Research has shown that organizational strategies aid in memorization tasks such as word recall. Several studies have shown the effectiveness of using organizational strategies such as hierarchical categorization in aiding in word recall. Our experiment, a partial replication of the study conducted by Bower et al. (1969), examined the impacts of hierarchical word lists on word recall. College students were presented with word lists that were arranged either randomly or in categories. The number of words correctly recalled was measured for each participant. While our results were not as definitive as Bower et al. (1969) study, they do yield implications for further research for additional age groups.
The Impact of Categorization on Word Recall
Research has shown that organizational strategies aid in memorization tasks such as word recall. Matlin (2002) presents four such organizational strategies: chunking, first-letter technique, narrative technique, and hierarchy technique. In chunking, small bits of information are combined into larger, more meaningful units. An example of this is telephone numbers. A seemingly random set of ten numbers (2125076573) may be difficult to remember, but when organized into chunks -- [HIDDEN] , it becomes much easier to remember. Similarly, a method frequently used by students to organize information is the first-letter technique. For example, most elementary school students learn the order of mathematical operations using the acronym PEMDAS (Please excuse my dear Aunt Sally) which stands for Parentheses, Exponents, Multiply, Divide, Add, and Subtract. The third technique described by Matlin (2002) is the narrative technique. With this technique, people are instructed to create a story to remember the words. Matlin (2002) cites a study where participant using the narrative technique recalled six times as many words as those who did not, but cautions that the technique is only effective if the story is easily generated in both learning and recall. Finally, constructing a hierarchy is an effective way to remember information. Matlin (2002) explains that a hierarchy is "a system in which items are arranged in a series of classes, from the most general classes to the most specific" (p. 166). Several research studies have demonstrated the effectiveness of the hierarchical technique.
One of the first studies was conducted by Cohen (1963) to evaluate the hypothesis that "there should be no significant difference between the number of words recalled from a list of 20 unrelated words compared to the numbers of categories represented by the recalled words of a list of 20-word categories" (p. 227). To prove this hypothesis, he conducted two experiments in which nine groups of participants were presented with lists of words for recall. Participants were divided into three groups. The first group was presented with lists of 10, 15, or 20 unrelated words. The second group was presented with lists of 10, 15, or 20 non-exhaustive word categories, and the third group was presented with 10, 15, or 20 exhaustive word categories. Exhaustive word categories include groups of words where three or four words represented all or nearly all of the words in the category (e.g. spring, summer, autumn, winter). Non-exhaustive categories include word lists where three or four words were selected from a larger sample (e.g. arm, leg, head, and hand). In the first group, credit was given for words recalled which were on the list. In the second and third groups, credit was given for a category if at least one word in the category was recalled. Cohen (1963) found that the number of chunks recalled from the 20-category lists did not differ significantly from the number of words recalled from the list of 20 unrelated words.
Bower et al. (1969) built upon the work by Cohen, but made three significant changes. First of all they added the category label to the categorized list. Secondly, word lists selected were organized as a hierarchy with higher and lower levels of categories. For example, in the category of minerals, there are subcategories of metals and stones, and those subcategories are broken down further into more subordinate categories. Finally, rather than presenting the words one at a time, Bower et al. (1969) presented a complete set of words all at once. Participants in the experiment were presented with sets of words either blocked or randomized. Participants receiving the blocked lists were presented with words organized in hierarchical trees similar to Figure 1. Participants receiving the randomized word lists were presented with the same words which were scrambled but were arranged into a similar-appearing spatial tree, as shown in Figure 2. Bower et al. (1969) found that the group receiving the blocked lists (WB) recalled significantly more words than did the group receiving the randomized list (WR). On the first trial the WB group correctly recalled 73 words while the WR group correctly recalled 20.6 words. Additionally, performance improved over additional trials. On the fourth trial, the WB group correctly recalled all 112 of the words, while the WR group correctly recalled 70.1 words. Bower et al. (1969) conclude that "If S. can discover or learn a simple rule or principle which characterizes the items on a list and which relates them to one another, then he uses that rule as a retrieval plan in reconstructing the items from memory, with a consequent improvement in his performance" (p. 340).
A more recent experiment was conducted by Longnecker et al. (2010). This study examined 200 participants, half of whom were healthy and half who had been diagnosed with schizophrenia. Participants were selected from the National Institute of Mental Health's Schizophrenia Sibling Study based on their scores on the California Verbal Learning Test (CVLT). The highest 50 and lowest 50 performers were selected. All were between the age of 21 and 55, and all were free from any issue that might interfere with test performance such as alcohol or drug abuse, suffering from loss of consciousness, or other medical or neurological issues. Participants were orally presented with a list of 16 words and then asked to freely recall them. This process was repeated 5 times. Longnecker et al. (2010) found that "Regardless of diagnosis, low performers were more likely to recall the first 2 and last 4 items from the list; high performers increased engagement of semantically-based transitions across the 5 trials, whereas low performers did not" and concluded that "the number of words recalled is predictive of the strategy used, with a greater number of words recalled following a greater reliance on a semantically-based strategy" (p. 634).
As these studies have shown, the use of cognitive strategies and mnemonic techniques such as hierarchical organization improves the ability to remember a given set of information. Our experiment will partially replicate the work of Bower et al. (1969) to demonstrate how the organization of sets of words influences the rate of word recall. Our hypothesis is that words organized into hierarchical categories should result in a higher rate of word recall that words that are randomized.
Participants (n=20) were undergraduate college students enrolled in a psychology course. Participants completed the experiment voluntarily as part of an assignment for the class. Participants were instructed to study four lists of 25 words which were presented for one minute on a computer screen. The same lists of words were presented to all participants, but half of the participants (Group A) were presented with lists of words that were arranged in hierarchical categories. Figure 1 is an example of a categorized list. The other half of the participants (Group B) were presented with lists of words that were not organized into categories. Figure 2 is an example of an uncategorized list. The computer randomly assigned either categorized or non-categorized lists to the participants. Once the four lists were presented, participants were given five minutes to recall as many words as they could. The number of true recalls (words recalled correctly from the list) was totaled. Plurals as well as words in upper or lower case were scored as true recalls. Additionally, the number of intrusions (words recalled that were not on the list) was recorded.
Group A, the group of participants presented with the categorized word lists correctly recalled an average of 41.9 words out of 100 while Group B, the group of participants who were presented with the uncategorized word lists, correctly recalled an average of 37.9 words out of 100. Additionally, Group A had an average of 5.3 intrusions while Group B. had an average of 4.4 intrusions. Results for all participants are presented in Figure 3.
Our experiment was a partial replication of the one completed by Bower et al. (1969) which compared the free recall of hierarchical word lists presented either in categories or in randomized lists. However, our results did not produce the differences in the two tested groups that were present in the Bower et al. (1969) study. Our study only showed a difference of 4 more words recalled between Group A and Group B. Additionally Group A had a higher rate of intrusions than did…