Memory research is incredibly valuablethe modern world where we have more and more to remember, such as revision at school and university, shopping lists, prospective memory and knowledge of the world. We have generally become more interested in the ways we can improve our memory, including looking at the meaning and structure of how and what we remember and how we can reduce forgetting.
Teachers and cognitive psychologists have come up with theories for improving memory but there is limited research. For example they have talked about Spacing Effect, Modality Effect, Bizareness Effect, Encoding-Difficulty Effect and Encoding-Context Effect and more. Strategies like mnemonics have been thought to help deeper processing and storage but they can only be used for unrelated material.
Slamecka and Graf (1978) talked about the Generation Effect (in particular the Self Generation Effect). This effect happens when mnemonic aids are self generated which leads to deeper processing, increased meaning and more links to schemas which evidently leads to better memory. They presented subjects with words that were to be generated by themselves and then the same words that were just to be read.
They also tried variations, such as timed and untimed, but across them all they found that performance was better in the generated word condition than the read condition. This investigation will replicate the study by Slamecka and Graf with a few differences. It will be interesting to see if the findings are reproduced. Rogers, Kuiper and Kirker (1977) studied a similar topic: they came up with the Self Reference Effect which is where material can be made more meaningful or memorable by reference (generation) to the self.
Craik and Lockhart (1972) did some research on depths of processing. They said that long term memory traces are formed at the time of learning depending on the processes that occur. Deeper levels of analysis will contribute to longer lasting, stronger traces (also developed in Atkinson and Shiffrin’s Multistore Model of Memory).` This would support the research by Slamecka and Graf and indicate that generation of material will lead to better memory of it. Miller (1956) mainly spoke about how chunked words will be remembered better but he also stated that people can remember around 7 words or chunks of information at a time.
His research was very important so it will be interesting to see if the participants in this study remember around 7 words too. Participants will be asked to think about the opposite of a word and remember it, then these results will be compared with those when they simply have to remember a given word. Perhaps they will remember more of the opposites than a normal list of words because a lot more thought is involved e.g. deeper processing. It must be controlled, for example similar words need to be used in each condition. Counterbalancing will also need to be carried out.
It is hoped that Slamecka and Graf’s findings will be replicated using a within subjects test the generation method of opposites with equal numbers of words in each list. The experimental directional hypothesis is that the self generation of words will significantly increase recall ability when compared to words that are just read. The null hypothesis is that there will be no difference in memory for generated words and memory for words that are just read.
Method Participants 20 participants took part in the investigation in the form of an opportunity sample. Most of the participants were students. Materials Two lists of words were produced, one with a list of words with the first letter of its opposite as a trigger for the generated word and the other with a list of words deliberately similar to the generated list. Participants recorded the words they remembered on a plain sheet of paper (see appendices). Design An independent groups design was used. There was no independent or dependent variable as an exploratory study was carried out. It was counterbalanced by presenting the two lists in an alternate order to each participant to avoid order effects.
Procedure The investigation was carried out in quiet surroundings and participants were tested individually. They were informed briefly about the aim of the study, given standardised instructions (including having the right to withdraw at any time) and presented with the first list. After this, they were asked to do an interference task which involved counting backwards from 100. They then wrote down as many words as they could recall in as much time as they needed. The procedure was repeated for the second list. Participants were debriefed and data was collected.