimental design was independent groups

Question 25

why do people do worse in the semantically similar condition if ltm is good at semantic stuff?

Answer 25

- ltm is good at noticing **semantic differences** - it helps us remember things that stand out - **similar meanings** get grouped together and confused (eg screwdriver, hammer, axe) - **different meanings** go into different mental boxes (eg apple) - **odd things** are easier to remember because **they don’t fit**

Question 26

generalisability - strength + CA

Answer 26

- large sample = 72 ppts - **any anomalies** (unusual results) get averaged out - makes results more **reliable** - easier to **generalise** - **CA**: 4 conditions - each group had only **15-20 ppts** - small groups mean **less reliable results** - one **unusual person** may affect scores - harder to trust or generalise findings

Question 27

reliability - strengths

Answer 27

- **lab exp** with careful controls - same display time for each word - equal **recall time** for all ppts - reduces impact of **ev** - helps establish **cause and effect relationship** - **standardised procedure** - all ppts did same learning trials - same **interference task and retest** for all ppts - makes study **reliable**, easy to repeat and **replicable**

Question 28

application - strength

Answer 28

- shows differences between **stm and ltm** - helps us understand how memory works - useful for students when **studying** - **mind-maps** are a good revision tool - **organising** info into **categories** helps **ltm storage**

Question 29

validity - weakness + CA

Answer 29

- study was done in a lab, an **artificial setting** - doesn’t reflect real-life memory use - we don’t usually learn word lists and recall them in order - **low ecological validity** means it’s **harder to generalise** to real life - **CA**: added a **surprise "forgetting" trial** - ppts weren’t expecting it - this is more like real life, where **recall isn’t usually planned**

Question 30

validity and reliability - strength

Answer 30

- results were measured **scientifically and objectively** - **quantitative data** collected - number of words recalled correctly out of 10

Question 31

ethics - strength

Answer 31

- ppts were volunteers from the **Applied Psychology Research Unit** - gave **consent** to participate (but **not informed consent**) - could **withdraw** from the study at **any time**

Question 32

validity - strength

Answer 32

- improved **internal validity** - used **controls** to **reduce bias** - ppts recalled **word order**, not just words - words were **displayed the whole time** - reduced **risk of familiarity** or associations affecting recall