33.12 Recommendations For Future Research

Additionally, many other specific issues that might be pursued include the following:

1. What specific instructional events are most or least amenable to providing or withdrawing learner control? That is, of the many instructional strategies, methods, activities, and events from which designers may draw upon to build lesson designs, which ones are most promising? Theoretical work by Laurillard (1987), Milheim and Martin (1991), and Steinberg (1989) go a long way toward providing prescriptive guidelines for designers; however, more specific recommendations need to be explored. I also strongly concur with the suggestion of Milheim and Martin (1991) to conduct more empirical and theoretical work on the nature and role of learner motivations in learner-controlled CBI settings.
2. What exactly is the nature of a learner's mental processes as he or she proceeds through learner-controlled instruction? If we can better understand both the rational-cognitive thought processes and the emotional-motivational states of the learner, we might be able to devise means to encourage optimal processes and perhaps attempt to alter or at least compensate for dysfunctional processes. This type of investigation has been suggested before (Clark, 1984; Robson, Steward & Whitfield, 1988) but has not yet been adequately pursued, perhaps because of the inherently qualitative nature of the data and the lack of comfort with such methodologies by many learner-control investigators. Reeves (1993) goes so far as to suggest that, "Perhaps a moratorium should be called on the types of quantitative studies described in this [Reeves's] paper, replacing them with extensive, in-depth efforts to observe human behavior in our field and relate the observations to meaningful learning theory that may later be susceptible to quantitative inquiry" (p. 44). (See also 40. 1.)
3. Related to the previous suggestion, it is time investigators more closely examined the social nature of learner-controlled activities. Anyone who has observed classroom situations where students navigate through instruction has informally noticed that there can be a great deal of discussion among students, both those sitting at separate computers and those working at the same computer. Rather than attempt to eliminate such interactions in order to investigate the "pure" effects of learner or program control, researchers may wish to adopt methodologies closer to field studies or naturalistic inquiry to study how learners can feed off each other's comments and actions during instruction. Although some experimental studies have been conducted to try to sort out the relative effects of learner control or program control for cooperative groups or for individual students (Hooper, Temiyakarn & Williams, 1993; Temiyakarn & McDonald, 1993), more work remains to be done before general conclusions, if any, can be drawn. (See also 35.9.4.)
4. A perhaps erroneous assumption undergirding most learner-control research is the implicit value placed on an individualistic and internally referenced system of control over instruction. That is, we tend to say that it is a good thing for learners to develop a capacity for intelligent control over their instructional experiences. This assumption might, however, be culture dependent. There is the additional question of whether the psychological bases discussed earlier (cognitive-rational and emotional-motivational) will operate similarly for students of differing cultural backgrounds. Wong (1988), for example, found that for students in a study conducted in Singapore, those who were under learner-controlled conditions selected more instructional options than students received under program control, a finding contrary to most learner-control research. Cross-cultural studies of learner control are sparse, but are needed to shed light on the question of whether learner control can be viewed with the same assumptions for children of different cultural backgrounds and values.
5. Learner control should be much more closely investigated under other common or developing types of computer-based environments, such as simulations, hypertext/hypermedia (including browsing through internet systems such as the World Wide Web), on-fine databases (such as electronic encyclopedias), online help and other support tools, and distance education. All of these contexts (possibly excepting distance education), by definition, intrinsically allow learner control to a greater or lesser degree. And it is likely that these types of computer-based experiences will soon be more frequent experiences for students than standard tutorials or "drills & practice." However, research to sort out the peculiar learner-control factors, each of which needs attention or support in these instructional systems, is still in its infancy (Trumbull, Gay & Mazur, 1992; McGrath, 1992; Saba & Shearer, 1993). Nevertheless, Chung and Reigeluth (1992) have given a jolt to this area of investigation by providing some useful instructional prescriptions for the use of learner control in hypermedia learning systems which they induced from the standard learner-control literature.
6. There needs to be a greater link made between learner--controlled CBI research and a growing body of literature on the topic of self-regulated learning. Briefly, this area of investigation, contributed to notably by McCombs and associates (McCombs, 1982, 1984; McCombs & Marzano, 1990) and by Zimmerman and associates (Zimmerman, 1990; Zimmerman & Martinez-Pons, 1986, 1988, 1990) conceptualizes students as "metacognitively, motivationally, and behaviorally active participants in their own learning processes" (Zimmerman & Martinez-Pons, 1988, p. 284). Although the research so far has primarily focused on understanding on a rather macro level the mental strategies occurring during successful self-regulated learning, this literature has clear implications for the inclusion of motivational variables in the design of learner-controlled instructional systems. In fact, some investigators have recently begun to explicitly address motivational variables operating in self--regulated learners (e.g., Schunk, 1990b; Zimmerman & Martinez--Pons, 1990). Additionally, there is beginning to emerge an interest in the application of self-regulated learning models to other formats of CBI, such as computer programming (Armstrong, 1989; Fischer & Mandl, 1988). Finally, a recent integrative paper by Kinzie (1990) provides a much-needed conceptual framework for discussion of the related areas of self-regulation, continuing motivation, and learner control.
7. The constructivist paradigm for learning would seem to have great implications for both the explanation of findings from the existing literature on learner control as well as to offer suggestions for new types of research questions based on the perspective it brings. Lebow (1993), in fact, suggests that constructivism provides a much-needed framework for interpreting the often confusing results from learner-control research. These reinterpretations of the learner-control literature from this point of view remain to be done, however. Additionally, he points out, as do others (e.g., Jonassen, Wilson, Wang & Grabinger, 1993; see also 7.4.5), that constructivist instructional perspectives are associated with a high degree of learner control and imply many new types of research questions. Indeed, Jonassen et at. (1993) discuss how constructivist approaches and learner control are inextricably linked, "The more learner-controlled the instructional systems are, the more generative they are; that is, they require learners to generate or construct their own knowledge" (p. 87). Certainly future investigators who adopt this. type of philosophy would ask questions that are far removed from "Which is better learner or program control?"