Psychologists have also noted that recall can be improved if identical cues are mirrored during recall and during the process of learning. The two types of cue-dependent forgetting are state-dependent forgetting and context-dependent forgetting. State-dependent forgetting is when a person is not in the same physical, emotional or mental state when asked to recall information, to the one they were in when they first learnt it. A key theory that supports this is Korsakoff’s syndrome which states that if we learn material while in a drunken state, then it will not be accessible when a person is sober. However, by being drunk, which provides internal cues that enable recall, a person is able to remember again. Further support is given by Goodwin et al. (1969) who found that excessive drinkers who learn things when is a drunken state are most likely to remember them when in a similar state of mind.
McCormick and Mayer (1991) believed that the internal cues to do with remembering and forgetting could be due to the mood that you are in. For example, you are more likely to remember happy things when you are feeling happy instead of when you are sad, thus indicating that physical and mental state does influence recall. Bower (1981) also found mood to have effects on retrieval and found that his subjects recalled more memories learnt when sad if he tested them again when hypnotised to be in a sad mood than a happy one.
Williams and Hollan (1981) undertook a study that involved imagining the context you were once in, in order to help recall details from a previous experience. Their results highlighted that when people are asked to recreate known environments, imagining the activities that they were involved in, and the surroundings, then people are in fact able to recall names of school friends that they may have be absolutely sure they had actually forgotten. This process is known as ‘redintegration’ and is based upon the principles of people retrieving whole memories when thinking about the whole context.
Context-dependent forgetting is when external cues from the environment are missing. One of the first studies that examined the effects of context retrieval by carried out by Greenspoon and Ranyard (1957). They tested learning material in one room and then compared the success rate of recalling it in the same room and in a different room. These findings were later supported by Smith (1970) who conducted an experiment based on context-dependent forgetting, where he tested the recall in the same environment where material was learnt, in a different environment but imagining to be in the same environment where the material was learnt (in the basement) and finally, in a totally different environment. Smith’s findings showed that the group that were in the same environment when learning and recalling information, were the most successful when later asked to recall the word lists. In support, Godden and Baddeley (1975) carried out an experiment where divers were asked to learn words either on land or fifteen feet underwater. Divers were then required to recall the words, either on land or underwater. Their findings supported those of previous studies, as there was a decline in recall of 30% of those divers who learnt and recalled the words in a different context, illustrating a lack of initial external cues. Moreover, Jerabek and Standing (1992) concluded from their own study, that recall success could be aided by simply imagining the original context of the room where the information had first been learned.
Chu, Handley & Cooper (2003) raised the idea that changing contexts can be equally as effective as reinstating them, with regards to the topic of interest of context-dependent forgetting. Standard context-dependent memory studies have shown that when information is encoded in a certain environment, recall performance is not as string when retrieval takes place in a different, rather than the same, environment. It is through the use of mental visualisation during the encoding stage that recall can be enhanced. The study to test this idea found that information can become linked with mental images of environmental contexts such as physical installation of an imagined environmental context can function as effectively as physical environmental context reinstatement and eliminate context-dependent forgetting.
Abernathy (1940) carried out another study into the effects of context-dependent forgetting where the recall of information was tested in the same room where the material was learnt and in a different room to where the material was learnt. There were two groups used in order to carry out this experiment. The groups who learnt and recalled the material in a different room were seen as inferior, due to low levels of recall, compared to the contrary group, who learnt and recalled the material within the same room. This showed that the absence of external retrieval cues made it difficult for the group tested in different rooms to recall words. These external cues aided the people who recalled in the same room from which they learnt.
It is the experiment that Abernathy carried out that I am going to adapt. As previous research suggests that context does affect recall, my alternative/experimental hypothesis is going to be directional (one tailed). Abernathy’s study is rather dated and learning styles in education have changed significantly so I am going to see if environment does still have an affect on the recall of a simple word list.
Aim
The aim of this experiment is to compare the recall of words within the same room that they are learnt and within a different room. This will then indicate if initial external cues play a pivotal role in recall and, if context-dependent forgetting exists as found in Abernathy’s work.
To test for significance a cut off needs to be chosen. The standard cut off p≤ 0.05 level was chosen as being acceptable and a Mann Whitney U Test will be applied to the results.
Alternative Hypothesis
Participants who recall words in the same room as they learn them will be significantly more successful in their recall of those words, compared to those participants who recall the words in a different room.
I have chosen a one tailed hypothesis because previous studies, including those of Abernathy (1940) and Smith (1970) suggest that the external cues present affect recall.
Null Hypothesis
Any differences in word recall scores between the participants who learn and recall words in the same room, and those who learn and recall in a different room, will be solely due to chance alone.
Method
Design:
A laboratory experiment was chosen to gain the most accurate results as researchers can intervene directly in the situation and have the highest level of control over variables. This makes it more likely to attain accurate results.
I have chosen to use independent measures design, preventing the order effect because none of the participants taking part have carried out this experiment prior to this study. Using this design means, half the number of participants learnt and recalled the words in the same room and, the remaining half of the group, learnt the words in one room and recalled them in a different room.
The independent variable was the room where recall took place, e.g. either in the same room as where the words were learnt or in a different room. The dependent variable was the amount of words recalled by participants. The controlled confounding variables include the wordlist given to all participants. Wordlists have to be the same for all participants to ensure that the results of the study are not affected by any differences in the wordlists. Furthermore, the task to prevent rehearsal for the two groups had to be constant, as did the noise and temperature within the environment. Consequently, this would make any impact on the results unlikely.
Participants:
Thirty-two participants were involved altogether, consisting of 16 work colleagues in each condition. Each group was divided equally into eight men and eight women, ensuring that the results were not influenced by any difference in performance between the sexes. Participants’ ages ranged from 25 – 45 and were accordingly divided fairly amongst the two test groups.
The sampling method was an opportunist sample, whereby selection was based on anyone who was available for testing at the required time.
Apparatus/materials:
- Consent forms (Appendix A)
- Task instructions (Appendix B)
- Wordlists (Appendix C)
- Crossword puzzles (Appendix D)
- A stopwatch
- Plain paper & Pencils
Procedures:
All participants were taken to Room 1, where all equipment had been previously given out. Wordlists were faced down. There were two pencils on each desk and a piece of plain paper. Participants were asked to enter the room, sit down, not to touch any of the equipment on their desk and not to communicate with anyone until told to do so. Instructions were then read to participants and I made sure that all involved fully comprehended what was being asked of them. As soon as participants were ready, they had two minutes to remember as many words as possible from the given wordlist. As soon as the two minutes were up, all wordlists were collected. Participants were then given a crossword puzzle in order to prevent rehearsal of the learnt words. Like in the first instance, all participants began the crossword puzzle task at the same time and were all given five minutes to complete as much of the task as possible.
When the crossword task came to an end, participants were then asked to turn over their crosswords and, using the plain paper and pencil provided, recall and write down as many of the words from the wordlists as they could, within an additional two minute period. Group A recalled their words from the wordlist in the same room where they had learnt them and Group B recalled their words from the wordlist in Room 2, a different room to where they had originally learnt the words from the wordlist. Group B were moved quickly and were supervised by a fellow colleague who was present when instructions were being given out.
When participants had completed the study, they were fully debriefed and told the objective of the experiment for a second time. They were reminded that their results were confidential; hence their names were not essential. Also, they were informed that they would be able to read the final report, which includes participants’ results.
Treatment of results
A table to show a comparison between the two conditions
The results from this experiment support the alternative hypothesis, and more importantly, reflect what was initially thought at the outset of the experiment. Participants in Group A on average recalled more words than participants in Group B. This reinforces those research findings of Abernathy (1940) and Smith (1970) by strengthening claims that external cues present does affect recall.
The highest score in Group A was 10 and was gained by 6 participants. When compared to Group B, no-one scored 10 and only one participant scored 9 words during the recall exercise. When speaking to participants after the experiment, those in Group A said that they tried to associate objects in the room to the words on the list, e.g. Balloon was associated to the bin, photograph was associated a painting and cloud was associated with the whiteboard. Those in Group B also commented on doing the same. However, when they were removed from the room where they learned the words originally, they were unable to remember these associations due to not having the same cues present as Group A. Participants did not report finding any of the words difficult to recall, it was solely the lack of cues to aid the recalling process that primarily affected their recall score.
The results from my experiment show a difference in recalling words from a wordlist when recalling them in the same room as they were learnt and recalling them in another room. I therefore decided to carry out the Mann Whitney U Test, in order to discover if the difference in recall between the two conditions was significant at the level of p≤ 0.05. I used the Mann Whitney U Test as it identifies reasons for any visible differences within a set of data, inferring that this statistical test is appropriate when using an independent measures design, when two conditions are unrelated. Furthermore, this statistical test deals with data on an ordinal level, which involves ranking data into a particular order. Finally, when testing a small sample of people, the Mann Whitney U Test is the most practical. (See Appendix H) The present data meets these requirements when ranked by the test.
A table to show the results from the Mann Whitney U Test
Since the calculated value of 9 for ‘U’ in the Mann Whitney U Test is less than the critical value of 83, it provides the confirmation that the difference in recall between Condition A and Condition B is significant at the p≤ 0.05 level. This eliminates any possibility that the difference many be due to chance, alone. Subsequently, the null hypothesis was rejected, and the experimental hypothesis was accepted, giving reason to believe that the significant diversity between the two sets of data was due to factors, rather than chance.
Discussion
Explanation of finding:
Through scrutiny of the collected data, the aim of the experiment had been fulfilled, and a conclusion could be reached. There were noticeable differences when analysing and comparing the data in the tables and graphs. It is more apparent that participants in Condition A were more successful in their word recall, when compared to those in Condition B, who appeared to be lacking retrieval cues, which were no longer present when they were moved to another room. This suggests that the environment you are in, does impact on your retrieval success.
Using the inferential statistics that the Mann Whitney U Test provided, I was able to see a significant difference between the two conditions, at the significance level of p≤ 0.05, thus supporting the existence of context-dependent forgetting. Furthermore, this lead to the rejection of the null hypothesis and the acceptance of the experimental/alternative hypothesis. To sum up, this indicates that the difference in recall is caused by something other than chance, leading us to believe that retrieval of information could possibly be dependent on external cues.
Relationship to background research:
These results support the findings of Abernathy (1940), along with findings from Godden and Baddeley (1975) and Tulving and Osler (1968), suggesting that initial retrieval cues enhance recall. Brown and McNeil (1966) proposed the idea of the ‘tip of the tongue’ phenomenon in order to explain accessing information problems. They believed that prompts and cues play an integral role when accessing information. Forgetting can occur in both STM and LTM. Processing and locating information on demand is not always easy and people will often say they don’t know if the information can not be recalled immediately. However, as noted previously, the information is there and waiting to be recalled, it just simply needs the right prompt or cue to be successfully recalled.
Although my work can be related to that of Abernathy (1940) and Godden and Baddeley (1975), it differs slightly, in that, their research did not involve carrying out a task to prevent recall. Moreover, differences in levels of recall in these studies are more significant, e.g. 30% difference in recall between the two groups of divers, compared to those in the experiment that I carried out, which depicts a difference of twenty-five words between the two groups. Furthermore, the current investigation seems to have produced findings which relate to previous research, and so therefore, strengthens the claims that external cues play a pivotal role in when recalling.
This clearly indicates that context-dependent forgetting is an issue in everyday life and that by being in the same context when encoding takes place and recalling, it can enhance and be beneficial, as external cues are constant in this condition, as opposed to where external cues differ between the two different contexts, likewise respectively as with the two conditions in my study.
Brown and McNeil (1966) proposed the idea of the ‘tip of the tongue’ phenomenon in order to explain accessing information problems. This theory was not tested by this research. Also, McCormick and Mayer (1991) found that mood affects recall. This idea too was not tested by this research.
Research limitations and modifications:
When undertaking research, it is important to consider the limitations and possible modifications, in order to increase the reliability and credibility of the research. The participants of the current study may not be representative to the wider population, thus not allowing the findings to be generalised to the population, due to the small sample size of people used in this investigation. However, if a larger sample size had been used, it could have improved the accuracy and credibility of my results. Moreover, the larger time gap in which participants from Condition B encounter when moved from one room to another, after completing their crossword puzzle, may have affected their results, as they recalled the words after a longer period of time, as opposed to those in Condition A, who remained within the same room, throughout the entire investigation. Therefore, participants in Condition A could have achieved better results because they recalled the words from the wordlist after a shorter period of time. After all, according to the Decay theory, information is rapidly forgotten, due to the passage of time, rather than some form of displacement, implying that the longer the period of time before information is recalled, the less successful the result will be. In order to overcome this limitation, if the investigation was repeated, the space of time before recall would have to remain the same in both conditions, encouraging greater accuracy from data collected, and in turn, preventing Trace Decay from occurring.
As this was a laboratory experiment, it could be criticised for lacking validity, as such high levels of control were imposed. This artificial setting is noticeable different to a real-life situation. Ecological validity also tends to be lacking in this experiment because the results may differ, depending on the setting or situation. This problem could be addressed in the future, by carrying the research out in a situation that relates to everyday life. Perhaps a field experiment could be a possibility; this way, a lower degree of control is required, allowing participants to feel more comfortable and at ease in ordinary and familiar surroundings.
Finally, I failed to use participants who did not know the researcher and they may not have taken it seriously enough, thus behaved in a way that would impress the researcher or even given results that they thought were necessary. If this experiment was re-run, I would endeavour to use participants who I don’t know and, as a result, eliminate any confounding variables.
Implications:
The main implication from this investigation is that people are dependent on external cues and, in the absence of external retrieval cues, it makes it more difficult for people to successfully recall words, when compared to those who are given external cues to aid the recall process. Context-dependent forgetting raises serious issues for eyewitness testimonies. Recalling details of an event can be more successful when you are at the specific area where the incident took place. Consequently, none of the initial external cues are missing, reducing the possibility of context-dependent forgetting. In order to increase reliability of eyewitness testimony in the future, witnesses could be taken to revisit crime scenes, for locations to aid recollection of information.
Context-dependent forgetting could also be a possible problem during pilot training programmes. People tend to undergo pilot training in simulators to learn the qualities of a pilot, e.g. aircraft control, emergency landing procedures etc. This many not be effective because a simulator differs to a real aircraft. Consequently, when moving from a simulator to a real aircraft, performance may deteriorate, due to the variation between the two different settings. However, although this is the case, training cannot take place in a real aircraft for obvious reasons, mainly due to cost implications.
Suggestions for further research:
Schools are beginning to recognise and see the benefits of understanding the importance of research which is centred on context-dependent forgetting. Schools are now in better positions to coach students for forthcoming tests by enabling them to sit them where enriched revision sessions took place or in rooms that were once filled with informative wall displays. Students are encouraged to recreate an image of the learning that has previously taken place, in order to aid their recollection of information during test conditions. It is from this idea, that a future study could be designed in order to test performance in schools. A group of students, who share a similar level in ability, would be tested in the same subject area. The group would be divided into two equal sub groups, taking gender into consideration. All students would take their test in the same room, but one of the groups would be asked to imagine the room in which they had been taught. Additionally, they would be advised to bring their own writing materials. Although this does not hold a problem for students who sit tests in Primary Schools, those who sit tests in Secondary Schools and Colleges, now have them lead by a team of unknown adults. Again, students would have their own teachers present, particularly the teacher who delivered the work that they are being tested on. The occurrence of initial external cues would be expected, reducing the possibility of context-dependent forgetting. On the other hand, the remaining group of students would sit the tests as normal. They would not be required to bring with them, any writing materials, which could encourage the intervention of initial external cues. This would add further support to the research carried out by Williams and Hollan (1981) who undertook a study that involved imagining the context you learnt in and recreating known environments and surroundings to aid recall.
To sum up, great improvements could be made to future research carried out in this area of psychology, as we are able to gain a deeper and enriched understanding to the extent to which context plays an important role in real-life situations. If a well published and credit worthy reliable research study was carried on this topic, then it would help to raise the profile of this important subject matter and, would be extremely beneficial to the world of education and may instigate a positive change to the testing of students.
Conclusion
The null hypothesis, which stated that there would be no difference in word recall between the two conditions, but any difference would be due to chance alone, was rejected and the alternative hypothesis was accepted. This supports the fact that any absence of external cues would result in context-dependent forgetting, meaning poorer recall.
References
Abernathy, E.M. (1940) The effects of changed environmental conditions upon the results of college examinations. Journal of Abnormal Psychology, 10, pp.293-301.
Brown, R. and McNeil, D. (1966) The ‘tip of the tongue’ phenomenon. Journal of Verbal Learning and Verbal Behavior, 5, pp.325-337.
Chu, S., Handley, V. and Cooper S. (2003) Eliminating context –dependent forgetting: Changing contexts can be as effective as reinstating them. British Journal of Psychology, 53, pp.549-559.
Godden, D and Baddeley, A. (1975) Context-dependent memory in two natural environments: On land and under water. British Journal of Psychology, 66, pp.325-331.
Goodwin, D.W., Powell, B., Bremer, D., Hoine, H. and Stern J. (1969) Alcohol and recall: State dependent effects in man. Science, 163, 1358.
Groome, D. (2004) An Introduction to Cognitive Psychology: Processes and Disorders. Sussex, Psychological Press Ltd
Hill, G. (2001) A Level Psychology through diagrams. Memory: Forgetting in long-term memory, pp.110-111.
McCormick, L.J. and Mayer, J.D. (1991) Mood-congruent recall and natural mood. Poster presented at the annual meeting of New England Psychological Association, Portland, ME.
Smith, S.M. (1979) Remembering in and out of context. Journal of Experimental Psychology: Human Learning and Memory, 5, pp.460-471.
Waugh, N.C. and Norman, D.A. (1965) Primary memory. Psychological Review, 72, pp.89-104.
Williams, M.D. and Hollan, J.D. (1981) The process of retrieval from very Long-term memory. Cognitive Science, 5, pp.87-119.
Bibliography
Cardwell, M. and Clark, L. and Meldrum, C. (2003) Psychology for AS-Level, 3rd Edition. London: Collins.
Cardwell, M. and Clark, L. and Meldrum, C. (2004) Psychology for A2-Level, 3rd Edition. London: Collins.
Cardwell, M and Coolican, H. (2001) The complete A-Z Psychology Coursework Handbook. Kent, Hodder and Stoughton.
Cardwell, M. and Flanagan, C. (2005) AQA ‘A’ Specification – Revised Edition Psychology AS: The Exam Companion. Cheltenham, Nelson Thornes Ltd.
Cardwell, M. and Flanagan, C. (2004) AQA ‘A’ Specification – Psychology A2: The Complete Companion. Cheltenham, Nelson Thornes Ltd.
Coordination Group Publications Ltd (2003) AS-Level Psychology: The Revision Guide. Newcastle-upon-Tyne, CGP.
Gross, R. D. and Malveen, R. and Coolican, H. (2000) Psychology: A New Introduction for AS- Level. Kent, Hodder and Stoughton.
Eliminating Context-dependent Forgetting: Changing contexts can be as effective as reinstating them. Accessed 5th January, 2007.
Eliminating Context-dependent Forgetting: Changing contexts can be as effective as reinstating them. Accessed 5th January, 2007.
Appendix A: Consent form given to participants
Consent form
I am researching the effects of learning words in one room and recalling them in the same room or recalling them within a different room. I have based my experiment on a piece of research carried out by a psychologist called Abernathy (1940) entitled ‘Context-dependent forgetting’.
I give my consent to take part in this experiment yes / no (please circle as appropriate)
The experiment will involve learning a list of 10 words in one room, then you will either recall them in the same room as where your learnt them (this will be people in Group A), or within a different room (this will be people in Group B).
Please answer the following questions
- Have you studied psychology before? If yes, which aspects have you covered?
…………………………………………………………………………………………………………………………………………………………………………
- Have you taken part in an experiment identical to this one, or similar to it, before?
………………………………………………………………………………………………………………………………………………………………………………………………………………………………………...................................
Your rights
As a participant, you have the right to withdraw at any time during the investigation, if you choose to do so. You will be fully debriefed at the end of the experiment, when any questions you have will be answered and the aim of the research will be revealed.
Your right to confidentially will be fully respected and your results will not be used without your consent. Also, you will not be asked to write your name on any of the task sheets provided and therefore will remain anonymous. Also, your scores will not be divulged to any of the other participants unless done so by yourself or with your permission.
Consent: In signing this consent form, I agree to my results being used in this investigation and
I understand that I have the right to withdraw at any point during the experiment.
Signature: …………………………………………. Date: ……………………………..
Appendix B: Instructions for participants
Task instructions
Firstly, many thanks for kindly agreeing to take part in the investigation based on memory in humans.
On your desk are a number of items, please do not touch any of them until told to do so.
In a moment, you will be asked to learn the ten words from the wordlists placed on the left hand side of your desk. You will have two minutes to undertake this task. You may not write anything down but you can look at the wordlist as many times as you need to. Please remember that other people are also performing the same task and you should not talk to anyone else at any point during this experiment.
Once the two minutes are over, you will be asked to turn over your wordlists. You will then be given a crossword puzzle to complete, which involves solving clues and writing your answers down. You will be given a further five minutes to complete this task. After the five minutes, you will be asked to stop the task and will be given a blank piece of paper, on which I would like you to recall as many words as possible from the wordlist as you can, within the space of two minutes. Those who have a Group A sticker will stay in this room and those who have a Group B sticker will go and recall the words from the wordlist in Room 12.
Does anyone have any questions regarding this task and what has been asked of you? If you do have a question, please ask know, as once the task begins, I will be unable to answer any queries.
You can withdraw from the task at any time if you wish to do so. The answers you give will be anonymous and be treated confidentially; hence you are not required to write your name on any of the task sheets.
Once the investigation is over, you will be debriefed and told the significance of this experiment. Also, I will explain what research findings have previously shown in this area of psychology and, my own results and findings, will be available for anyone to read at a later date, if you wish to do so.
If there are no questions, then you may begin.
Appendix C: Wordlist given to participants to learn
Below is a list of ten words. Please learn as many of them as you can within the next two minutes.
Balloon
Apple
Crayon
Kite
Photograph
Street
Bus
Trampoline
Cloud
Kangaroo
Appendix D: Crossword task
Crossword task
Please use the allocated time of five minutes to try and complete as much of the crossword as possible, by solving the clues. As your work is confidential, your name is not required. Thank you and good luck!
Simple find any crossword puzzle and put it here. I photocopied mine once I had stuck it on this sheet!
Appendix E: Results
The above Bar Chart shows the mean average words recalled in each of the two conditions and the standard deviation is also illustrated on each condition on the bar chart.
The Mann Whitney U Test, used to provide the inferential statistics, indicates clearly that the difference in recall score between the two conditions is significant. The significance level is p≤ 0.05 level.
Results continued:
A table to show the score of mean word recall
and the distribution of results within the two conditions
The above results illustrate the mean average scores from both conditions and the standard deviation of those scores. I chose the mean average because it uses all the scores individually and it is quick to calculate. Also, it is appropriate to use the mean average because there are no extreme outlying values e.g. there are no high or low scores to skew the overall results. The median, mode and range can be found in Appendix G.
The standard deviation was also selected in order to measure the variability of the attained scores from its mean. Like the mean, it takes into account all the scores in a set of data, so is more accurate than the range and it can be used in further analysis. The standard deviation also enables me to formulate statements of probability about how likely, or how unlikely, a given value is to occur.
Since the standard deviations are so similar, e.g. a difference of 0.2 of a word, it would be valid to carry out a stats test non-parametric.
A mean of 8.9 for the same room condition as against 6.4 for recall in a different room, does support the hypothesis, although a stats test may be needed to test for significance in this difference.
Appendix F: Findings
The table below shows the raw data collected during the investigation, illustrating the difference in recall between the two group conditions, and in turn, reinforcing the importance of context on recall.
Appendix G: Descriptive statistics
Descriptive Statistics continued: Workings out of Standard Deviation
Appendix H: Workings out of Inferential Statistics (Mann Whitney U Test)
Inferential Statistics continued: Workings out of Mann Whitney U Test
Calculation of U Calculation of U1
U = n1 x n 2 + n1 (n1+1) ÷2 – T1 U1 = n1 x n2 – U
= 16 x 16 + 16 x (15 ÷ 2) – 367 = 16 x 16 - 9
= 256 +120 -367 = 256 -9
U = 9 (calculated value) U1 = 247
As 9 is the lowest of the two calculated values for U, this becomes the calculated/observed value, which will be compared to the critical value, in order to see if the difference in word recall between the two groups is significant. In doing this, I will be able to discover if the difference between the two groups was simply due to chance, or if there were other factors involved, supporting the existence of Context-dependent forgetting.
