An experiment to see if interference affects recall in short-term memory.

Tom Clarke

An experiment to see if interference affects recall in short-term memory

Introduction

One of the most influential models of memory was the two-process model by Atkinson and Schiffrin (1968); incoming information enters the short-term memory (STM) as a result of applying attention to a stimulus. The brain makes sense of the incoming sensory information by perception, and then in the rehearsal loop, the information is rehearsed before it is stored in the long-term memory (LTM). The rehearsal loop is an important part of this model as it is what enables information to be remembered more easily and stored in the LTM.

Peterson and Peterson (1959) made an experiment where they researched the capacity of the STM. The experiment was called ‘The Brown-Peterson technique’, which involved participants hearing various trigrams (such as XPJ). Only one trigram is presented on each trial. Immediately afterwards, the participants are instructed to recall what they heard or to count backwards, in threes, out loud, from some specified number for 3,6,9,12,15 or 18 seconds (the retention interval). The function of this interference task is to prevent rehearsal. At the end of the time period, the participants try to recall the trigram.

Peterson found that nearly 70 per cent was forgotten after only a nine-second delay, and 90 per cent after 18 seconds. The STM is believed to hold information for about 20 seconds without repeating it (rehearsal). If information is sufficiently rehearsed it will be transferred to the LTM and can be recalled. It is also said that unattended information may also enter the STM, and this may be due to not applying enough attention to the information that needs to be rehearsed and may affect recall. In other words it may be due to interference and this experiment is to see if interference affects recall in STM.

Rationale

The two-process model assumes that incoming information enters the STM and it is rehearsed in the rehearsal loop before it is stored in the LTM. This is related to this study, because various objects are looked at by participants and then rehearsed. The objects being rehearsed is the sensory information entering the STM, and the participants must rehearse these objects in the rehearsal loop to store it in the LTM. The ‘Brown-Peterson technique’ experiment also relates to this study. Participants are instructed to recall various trigrams of letters, such like the objects in this experiment, with and without interference, where participants have to count backwards in threes from a specified number. The interference task used in the present experiment is based on this, instead having to put books in alphabetical order. The ‘Brown-Peterson technique’ focused more on how the amount of time to rehearse affected recall, whereas in this study, the focus is purely on whether interference actually affects recall in the STM.

Aim

The aim of this experiment is to see if interference affects recall in STM. In one condition participants rehearse and in the other an interference task blocks rehearsal.

Hypotheses

H¹ – The participants who rehearse the objects without interference will recall more of them than the participants who have an interference task blocking rehearsal.

Hº – If there are any differences between the participants who rehearse the objects without interference and the participants who have an interference task blocking rehearsal, it will be because of chance.

Method and Design

The experiment is a laboratory experiment, therefore being in a controlled environment. Conditions must be controlled in order to make it a fair test. Each participant must carry out the experiment under the same conditions. This included keeping the setting the same, making sure no extraneous variables such as external noise, can disrupt it. However, because it is an independent groups design there may have been participant variables confounding the results. These are individual differences between the participants; for example, the emotional state of an individual or whether the memory of an individual is better than another. Psychology students may have an advantage over non-psychology students because they already have an insight into the same or similar types of experiments. It is an independent groups design because two separate groups of participants are being tested. One group receives the experimental treatment, which in this case, is the interference task. The other group receives no treatment, acting as a control group. Both groups are tested once and their results are compared.

Procedure

- Each participant is tested in a corridor away from any interference from other students.

- 12 objects are placed on a table including a skull, football boots, board pen, spectacles, fork, mobile phone, can of coke, car keys, screw driver, school tie, watch and a chocolate bar.

- 20 participants from the sixth form in the age range of 16-18 years will take part. The standardised instructions given to each participant will be to look at the objects for 20 seconds, as this is believed to be the amount of time the STM can hold information ...

This is a preview of the whole essay

Procedure

- Each participant is tested in a corridor away from any interference from other students.

- 12 objects are placed on a table including a skull, football boots, board pen, spectacles, fork, mobile phone, can of coke, car keys, screw driver, school tie, watch and a chocolate bar.

books in alphabetical order.

- 10 participants will be under one condition and the other 10 will be under another. In one condition there will be no interference during rehearsal. In the other there will be an interference task blocking rehearsal.

- This task will be to stack 10 books in alphabetical order.

- The participants will write down the objects they can recall, and this will be recorded.

Variables

Independent variable: Condition 1- interference task during rehearsal of the 12 objects.

: Condition 2- no interference task during rehearsal of the 12 objects.

Dependent variable: this is the number of objects the participants can recall.

Controls

The controls in the experiment that need to be kept the same for each participant is the setting, making sure there is no extraneous variables such as any external noise. The actual objects and time period for looking at and rehearsing the objects with and without interference must be the same. The books used in the interference task should also be the same for each participant. The participants will be debriefed after the experiment and will be asked for their consent before. They are also told that they have the right to withdraw at any time during the experiment. The confidentiality of the experiment is important. The results of the participants are not revealed to anyone. The results obtained are only kept if informed consent is received.

Participants

The participants chosen for the experiment are in the age range of 16-18 years. The target population is students who attend sixth form. They include psychology and non-psychology students, and they are chosen by opportunity sampling. This sampling was used because it is the most practical in the circumstances. It is a quick and easy way of choosing participants, and this was important for this study because it was carried out in school hours, when students had lessons to go to and therefore students were chosen if they were available at the time. The number of participants chosen was 20 and the ratio of girls to boys was 9:11. The ratio of psychology to non-psychology students was 7:13.

Apparatus

The apparatus in the experiment includes:

- Three tables, one with 12 objects on, which include a skull, football boots, board pen, spectacles, fork, mobile phone, can of coke, car keys, screw driver, school tie, watch and a chocolate bar. Opportunity sampling was used to choose these 12 objects found around the school and at home, finding any type of object of similar sizes to fit on a table, suitable to use in a school environment. This type of sampling enabled the objects to be varied and not just from one particular environment, preventing schemas to make recall easier.

- The other table is used to write down how many objects the participant can recall.

- The Third table is used for the interference task (10 books in alphabetical order). Opportunity sampling was used to find the books, finding them within the psychology classroom.

Results

Summary Table

Summary Table Commentary

From the summary table, the results support the hypotheses. The mean, mode and median without interference were higher than with interference. There was not a large difference between these, which shows that the hypotheses were not as convincing as expected. From the results recorded there were some results that seemed anomalous, 11 objects were recalled with interference, and this was the highest result. This participant however, was a psychology student, and from the results a pattern could be seen where the psychology students recalled more objects than non- psychology students.

Descriptive Statistics Commentary

From the bar graph, the mean, mode and median for the objects recalled with and without interference can be identified a lot more clearly. The obvious trend is the fact that they are higher without interference. However, a pattern can also be seen in the graph. The mean, mode and median results with interference are in a reversed order of size compared with the mean, mode and median results without interference. In other words, with interference the mode is the smallest result at 7, and without interference the mode is the highest result at 9. This is the same pattern for the mean and median.

Relationship of Results to Hypotheses

From the results the experimental hypothesis can be accepted because the mean objects recalled were higher without interference than with interference. However, this was not the case for all the participants. There were some participants who recalled more objects with interference. This relates back to the null hypothesis where any results such as these were said to be due to chance. This may be true to an extent but another factor that may have affected these results is whether the participants were psychology or non-psychology students. The pattern that was identified showed that psychology students tended to recall more objects than non-psychology results. However, overall from the results, the null hypothesis has to be rejected because the experimental hypothesis can be seen quite clearly, and the reasons for the few anomalous results would most likely be due to participants having a better memory or a better understanding than others. It is not due to chance as it says in the null hypothesis, and so must be rejected.

Discussion

Validity

The test had concurrent validity as the phenomenon being measured actually exists. There has been research and other similar tests designed to measure interference on recall in the STM. For example ‘The Brown-Peterson technique’. It also had construct validity because the test items, this being the recall of objects with and without interference fit the underlying theoretical constructs of the test. The theory of interference affecting recall fits in with the method of rehearsing objects with and without interference in the experiment.

There is no ecological validity however, because the test does not really reflect reality. It is not natural in real life situations to be asked to memorise various objects in a certain time or put books into alphabetical order whilst rehearsing the objects.

It is a good way to test interference however, because of the fact that the interference task of putting 10 books into alphabetical order would not be expected by the participants, as it is not ecological. It is also made even more difficult because most of the books begin with ‘P’; therefore the participant is not just looking for the first letter in the title.

Improvements of Validity

Another more valid test could have been used. This could have been to shorten the time limit when looking at the objects to make the recall harder so that the test can focus more clearly on whether interference affects recall in STM. Using only non-psychology students could also make the study more valid as there would not be as many anomalous readings due to the psychology students having a better understanding of the experiment, because of demand characteristics, and participant variables (see method and design). To improve this the psychology and non-psychology students could have been separated into groups, and not mixed in with each other, or the test could have been only targeted for non-psychology students.

Reliability

The reliability of this study was relatively good because the mean, mode and median results were all particularly close together both with interference and without interference. However, more tests could have been carried out so that these results were even more distinctive, and the difference between the amount of objects recalled with interference than without interference would have been more apparent.

The controlling variables in the experiment make it reliable because the location was kept the same, providing a quiet environment for each participant. This reduced the introduction of extraneous variables. The quiet setting prevented any external noise disrupting the experiment, thus affecting the participants thinking.

The independent groups design was used allowing no order effects, therefore participants in each group did not know what the other condition was, or whether there actually was one. However, there were demand characteristics due to the psychology students having an idea of what the experiment was about and therefore guessing what was required, whereas the non-psychology students could not compare the different parts of the study and do this. There are participant variables as well. When comparing the 2 sets of results, you are also comparing different people. For example, one person might have a better memory, or be in an emotional state.

Improvements of Reliability

An alternative method could be to change the sampling of the participants. The experiment could only involve non-psychology students so that no participants would have an idea of what to expect. Another way of preventing the participant variables could be to use a matched-pairs design, for instance, matching pairs of participants with their gender, age and whether they are psychology or non-psychology students. This would produce a fairer and more reliable range of results. Another alternation could be to have more experiments involving a larger amount of students so that more results could be obtained, and therefore getting a more accurate mean, median and mode. Using the matched-pairs design, there would be a wider age range and people of different occupations could be used so that the results can be generalised to other populations.

Implications of the Study

The implications for the research outlined in the introduction to the study, given the results of the study are the ideas of Peterson and Peterson of the chunking in STM. It is believed that up to seven chunks of information for about 20 seconds can be held in the STM. In the experiment the time to look at the objects was for 20 seconds, relating to this idea. From gathering research from Atkinson and Schiffrin, the idea of the rehearsal loop taking place, is very relevant to this study. The rehearsal loop provides evidence for this study because the fact that when trying to remember things such as the objects in this experiment, rehearsal is very much needed, and that disrupting this, causes less recall.

Generalisation of the Findings

The problems with generalising the results to populations outside the target population are that the ages of people will be widespread in the entire population and this would affect the validity and reliability of the results. The older people are, the less capacity they have in the STM, and the younger people are, the more capacity they have. Disabilities among people would also affect the results and so it is particularly difficult to generalise the results to the entire population. The method that would need to be used to generalise this would be to match the different age groups, gender, disability and perhaps the occupational status in order to gather reliable results. It would be difficult to use any population, and produce reliable results because of the large amount of individual differences there are amongst people.

Application of the Study to Everyday Life

One real life application of the results could be when watching television whilst doing homework. The television would interfere with the rehearsal of information from the homework. The attention of an individual would be attracted to the noise the television is making and the picture it is displaying. Therefore, when doing homework, particularly revision, the individual’s rehearsal of information in the STM will be disrupted, and less information will be stored in the LTM. This would therefore cause less recall when in a school exam.

References

Atkinson, R.C. & Schiffrin, R.M. (1968) Human memory: A proposed system and its control processes. In K.W. Spence & J.T Spence (Fds) The Psychology of Learning and Motivation; Volume 2. London: Academic Press. Cited in Jarvis, M., Russell, J., Flanagan, C., Dolan, L., (2000), Angles on Psychology, Cheltenham; Nelsen Thornes.

Peterson, L.R. & Peterson, M.J. (1959) Short term retention of individual items. Journal of Experimental Psychology, 58, 193-198. Cited in Gross, R., (2001), Psychology, The Science of Mind & Behaviour, 4th Edition, London; Hodder & Stoughton Educational.

Appendices

The ten books used for the interference task were:

Angles on Psychology

Advanced PE for Edexcel

Changing Environment

Cognitive Psychology

Exchange and Transport Energy and Ecosystems

Human Biology

Molecules and Cells

Psychology for A/S Level

Psychology-The Science of Mind and Behaviour

Sociology in Perspective

An experiment to see if interference affects recall in short-term memory.

Tom Clarke