Aims:
Hypothesis:
Null Hypothesis:
METHOD
Design:
This was an experiment using independent measures, 10 subjects per group/condition, 1 trial per subject. One condition used recognition and one used free recall. This design was chosen to eliminate the possibility of re-learning and saving scores polluting the results.
Variables:
Independent variable - Provision of recall cues.
Dependant variable - Amount of syllables correctly recalled.
Controlled variables - Number of words to be memorised.
Time allowed for memorizing.
Time allowed for recall.
Time between memorizing and recall.
Instructions to subjects.
Amount, form and position of distractor items.
Participants:
20 subjects were used. These were friends and family and therefore not a random sample. They were between the age of 15 and 72. Although age was not strictly controlled for, a balance of under 35’s and over 35’s was achieved in both conditions. 12 subjects were female and 8 male.
Materials:
1.) List of nonsense syllables to be learned, headed with
standardised instructions. The syllables were selected
randomly. See Appendix 1.
2.) Blank paper for free recall condition.
3.) Sheet with target syllables mixed with distractor syllables,
for recognition condition. See appendix 2
4.) Watch.
5.) Pen.
Procedure:
Free recall condition
1. The subject was given the list of 20 syllables (App. 1), and
asked to read the instructions.
2. After two minutes the paper was removed.
3. They were asked to write down as many syllables as they
could remember. They were given minutes to do this.
4 This procedure was repeated with 9 more subjects.
Recognition Condition
1. As in the free recall condition.
2. “ “
3. Subjects were given a sheet with the target and distractor
items mixed together, (app.2). They were given 5 minutes
tick off the recognised syllables as directed at the head of
the sheet. After 5 minutes the paper was removed.
All subjects were debriefed on completion.
RESULTS
A Table Showing the Number of Words Recalled Correctly by Participants in the Free Recall and Recognition Conditions.
Results continued
For graph see appendix 4
Treatment of Results
The Mann-Whitney U Test was used to test for a difference between the two conditions, which had not arisen by chance. This data is suitable because it is at least ordinal level and from Independent Measures.
For calculations see Appendix 3.
U1 = 17
U2 = 83
Interpretation of Results
As U1 is less than the critical value of 27, the null hypothesis is rejected at the 5% level of significance.
DISCUSSION
Even the most superficial examination of the results shows an obvious difference in performance. The total number of syllables recalled in the free recall condition was 52 with a mean average if 5.2, compared to 107 in the recognition group with an average of 10.7. The lowest figure for free recall was 1; in the recognition condition it was 8. The Mann-Whitney U Test produced a U1 of 6; this resulted in the H1 being accepted at the 5% level of significance. This indicates that a difference did occur that was not due to chance. In this case it supports the view that recognition is a more effective way of remembering than free recall.
Tulving’s failure to retrieve theory is also supported by this, as it is fair to assume that the target syllables had acted as retrieval cues. Although the different conditions were discouraged from processing the material deeply by giving them nonsense syllables, the recall of certain items was much more frequent. Comments from the subjects, suggests that some syllables were the same or similar to known abbreviations. These could be linked to existing schema and therefore processed more deeply. It was reported by subjects of both conditions that such associations were actively sort after at the time of learning and syllables of this nature were the most easy to remember. With this in mind the results supported Craik and Lockhart’s depth of processing model of memory.
The results also supported the pattern recognition theories which would predict that re-presenting the arrangements of parts would allow a process to take place that simply could not in free recall.
Whilst conducting the experiment the need for further controls became apparent. These included isolation of the subject whilst learning and recalling, as interruptions and knowing they were being watched and possibly judged by others was at times obviously distracting.
The age and gender of subjects was not controlled for. It would be improved by having equal number of each gender and subjects from age categories in both conditions.
This has implications regarding how witnesses are questions and assisted with recall. Approaches such as taking them back to the site of an event, producing pictures or other material in order to tease out memories that are temporarily lost, and not readily accepting their genuine belief that they cannot remember, should all be adopted according to these findings.
The way exam questions are presented and where they are taken is also crucial according to these findings. If an exam requires total free recall it is setting limits of only measuring retrieval under cueless conditions. This could produce strikingly different results from the same student at any given moment than in a condition providing even a minimal amount of cues.
Future experiments might include varying the time elapsing between learning and recall or to include performing an interference task. It would also have been interesting to give the free recall group the recognition sheet immediately after finishing free recall, to see how many new syllables they recognised.
REFERENCES
Psychology for A Level Cardwell, Clerk & Meldrum
Psychology, the science of mind
And Behaviour. R. Gross
APPENDIX
Mann-Whitney U Test
Scores
1 4 4 4 5 5 5 6 7 8 8 9 10 10 10 11 12 12 12 16
Recognition scores
- = 18 T = 149
- = 18 N1 = 10
- = 18 N2 = 10
- = 20
-
= 14 U1 = N1N2 + N1(N1+N2) -T
- = 14 2
- = 14
-
= 10.5 U1 = 10 x 10 + 10(11) - 149
8 = 10.5 2
9 = 12 U1 = 100 + 55 - 149
T = 149
U1 = 6
U2 = N1N2 - U
U2 = 100 – 6
U2 = 94