Like Richardson, Bower (1972)* conducted a study into the use of imagery as a learning aid. The experimenter showed two groups of participants 50 pairs of un-related nouns e.g. ‘dog’ and ‘hat’. The experimental group was asked to form a mental image whilst learning each of the pairs of words. The control group was asked to remember the words (with no reference to imagery being made). Both groups were given the same amount of time to remember the words. Later, each subject was shown first the first word in each pair and asked to recall the second word. The imagery group recalled 80% of the pairs of words, compared with only 33% recall amongst the control group, showing that forming images aided the learning.
* Cited – ‘Memory in the real world’
* Cited – ‘Psychology – The science of Mind and Behaviour’
* Cited – ‘Understanding Psychology’
Luria (1968)* was asked to complete a psychological study on a reporter that their editor had noticed could remember the smallest detail, without ever taking any notes. The reporters memory capacity was seemingly unlimited, and he could remember the lists that he learnt with Luria 30 years later. Luria found that the reporter would associate lists of things which he wished to remember with familiar features of a scene (e.g. a street). He mentally placed each object at a certain point on the scene. He then just had to recall the mental image of the scene to recall the list.
Paivio (1986)* believes we can store nouns in a verbal and an image form. If there is an image to go with a word it means it is stored in the memory in both the verbal and the image form, thus meaning it is more likely to be recalled. This is known as the dual-code hypothesis. Paivio would predict that words accompanied by images would have a higher recall rate. Richardson (1972)* has a similar theory to Paivio. He regards imagery as a process of organisation that is an aid to memory. Parkin (1993)* also believed this theory. He thought that pictures are easier to remember than the words they represent, because a picture is more likely to activate a verbal code as well as an imagine one, while words are less likely to evoke a picture.
Aims: To see whether words are remembered better when they are accompanied by pictures.
Experimental Hypothesis: (one tailed test)
There will be significantly more words recalled from a list of words accompanied by pictures than from a list of words not accompanied by pictures.
Null Hypothesis:
There will be no significant difference in the amount of words recalled from a list of words accompanied by pictures and a list of words not accompanied by pictures.
* Cited – ‘Psychology – The Science of Mind and Behaviour’
* Cited – ‘Psychology – A New Introduction’
Method:
Design: The laboratory experiment was used so that we could manipulate all of the independent variables and also control the extraneous variables as much as possible.
To reduce ordinal practice effect the independent measures design is needed to be used.
The independent variable in the experiment was whether pictures accompanied the words or not. Sixteen participants were individually shown a list of 20 words without pictures on individual cards. The other 16 participants were also individually shown words on individual cards, but this time a picture accompanied the words. The dependent variable was the number of words the participants could recall when asked to write down all of the words they could remember.
The experiment environment wasn’t controlled in that it was carried out in the college library, where participants were likely to be distracted, although both of the conditions were carried out during the afternoons of consecutive days. For the experiment we had to control some of the variables. As time and money was very limited we controlled the age of the participants to 16 – 18 years olds, as this was the most accessible age range. Standardised sets of instructions were used to eliminate researcher bias (see appendix, item 1). We also controlled the sex of the participants by ensuring that half of the participants were male and that half were female.
In the first set of participants (the participants were tested separately) there were 16 people. In the second set of participants (also tested seperately) there were 16 people.
This independent measures design produced data that was at least ordinal and therefore the Mann Whitney U test was used to analyse it.
Participants: The total number of participants used was 32 with there being 16 participants in each of the groups, all aged between 16 and 18. There were 8 females and 8 males in each of the groups. Participants were taken from college using an opportunity sample. Only two people approached refused to participate.
Materials and Apparatus: Two sets of 20 cards were used, all of the word cards produced had a single word (and picture) on them, all typed in upper case, with the font used Times New Roman. The font size between words was not kept the same, as some words were harder to fit onto the page than others. On one set of cards each of the words were accompanied by a picture. Each of the participants were given an A6 blank piece of lined paper to write the words they could remember, and a pen. A watch with a second hand was used to time intervals, time to remember words, and so forth.
Procedure: We approached participants and asked if they wouldn’t mind participating in a psychology experiment for 15 minutes. Once their permission had been obtained we allocated them to one of the two conditions by deciding before the experiment was carried out that the first 16 participants were to be shown words with pictures, and the second 16 participants were to be shown words without pictures. Each participant was read standardised instructions (see appendix) and then were shown either the set of words with pictures, or the set of words without pictures. The participants could withdraw if they wished to. Once the participants had seen all of the words they had a one minute break in which they sat silently, whilst the experimenter observed. After this time they were given a piece of blank A6 lined paper with their group – with pictures (group A) or without pictures (group B) written at the top, and a pen. They were then asked to write down as many words as they could remember in two and a half minutes without help from anyone around them. When this time had elapsed they were debriefed about what the aim of the experiment had been. Once all of our data had been collected it was analysed using the Mann Whitney U test.
Materials:
Watch
Copy of standardised procedure
Copies of words (and pictures)
Paper
Pen
Results: (For Table of Results see Appendix item 2)
Mann Whitney U test observed value: 38.5
Critical value of U (at 0.05 significant level): 45
Mean A (with pictures) = 14.56 (2.d.p)
Mean B (without pictures) = 10.56 (2.d.p)
Mode A (with pictures) = 16
Mode B (without pictures) = 10
Median A (with pictures) = 16
Median B (without pictures) = 10
Range A = 9
Range B = 5
Standard Deviation A (with pictures) = 2.99 (2.d.p)
Standard Deviation B (without pictures) = 1.67 (2.d.p)
(for calculations see appendix items 3, 4 and 5)
Statistical analysis: The participants scores were of at least ordinal level and an independent measures design was used meaning the Mann Whitney U test was used to analyse the data collected. The observed value of U was 217.5 and the observed value of U` was 38.5. The lower value of U is used, therefore U` was taken as the observed value for U. The critical value for U was 45 at the 0.05 significance level on a one tailed test. As the observed value for U (38.5) was less than the critical value (45) the null hypothesis has been rejected in favour of the alternative hypothesis. This means that when images are presented with words the retrieval of these words is greater (as is shown by a mean of 14.56, mode 16, and median 16) than if these words were not presented with pictures (mean 10.56, mode 10, and median also 10). The standard deviation of condition A (with pictures) is 2.99, whereas the standard deviation of condition B (without pictures) is 1.67. This shows how the results of condition A were more widely spread out around the mean of A than B. This means that the mode and median are better measures of central tendency for the results of condition A, as they are affected less by extreme numbers, than the mean is. The range reflects this. The range for condition A is 9, showing that the results for these participants was widely spread, compared to a range of 5 for condition B, showing that these results are not as widely spread apart.
Discussion:
As you can see from graph 1, it is evident that the participants shown the words accompanied by pictures recall more words, whereas the participants shown words unaccompanied by pictures could recall fewer words. This shows that memory is aided by items being accompanied by pictures. The mean, median and mode of the results also support this. The mean for condition A (with pictures) was 14.56, whereas the mean for B was 10.56. Those who were shown pictures with the words could remember on average 4 more words than those who weren’t. The median and the mode for condition A were the same at 16. This was also true for condition B with the median and mode both being 10. These two measures of central tendency show that on average those shown pictures could remember 6 more words than those who weren’t shown pictures. The range for sample A (9) shows how much the results were spread out compared to condition B (5). This could be due to the participants with the lower scores (in both conditions) having a poor short-term memory – some people with specific learning difficulties have poor short-term memories, and this hasn’t been taken into account. The histogram drawn shows that the majority of those who could remember more words had been shown pictures with the words.
These results support the alternative, one-tailed hypothesis, therefore the null hypothesis is rejected. The Mann Whitney U test showed that the data was significant at the 0.05 level, thus showing that the use of pictures aids the recall of words.
Although it may be true that pictures aid the recall of words, some of the participants did not show this as strongly as the others. In condition A participant number two recalled 9 words, and participant number nine scored 10, which were both similar scores to the participants’ scores in condition B. This could be due to less attention being paid by some participants, fatigue or boredom towards the end of the experiment, distractions from others around whilst the experiment was taking place, or just due to a poor short-term memory.
The results from this experiment support Paivio, Richardson and Parkin’s theories that words are remembered better when they are accompanied by pictures.
These theories, about imagery as a technique of remembering have been criticised. Craig and Lockhart (1972)* in their levels-of-processing model show that they do believe, like Pavio, Richardson and Parkin that images are remembered better, but they believe that it is not because we have a better memory for images, but because when an image is formed we increase the number of associations the words have, and it therefore being more likely for someone to remember the word. This is a form of semantic processing (memory for facts). It may well be that Craig and Lockhart’s theory happened in our experiment.
Paivio’s theory in particular has been criticised saying that strong images would need a lot of memory space to be retained, and also a system by which the brain could decode these stored images.
* ‘Psychology – A New Introduction’
Summary:
From this experiment we can see that our hypothesis was supported that using images to accompany words in learning is an effective method of doing this.
It could well be that it is not necessarily the images that are stored better, but the way in which the brain does this.
Limitations of the design and modifications:
The participants’ performance may have been affected by other students around them, as the experiment was not conducted in a secluded environment, this condition was not controlled. The time of day (in the afternoon) may have affected the participants performance, as some participants may perform better in the morning than in the afternoon, but for other participants it may be the opposite.
Although the design did ensure that the ordinal practice effect was not present, the abilities of the participants were not matched. Ideally matched pairs could have been used, as the different characteristics of the participants may have effected the results of our experiment, but this would have been extremely time consuming.
The sample size from our experiment was far too small to generalise to the population. A quota sample of the population would have been far more representative, as the age and sex was not representative and there was no mention of the ethnic minority of the participants and how representative of the population this was. Using a quota sample of the college would have been far more representative .
Some of the words in the word list may be easier to remember than others, as this was not investigated, and all the pictures were not taken from the same software, and some were drawn by hand due to restricted time for searching for suitable pictures. This may have effected the recall of the words accompanied by pictures.
Further study:
Although an equal amount of males and females were used in this experiment a comparison between them was not carried out. This could be done as a further study to see if retrieval is any different between the two sexes by repeating the experiment, but this time recording the results of the different sexes as well as recording the results between the two conditions.
Research could also be carried out to see whether the age of the participants effects the results of this research.
The experiment could be repeated with the words being read out instead of being held up on cards. The results could then be compared to those of this experiment to see if this was to affect the results, and if so research into the reasons why.
Different length word lists could be used to see if this affects the recall rate.
Instead of using free-recall participants could be asked to recall the words in the order they were presented to them.
A comparison could be made between the effectiveness of imagery to remember words and the effectiveness of rehearsal to remember words. Two groups of participants could be given the same words to memorise but each group using either the imagery or rehearsal techniques.
The participants could also be given time to memorise the words, with the experimental group being told to memorise using imagery.
A comparison could be carried out between left-handed and right-handed people to see if this effects results.
Lots of different variations on this study could be carried out by changing the words to see if some words are easier to remember. Word pairs could be used (does it make a difference if the word pairs rhyme?), are everyday words easier to remember than words that a person rarely uses?, and words with familiar letter strings could be used.
Implications:
This study shows how using words accompanied by pictures could be useful to teachers, as we can see that things are remembered far better if an image is used, therefore in teaching if visual aids are used recall in pupils should be higher.
This study explains as to why a lot of advertising uses pictures rather than words as the product is far more likely to be remembered when pictures are used. Businesses also work in this way by giving presentations that incorporate visual aids into them to try and get clients to remember the message they are trying to put forward.
Many large companies have logos to help their customers remember them by.
Even something everyday like a shopping list could be remembered far more easily if it was converted into images.
Information leaflets carry images on them to try and get people to remember the message they are trying to put across.
References:
Rita L. Atkinson et al. – Introduction to Psychology (11th edition) (1953) Harcourt
Brace College Publishers
Alan Baddeley – Your Memory – A User’s Guide (1993) Penguin Books
Alan Baddeley – Human Memory – Theory and Practice (1990) Lawerence Erlbaum
Associates Ltd: Publishers
Mike Cardwell – The Complete A –Z Psychology (1996) Hodder and Stoughton
Annette Cassells – Remembering and forgetting (1991) The British Psychological
Society
Gillian Cohen – Memory in the Real World (1996) Psychology Press Ltd
C.B. Dobson et al. – Understanding Psychology (1981) Weidenfeld and Nicolson
Richard Gross – Psychology – The Science of Mind and Behaviour (third edition)
(1996) Hodder and Stoughton
Richard Gross et. al. – Psychology a new Introduction (1999) Hodder and Stoughton
Alan J. Parkin - Memory and Amnesia – An Introduction (1987) Basil Blackwell
Standardised Instructions (Item 1) – read to all participants
Only read to participants the instructions in italics:
In this memory test you will be presented with 20 words (with pictures) each for 7 seconds. Please try and remember these words so you can recall them when asked to.
List of words (for reference)
- Cloud
- Frog
- Bucket
- Bridge
- Hand
- Crown
- Train
- Eye
- Tree
- House
- Castle
- Cat
- Brush
- Fire
- Boot
- Book
- Seagull
- Glasses
- Chimney
- Dinosaur
Show each of the words (with / without pictures) for 7 seconds each, in the order shown above. Do not interact with the participant during this time.
Once all of the words have been shown tell the participant they are to have a short break of one minute.
Give the participant a piece of paper and pen and ask them to write down as many of the words as they can recall. Tell the participant they have two and a half minutes to do this.
Once this time is up collect the paper off of them, and thank them for their time and help. Then debrief them…
Debriefing:
You have just participated in a memory test. I am trying to find out if a list of words can be recalled more easily if they are accompanied by pictures.
(If the participant saw the words without pictures then add…)
You have participated in the first part of the experiment where words are not accompanied by pictures.
Table of Results: Item 2