The investigation into whether images aid memory recall has been fairly successful. The experimental hypothesis: participants in condition 1(words with images) would perform better than those in condition 2 (words without images), has been rejected therefore the null hypothesis: there will be no difference in condition 1 (involving grid of random words supported by images) and condition 2 (only consisting of random words) has been accepted.
Only to some extent from the line graph (Condition 1 vs. Condition 2), can we say condition 1 (words with images) recalled more words than those in condition 2 (words without images) because the line for condition 1 is generally above the line for condition 2 and this fairly supports the experimental hypothesis. The results measured using central tendency and measures of dispersion moderately support the experimental hypothesis.
The mean and median for condition 1 (14. 1 & 14. 5) was greater than of condition 2 (12. 6 & 13. 5) indicating that there is a better recall when words are companied by visual aid and this supports the experimental hypothesis. The range (11) illustrated that data for both conditions were equally spread out, which does not necessary support the experimental hypothesis. Standard deviation, a more reliable measure of the dispersion than range has shown data in condition 2 (3. 977715704) is slightly more spread out than in condition 1 (3. 604010112).
Descriptive statistics (numerical & graphical) have illustrated that condition 1 (words with images) performed better and this supports the experimental hypothesis. Not only do the numerical statistics support the hypothesis but the results are fairly pronounced, simply by looking at the line graph (graphical statistics); there is definitely a difference between both conditions although the Mann-Whitney U-test shows this to be below the level of significance.
Inferential statistics, which enable us to draw clear conclusions about the likelihood of the hypothesis being true, is evidence for accepting the null hypothesis. The implication of the results measured using a non-parametric method, The Mann-Whitney U-test clearly shows that at the significance level of 0. 05, the results were very likely to have happened by chance. Though the observed value, 35 is greater than the critical value of 23 but only by 10. Comparatively, the findings of this study support previous theory and research but also are contrary to them.
Bower, 1972 imagery recall experiment found that participants, who used imagery, recalled 80% of the words compared to only 45% by the non-imagers. In this investigation, participants in condition 1 (words with images) recalled 70. 5% of the words compared to 63% by the non-imagers. My investigation does support Bower’s findings, since there is a difference however the differences between the two conditions is not significant. Previous research does suggest that data in STM is stored in an acoustic manner this was demonstrated by Conrad (1964).
Illustrating people may not always use visual codes to remember data and this supports the null hypothesis; in addition Baddeley’s (1966) study suggests that data is also stored semantically and this again illustrates that people may not always use visual codes. A possible problem with this experiment is the experimental design which lacks ecological validity and also does not account for individual differences. The study used independent groups, which meant it lacked control of participant variable and needed more participants.
A matched participant design could have been used instead to deal with participant variables as participants are matched on key variables such as age and memory ability. Another problem of the study was the sample size. The study only used 20 participants, a small sample; therefore findings cannot be generalised to the rest of the population. So, a larger sample size of 50+ would be more representative and can be generalised. In addition, the investigation used opportunity sampling, which is very biased, and again it cannot be generalised.
Random sampling would be the best method to select participants since it is potentially unbiased. An additional problem was the images used; some of the images were cartoon images (i. e. bible & pizza) and others were actual photographs (real life images, i. e. greenhouse). This may have been a problem since real life images may possibly be easier to remember than cartoon images or vice versa. To resolve such problem, would be by using only cartoon images alone or only real life images.
The recall between genders may have wider implications: on the whole men have performed better than women, when studies have shown women should perform better than men. For example; women perform better than men in tasks such as verbal learning remembering tasks, name face association, and first last name associations learning (Larrabee and Crook, 1993). Ideas for a follow up research study is the effects of gender recall and discover if there is a significant effect for the sex of a participant on the types of gender associated images recalled and to compare the effects of visual and semantic codes in depth.
There is evidence to illustrate that people do remember more with the aid of visual representation and evidence to illustrate people remember more without any visual representation. This itself suggests people do remember data and information differently (individual differences), i. e. acoustically, semantically. Therefore it is important to recognize these findings as they have strong insinuations for its involvement in everyday life, for example with teaching, revision and marketing.
References & Bibliography
Atkinson, R. C. & Shiffrin, R. M. (1968) Human memory: A proposed system and its control processes. In K. W. Spence and J. T. Spence (Eds. ), The psychology of learning and motivation, vol. 8. London: Academic Press Baddeley, A. D. (1966) Encoding in LTM: The influence of acoustic and semantic similarity on long-term memory for word sequences. Quart. J. exp. Psychol. , 18, 302-9 Bahrick, H. P. (1975) The nature of LTM: ‘Fifty years of memory for names and faces: A cross-sectional approach’, Journal of Experimental Psychology: General, vol. 104, pp. 54-75