top of page

Coding, Capacity & Duration

Research on Coding

Coding in STM


Alan Baddeley's 1966 study aimed to demonstrate that short-term memory encodes information acoustically, while long-term memory encodes it semantically.


He presented participants with four types of word lists:

Group 1 – Acoustically similar words, such as cat, cab, can.

Group 2 – Acoustically dissimilar words, such as pit, few.

Group 3 – Semantically similar words, such as great, large, big.

Group 4 – Semantically dissimilar words, such as good, huge, hot.

Participants were asked to recall the words in the correct order after immediately or after a delay of 20 minutes.


Group 3 performed worse in delayed recall, suggesting long-term memory encodes semantically. In immediate recall, Group 1 performed worse, indicating short-term memory encodes acoustically. Baddeley concluded that short-term and long-term memory stores code information differently.


1.One strength of Baddeley's research is that he used a laboratory environment, allowing for the control of extraneous variables. This research has enhanced our understanding of how short-term and long-term memory stores differ.

2. A limitation of Baddeley's research is the artificial nature of the stimuli and lack of task realism. In everyday life, people may process information differently, possibly using semantic coding for short-term memory tasks.


Research on Capacity


Research into he capacity of STM was completed by Jacobs in 1887 using a technique called the digit span technique. The digit span technique involves giving a list of either digits or letters to participants, which they have to recall in order. The list is increased by one digit or letter each time it is presented until the participants can no longer record them in the correct order. This is their 'digit span'.

Jacobs demonstrated that the capacity of short-term memory appears to be somewhere in the region of 5 to 9 items. He found that the span for digits was approximately 9.3 items and the means span for letters was approximately 7.3 items.

George Miller in 1956 also studied the capacity of short-term memory and suggested that the capacity of short-term memory is roughly 7 items, plus or  minus 2. He argued that it was possible to increase the capacity, short term memory by chunking information. This is seen in everyday life when someone is trying to remember something, such as a telephone number they appear to chunk the information. For example, when remembering an 11 digit mobile phone number people will tend to chunk the numbers into smaller chunks to learn, e.g. 0784 - 3605 - 429



1. one strength of the research of Jacobs and Miller is that it has been replicated numerous times, even though Jacob's study is an old study and would not have been conducted in a controlled laboratory setting. Therefore the research has good validity.

1. One limitation of Miller's research is that he might have overestimated STM memory capacity. Later research by Cowan in 2001 suggested that short term memory might only be around 4 plus or minus one chunk.

Research on Duration

Duration of STM


The duration of short term memory has been investigated by Peterson and Peterson in 1959.


They tested 24 participants by giving them nonsense trigrams to recall. A trigram is a three consonant group of syllables which can be meaningful (e.g. FBI, ITV, BBC) or in this case meaningless (FJW). Once they were presented with the trigram, the participants were then given a distraction task which involved being given a large three digit number and being asked to count backwards in threes until they were told to stop. This was in order to prevent them from rehearsing the trigram. Peterson Peterson varied the time period over which participants were asked to recall the trigram. It was either after three seconds, six seconds, 9 seconds, 12 seconds, 15 seconds, or 18 seconds.


Peterson & Peterson found that the recall of the trigram became increasingly more difficult for the participants. The average recall after three seconds was roughly 80% but after 18 seconds it was approximately 3%.


Peterson and Peterson concluded that the short term memory duration appears to be approximately 18 seconds with a maximum of 30 seconds.


1. One limitation of Peterson and Peterson's research is that the stimulus material is very artificial. Trigrams are not something that a person would be expected to recall in every day life, therefore this study lacks external validity. However, as it was a laboratory study, all variables were controlled and therefore the study has good internal validity.


Duration of LTM


The duration of long-term memory was studied by Harry Bahrick in 1975.


Bahrick studied 392 participants, all American aged between the age of 17 and 74. Participants were tasked with recalling individuals from their high school yearbook.


Initially, they underwent a free recall task, and as expected, those who had graduated 48 years prior showed about 30% recognition, while recent graduates (within the past 15 years) demonstrated approximately 60% accuracy. When given a recognition task with 50 photographs, participants' recall after 48 years was around 70%, compared to about 90% for those who had graduated more recently.


1. One strength of Bahrick's research, in contrast to other studies like those on the duration of STM, is the higher validity of the stimulus material. Participants were asked to recall personally meaningful information, suggesting that when stimuli are significant, recall rates improve, thereby indicating good external validity in Bahrick's findings.


Coding, Capacity & Duration Exam Questions

Coding, Capacity & Duration Mark Scheme

bottom of page