Experiment 2. Organization in memory
19 participants took part in the first experiment. They had previously reported no memory issues or complaints, and did not have mental diseases.
Apparatus and procedures
There were two trials in the experiment, using organized and unorganized lists. In each trial, the participants were provided four lists of 26 words, and were given one minute for studying each list. After the study time, the participants were asked to recall as many words as they could in any order during the next 5 minutes. Next, the participants reviewed a set of all words in alphabetized order, so that they could mark correctly recalled words and false recalls.
Both organized and unorganized lists were presented using hierarchical categories, with subheadings and subcategories. The difference between two conditions is the following: in “organized”ť lists the words were meaningfully arranged into categories and subcategories, while in the “unorganized”ť lists the arrangement was random.
The number of correct recalls was measured for every participant in both trials (the dependent variable). Independent variable was the list condition (organized or unorganized). Counterbalancing in this experiment was performed basing on subject number and on the condition assigned to the subject.
Mean number of correctly recalled words was 41.92 (SD=23.11) for organized lists and 34.40 (SD=9.40) for unorganized lists for the first recall, and 53.46 (SD=30.75) for organized lists, and 44.77 (SD=14.85) for unorganized lists in the second recall. Overall, mean number of correctly recalled words for organized lists was 47.69, and 39.58 for unorganized lists. Figure 2 illustrates the differences of average number of correctly recalled words grouped by list type and number of recall.
Figure 3 shows the dynamics of changes of average number of correctly recalled words for organized and unorganized lists for first and second recall. Basing on the analysis of graphical data, it is possible to suggest that there was main effect of recall and main effect of list type on the mean number of correctly recalled words, with almost no interaction effect. However, ANOVA analysis of the number of correctly recalled words showed that there was main effect of recall, F(1, 17) = 15.885, p=.001, but there was no main effect of list: F(1, 17) = 0.726, p=.406. There was no interaction effect witnessed: F(1,17) = 0.046, p=.833.
It is possible to state that hierarchically organized lists have a statistically significant positive impact on information organization in memory; this result is aligned with the findings of Bower, Clark, Lesgold and Winzenz (1969). At the same time, the absence of statistically significant effect of list organization does not correspond to previous research findings, and is most likely related to the limited number of the participants, and not equal number of participants for organized and unorganized lists (n=10 and n=9 accordingly). The experiment can be refined by using lists of different length and by adding more participants, with equal groups for both types of lists.
Both experiments provided results corresponding to the findings of previous studies, but in both cases ANOVA analysis showed that the results were not statistically significant. Therefore, it is reasonable to organize further experiments so that possible confounding variables and other factors affecting statistical significance will be removed.
The experiment on iconic memory can be modified using the suggestions of Averbach and Coriell (1961): adjustment of brightness and use of visual indication instead of audio indication. It might be particularly interesting to investigate the differences of iconic memory for people with focus on different senses: visual, audial and kinesthetic leading channels. The effects of interaction can be studied using the model provided by Eriksen & Collins (1967) where brightness of background and image as well as the interaction of dotted images were considered; again, in this context it would be particularly interesting to study the differences for participants with different main sense focus.
The experiment on information organization showed that there was main effect of recall type on the proportion of correctly recalled words, but there was no statistically significant main effect of list organization, despite appropriate suggestions associated with graphical representation, and the fact that the difference between the means supported initial hypothesis. Further development of the experiment should include involvement of a larger number of participants, equal number of participants for organized and unorganized lists, and exploration of retroactive facilitation of words contained in the hierarchic list during the second recall.