Introduction
Psychology is the scientific study of the mind and its functions and the mental processes. Memory is a function of the brain which is studied in depth by Psychologists, looking into the function of the brain in which information is interpreted, encoded, stored and retained until it is needed. In psychology, memory is broken down into 3 subsections: encoding, storage and retrieval. Encoding is the process of retrieving and processing information, before our brain moves it into storage for a certain amount of time before we can ‘retrieve’ the information at will. Short-term memory has a very small capacity of an average of 7 words/objects +2 or –2, depending on the individual (according to Miller, 1956).
Many studies have been conducted regarding the performance of short-term memory (STM) in relevance to our ability to remember words or images more efficiently. Paivio and Csapo (1971) found that the recall is much higher for words presented as images compared to them being presented as words. This involved a task providing picture-picture, word-word and picture-word repetitions. All of these results had a higher recall rate for images than words, proving his hypothesis that image and verbal memory codes were different. Paivio also discussed this in his ‘dual-coding theory’ (1986), which suggests the existence of two cognitive subsystems, one that processes non-verbal (imagery) and another that processes verbal (sound, written words). These two systems operate separately, meaning that people will have different abilities to remember images rather than words.
Another experiment examining this was Bower (1972), who showed that images had a profound effect on a person’s ability to remember words. Participants were given 100 different cards, one at a time, each card containing 2 unrelated words. Participants were asked either to memorise the pair or create a visual image containing both words. Bower found that by asking some participants to create a visual image caused them to recall with 80% accuracy. Participants who were asked to only remember the words had 45% recall. The results showed the increased ability to remember words associated with images – and that images were recalled better than words on their own.
Aim:
To see if participants would remember more images or words when asked to recall them in timed conditions.
Hypothesis:
Experimental: Participants will recall more images than words overall.
Null: Participants will recall more words than images overall.
Variables
IV: The presentation of 15 individual words and then 15 individual images on a set timer of 2 seconds per word/image.
DV: The participants ability to recall words or images.
Methods section
Design
This experiment was conducted in a classroom in a college, using a laboratory environment which was easily to replicate as it was a standardised procedure being used. This allowed for high control of extraneous variables, ensuring that any changes to the DV were caused by the IV – establishing a cause and effect relationship. However, this setting possibly increased demand characteristics from participants. This type of environment was controlled and reduces the ecological validity of the overall result as it was not a real-life situation. Both the words and images were given to the same group of participants, using repeated measures to ensure that the conditions were controlled and there was no disparity between variables.
The two independent variables in this experiment were the two conditions, one being the words section and the other being the images section. These were 15 words/images shown individually for a duration of 2 seconds per word/image. The dependent variable was the amount of words and images the participants were able to recall. Repeated measures were used to test the same sample group in both conditions, controlling participant variables. Repeated measures meant fewer participants were needed to conduct the experiment, making it time and cost-effective. However, this caused order effects meaning the participants could perform better in the second condition because participants are aware of what is expected of them.
All words were in capital letters, in bold and large to improve visibility from the back of the room. Despite this, some participants struggled to see the PowerPoint due to the harsh lighting of the classroom, and other participants were talking throughout the experiment which went unnoticed by the experimenters until the end – which were two extraneous variables that were unprepared for. The Hawthorne effect may have had an effect on the outcome due to the participants knowledge of being observed, but not explaining the true aim of the experiment until the debriefing helped lessen this and also lessen demand characteristics from the participants.
Participants
The experiment was conducted using a sample group of 14 people, consisting of 11 girls and 3 boys all around the ages of 16-22. This class was sampled through opportunity sampling – in which the researchers were able to quickly and easily collect a group of participants in order to study them. The experimenters were able to gain easy access to the room in which the experiment was conducted as it was readily available at any moment, making it quick and cost-effective way to collect data.
Materials
20 copies of score sheet 1 (Appendix 4), and 20 copies of score sheet 2 (Appendix 5) were printed out to guarantee every student had 1 copy of both. Each sheet contained 1 table numbered 1-15 to record results and the participants were informed what sheet should be used for each section. Standardised instructions (Appendix 3) were printed out and read to the participants to clarify was happening at every process in the experiment. Consent forms (Appendix 1) were printed out and given out to the participants to attain written consent, to confirm all participants were over the age of 16 and understood the aim of the experiment (a test on Memory). A debrief (Appendix 2) was printed out and read after the experiment had ended to again explain the purpose of the test and remind participants of their ability to withdraw. A PowerPoint (Appendix 6) was used to display the words and images. A debriefing (Appendix 2) was read out at the end to conclude the experiment, reveal the aim of the memory tests and remind participants of their right to withdraw.
Procedure
A class of 14 students consisting of 11 females and 3 males were asked to participate in a psychology experiment. The participants were brought into the laboratory (empty classroom) and introduced to the experiment and read the first part of the standardised instructions, explaining that the purpose of this experiment was to study memory. Consent forms were then handed out, which all participants read and signed, attaining gained consent. The consent forms were collected in and score sheet 1 was handed out. Before beginning the words section the experimenters read out part of the standardised instructions that informed participants that 15 words would be displayed individually on the screen for a duration of 2 seconds per word. Once all 15 words were shown, participants were given 60 seconds to recall as many words as possible. Participants were told that the order of words was not necessary to the findings.
Score sheet 1 was then collected and score sheet 2 was handed out. Before beginning the images section, the experimenters read out the part of the standardised instructions that informed participants that 15 images would be displayed individually on the screen for a duration of 2 seconds per image. Once all 15 images were shown, participants were given 60 seconds to recall as many images as possible. Participants were told that order of images was not necessary to the findings, the same as in the previous condition.
During both the word and methods section the participants were observed carefully to make sure none of them were discussing answers. Once the image section was completed, the second score sheet was collected and the students were read out the debrief. This explained the true aim of the experiment was to find out if visual images or visual words would be recalled better through short-term memory – and reminded them once again that participants were able to withdraw at any time after the experiment by contacting their Psychology department.
Ethics
This experiment was conducted in accordance with the BPS code of ethics. The consent forms and standardised instructions were written to ensure no physical or mental harm was done to any of the participants in the memory experiment and all participants could withdraw at any time. All 14 students were over 16 and fully consented. The consent forms informed participants that the study was on memory recall and their answers will be kept completely confidential and anonymous. All score sheets after the memory experiment would be destroyed after research was over. Participants were told the experiment would last around 20 minutes and if they had any questions they could ask any of the experimenters during or after the experiment.
Once the participants completed the memory experiment, the debriefing was read which repeated to all participants the right to withdraw from the study. This also explained that the aim of this experiment was to prove the hypothesis that, using short term memory, images will be recalled better than words. This was only revealed in depth at the end to ensure the experiment was fair and participants were unaware of what the exact aim of the experiment was. The debriefing was read to answer any questions they may have and guarantee all participants wellbeing’s were protected and left the experiment in the same frame of mind they entered it in.
Results
The results were collected by the experimenters and the mean, median and mode were calculated as it was the most efficient way to calculate the measures of central tendency. The range was calculated to find the difference between the highest and lowest results. The mean calculated the average of the total results, and was useful as it took each participants’ individual score into account. The median and mode were useful for finding both the midpoint for the results and the most commonly occurring number of correct answers given.
Below is a table and graph of the results.
(Calculations can be found in appendix 7).
Table 1: A summary table of participant recall – Conditions 1 & 2
Mean
Median
Mode
Range
Words
(Condition 1)
7.1
7
7.5
6
Images
(Condition 2)
8.8
9
9
7
Graph 1 – Statistical findings of participant recall
Summary of results
Table 1 and graph 1 both show that the overall range of the words was lower than the images, with the words having a range of 6, while images was 7. This shows that images were recalled better. The mean for images was 8.8 and 7.1 for words, again proving that people recalled images better. The median and mode calculations also show this, with words being 7 for median and 7.5 for mode, meanwhile images were much higher at 9 for both the median and mode. Across the table we see higher results for the images section – proving the experimental hypothesis that visual images would have a higher recall rate than visual words. These results refute the null hypothesis that visual words would be recalled better than visual images, due to the lower overall results after calculating the mean, median, mode and range of them all.
Discussion
The results support the hypothesis that visual images would have a higher recall rate than visual words, supported by the mean, median mode and range calculations. In condition 1, the table shows that the mode of correct words recalled was 7.5 – compared to 9 for condition 2. This is a stark difference between both conditions – with the graph showing the mean of words and images being 7.1 to 8.8. The median of the words was 7 for words and 9 for images, again supporting the experimental hypothesis. The range of all results, which took raw data into account, also supported the hypothesis. The range for words was 6, while the range for images was 7. Although some of the calculations didn’t have a particularly large or notable difference, it is clear that overall, the visual images shown to participants were recalled more accurately than visual words. These results therefore disprove our null hypothesis that visual words would be recalled more accurately than visual images.
The majority of the experiment ran smoothly and efficiently, as it was a lab experiment in order to ensure it was easily replicated and both conditions were conducted in the exact same environment. This method was effective in ensuring the dependent and independent variables were controlled. There were some extraneous variables that were accounted for beforehand, such as ensuring the words and images were large and bold enough to see by participants sitting at the back of the room. The examiners ensured the environment was in a quiet exam condition to ensure no participants were distracted by outside noise. However, some EVs were unprepared for – such as participants finding some of the images unclear. In the pilot study, it was found that some of the participants confused certain images (e.g. some wrote ‘computer’ instead of ‘TV’). This was corrected before the final experiment, but it was found afterwards when examining the answers that a small number of participants still experienced confusion between certain images. A way to have avoided this would have been to use clearer and more universally understandable images. Another EV was the harsh lighting in the room affecting visibility of the PowerPoint for some participants at the back of the room. This affected visibility and possibly impacted their results if they were unable to see every word or image clearly. To combat this, we could have altered the lighting of the room or performed the experiment earlier on in the day during broad daylight to avoid harsh indoor lighting. A confounding variable could have impacted the results, for example, some students not being interested in the experiment and therefore not trying to remember as many words as they possibly could, compared to others who might have used their full capacity for recall.
The small sample size used in the experiment was not a representative of the population and is therefore inaccurate to use as a broad discussion of everyone’s memory capabilities.
The experiment had high face validity as the participants had clear instructions and were aware of what the experiment entailed. However, there is lack of ecological validity in this experiment because it is unlikely to be replicated in the real world and it is not possible to collect these findings in a real-life situation due to the laboratory condition and controlled nature of the experiment.
Conclusion
Overall, the research conducted in this experiment not only confirmed the experimental hypothesis and refuted the null hypothesis, but also supported research conducted by other notable psychologists such as Paivio and Csapio (1971) and Bower (1972), along with Paivio’s 1986 ‘dual-coding theory’. Therefore, this accepts the theory that visual images are recalled better than visual words through short-term memory.