Acoustic confusions in STM
The aim of the study was to investgate the nature of coding in STM.
Conrad presented participants with several series of 6 letters visually at a rate of 0.75 seconds per letter. The letters used were selected from B, C, F, M, N, P, S, T, V, & X. Participants were told that they had to write down each list in the order the letters were given immediately after the presentation (thus making this a STM task). Conrad analysed the type of errors made, rather than the number of letters recalled.
His findings were that participants tended to make errors by substituting letters that sounded the same, rather than those that looked the same (acoustic confusions). For example they would substitute B for V, F for S, P for B and so on.
The conclusion was that the letters had been encoded by sound even when presented visually. Thus, Conrad’s experiment is considered to be an effective demonstration of acoustic coding in STM.
The study does not rule out other methods of coding operate in STM, such as the use of a semantic code. However, Baddeley (1966) showed that coding was primarily acoustic. He compared memory for acoustically confusable words such as MAP and CAP with memory for semantically similar words (BIG, HUGE, WIDE etc.). Recall was much worse for the visual presentation of words that were similar acoustically.
Back to Short Term and Long Term Memory
Back to Summary of Key Studies