A. purpose or goal orientation: Most of respondents think learning strategies are goal-directed.
B. sequence of actions: There was general consensus among respondents strategic behavior could fall along a continuum from a single action or a sequence of actions.
C. focus of attention: There was relatively solid consensus for including attention as a feature on a continuum from focus on to focus away.
D. degree of monitoring: There was relative consensus that monitoring deserves status as a prototypical feature of a strategy. The concern was for the extent of monitoring likely to be found in actual strategy behavior.
E. level of observability: This feature received less acceptance because strategies are at various levels of observability along a continuum. There may be numerous strategies that involve mental processes.
F. deliberateness: This proved to be a problematic distinction between deliberateness and automation. Some respondents strongly hold that once a process is automatic, it can no longer be a strategy since in this context “automatic” means habitual and unconscious.
G. extent of evaluation: Only some respondents recognize evaluation as a necessary dimension for a strategy because in fact learners do not often reflect on how effectively they use the given strategy.
F. self- initiation: What are the source of intiation? From the teacher, a peer or learners themselves? Tracking the source of types of strategies might provide useful insights about the value of strategy instruction.
H. degree of planning: There are mixed views about the item. Some respondents hold the view that some intuitive students are not planning as much as they are instantly understanding the task and knowing the strategy to use, such a planning continuum might be a bit simplistic and require to really understand what planning the use of a strategy actually means.
Cohen’s attempt to describe the prototypicality of strategies is a step forward concerning defining learning strategies. It might reflect the nature of learning strategies to a large extent because the answers to the questionnaire come from strategy experts. It is true that some problems still exist. For example, how does Cohen select some features as prototypicality? In the item F, self-intiation maybe has not much to do with strategies. However, the prototypicality sheds light on the further study. Greater clarity can serve the action along, and bring about greater consensus.
2．Classification of learning strategies
2.1 General introduction to its classifications
Ellis (1994) notes that learning strategies differ in a number of ways, reflecting the type of learners under study, the setting and the particular interests of the researchers. Therefore, different researchers form different frameworks to classify learning strategies. The categories that have been established are high-inference in nature, which often needs considerable explanation on the part of the researchers (Ellis, 1994).