MEMORY
Coding, capacity and duration of memory
Coding
The format in which information is stored in memory.
Sensory memory: Iconic memory (visual information from the eyes stored as images). Echoic
memory (auditory input from the ears stored as sounds).
STM: Baddely visually presented students with letters one at a time. He found that letters which
were acoustically similar were harder to recall. This suggests that STM mainly encodes things
acoustically.
LTM: Baddely presented lists of 10 short words one at a time, some being semantically similar.
They were tested on their recall immediately then after 20 minutes. After this delay, they
recalled worse on semantically similar words. This suggests that coding in LTM is done
semantically.
Evaluation
Baddely’s study identified a clear Used artificial stimuli rather than
difference between two memory stores. meaningful material. For example, the
Despite some exceptions, Baddely’s word lists had no personal meaning to
ideas that STM codes acoustically and participants, so his findings may not tell
LTM codes semantically has not failed us much about coding in other
over time. This is important in situations like everyday situations.
understanding the memory system, Findings from this study have limited
which led to the multi-store model. application.
,Capacity
The amount of information held in a memory store.
Sensory memory: unlimited.
STM: Jacobs used digit span to measure capacity. The researcher, for example, gave four
digits, and if participants recalled these correctly in the right order, this would increase to five,
and keep increasing until the participant can’t recall correctly. The mean span digit span was
9.3, and the mean letter span was 7.3.
Miller, after observing everyday practices, concluded that most things come in 7s. He
concluded that STM can hold ‘7 plus or minus 2’. On average, the capacity of STM is
5-9 items.
Miller found that the capacity of STM could be extended by chunking – combining
separate bits of information into larger chunks.
LTM: Unlimited.
Evaluation
Jacobs study has been replicated. Miller’s research may have
Jacobs findings have been confirmed overestimated STM capacity. Cowan
by other controlled studies such as reviewed this and concluded that
Bopp and Verhaeghen, which is good capacity may only be 4, plus or minus
as it is an old and early study so the one, chunks. The lower end of Miller’s
results may have been estimate is more appropriate.
underestimated. Jacobs study is a valid
test of digit span in STM.
, Duration
The length of time information can be held in memory.
Sensory: Sperling gave participants a grid of digits and letters for 50 milliseconds. They were
either asked to write down all 12 items or write down an indicated row after a tone immediately
after exposure. When asked to recall the whole thing 42% recalled 5 items, whereas when
asked to recall one row 75% recalled 3 items. This suggests information decays rapidly in the
sensory memory store and therefore has a limited duration of milliseconds.
STM: Peterson and Peterson got students to recall 3 letter combinations after longer and
longer intervals, in which they were prevented from rehearsing the letters by doing a counting
task. After 3 seconds, the average recall was 80%. After 18 seconds, the average recall was
3%. As the delay got longer, the recall got worse. Findings suggest that duration of STM is 18-
30 seconds unless rehearsed.
LTM: Up to a lifetime. Bahrick studied 392 American participants aged 17-74. They tested
recall using high school yearbooks by either a photo recognition test consisting of 50 photos, or
a free recall test of recalling names of all their graduating class.
Photo recognition 15 years after graduation: 48 years after graduation:
90% accuracy 70% accuracy
Free Recall 15 years after graduation: 48 years after graduation:
60% accuracy 30% accuracy
Evaluation
High external validity. Researchers Peterson and Peterson’s study used
investigated meaningful memories. stimulus material that was artificial.
When studies on LTM were conducted Recalling consonant syllables does not
with meaningless pictures to be reflect most everyday memory
remembered, recall was lower. activities where we are trying to
Bahrick’s findings reflect a more ‘real’ remember meaningful information.
estimate of the duration of LTM. Lacks external validity.
Coding, capacity and duration of memory
Coding
The format in which information is stored in memory.
Sensory memory: Iconic memory (visual information from the eyes stored as images). Echoic
memory (auditory input from the ears stored as sounds).
STM: Baddely visually presented students with letters one at a time. He found that letters which
were acoustically similar were harder to recall. This suggests that STM mainly encodes things
acoustically.
LTM: Baddely presented lists of 10 short words one at a time, some being semantically similar.
They were tested on their recall immediately then after 20 minutes. After this delay, they
recalled worse on semantically similar words. This suggests that coding in LTM is done
semantically.
Evaluation
Baddely’s study identified a clear Used artificial stimuli rather than
difference between two memory stores. meaningful material. For example, the
Despite some exceptions, Baddely’s word lists had no personal meaning to
ideas that STM codes acoustically and participants, so his findings may not tell
LTM codes semantically has not failed us much about coding in other
over time. This is important in situations like everyday situations.
understanding the memory system, Findings from this study have limited
which led to the multi-store model. application.
,Capacity
The amount of information held in a memory store.
Sensory memory: unlimited.
STM: Jacobs used digit span to measure capacity. The researcher, for example, gave four
digits, and if participants recalled these correctly in the right order, this would increase to five,
and keep increasing until the participant can’t recall correctly. The mean span digit span was
9.3, and the mean letter span was 7.3.
Miller, after observing everyday practices, concluded that most things come in 7s. He
concluded that STM can hold ‘7 plus or minus 2’. On average, the capacity of STM is
5-9 items.
Miller found that the capacity of STM could be extended by chunking – combining
separate bits of information into larger chunks.
LTM: Unlimited.
Evaluation
Jacobs study has been replicated. Miller’s research may have
Jacobs findings have been confirmed overestimated STM capacity. Cowan
by other controlled studies such as reviewed this and concluded that
Bopp and Verhaeghen, which is good capacity may only be 4, plus or minus
as it is an old and early study so the one, chunks. The lower end of Miller’s
results may have been estimate is more appropriate.
underestimated. Jacobs study is a valid
test of digit span in STM.
, Duration
The length of time information can be held in memory.
Sensory: Sperling gave participants a grid of digits and letters for 50 milliseconds. They were
either asked to write down all 12 items or write down an indicated row after a tone immediately
after exposure. When asked to recall the whole thing 42% recalled 5 items, whereas when
asked to recall one row 75% recalled 3 items. This suggests information decays rapidly in the
sensory memory store and therefore has a limited duration of milliseconds.
STM: Peterson and Peterson got students to recall 3 letter combinations after longer and
longer intervals, in which they were prevented from rehearsing the letters by doing a counting
task. After 3 seconds, the average recall was 80%. After 18 seconds, the average recall was
3%. As the delay got longer, the recall got worse. Findings suggest that duration of STM is 18-
30 seconds unless rehearsed.
LTM: Up to a lifetime. Bahrick studied 392 American participants aged 17-74. They tested
recall using high school yearbooks by either a photo recognition test consisting of 50 photos, or
a free recall test of recalling names of all their graduating class.
Photo recognition 15 years after graduation: 48 years after graduation:
90% accuracy 70% accuracy
Free Recall 15 years after graduation: 48 years after graduation:
60% accuracy 30% accuracy
Evaluation
High external validity. Researchers Peterson and Peterson’s study used
investigated meaningful memories. stimulus material that was artificial.
When studies on LTM were conducted Recalling consonant syllables does not
with meaningless pictures to be reflect most everyday memory
remembered, recall was lower. activities where we are trying to
Bahrick’s findings reflect a more ‘real’ remember meaningful information.
estimate of the duration of LTM. Lacks external validity.
