Recall in Memory Using Mnemonics

Psychology A2 Coursework

Introduction

Mnemonics are aids to a person in remembering data. This was proven in Bower (1973)’s experiment in which two different groups of students were given five lists of twenty words. Of these participants, those who used mnemonics remembered 72% of the items while the non-mnemonic group only averaged 28% recall.

This however, was not the first research into memory. Ebbinghaus used nonsense syllables to test recall and tested himself many times. He found that memory declines quickly at the outset but then levels off soon after. This experiment has been duplicated many times with the same results.

Ebbinghaus’ recall method was relearning; although there are others such as serial recall, free recall, paired associates and recognition tasks.

Memory processing can be split into three procedures: registration, encoding and storage – followed by retrieval. Encoding is how the information is changed into a storable form to be kept in memory. Storage is the actual keeping and retaining of the information in memory. Retrieval is how the information is taken from storage and brought to light (i.e. remembering). Forgetting may also occur at any of these stages causing the person to be unable to recall the information stored. However, not all information encoded is actually stored and not all stored information can be retrieved.

Memory is commonly split into three forms – Sensory, Short term (STM) and Long Term (LTM).

From research into STM we find that it has a capacity of 7 +/- 2 items (Miller 1956). Also if unrelated information is chunked (i.e. linked together) recall is improved.

Sensory memory has a duration of about 1 / 2 seconds in order for a person to decide whether further processing is necessary. This form of memory is modality-specific and although research in this area is rare, the current research focuses on the visual and auditory stores (iconic and echoic stores).

LTM is seemingly limitless in capacity and duration. Memory stored is usually via semantic encoding (e.g. verbal material) but visual and auditory encoding is also used.

Returning to recall aids (Mnemonics), imagery is a mnemonic device to help increase recall of seemingly unrelated items. For instance verbal material is easily remembered if associated with a visual image. Bower’s experiment mentioned above, asked participants to pair words with visual images in order to increase recall. For example if someone paired the items with things in his house, he would take a walk around his house and mention the items as he ‘passed’ them.

This method, first used in Ancient Greece, is also known as the ‘Loci’ method of recall. By building a picture in your mind of familiar objects, one can use this familiarity to aid you in recalling unfamiliar items of information.

Also, Paivio in 1971 found the same results and concluded that the reason for the improved recall when associated with visual images is that the information is encoded in two separate ways, both verbally and visual which gives the person double the chance of recalling the information later.

In addition to this research it was found (Bower 1972) that if the items were in chunks or clusters recall was further improved.

Luria (1968) studied a Russian reporter Sherehevski who was able to recall strings of words up to fifty words in length after hearing them just once. He associated each item with a visual image on a street in Moscow. When recalling the list he would take a mental ‘walk’ down this street and would recall each item as he passed it. Luria claimed that Sherhevskii could remember all fifty words in the string even 15 ...

This is a preview of the whole essay

In addition to this research it was found (Bower 1972) that if the items were in chunks or clusters recall was further improved.

Aims

My aim in this research is to further study the effects of mnemonic devices on recall. I will be using the Loci method mentioned above in my investigation on the experimental group and there will be a control group who will not be using any recall aids. The experiment will be similar in method to Luria’s work but will incorporate more participants to gain more accurate and standardized results on a wider scale. I will be using an experimental method to avoid confounding variables disrupting the results.

The experiment will take place in a field setting to reduce any bias that would be found in a lab experiment and Independent groups will be used to avoid order or practice effects.

Hypotheses

As the aim is to see how mnemonics help recall there should be a significant difference in the amount of words recalled in the two groups. My hypotheses will reflect this. The null hypothesis is used to predict no difference in the two groups, and any difference will be due wholly to chance factors. The experimental hypothesis will be one tailed (directional) I chose this as previous research seems to point towards the fact that mnemonics improve recall.

H1 (experimental): The experimental group will achieve higher scores in the recall test than the control group as previous research points to this conclusion.

H0 (null): There will be no difference between the two groups, and any difference that may occur is due wholly to chance factors.

Method

An experimental method will be chosen since maintaining control over the variables is easier than in a non-experimental method, thus decreasing the chances of extraneous variables confounding the results. It is also much easier for the researcher to directly manipulate the Independent Variable.

A field setting will be used as this decreases bias. In addition, the participants feel much more relaxed in their natural environment then in a laboratory.

An Independent Measures design will be chosen as it avoids order and practice effects. (i.e. those participating do not have to forget their first condition). Participants are unaware of the true purpose of the study so they are not likely to behave in a certain way to please the experimenter – i.e. Demand Characteristics are reduced.

The two independent groups used are the Experimental Group and the Control Group. The former will be taught the Mnemonic method of recalling information, whereas the left will be left to their own devices to recall the word list.

There will be an interference task (duration 30 seconds) to eliminate Primacy and Recency Effects.

Participants

Number: 20 – decreased bias compared to a smaller group. Equal numbers of participants in each condition and equal numbers of males and females in each group to decrease gender biases.

Selection: The Opportunity Sampling method will be used, as it is the easiest and quickest method.

Target Population: School students aged 15-18

The sample population chosen was naïve to the true nature of the experiment and had no Psychology background, as this would affect their responses to the tasks.

The Participants were allocated top their groups by choosing number cards (1=control, 2=experimental). This random selection was chosen to eliminate bias arising when participants chose their own groups.

Apparatus

A list of instructions on how the experiment will work.

A word list consisting of ten words to memorize (words were chosen at random and were unconnected).

Stopwatch with which to time the one minute participants had to remember the words.

All of these were standardised – i.e. every participant had exactly the same in every case.

The score of each participant was taken down on a sheet hidden from view.

Procedure

Participants were first informed of their right to withdraw from the experiment and that their personal information would not be used at all during or following the experiment.

Also, they were informed that they could view their results at any time.

Participants gave informed consent and were again told that they had the right to withdraw at any time.

Participants were then told about the experiment and what to do, (these instructions differed between the two groups, but within the groups, the information was standardised).

There were two experimenters present at each experiment to ensure that each one was using the standardised procedures and to decrease experimenter bias.

Participants in the Experimental group were instructed how to use the Mnemonic device both via a written list of instructions and verbally.

Control group members were not taught this method and were instructed to begin memorizing the word list immediately after being introduced to the experiment.

The experiment begins after the participant completely understands what to do.

Participants were required to memorise a list of ten words in no particular order, within the space of a minute.

Then, the participants had to perform an interference task to prevent rehearsal – this would involve reciting their 6 times table backwards from 300. The task would last 30 seconds.

They then had to perform the recall task – remembering the ten words in any order within a minute.

Their score was recorded in a raw data sheet hidden from view.

All participants were told that their scores were within the norms so as not to harm their confidence.

Following the experiment, participants were fully informed to the true nature of the experiment and debriefed both individually and as a group and all questions were answered.

Results

Results Table

Looking at the above table it is possible to see whether our hypothesis that mnemonics aid memory was proven or not. The data shows that the experimental group recalled more words than the control group. Descriptive statistics were used to examine the data further. The central tendency was found and the mean was calculated. The mean was used instead of the mode or median because it gives a more specific score and is more representative, however, it can be distorted by freak values.

The mean for the experimental group (7.9) is higher than that of the control group (6.5). This indicates that mnemonics do in fact have an effect on recall.

The standard deviation was calculated to show the dispersion of the scores. A low S.D. shows that the results are closer the mean i.e. a smaller variation.

The S.D. for the experimental group was lower than that of the control group

A histogram was used to present the data because it gives a good general overview of patterns of recall.

This type of graph looks at how frequent certain scores were and how they were distributed. The scores

varied from five to ten words recalled for both groups. The most frequent number of words recalled.

for the experimental group are seven and eight and the least frequent are five and six. There are no less

than six words recalled. For the control group the most frequent numbers of words recalled are five

and six and the least frequent are nine and ten. This group has numbers recalled varying from five to nine.

The Man Whitney U test showed that UA was 23 and UB was 77. The smallest of these value becomes U and so the observed value was U 23. The critical value of U at the 0.05 level of significance on a two-tailed test is 23. The U is equal to this level and so is significant. The statistical test shows that the experimental group recalled significantly more words than the control group.

A significance level of p< 0.05 was used because it is the most conventional significance level. At this level the chance that the null hypothesis will be retained and that there will be no real difference between groups under these conditions is about one in twenty. Therefore there is a possibility that the results were obtained by chance.

The null hypothesis can be rejected and therefore the alternative hypothesis is retained. There is never certainty about whether the results were simply a fluke and so a type one error may have occurred. This means that a mistake may have been made in rejecting the null hypothesis when it is true. Variations in the results may have been caused by random variables.

Discussion

After performing my experiment, researching how the Loci method of recalling data affected the recall of a word list, I found that my results supported all the previous research I had looked into. They showed that memory was improved with the aid of mnemonic devices.

Firstly, Bower’s study, which found that in the recall test those who used mnemonics remembered 72% of the items while the non-mnemonic group only averaged 28% recall. Although his experiment used 5 word lists each containing 20 words and in this experiment only 1 list of 10 words was used, the findings were similar to Bower’s.

Secondly, Luria’s experiment (1968) which studied a Russian reporter Sherehevski who was able to recall strings of words up to fifty words in length after hearing them just once. He associated each item with a visual image on a street in Moscow. When recalling the list he would take a mental ‘walk’ down this street and would recall each item as he passed it. Sherehevski used the Loci method as the experimental group did here – and the results in this experiment show that his method works better than recalling without any mnemonic devices. The difference however was that Luria’s subject could recall up to 50 words at a time, obviously way above the number available here.

Baddeley’s experiment in which he gave one group a list of words and asked them to recall them straight after learning, and gave another group a 30 second delay after learning. He found that recency effects were eliminated after the 30 second delay. In this experiment, there were signs of recency effects with some participants remembering the last 3 words (recency effect) and some remembering the first 3 words (primacy effect).

This experiment had an unrepresentative sample – only participants aged 16-18 were used and all were from similar educational backgrounds. The experiment could be improved by using a few different methods. One could have selected a target population that included people without any educational background (and so would be unused to learning rules) and one could choose participants from different age groups.

Also, the sampling method used was not random so there would definitely an element of bias in the sample chosen. A computer could have selected participants randomly but this would not be necessary in a study of such a small scale.

As in Bower’s study, the word lists could have been longer (20 words or more) and there could have been more of them (Bower used 5 lists). In this study, with only 10 words in the list, participants would be able to memorise the words without using the mnemonic device. Longer word lists would have made it more difficult for the participants to use repetition to recall the items. In addition, the interference task may not have been effective enough as recency effect was still present, as mentioned above.

This experiment could be used both in everyday life (e.g. remembering shopping lists) and in medical cases (e.g. Amnesiacs or sufferers of brain trauma) to aid memory and in the case of Amnesia, to slowly reconstruct the client’s memory.

Further to this study, one could research the gender differences in depth regarding mnemonic aids in memory. Also, one could study the effects of mnemonics across different cultures. This would help as different things can be thought of in different ways by different cultures e.g. the items on the word list or the places in the Loci method may mean something to one culture but mean nothing to another culture.