My experimental hypothesis is:
‘It is predicted that Ps will recall more words from the list with associated questions related to the meaning of the word (deep processing) than from the list with associated questions related to the case of the word (shallow processing).’
My null hypothesis is:
‘It is predicted that there will be no difference in the recollections of words from the list with associated questions related to the meaning of the word (deep processing) than from the list with associated questions related to the case of the word (shallow processing).’
Method
Design and Overview
My investigation will be an experiment. I chose the independent groups design because there is an element of surprise in the task – the participants do not know beforehand that they will have to recall the target words – and to use repeated measures would mean the participants would know by the second condition what is coming, and would therefore make an effort to remember the words (showing demand characteristics). To use matched participants would not be practical either because it is too time consuming, and it would be difficult to try and match people on their memory (especially as this would give away the aim of the study).
I am only using the two conditions, to show the difference – if any - in the remembrance of the words, and also because it is simpler and less time consuming. My IV will be the level of processing involved in thinking about the word, either shallow or deep. This was operationalised by the questions given to the participants:
Condition A. ‘How useful would this item be in the desert?’ – This involves deep processing (semantic)
Condition B. ‘How many letters are in this word?’ – This involves shallow processing (physical)
The DV was the amount of words remembered by each participant. This was operationalised by collecting their results on answer sheets (in the appendix) and working out a mean from both conditions.
One EV of my experiment that I had to control was using words that do not sound similar to other words or could be spelt in two different ways. For example, if I had used the word ‘Hole’. This could have been a problem in my experiment because if I had used words like this, which I would have said to the participants, it may cause confusion. It would affect the two conditions in different ways, as those thinking about a word’s usefulness would have realised when I said ‘Hole’ that I meant ‘Hole’, the noun, as opposed to ‘Whole’, as with their question it only made sense to use nouns. However, if I used this word with the group counting the letters in the words (shallow processing) they would not know whether I meant ‘Hole’ or ‘Whole’. Thinking about this would mean they would process the word deeper, or perhaps question which was meant therefore meaning they had more time to think about it. Therefore I made sure I used words of which there would be no confusion in meaning.
I will make sure that the conditions of the experiment are controlled, by ensuring all participants take part on their own so there is no influence from other people. I will time the first part of the experiment (the rating of the words) to make sure everyone has a equal chance to process each word, as well as using the same words for each participant. I used a variety of words, none that sound similar, to make sure there isn’t a problem of interference.
Subjects and Participants
My target population was the students at Exeter College. I decided only to use teenage students, as mature students may have a different memory capacity. The age range of the target population was therefore 16 – 19. I also checked before the experiment with those I asked that they were not psychology students, in case they know what is involved in the study.
From this I used opportunity sampling to find 20 participants – 10 for either condition. For this I stood in the main college building and found people available and willing to partake at the time. Opportunity sampling is quick and easy, although can be seen to be biased as participants often have the same sort of qualities. However, as I was testing memory I did not need a vast range of personal qualities, and therefore my sampling method did not greatly limit my research. I did not ask for any personal information e.g. name as I didn’t think this was relevant to my research. Therefore my participants will be anonymous.
Ethical Considerations
It was important to make sure I had consent for ethical reasons. Although I gained their consent, it wasn’t possible to gain their fully informed consent, because of the nature of the task and the element of surprise. If I was to reveal all the details of the experiment beforehand, there would be a possibility of demand characteristics. Before each person takes part in the experiment I will explain to them the basics of what is involved in my research, before getting their consent to participate. To overcome the problem of the consent being fully informed, I will have to debrief the participants after the experiment. This will be done by explaining afterwards the purpose of the task and why they were not informed of the whole process beforehand. If this is a problem for the participant they will be able to ask to be omitted from the study. Consent forms, my brief and debrief will all be included in the appendix.
Materials and Apparatus
I did not need a lot of materials for my experiment – only my word list and forms for the participants. These forms are included in the appendix, and consist simply of a title, basic instructions and a bulleted list to write either the answers to the questions (e.g. ‘how many letters does this word contain?’) or to recall the words. I created the forms/answer sheets on Microsoft word.
The word list I used was made by flicking through the dictionary and finding nouns with a variety of length. I only picked nouns that were used in general conversation – I couldn’t use words that people would not know, as this would confuse them and perhaps affect the results. The words used are listed in the appendix. I chose the questions based on complexity – I needed a simple question for the shallow processing condition and therefore chose to ask about word length because it did not involve much thinking for the participants. For the deeper processing question, I had to think of a question that would make the participants think more about the meaning of the word. I chose the theme of the desert because the P’s would think about the noun in a different context and therefore process it deeper. The only other equipment I used was my watch, to time the intervals between the readings of each word.
I kept all the forms in a folder and made sure they all looked the same, so that those taking part wouldn’t realise the other parts of the experiment.
Procedure
I conducted my experiment in the following way:
- Approached target population – those who didn’t look in a rush.
- Asked them if they’d be willing to take part in a psychology experiment, ensuring they were in the age range I wanted and also checking they were not psychology students. If they agreed I briefed them, using the brief contained in the appendix.
- After the brief, if they continued to agree, I asked them to sign the consent form.
- I then gave them a form for the first part of the experiment. I let them read the question at the top of the form, made sure there were no distractions, asked them if they were ready, and then read them the words from the word list.
- When reading them the word list, I made sure they only had 5 seconds to process each word by timing, using my watch. I also made sure I stood facing them with my folder up. This ensured they couldn’t see the words.
- After they had completed the first part of the experiment, I took their form and put it in the folder. I thanked them and then told them the experiment wasn’t finished, and asked them whether they’d mind completing it. If they agreed I briefed them for a second time. I made sure this took exactly a minute.
- If they understood and were willing to continue, I gave them the second form and 3 minutes to recall as many words as they can.
- If they did not agree, they had the choice to withdraw themselves and their results from the experiment.
- After the 3 minutes, I collected their second form and then thanked them again.
- Finally I debriefed them, using the debrief form contained in the appendix. I thanked them for taking part and asked finally if they were willing for their results to be used.
Results
My results from the experiment are shown in the table, and also in the graph below.
The mean number of words recalled in Condition A (Deeper processing) was 12.1 out of 20, while in Condition B (Shallower processing) the mean was 7.8. This is a difference of 4.3 words between the two conditions – with participants in condition A recalling 20% more of the words than those in condition B. Although the sample was only small, this seems to show a big enough difference to come to a conclusion. This conclusion is that the mean results clearly support the experimental hypothesis, as the participants in Condition A scored much higher in remembering the words than those in Condition B. If the theory behind my hypothesis is correct, this is because they had to process the words more when rating them and therefore the memory trace for each word was deeper, hence recalled more easily.
The raw data of my experiment (e.g. the results out of 20 for each participant) are included in the appendix, as well as the calculations for the mean. I did this for both A & B by adding up each score in that condition and then dividing by 10 (the number of P’s).
Discussion
The results of this experiment are similar to the results of Craik and Tulving’s experiment. There was not only a difference in the means, but a significant one – this supports the experimental hypothesis and implies the type of processing involved when thinking about something does affect how well that information is remembered.
My results are similar to that of Craik and Tulving’s experiment, not only by the conclusion but also from the actual results. When converted to percentages, participants in condition A of my experiment (deeper processing) remembered 60.5% of the words, compared to 65% of those in Craik and Tulving’s study, which led too their Levels of processing theory . However, those in condition B of my experiment remembered 39% of the shallowly processed words – much more than the 17% in C&T’s study. The difference in these results could be because of many reasons e.g. small sample size, different age in participants or changes to original study.
Limitations
One limitation of my study is the target population. I used Exeter College students, aged 16-19, who weren’t psychology students. Although this was practical for my study, Craik and Tulving used a more varied sample of people from different ages. Using only 16-19 year olds meant I may have got an unrepresentative selection of people, with regards to memory. This limits my study because comparisons to Craik and Tulving are not entirely valid. However, by using this sample, I could build on my study by comparing different age groups.
Improvement
As mentioned in limitations, I could improve on my research by comparing this experiment to another carried out in the same way, but with different age groups. This would see how this study, along with the Levels of Processing theory, compares with a different target population, and whether age makes a difference on memory function.
Further Research
One way I could expand on this research is to repeat the study but using a third condition, acoustic processing, which was used in Craik and Tulving’s study. For example, I could use the question ‘how many syllables are there in this word?’ and use the same format of procedure and answer sheets. This would provide a third results to compare with the experimental hypothesis.
Conclusion
From my results, and comparisons to the previous theory and study, I believe the Level of Processing theory can be supported enough to be assumed true. This has implications for the real world, not just in psychology,
For example, anyone trying to remember things, such as students revising, could take the Levels of Processing theory to adapt their revision programme to remember more, using semantic processing of information instead of perhaps just reading information.
Reference: Craik, F. & Tulving, E. (1975) Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104, 268-94.
Appendix
- Brief – this was used to prepare participants for the experiment.
- Consent Forms – these were an ethical consideration.
- Word list, raw data, calculations for means
- Answer Sheets – these are a sample of the data collected from the participants.
- Debrief – this was used to inform participants after the experiment the full purpose of it.
Word list
Raw Data – Words remembered in each condition
Calculating mean results (Numbers in brackets represent individual results, as shown in table above. These were divided by 10 as this was the number of participants for each condition.)
Condition A
(13+15+12+10+11+13+7+16+13+11) / 10 = 12.1 (Mean number of words recalled out of 20 in the deeper processing condition)
Condition B
(5+7+9+8+10+6+7+7+11+8) / 10 = 7.8 (Mean number of words recalled out of 20 in the shallower processing condition)