Baddeley

Baddeley (1966) Short-Term Memory

Before developing the Working Memory Model, Baddeley studied both LTM and STM.

Before 1966, Baddeley had already studied memory. He was particularly interested in encoding, which refers to how information is coded or converted to be stored in memory.

Baddeley was interested in whether encoding happened semantically or acoustically. Hypothetically, semantic memories would be stored based on the meaning of the information, coding the concepts and abstract parts of an object or thought. Contrarily, acoustic memories would be stored based on coding sounds.

Baddeley (1966) Study

The study that Baddeley (1966) designed aimed to test whether LTM was encoded acoustically or semantically. To test this theory, participants were provided with different word lists that either sounded similar or meant similar things. Baddeley predicted that if participants could remember more similar-sounding words, LTM would be encoded acoustically. On the contrary, if participants remembered more words with similar meanings, LTM would be encoded semantically.

Research design

The research presented a between-groups comparison. The 72 participants were recruited through opportunity sampling and were randomly allocated to four groups.

Variables

The independent variable (IV) that was investigated had two levels:

1. Acoustically similar words versus acoustically dissimilar ones.

2. Semantically similar words versus semantically dissimilar ones.

The dependent variable was the score on a recall test of 10 words (calculates as a percentage). Participants did not only need to recall the words themselves but also the order in which these were presented.

Procedure

Each of the four groups was shown a slideshow including 10 words. The words were displayed for 3 seconds each.

The four groups emerged based on the four levels of the IV. In this way, the groups were the following:

1. Acoustically similar words, which included words such as cap, hat, or sack.
2. Acoustically dissimilar words which included words such as cow, dad, or led.
3. Semantically similar words included words such as big, large, or huge.
4. Semantically dissimilar words which included words such as hot, pen, or man.

As you can see from the examples, all words were simple and had one syllable. This was done to avoid that word difficulty would impact the results.

The learning phase was separated from the testing phase by an interference test in which participants had to hear and write numbers.

In the testing phase, participants took part in a recall test. For the recall test, participants needed to recall the words and the order in which they appeared.

This procedure was repeated four times. After the fourth trial, participants took a break and completed an unrelated interference task. To their surprise, they were asked to recall the list of words again after the break.

Baddeley (1966) Results

When calculating the results, the performance of the individuals in groups 1 and 2 was compared, as well as the performance of groups 3 and 4, separately. Further, participants' performance on the 5th trial was analysed to test for forgetting.

The results indicated that:

• Acoustically similar words were harder to recall than acoustically dissimilar words. Remembering the words cap, hat, and sack is more difficult than remembering cow, dad, and led.

• Semantically similar words were harder to recall than semantically dissimilar words. Remembering the words big, large, and huge is more difficult than remembering hot, pen, and man.

• When assessing STM, this is, when comparing trials before the break, the worse performance was for the acoustically similar words.

• Performance was overall better on the semantic condition than on the acoustic one.

Fig. 1. Results graph from Baddeley's (1966) original paper

Baddeley's (1966) conclusion

The main conclusion that Baddeley drew from such results is that LTM is encoded semantically. This comes from the fact that performance on the firth condition was better for the semantic condition than for the acoustic one. Further, Baddeley (1966) concluded that STM encoding is acoustic. This was deduced by the fact that performance on the short-term conditions (trials 1-3) was worse for semantically similar words.

Baddeley Findings

Taken together, through the pioneering work of Baddeley, psychologists learned that the encoding of STM and LTM differed from one another. Baddeley's main contribution to our knowledge of memory is that while LTM is encoded semantically, STM is encoded acoustically.

Baddeley (1966) Strengths and Weaknesses

Here there is a critical evaluation of Baddeley's (1966) study.

Baddeley (1966) Strengths

The strengths of the study are:

• Since the study was standardised and can easily be replicated, it presents high reliability.

• Baddeley used interference tasks to make sure that he was actually measuring long-term memories. This, in turn, increases the internal validity of the findings.

• Baddeley and Hitch used this study as supporting evidence for the proposal of the working memory model in 1974, which provides evidence of the implications that these findings had.

• It has beneficial implications for real-life scenarios; for instance, students can use these findings to strategise their revision techniques better).

Baddeley (1966) Weaknesses

The weaknesses of the study are:

• It was carried out on British students, which makes it ethnocentric. Therefore the research does not consider cross-cultural differences and limits the generalisability of the findings.

• The sample included 72 participants, which is not representative of the population and limits the generalisability of the findings.

• Given that it is a laboratory study, has low ecological validity; it is unlikely that the procedure is used in everyday life.