基于价值的议程对学习时间分配影响的眼动研究

通过对比不同梯度下, 分值激发的议程与习惯性反应作用一致、不一致条件中, 优先选择项目和学习时间的差异, 考察基于价值的议程对学习时间分配的影响及其动态过程。结果发现：(1)等分值条件下, 汉语为母语被试存在从左到右的习惯性反应。(2)分值梯度对基于议程的学习时间分配的有效性具有调节作用。小分值梯度(1分、5分)激发的议程能够克服习惯性反应对学习时间分配的影响, 但不能使被试建立起优先学习高价值项目的议程; 大分值梯度(1分、10分)能够克服习惯性反应对学习时间分配影响, 且能够使被试建立起优先学习高价值项目的议程。(3)基于议程调节的学习时间分配在时程和阶段上具有动态性和情境特异性。

How value-based agendas affect study time allocation: An eye tracking study

Allocation of study time is generally considered as one of the most important indicators to show how learners adjust and control their learning processes efficiently, and it is influenced by habitual responding or an agenda-based process. In the agenda-based regulation (ABR) framework, agenda is a type of plans which can help learners finish their study in some situation. Generally, multiple factors, such as item difficulty and value, have effects on agenda-based regulation. Under the condition of limited study time and equal rewards for each item, learners often choose the item at their reading habit preferred position (left for Chinese) first to remember and then shift to another position (right for Chinese) when they feel they can fully recall the first one in the later test. However, if the rewards are unequal, especially when the item at opposite preferred position has higher rewards than those at the preferred position, habit-oriented learning may become counter-productive and learners tend to adopt a value-based agenda in item selection and study time allocation. The present study controlled the item difficulty in moderate level to observe whether the item value can influence the study decision shifting from habitual bias to value-based agendas by using the eye tracking technique. Three experiments were conducted with a paradigm devised by Metcalfe and her colleagues. Experiment 1 evaluated the habitual responding of native Chinese readers (mainly select items in a left-to-right fashion) in Chinese and provided a baseline to compare to Experiments 2 and 3. Participants were asked to learn the associations of 24 pairs of nouns. Two pairs of nouns were presented on a computer display for a total 3 s, one pair to the left and the other one to the right of the fixation. With the observation of clear bias of study time allocation to items on the left side, Experiments 2 and 3 were carried out to clarify if this habit oriented strategy was modulated by rewards, and if so, could rewards override such habit-oriented strategy and encourage the adaptation of a value-based strategy? The task procedures were the same across Experiments 2 and 3, and were similar to that of Experiment 1, only with the following two exceptions. First, the two pairs of nouns were associated with different rewards (1 vs. 5 points in Experiment 2; 1 vs. 10 points in Experiment 3). Second, participants need to complete two experimental blocks. In one block, the pairs of nouns with high rewards were presented on the left. In the other block, the pairs of nouns with high rewards were presented on the right. The results showed that: a) without rewards, participants tended to process items on the left side first, which give them more study time (Experiment 1); b) this left side bias remained when items of high rewards were presented on the left side of the display (Experiments 2 & 3); and c) when high rewards items were presented on the right side, the left side bias was disappeared if the reward contrast between left and right side items was moderate (Experiment 2), and was reversed if the reward contrast is high (Experiment 3). These observations suggest that value-based agenda can counteract or even override habitual processes in self-paced learning. The eye movements data of Experiments 2 and 3 further showed that the magnitude of reward contrast between items modulates the time course of learning decisions. High reward contrast triggers immediate selection of items associated with higher rewards, whereas the selection of higher rewards items takes a bit long time to resolve, when the reward contrast was moderate. The findings are in general consistent with the ABR framework. Study time allocation in self-paced learning is dynamic and context-specific. Item selection and study time are modulated by both habitual and value-based agendas, and the strength of value-based agenda is contingent on reward contrast.