Free recall experiments where the subjects are given a number of words and then asked to recall them in any order and the results fall into a pattern known as serial position curve or the “U curve”. Basically the “U curve” shows that the participant will show greater recall of words at the beginning and at the end whereas the middle part will show lower level of recall.
Free recall was investigated by Murdock (1962) where he presented a group of participant with 40 words, he found out that the participant tended to remember the first few words and the last few words of the list which also supports the multi store model theory which states that the primary effect is due transfer of the words to the long term memory store through rehearsal and the Recency effect was due to still the words being in the short term memory store.
Aim: the aim of this experiment is to investigate the Primacy and Recency effect by conducting an experiment to show whether participants tend to remember the first few words and the last few words better than the middle words.
Hypothesis: one tailed hypothesis
The first 10 words of a list of 30 (positioned from 1-10) and the last 10 words (positioned from 21-30) of a list will have a higher recall than the middle 10 words (positioned from 11-20)
Null Hypothesis: There will be no significant difference in the words recalled wherever the words are positioned in a list. Whether they are the first 10, second 10 or the last 10. Any difference will be due to chance
Design
This investigation was a controlled experiment due to the fact that controlled experiment is a wide spread method used and a suitable experiment for investigating memory with a controlled experiment because with this form of experiment there is a high chance of controlling variables which may effect on the recall of words. The reasons for choosing a controlled experiment is that it allows a better chance of controlling extraneous variables which therefore makes replication of the experiment possible, quantitative data can be collected and easily analysed.
A repeated measure design was used in this experiment as each participant individually carried out the task, where all the participants were carrying out the same task under same conditions. A repeated measure design allows the variables for example participants to be kept constant, statistical testing can be done more accurately because there is less variation, fewer participants are required and therefore more economical method of carrying out an experiment, however some problems can occur by using this method such as demand characteristics where the participants can easily guess the aim which can have an effect on the results and therefore making the results bias, to overcome this problem each participants carried out the experiment individually and only did the experiment once but all the participants were in the same condition.
The independent variable is the serial positioning of words (1-10), (11-20) and (21-30) on the word list of thirty words. The Dependent variable is the number of words recalled from the different positions on the list.
This experiment was taken under a controlled condition where all the participants were being given the exact same amount of time to recall the words in the list for each participant (one minute) and the same amount of time given to view the words. This was crucial because if time amount given was not controlled the results produced would not be reliable because if participants had more time to look at the words they would try to memorise all the words or try to make a short story out of the words, make a mnemonic device which would enable them to recall all the words found on the list. This would not show the true reflection of the experiment to find out the primacy and recency effect.
One extraneous variable in this experiment could be the length or the number of syllables in each of the 30 words list, as this could affect the results of the experiment. This variable can affect the results as if the words are 10 syllable words long most of words would be displaced as we know that our memory can deal with (7+ or – 2) chunks of information at one time. Therefore by having a 10 syllable word would not be beneficial as they results might not show was expected. To control this extraneous variable each of the 30 words were a 6 lettered word so that each word takes the same amount of capacity as the other words as longer words might take more space than shorter words.
Another extraneous variable that was controlled in the experiment was any participants having the knowledge of psychology were not used in this experiment. This was due to the fact that they might cause demand characteristics as they know what the experiment is about therefore will either perform in a way that they researcher might want them to do, this would mean that the results would not show a true reflection of the primacy and recency effect. This variable therefore was controlled by not letting any student having the knowledge or studying psychology took part in the experiment. This was ensured by asking them before the experiment. Any students that had knowledge of psychology were eliminated from the experiment.
Ethical issues
The BPS guidelines were all followed such as consent, confidentiality, right to withdrawal, deception and debriefing were fully taken into considerations when the experiment was taking place.
Before commencing the experiment the participants were asked for their consent in the form of initialising their signature in a consent form that was provided (See appendix 1). This was crucial so that to ensure that they know they are taking part in an experiment.
The participants were also verbally informed and on writing that their results would be kept confidential. This was taken under controlled by numbering the participants e.g. participant number 2, participant number 19 e.t.c. so that nobody was aware whose results were whose.
The participants were notified before the experiment that they had the right to withdraw from the experiment if they insisted at any time during or before stating the experiment. Specifically in the debrief the participants were told specifically that they had the right to withdraw their results (see appendix 3)
In the matter of deception concerning the experiment the participants were told after the experiments in the debrief section (See appendix 3) what the experiment was about, the main use of results once achieved which means that the participants were not lied to. This was crucial as if they participants were told what was going to be found and was exactly the experiment was about in detail then this would not produce results in a reliable way as they would not be a true reflection as the participants know what the experiment is about the will act in a way to please the researcher.
Participants
The target population for this experiment were all students of year 12 attending St. Mary’s College in Blackburn at an age group between 16-18years old.
The sample consisted of 20 participants and the sampling method used was opportunity sampling where first available participants were chosen from the college canteen. Opportunity sampling method was chosen due to the fact that this method of sampling is quick, convenient, less time consuming therefore the most economical method of sampling compared to other sampling method.
Procedure
Part of the target population, which was males and females students attending St. Mary’s College, Blackburn between the age group of 16-18, where 20 participants were chosen to participate
The participants were approached in the college canteen and asked if they would spare few minutes to take part in an AS psychology coursework if they agreed, they were politely taken into library in one of the quiet room individually.
After being taken to the room the participant was given a brief statement, consent form to be read and signed and they were verbally informed what will happen in the experiment and that they have the right to leave the experiment if they wished.
Participants were then given a standardised instruction sheet, which informed them on how to carry out the short experiment.
After reading the standardised sheet the participant was given the list of words sheet which was facing down on the desk so that the participants do not look at the words and they were once again verbally asked to look at the words on the sheet for 1 minute and the duration of 1minute was timed using a stopwatch. After the one-minute duration of time ended the word list was taken away and then they were given a blank piece of paper and told to scribble down as many words as they can remember in any order immediately after the word list was taken away from the participant. The participant had one minute to write down as many words as they could remember and this duration of time was also timed using a stop watch, after the duration ended the answer sheet was taken away and once again the participant were reminded about what will be done to their results and they had the right to withdraw their results if they wanted.
After the experiment they were fully told verbally and also in the debrief section about the main aim of the experiment, what the experiment was testing, and what the results were to be used for, they were also fully informed about all the ethics, to ensure the participants are aware. They were also informed that they had the right to withdraw from the experiement.
The answer sheets were taken away and the participants were thanked.
Results
In this experiment, the Primacy and Recency effect was investigated. The aim of the experiment was to find out whether people remember the first few words and the last few words in a word list.
Table to show the mean recall rate of words from the first 10 words (1-10), the recall rates for words between (11-20) and the last ten 10 words (21-30) in a word list containing 30 words
See appendix 6 for calculation of the mean
Looking at the mean values above the differences between the Primacy and Recency was:
9.4-7.9= 1.5 which shows that there was not a great difference in the recall rate of the words being recalled from the first ⅓ of the words and the last ⅓ of the words from the word list.
7.9-5.5=2.4 which shows that the difference between the first ⅓ of the words and the second ⅓ of the word list there is a quite big difference in the recall rate
Whereas the difference between the second ⅓ and the last ⅓ of the words from a word list 9.4-5.5=3.9 there is a huge difference in the recall rates.
This is because the difference between the Primacy and Recency has a smaller difference due to the words in Primacy effect being stored in the LTM as they were rehearsed whereas the middle words were displaced by the last incoming words which in this case were the last ⅓ of the words from the word list.
The results shown in the table show that the Primacy and Recency have occurred in this experiment, where the Primacy has a mean recall rate of 7.9 and the Recency has a mean recall rate of 9.4, looking at the mean recall rate values it seems that the Recency have a higher recall rate than the Primacy and hence the difference is quite small between them. This overall shows us that the recall rates for the first few words and the last few words are better recalled than the middle words in a word list.
The bar chart above also shows us that both the Primacy and Recency effect had occurred in this experiment as the bar chart for the first ⅓ of the words
(1-10) and the last ⅓ of the words (21-30) have a higher bar chart than the middle ⅓ of the words in the word list.
In this experiment, the Primacy and Recency effect was investigated. The aim of the experiment was to investigate the Primacy and Recency effect basing on the Glanzer and Cunitz theory. The experimental hypothesis for this experiment which stated that: The first 10 words of a list of 30 (positioned from 1-10) and the last 10 words (positioned from 21-30) of a list will have a higher recall than the middle 10 words (positioned from 11-20)
The null hypothesis for this experiment which stated that: There will be no significant difference in the words recalled wherever the words are positioned in a list. Whether they are the first 10, second 10 or the last 10. Any difference will be due to chance
The results accept my experimental hypothesis and therefore reject my null hypothesis.
Discussion
In this experiment, it was found that both the Primacy and Recency effect had occurred therefore the null hypothesis was rejected, which stated that there will be no significant difference in the words recalled wherever the words are positioned in a list. Whether they are the first 10, second 10 or the last 10. Any difference will be due to chance. Looking at this experiment the positioning of the words did have an effect on recall rates. The experimental hypothesis was accepted which stated that the first 10 words of a list of 30 (positioned from 1-10) and the last 10 words (positioned from 21-30) of a list will have a higher recall than the middle 10 words (positioned from 11-20). Atkinson and Shiffrin suggested that incoming information is automatically and immediately transferred to sensory memory, which can hold information for up to one second (Sperling, 1960) before being lost. If this information is attended to it is coded in short-term memory. Similarly, information rehearsed adequately in STM is coded in long-term memory. This theory is widely accepted and shows the existence of STM and LTM. Therefore, it also provides reason for the serial position curve as the two peaks in the curve as shown in the diagram below which highlight the effect of STM and LTM; whereas information not attended to or rehearsed enough is lost. This explains the unique shape of the Serial position curve as shown on the graph
The graph above shows which is similar to what was found in this experiment where the first few words and the last few words are remembered more than the middle words.
The results found from this experiment also support the Glanzer and Cunitz multi store model theory where they also found that the first few words and the last few words are tended to be remembered more than the middle words. Which as seen from this experiment from the bar chart that the first few words (1-10) had a mean recall rate of 7.9 and the last few words (21-30) having a mean recall of 9.4 which was better than the mean recall rate found for the middle words (11-20) having a mean recall rate of 5.4. As seen from the results it seems like the theory of Primacy and Recency effect does occur.
These findings can be applied to real life such as when revising, this technique could be used to improve revision skills by having to remember more things. As shown from the experiment that through rehearsal the items that you have learnt are stored in the Long Term memory. The positioning of the items also plays a crucial role where the first few things and he last few things are tended to be recalled more than items found in the middle which is due to the fact and evidence shown that the middle words are displaced from the incoming words whereas the first few words are remembered more because they have had time for rehearsal which has lead the items being transferred into the Long term memory which causes the effect of Primacy effect. The final items are also tended to be remembered more since they are the last items to enter the brain which means that the retrieval of these words will be recalled firstly and more of these words will be recalled because they are still available in the sensory memory. This means that when revising perhaps by moving around the items on the page by having the notes needed to learn placed at the top and at the end of the list which will help recall of information to be improved.
As this experiment was a controlled experiment, it may be considered that the results collected are not ecological valid as the results are not collected from a natural environment. A controlled experiment was very crucial as it was an experiment on a memory recall which means that a noisy environment such as in this case the college canteen would not be a suitable place to test on a memory test because the environment is noisy. The noise levels can interfere with what is being learned. Therefore a controlled experiment was very crucial in this experiment involving memory recall.
The number of participants might be increased to increase accuracy of the results and also the more the participants are used the more the results can be generalisable to the population. As from this experiment 20 participants were used so if this experiment was to be done again the participants used would be increased to 40. It will increase reliability significantly and give more generalisability to the target population. Then again, the number of subjects participating is and extraneous variable as it is difficult to determine how many will show voluntarily. Having a larger sample would inevitably lead to more participants, which would help improve accuracy.
An idea of further research in this area could be to investigate whether gender difference has an effect on the recall of words when testing for the Primacy and Recency effect. This would be interesting as it would show the experimenter whether the accuracy of memory differs between boys and girls. These variables could be altered in order to make sure that equal number of boys and girls are chosen to participate in the experiment.
Another idea of further research could be study whether age groups affects the recall of words when testing the Primacy and Recency effect. This could be carried out by using different age groups of participants for example comparing the words recalled from an age group of around 10-15 years old, 20-25 years old, 30-35 years old and 40- 45 years old. And see as to whether age has an effect on the words recalled.
In conclusion the results of this experiment showed that the Primacy and Recency effect had occurred. The results of the experiment could be used to show that people tend to remember the first few words and the last few words better than the words found in the middle.
Appendix 6
Calculations
Mean calculation for the first ⅓ of the words (1-10)
14+8+5+12+2+5+3+0+13+17=79
79/10= 7.9
7.9 is the calculated mean value for the first ⅓ of the words (1-10_
Mean calculation for the second ⅓ of the words (11-20)
7+8+2+11+1+1+3+2+9+11= 55
55/10= 5.5
5.5 is the calculated mean value for the second ⅓ of the words (11-20)
Mean calculation for the final ⅓ of the words from the word list (21-30)
12+11+13+1+3+14+6+2+17+15= 94
94/10= 9.4
9.4 is the calculated mean value for the final ⅓ of the words (21-30)
Materials and apparatus
- Consent form for participants in order to gain permission to carry on with the experiment. (See appendix 1)
- Debriefing sheet to make sure that the participants are kept fully informed of what the experiment is about and what will be done after the experiment i.e. what will be done with the results. ( See appendix 3)
- Standardised instructions to let the participant know what to do once the experiment in on or what is expected from them (See appendix 2)
- Pens so that the participants can write down the words recalled
- Paper so that the participant can write down the words recalled
- A quiet, well lit room with a sign on the door which will state that “ do not disturb”
- 20 participants needed so that I can test my hypothesis
- A stop watch to time the participants
- Word list which will include the words which the participants will need to memorise and recall. The word list having 30 words which are all 6 lettered words so that they have the capacity of storage (see appendix 4)
- A tabulated record sheet to record the result for analysis and discussion (See appendix 5)
- A calculation sheet showing how I calculated the mean (see appendix 6)
- Sample results and comment forms from the participants who did the experiment (See appendix 7)
References
- Cardwell, M, (1996) Psychology for A Level, Collins Educational
-
Gross, R, (2005) Psychology The Science of Mind and Behaviour (5th edition), Hodder Arnold
- Hayes, N, (2000), Foundations of Psychology, Thomas Learning
- Henderson J, 1999 Memory and Forgetting, Routledge
- Pennington, D, (2002) Introducing Psychology, Hodder and Stoughton
Website used to get the serial positioning curve graph
www.kgv.net/psychology