This study investigates the Involvement Load Hypothesis proposed by Laufer and Hulstijn (2001). The involvement load hypothesis posits that vocabulary learning is determined by involvement load or mental effort. Involvement load has three components, need, search, and evaluation and each component is scored for three levels: index 2 for the strongest, index 1 for a moderate degree, and index 0 for none.

Each participant learned six words with graphic organizers at the high involvement load (need index 1, search index 1, evaluation index 2, total index 4) and six at the low involvement load (need index 1, search index 1, and evaluation index 0, total index 2). Immediately and one week after completing the graphic organizer task, vocabulary knowledge was measured using three vocabulary tests that tested different levels of vocabulary knowledge: a translation test, a difficult multiple-choice test, and an easy multiple-choice test. Quantitative analyses of data from 291 university and college students in Japan were conducted, and audio-recordings from five pairs were analyzed to examine learning processes.

Repeated measures MANOVA and ANOVAs revealed significant differences between the conditions of the two involvement loads in the translation test and the easy multiple-choice test, but not in the difficult multiple-choice test. The effects of Task and Time were statistically significant, but there was no interaction. There were significant differences between the immediate test and delayed test observed in the translation test and the easy multiple-choice test, but not with the difficult multiple-choice test.

The current study supports the involvement load hypothesis, but caution is advised. Even though the high involvement load graphic organizers yielded more vocabulary retention than those with less involvement load in two out of the three vocabulary tests, the differences in mean scores were small and extensive differences were not observed in the participants' discussions. Additional statistical analysis indicated that the three vocabulary tests measured three levels of vocabulary knowledge. Determining the effectiveness of graphic organizers for vocabulary learning was only mildly successful as forcing greater involvement load proved to be challenging.