Latin vocabulary acquisition: an experiment using information-processing techniques of chunking and imagery.

This study investigated the effects of chunking and imagery, two information-processing techniques, on Latin vocabulary acquisition and memory retention. A total of 121 students enrolled in high school Latin I classes participated. Following a pretest, study participants were divided into four groups and provided a list of 21 Latin vocabulary words to study. Comparison Group A received the list of words grouped randomly into three groups of seven. Comparison Group B and C received the list of words categorized by related definition into three groups of seven. Comparison Group C received a five minute imagery treatment prior to the immediate posttest. The Control Group X received neither a chunking technique nor an imagery treatment. Immediate and delayed posttests revealed significant differences between the groups. Imagery and chunking instruction were found to improve student performance among study participants.

**********

The purpose of this study was to investigate chunking and imagery, two information-processing techniques, on students' Latin vocabulary acquisition and memory retention. Much of the research in foreign languages has focused on linguistics and the systematic study of phonology. While this has proven useful for a better understanding of teaching foreign language sound systems and grammatical structures, the manner in which students acquire foreign language vocabulary has received little systematic attention.

One of the more interesting questions in Latin instruction has been the process by which the reader focuses attention on word meanings, word order cues, and cues given by the inflectional endings. How does a skilled reader recall the meanings of Latin words before transferring those words into comprehensible sentences? Classicists need to answer this question before they can understand how one goes about reading Latin as Latin. Given the limited working memory of human beings, skilled reading simply cannot take place until lower level skills such as decoding, lexical access, and parsing become automatic (Hamilton, 1992). The present study addresses the following question. Can the information-processing techniques of chunking and imagery positively affect the automaticity of Latin vocabulary acquisition and retention?

The information-processing model focuses on transformations occurring as sensory impulses enter the human information-processing system. Information is either lost or stored in short-term or long-term memory. This model has proven to have wide applicability to learning, memory organization, and teaching strategies. The theory supports the techniques of imagery and chunking to enhance vocabulary acquisition and memory retention in the learning process (Joyce & Showers, 1991).

Inherent in the information processing model is the assumption that the human mind can process and retain on average seven bits of information at one time. Therefore, the word "chunking" as used in the study refers to the grouping of Latin vocabulary words into chunks or groups of seven. The imagery component of the study refers to the visualization of meanings of Latin words through verbal mnemonics prompted by the teacher (Gredler, 2001).

Methodology

A total of 121 subjects enrolled in eight Latin I classes from three North Texas suburban high schools with a population of more than 1600 students participated in the study. Each subject first took a pretest on 21 Latin vocabulary words taken from Jenney's Fourth Year Latin (Jenny, Scudder & Coffin, 1990) vocabulary list. All subjects were given 10 minutes to complete the pretest. Following the pretest, each subject then received a list of these same 21 Latin vocabulary words to study. Comparison Group A received a list of 21 words grouped randomly into three groups of seven. Comparison Group B and C received the sane list of 21 words grouped into three groups of seven based on related definitional categories. In addition, Comparison Group C received a five minute imagery treatment prior to the immediate posttest The Control Group received neither a chunking technique nor a imagery treatment.

After each comparison group participant was given 15 minutes to study, an immediate posttest was administered. A delayed posttest was offered each participant two weeks later.

Findings

To determine the equivalency of the pretreatment groups on Latin acquisition, an AVOVA was performed on the pretest means. No significant difference was found between the pretest means, F (3, 117) = 1.54. The mean number of words correctly identified by each group ranged between 1.5 and 2.8.

To assess the impact of chunking and imagery on Latin acquisition, immediate posttest means were compared using the ANOVA statistical procedure. A significant omnibus F was obtained, F (3, 117) = 6.91 (p<.01), indicating a difference between the comparison group means. Group means ranged from 15.69 (Comparison Group A) to 20.27 (Comparison Group B). The Fisher LSD test was subsequently employed to examine each pairwise comparison. Mean differences and multiple comparison test results are found in table 1., and reveal the following: 1) a significant difference between the mean achievement of students who received chunking with related categories and students who received no treatment 2) a significant difference between the mean achievement scores of students who received chunking with related categories and students who received chunking without related categories 3) a significant difference between the mean achievement scores of students who received imagery and chunking and students who received no treatment, and 4) a significant difference between the mean achievement scores of students who received imagery and chunking and students who received chunking without related categories.

Analyses of delayed posttest scores also revealed differences between comparison groups. A significant ANOVA omnibus F ratio was again calculated, F (3, 117) = 6.46, p<.01. Delayed posttest means ranged from 4.52 (Control Group X) to 9.47 (Comparison Group C). Results from subsequently performed Fisher LSD multiple comparison procedures, along with mean differences are provided in table 2. Pairwise comparisons revealed the following: 1) a significant difference was found in the mean achievement scores of students who received related chunking and students who received no treatment 2) a significant difference was found in the mean achievement of scores from students who received related chunking and imagery and those who received random chunking and students who received no treatment.

Conclusions

Based on the findings, it is possible to conclude that imagery and chunking instruction significantly improve student performance among high school Latin I students. This has implications for both students who need to memorize, retain, and retrieve vocabulary and language facts and to the teachers who teach them.

In order to fully utilize chunking and imagery techniques, teachers must first be exposed to, and develop an understanding of the two methods. Teacher training can be addressed through staff development on such topics as brain research, models of teaching, and more specifically, information-processing techniques of chunking and imagery.

Table 1

Immediate Posttest Mean Differences
Fisher LSD Test Significant Pairwise Comparisons

    X         A        B          C
    Control   Random   Related    Chunking and
                       Chunking   Imagery

X             .64      * 2.49     * 3.94
A                      * 3.13     * 4.58
B                                   1.45

Note: An asterisk * represents a pair of groups significantly
different (p<.05).
Table 2

Delayed Posttest Mean Differences
Fisher LSD Test Significant Pairwise Comparisons

    X         A        B          C
    Control   Random   Related    Chunking and
                       Chunking   Imagery

X             .83      * 3.01     * 4.95
A                        2.18     * 4.12
B                                   1.94

Note: An asterisk (*) represents a pair of groups significantly
different (p<.05).

References

Gredler, M. (2001). Learning and instruction. New Jersey: Prentice Hall

Hamilton, R. (1992). Reading Latin. The Classical Journal, 87, 165-174.

Jenney, C., Scudder, R.V., & Coffin, D.D. (1990). Teachers guide: Jenney's fourth year Latin. New Jersey: Prentice Hall.

Joyce, B. & Showers, B. (1991). Information-processing: Models of teaching. California: Booksend Laboratories.

Dr. Terri Carter, Assistant Principal, Grapevine High School, Grapevine, Texas. Dr. C. A. Hardy, Professor of Secondary Education, University of North Texas. Dr. James C. Hardy, Assistant Professor of Higher Education, Texas A&M University -- Commerce.

Correspondence concerning this article should be addressed to Dr. James C. Hardy, Department of Secondary and Higher Education, Texas A&M University -- Commerce, P.O. Box 3011, Commerce, Texas, 75429.