Duration

• This concerns how long information can be held in each store. The duration of the STM has been investigated by Peterson and Peterson (1959). They gave participants trigrams to remember. They were given one trigram at a time and then asked to count backwards in threes from a random three digit number. This was to prevent rehearsal. They did this for either 3, 6, 9, 12, 15 or 18 seconds. They found that recall became poorer the longer the retention interval. After 3 seconds recall was 80% accurate, this dropped to 50% after 6 seconds and less than 10% after 18 seconds.
• duration of the LTM researched by Bahrick et al (1975) who investigated memories of school friends. They asked 392 American participants to complete no. of tasks designed to assess who they could remember from high school (such as free recall of as many names as possible or looking at photographs to say who they recognised).
• Those who had graduated in last 15 years had an accuracy of 90% when recognising photographs, but even those who had graduated after 48 years had a 70% accuracy rating. Free recall was not as accurate as photo recognition. This suggests that LTM has a potentially unlimited duration for some material (particularly if there is a cue).
• A limitation of Peterson and Peterson's (1959) study into the duration of the STM is that it lacks external validity due to its low ecological validity. This is because the stimuli used to assess the duration of the STM did not have any relevance to everyday life. They were just nonsense trigrams. This is a limitation because the findings may not be relevant to the STM of stimuli that we encounter in everyday life.
• On the other hand, a strength of Bahrick et al's (1975) research is that it has higher external validity. This is because they were investigating memories that we may actually try to recall throughout various points in our lives. These memories can be considered as more meaningful and therefore more representative of things we would try to recall in our day-to-day life. This is a strength because the findings are more applicable to scenarios outside of the research context.