There was a significant increase in recall of the two-syllable list compared to the five-syllable list. This supports the hypothesis that the word recall would be higher when the number of syllables was lower as more words were recalled from the word list with two-syllable compared to five-syllables.
From the results we can see that the limit of working memory is capacity. This is because the two-syllable words were recalled easier than the five-syllable words. Baddeley (1975) suggested that one syllable could in fact be one item and so words with less syllables would be recalled better as they take up less room in the working memory and thus more could be stored there than words with more syllables as they would take more ‘slots’ up. Baddeley found this and suggested that the working memory’s capacity limit was 4 +/- 1. The results found that participants could recall on average around five words (from either syllable list, despite the slight increase in recall of the two-syllable list) suggesting that the capacity limit found by this experiment was 4-5 words which lines up with Baddeley (1975) findings.
This finding of a capacity limit of 4-5 words also corroborate Cowan (2001) finding of a capacity limit around four items. Though his definition of an item was not specified.
However, the findings do not agree with miller (1960) or Atkinson and Shiffrin (1968) capacity limit of 7±2. As mentioned in the introduction, Mathy & Feldman (2012) argue that the 4-5 item capacity limit is when multiple items have grouped together to fit into one ‘slot’, therefore this capacity limit is the limit for when all information has been compressed. Thus, Miller (1960) and Atkinson and Shiffrin (1968) is the uncompressed information. So, these results may show the capacity limit for compressed information or ‘chunked’ information. However, this cannot be shown by this experiment’s results without further investigation.
One limitation is the use of opportunity sampling, which meant that a large sample could be obtained which was useful. However, opportunity sampling has been known to be unrepresentative as it is not random (Farrokhi & Mahmoudi-Hamidabad, 2012). Having unrepresentative samples may lead to bias in the results as there may be outliers or the group may simply be so similar that the general population would not respond the same. An improvement for future research may be to use stratified sampling to get a wholly representative sample of people from all education types and backgrounds.
There is also a gender bias in this study with 13 males compared to 61 females. This may have affected the results because males are better at spatial working memory tasks and females are better at verbal working memory tasks (Lowe, 2003). This would mean that the results of the study may be unrepresentative as if females are better at verbal working memory tasks, which this experiment is, then the recall scores may be higher than they are in real life due to the higher number of females in this study.
Individual differences such as participant’s background or processing speed were not considered during this study. For example, it has been suggested that people who are more anxious have smaller memory spans due to having their working memory taken up by worry and thus would recall less than an average person (Hayes, Hirsch & Mathews, 2008). Participants backgrounds in this sense were not controlled for, so anxiety or other conditions may have affected their ability to recall information and thus would reduce the reliability of the results.
Future research may be useful to focus more around the factors that could affect working memory, such as processing speed, intelligence and conditions such as anxiety. It may also be useful to have a more gender balanced sample to obtain more generalisable results to the general population. Overall, further research into the limit of working memory would allow a more generalisable decision on the limit of working memory and factors that may contribute to it.
In conclusion, this study’s results have shown that people recall words better when the number of syllables is lower which is consistent with most modern research into memory capacity. Thus, this study acts as evidence that limit of working memory still seems to be capacity today; and that said capacity is still around 4-5 items. However, more research needs to be conducted in relevance to other factors that may be the limit of working memory such as interference, trace decay and processing speed.