Peterson and Peterson (1959)

Aims: To prove that things only stay in short term memory for around 20 seconds and then, if it is not rehearsed, it disappears forever.

Procedures: Participants were given sets of trigrams to learn and then tested on their recall. They had to recall them after 3, 6, 9, 12 or 18 seconds. They also had an interference task, counting backwards, in threes from a random number. The independent variable was the time delay and the dependant variable was how good the recall was.


After 3 seconds: 80%

After 6 seconds: 50%

After 18 seconds: Less than 10%

Conclusions: They had proved their hypothesis, there was very little left of the trace after approx. 20 seconds. It also proved that there was a distinct difference between the LTM and the STM.

Criticisms: It lacks mundane realism because the likelihood of the recall of trigrams in real life probably wouldn’t happen. The trigrams are not meaningful. Other research has shown that more meaningful things are remembered.

Bahrick et al (1975)

Aims: They aimed to test VLTM. They wanted to see whether long term memory was infinite.

Procedures: Participants included 392 American ex-high school students aged 17-74. Recall was tested in four ways.

  1. Free recall of the names of as many of their former classmates.
  2. A photo recognition test.
  3. A name recognition test.
  4. A name and photo matching test.


90% accuracy in face and name recognition after 34 years.

80% accuracy for name recognition after 48 years.

40% accuracy for face recognition after 48 years.

60% accuracy for free recall after 15 years.

30% accuracy for free recall after 30 years.

Conclusion: Classmates were rarely forgotten once recognition cues had been given. This supports the idea that people have VLTM. Recognition was better than recall.

Criticisms/Comments: This is a field experiment and therefore it can be generalised to the real world and this means it has high ecological validity. There would have been a great opportunity for rehearsal, increasing the rate of recall and therefore the results cannot be generalised to other types of information.

Bartlett (1932)

Aims: To investigate the effect of schema of participant’s recall of a story. A schema includes expectations, attitudes, prejudices and stereotypes.

Procedure: 20 English participants took part in a natural experiment. They were presented with a range of stories, or folk tales, from different cultures, making it difficult for Western participants to understand their significance. They were then asked to recall the stories are periods of time.

Findings: Participants’ recall of the story got shorter and shorter after multiple presentations. Participants often added their own Western words to substitute words in the story and so ‘canoe’ went to ‘boat’.

Conclusion: Accuracy was rare in recall. Participants were actively reconstructing using their existing schemas. Memory is influenced by our existing knowledge, which in turn is created by the culture in which we live.

Join now!

Criticisms: The intervals in between when the participants were asked to recall the story was different and therefore not entirely reliable. As the experiment was a natural experiment it meant that it lacked control of other factors which may have affected the results.

Demand characteristics would have suggested that the participants were supposed to react and they may also have acted in a way that the experimenter wanted them to. These both add up to a lack of internal validity. Schemas are very vague.

Loftus and Palmer (1974)

Aims: Language used in EWT can alter memory. Leading ...

This is a preview of the whole essay