Coding, capacity and duration of memory

• What is coding?

  The format in which information is stored in the various memory stores.
• Baddeley's procedure on coding:

  Gave participants different lists of words to remember: Acoustically similar words or dissimilar. Semantically similar or dissimilar.
• Baddeley's findings on coding:

  Immediate recall worse with acoustically similar words, STM is acoustic. Recall after 20 minutes worse with semantically similar words, LTM is semantic.
• Jacobs' research on capacity:

  Researcher reads four digits and increases until the participant cannot recall the order correctly. Final number = digit span.
• Jacobs' findings on capacity:

  On average, participants could repeat back 9.3 numbers and 7.3 letters in the correct order immediately after they were presented.
• Miller's research on capacity:

  Miller observed everyday practice, noted that things come in sevens - days of the week, deadly sins, etc.
• Miller's findings on capacity:

  The span of STM is about 7 items (plus or minus 2) but is increased by chunking - grouping sets of digits/letters into meaningful units.
• Peterson and Peterson's research on duration of STM:

  24 students were given a consonant syllable to recall and a 3-digit number to count backwards from. The retention interval was varied: 3, 6, 9, 12, 15 or 18 seconds.
• Peterson and Peterson's findings on duration of STM:

  After 3 seconds - average recall was about 80%. After 18 seconds it was about 3%. STM duration without rehearsal is up to 18 seconds.
• Bahrick et al.'s research on duration of LTM:

  Participants were 392 Americans aged between 17 and 74.

  1. Recognition test - 50 photos from high school yearbooks.
  2. Free recall test - Participants listed names of their graduating class.
• Bahrick et al.'s findings on duration of LTM:

  Recognition test - 90% accurate after 15 years, 70% after 48 years.

  Free recall test - 60% recall after 15 years, 30% after 48 years.
• AO3 - Baddeley's research identified two memory stores:


  Later research showed that there are exceptions to Baddeley's findings. But STM is mostly acoustic and LTM is mostly semantic. This led to the development of the multi-store model.
• AO3 - Baddeley's study used artificial stimuli:

  The words used had no personal meaning to the participants so tells us little about coding for everyday memory tasks. When processing more meaningful information, people use semantic coding even for STM. This means the findings of this study have limited application.
• AO3 - Jacobs' study has been replicated:


  This is an old study and may have lacked adequate controls. Despite this, Jacobs' findings have been confirmed in later controlled studies. This shows that Jacobs' study is a valid measure of STM digit span.
• AO3 - Miller's research overestimates STM capacity:

  For example, Cowan reviewed other research. He concluded that the capacity of STM was only about 4 (plus or minus 1) chunks. This suggests that the lower end of Miller's estimate is more appropriate than 7 items.
• AO3 - Peterson and Peterson's study uses meaningless stimuli:


  We sometimes try to recall meaningless things so the study is not completely irrelevant. But recall of consonant syllables does not reflect meaningful everyday memory tasks. Therefore the study lacked external validity.
• AO3 - Bahrick et al.'s study has high external validity:


  Everyday meaningful memories were studied. When lab studies were done with meaningless pictures to be remembered, recall rates were lower. This means that Bahrick et al.'s findings reflect a more 'real' estimate of the duration of LTM.