21
CLASSIC STUDY : Baddeley (1966b)
Aim: to investigate the influence of acoustic and semantic word similarity on learning
and recall in STM + LTM, through a lab experiment involving word sequences.
Procedure : pps men and women from Applied Psychology Research Unit subject panel,
were divided into 4 list conditions ( independent groups design)
List A : 10 acoustically similar
List B : 10 acoustically dissimilar
List C : 10 semantically similar
LIst D : 10 semantically dissimilar
1) Each of 10 words presented via a projector ( 1 every 3 s ) and copied
2) Distraction test : memorise 6 8-digit sequences, then recall word list in order
Repeated over 4 learning trials
3) 15-min interference task : self-paced 8-digit copying
4) Surprise retest on word list sequence
Results: recall of acoustically similar words was worse than acoustically dissimilar words in the
initial phase
Semantically similar words more difficult to learn and recall than semantically dissimilar words
in surprise retest
Conclusion : STM relies heavily on acoustic encoding, as acoustically similar words more
difficult to recall in the initial phase
21
, 22
LTM is largely semantic as later retest recall of List C ( semantically similar ) was impaired
compared to other lists, because :
- LTM is good at recalling semantic differences as it sorts things into meaningful categories
- In List C, there was no semantic difference between words that had the same meaning,
giving LTM “nothing to hold on to”.
Generalisability
Pps used 72 pps who were all British male + female volunteers
Any anomalies (people with unusually good or bad memories) “averaged out” in this
sample
However, many conditions in this study that each group only had 15-20 people in it
= any anomalies might skew results /make a difference to scores with small
numbers
LTM works the same for people from all countries, speaking all languages
= this sample is probably representative.
A volunteer sample might have more people with particularly good memories who enjoy
doing memory tests - not representative of people in general.
Experimental techniques used in lab research aren’t typical of the way memory is used
in an everyday context ( lists of random monosyllabic words not often learnt).
Ability to generalise these findings to everyday contexts is questioned, although to
understand memory and isolate the concerned aspects of memory :
22
CLASSIC STUDY : Baddeley (1966b)
Aim: to investigate the influence of acoustic and semantic word similarity on learning
and recall in STM + LTM, through a lab experiment involving word sequences.
Procedure : pps men and women from Applied Psychology Research Unit subject panel,
were divided into 4 list conditions ( independent groups design)
List A : 10 acoustically similar
List B : 10 acoustically dissimilar
List C : 10 semantically similar
LIst D : 10 semantically dissimilar
1) Each of 10 words presented via a projector ( 1 every 3 s ) and copied
2) Distraction test : memorise 6 8-digit sequences, then recall word list in order
Repeated over 4 learning trials
3) 15-min interference task : self-paced 8-digit copying
4) Surprise retest on word list sequence
Results: recall of acoustically similar words was worse than acoustically dissimilar words in the
initial phase
Semantically similar words more difficult to learn and recall than semantically dissimilar words
in surprise retest
Conclusion : STM relies heavily on acoustic encoding, as acoustically similar words more
difficult to recall in the initial phase
21
, 22
LTM is largely semantic as later retest recall of List C ( semantically similar ) was impaired
compared to other lists, because :
- LTM is good at recalling semantic differences as it sorts things into meaningful categories
- In List C, there was no semantic difference between words that had the same meaning,
giving LTM “nothing to hold on to”.
Generalisability
Pps used 72 pps who were all British male + female volunteers
Any anomalies (people with unusually good or bad memories) “averaged out” in this
sample
However, many conditions in this study that each group only had 15-20 people in it
= any anomalies might skew results /make a difference to scores with small
numbers
LTM works the same for people from all countries, speaking all languages
= this sample is probably representative.
A volunteer sample might have more people with particularly good memories who enjoy
doing memory tests - not representative of people in general.
Experimental techniques used in lab research aren’t typical of the way memory is used
in an everyday context ( lists of random monosyllabic words not often learnt).
Ability to generalise these findings to everyday contexts is questioned, although to
understand memory and isolate the concerned aspects of memory :
22
