3.Capacity= how it is stored (the way in which information is represented, eg visual, auditory).
Sperling 1960- Sensory:
Exposed PPs to 3 rows of 4 letters for 50 milliseconds
They remembered 4-5 on average but knew there had to be more than that.
PPs trained to use tones for which line to recall (tone was played after exposure).
PPs remembered average of 3 items, which means at this stage they would've been able to remember 3 from any row as the image was still available in the sensory memory.
Shows duration to be extremely brief fractions of a second- as the information decays & disappears so rapidly.
Is reliable as can be repeated with similar results, but lacks validity.
Peterson & Peterson- STM Duration:
Used nonsense trigrams, eg SPG365 & PPs had to count back in 3s or 4s for a duration of either 3,6,9,12,15,18 seconds.
Found that 90% could recall the trigram after 3 seconds, compared to 2% after 18 seconds- showed duration of STM was 18-30 seconds.
Easy to replicate, clear results (reliable), but lacks ecological validity (artificial) & lacks mundane realism.
Displacement may be a better explanation: new incoming information (counting) was simply displacing old information.
Jacobs- STM Capacity: 1
PPs were read lists of letters or numbers that they had to recall immediately after presentation.
Jacobs gradually increased the length of these digits/ letters until the PP could only accurately recall the information in the correct order on 50% of occasions. Recall had to be in the correct order.
Difference found between capacity for numbers & letters; PPs could recall 9 numbers & 7 letters, & noticed recall seemed to increase with age.
STM has a capacity of between 5 & 9 items of info & as age increases, we appear to develop better strategies of recall.
Jacobs- STM Capacity: 2
Miller & others have also discovered that chunking can increase capacity (BBC or 01858 becomes 1 chunk of info rather than 3 or 5 distinct chunks)- supports Jacobs' theory.
Lacks ecological validity; memory being used for artificial task.
Research is replicable; later studies have supported Jacobs' findings.
Bahrick- LTM:
392 high school graduates asked to remember former classmates, eg recalling names, recalling names from pics or matching names to pics.
PPs recalled remarkably well on recognition rather than recall- 48 years on still had 70% accuracy, but dip in performance after 48 years.
Shows that duration of LTM is unlimited.
Has good ecological validity as testing recall memories & good sample size so can be generalised to a wider population.
Difficult to generalise results to entire population as graduates used bias- no clear conclusions drawn to explain dip in performance.
Baddeley et al- Coding in STM & LTM: 1
PPs presented either: acoustically similar words (map, mad, cat, cap), acoustically dissimilar (pen, cow, day), semantically- meaning- similar (tall, high, broad, wide) or semantically dissimilar (foul, thin, safe).
STM- presented 5 words & recalled immediately, if sounded similar it was more difficult- concluded that STM must be acoustic.
LTM- presented 10 words 4 times, after 20 mins PPS recalled, & acoustically similar had no effect, but they performed worse with words similar in meaning- concluded that LTM is semantic.
Baddeley et al- Coding in STM & LTM: 2
STM= acoustic & LTM= semantic
Clear results, high internal validity, easy to replicate, so reliable.
There's lots of evidence suggesting different types of coding for both STM & LTM, so conclusion is actually inaccurate & memory is coded according to circumstance in which it was formed, rather than where it is stored.