Time-on-task and its Facilitative Effects on Word Recognition in

As the results shown in 4.1.2.1 and 4.1.2.2, the performances between the MCG-E group and MCG group reached significant differences in terms of both the immediate and delayed posttests, favoring the MCG-E group. The results implied that the extra three minutes for the MCG-E group did have some facilitative effects on vocabulary word recognition. Here, it was noteworthy that the extended three minutes (eights minutes for the MCG-E treatment in total) was determined by the pilot study, which allowed the participants to use as much time as they needed to process the multiple-choice glossed passage. Therefore, the extra three minutes for the MCG-E group were not supposed to be used to memorize the words in reading, but rather, they were used to fulfill the lexical inferencing processes.

According to the Depth of Processing (Craik et al, 1975) and the Involvement Load Hypothesis (Laufer et al., 2001), the time-on-task should not be specifically

controlled among treatments for it is an inherent characteristic. Higher involvement tasks might take more time to complete. The results from the comparisons of the MCG and MCG-E groups support the notion, revealing that there is a necessity to have sufficient time-on-task respectively in terms of different treatments. With longer processing time, the amount of word recognition increased in the MCG-E

group. The results suggested that the sufficient time-on-task could be a conducive factor for learners who use multiple-choice glosses in reading.

Other researchers, though not specifically pointed out in their studies, have recognized the influential role the variable “time-on-task” plays. For example, Mondria (2003) indicated that inferring unknown words in reading usually takes more time for learners than reading paragraphs with translations of the unknown words. Without intentionally controlling the time limit, Rott (2005) found that adult learners utilized more semantic strategies and performed more elaborately processing behaviors while reading articles with multiple-choice glosses. The participants used the time to look for cues and testify their hypothesis of word meanings in the context. The higher amount of word recognition in the immediate posttest of the MCG-E group in the current study implies that junior high learners could make good use of the extended time to tackle with multiple-choice glosses while reading.

In the focused-group interview for the MCG-E group, the participants reported that a large amount of given time was used to ‘think.’ According to them,

‘to think’ refers to ‘look for’ and ‘evaluate’ the clues in the context. For example, one of the participants reported that she figured out the meaning of the word

‘negative’ by finding out the key word ‘no’ in the preceding sentence. Then, ideas conveyed in the succeeding sentences further provided her cues to distinguish the meaning of ‘negative (負面的)’ from the distractor ‘猶豫的.’ The participants’

responses, corresponding to the involvement load constructs proposed by Laufer et al. (2001), revealed that the participants really exerted mental effort to decode the unknown words in the passage. The given extra three minutes seemed to assure the

quality (the depth) of the processing, resulting in higher performances on the immediate posttest from the MCG-E group. Referring to the performances of the MCG group, the scores were comparatively lower even though the MCG and MCG-E groups received identical reading materials and the same gloss type.

Therefore, it is reasonable to conclude that more time should be given to learners in order to assure the quality of the processing of multiple-choice glosses.

In terms of the retention effects from the two treatments, as revealed in Table 4.4, both of the mean scores of the two groups declined drastically. The decline was expected because there were no vocabulary interventions implemented between the immediate posttest and the delayed posttest. Nevertheless, the statistical data indicated that there were significant differences between the MCG and MCG-E groups, favoring the MCG-E. In other words, the participants in the MCG-E group retained significantly more vocabulary than those in the MCG group. As for the reason why the MCG-E group had a higher amount of vocabulary retention, it was likely that the elaborate vocabulary decoding processes had reinforced the length the vocabulary was kept in the memory. As reviewed in Chapter Two, the depth of the processing accounts for the length the new information stayed in the long-term memory (Craik et al., 1972). Since the participants in the MCG-E group reported that they looked for cues, chose between the two alternatives, tested the hypothesis, and finally made a decision, the high involvement load of the task and the participants’ full engagement in the processing explained the results in the current study. Rott (2005), who has great insights on glossing, indicated that the quality and quantify of word intervention tasks may be ‘a predictor for word retention (p.113).’

In the current study, the extra time given to the MCG-E group seems to ensure the

quality of the processing of multiple-choice glosses in reading.

To sum up the discussions in 4.2.1 and 4.2.2, the current study verified Laufer et al. claim (2001) that time-on-task should be treated as an inherent characteristic of learning tasks. Some may argue that directly comparing the STG, MCG, and MCG-E groups in the current study may not be acceptable in statistics for there were more than one variable among groups. However, if the time-on-task was not accounted for in the current study, the facilitative incidental lexical learning effect from the multiple-choice glossed task became apparent, as the results shown in other research comparing the learning effects of different involvement-induced activities (Hulstijn et al. 2001; Hill and Laufer, 2003; Rott, 2005; Webb, 2005).

Therefore, the current preliminary study attempts to suggest that junior high learners can have more benefits in terms of lexical learning and retention from the multiple-choice glosses when they are given sufficient time to process the unknown words in reading. So far, many studies have suggested that senior high or adult learners could benefit from the multiple-choice glosses in reading (Hulstijn, 1992;

Lin, 2010; Lin & Huang, 2008; Nagata, 1999; Rott, 2005; Rott & William, 2003). As the current study reveals, the fact that multiple-choice glosses generate more mental effort and semantic elaborations could be true for learners at junior high level, as long as sufficient processing time is given.