«Facilitating Higher-Order Thinking: Synthesizing Pedagogical Frameworks for the Development of Complex and Coherent Conceptual Systems Michael A. ...»
Fischer, Troendle and Mandl (2003) pointed out that, although elementary and high schools have benefited greatly from investment in web-based learning technologies, university education has "largely remained unaffected" (p. 194) by the advent of these tools, and (to aid the process) they have introduced five guiding principles derived from empirical studies on technology-based learning environments. In the context of "problem-oriented environments" (p. 195), these principles are authentic problem contexts, collaborative knowledge construction (including the integration of alternative perspectives), tools to represent the problem and the domain concepts involved, learning resources (including expert advice), and a tutor to guide the process (without supplying answers to the problem under discussion). The authors and their colleagues created a dynamic modelling and visualisation tool (called MUNICS) to aid in developing the design specifications for a virtual computer network, and graphics tools were provided to facilitate problem definition and resolution. A formative evaluation of the prototype was conducted, using a one-group pretest-posttest design, with the participation of eleven computer science students who worked in groups of two or three.
Dickey (2005) presented a comprehensive analysis of the use of 3D virtual worlds (interactive virtual reality combined with messaging in a desktop computing environment) as educational tools, with particular attention to design features that facilitate learning. In examining Active Worlds Educational University and Adobe Atmosphere, Dickey explored the notion that "learners construct understandings by interacting with information, tools, and materials as well as by collaborating with other learners" (p. 124). He compared the two software packages in terms of their interfaces and their educational implications, and he described the particular resources and tools that they each provide to facilitate the construction of communicative discourses and to support their users' experiences of learning. While the author's descriptions of these learning environments makes clear that opportunities for content creation, cognitive development, and collaboration are restricted by the practical constraints imposed on users (including the difficulty of constructing new objects, and the limits to the modalities of computer communication), nevertheless we may infer that there are advantages, especially for young learners, of creating a virtual identity and learning to collaborate in the interactive construction of knowledge in an environment reserved for educational (or combined educational and recreational) use. In particular, Dickey makes the important point that technological tools do not determine the effectiveness of educational relationships, which instead emerge from how the tools are used. "It is important to note that within a constructivist paradigm of learning, technology tools do not evoke the characteristics of a learning community, but rather these dynamics are the result of the interplay between content, the instructor, and the learners" (p. 132).
MacGregor and Lou (2005) studied the use of WebQuest by fifth-graders, in order to examine the effects of the use of concept mapping tools on recall and on the production of multimedia presentations. WebQuest is an online problem-solving environment, which provides introductions to problems, the tasks themselves, resources, procedures, assessment criteria, and conclusions.
The fifth-grade students used WebQuest to create presentations for secondgraders; they each selected an endangered species of animal, and gathered information about their subject to be organized in a slide show. Half of the students were assigned at random to receive an instructional aid, a concept-mapping template, which specified connections to be made between sub-topics and which served as a basis for designing the presentation; the other half were required to develop their own storyboards. Upon completion of the task (the production of which was closely monitored by the researchers, who analyzed each performance in detail), quantitative results demonstrated significantly higher scores awarded to the experimental (concept-mapping) group for both presentation content and organization; this group also demonstrated higher recall when reporting what they had learned. In reporting these results, MacGregor and Lou advise, "[I]t is important for teachers to be cognizant of the design features within a site and understand how they facilitate student use in achieving learning objectives. Design features that provided support for the students included appropriate discourse readability, high content relevance, easy navigation, user-friendly screen design, and multimedia" (p. 172; emphasis added).
WebQuest is apparently a very useful technological tool for supporting content learning using learner-centred methods and a problem-based approach to instruction. This application is designed as a "higher-order use of technology" (MacGregor and Lou, 2005, p. 172), in that it prompts students to search widely for information, and to integrate their findings into comprehensive problem solutions. While there is no question about the utility of WebQuest's application to well-structured tasks, as it can supply the means to solve algorithmic problems (and this capacity is quite useful to inexperienced learners), we might question its applicability to situations where problems and their expected solutions are not well defined. Creating a presentation from existing materials may be considered an open-ended task, in that the structure of the final solution is not fully predetermined; however, more ill-structured problems (which require a more heuristic approach), may require complex analysis and synthesis of information from a variety of sources even to define which parts of the problem are more or less amenable to resolution. While computer tools have not yet been devised to support the development and maintenance of the most highly complex and dynamic reflective conceptual frameworks, we may look forward to the production of more and more intelligent CBLEs.
Winn (1993) described the educational value of virtual reality (VR) applications, concluding that constructivism provides a basis for a theory of learning in virtual environments. Examining "immersive VR," where the interface between machine and user is transparent, Winn pointed out that learning in VR shifts from third-person (vicarious, objective, explicit) knowledge to the attainment of direct, personal and subjective first-person understandings. This type of experience allows for non-symbolic problem solution, which can later be integrated with third-person descriptions. Winn praises the educational capabilities of VR, claiming that it allows us to create knowledge from direct experience, bypassing the symbol systems, which represent other people's formulations while we construct our own learning from direct interaction with virtual objects. VR offers unique educational advantages (allowing us to resize objects for close study, to transduce information to perceivable forms, and to reify abstract objects and events), and the conversion of traditional third-person educational practices into firstperson events has demonstrated clear benefits in many applications (e. g., in complex simulations used to train airline pilots, astronauts and supertanker captains).
Notar, Wilson and Montgomery have developed a framework for instructional design (ID) that focuses on the development of higher-order collaborative cognition through distance education (DE). While we might question their premise that ID should make DE "no different than learning in the traditional classroom" (par. 1), their conclusions with regard to effective ID seem to be cogent. Notar et al. stress the value of hypermedia as a tool that allows for traditional educational activities (experiments, demonstrations, and personal participation) to be enacted in new ways (although one might argue that the resemblance of hypermedia applications to corporeal interactions is superficial, or even that hypermedia can sometimes allow for superior learning opportunities). Describing games and simulations, they remark on their value in engaging learners and thus promoting motivation, and they address how technological developments can facilitate the development of higher-order thinking by embedding instruction in the dynamic experience of social participation (always facilitated by an instructor who exemplifies excellent judgment in applying principles, using information, and executing procedures). Sharing and dialogue are instrumental in this process, which encompasses both synchronous and asynchronous opportunities for conversation. Notar et al. listed ten design factors which contribute to the effectiveness of the learning process and resulting cognitive growth; these include "rich" learning activities, presentation of multiple perspectives and multiple links between ideas, continual selfassessment by all learners, exposure to expert performance, and collaborative work on highly complex problem scenarios. Although these authors do not distinguish between ID work at different levels of education, and they do not consider the specific disciplinary content of educational objectives, the general principles that they promote are consistent with contemporary ideas about learner-centred and problem-based instructional methods.
Of course, instructors are always faced with figuring out how to increase their skills in applying such theoretical generalities to particular lessons.
Azevedo (2005a) described the relationships of metacognition, self-regulated learning (SRL), and hypermedia learning tools, claiming that SRL provides a useful theoretical framework for guiding the study of learning with hypermedia. Pointing out that SRL theory has been developed to describe how we learn to deal with the complex and dynamic functions which characterize higher learning, and that five psychological processes associated with SRL (planning, monitoring, strategy use, control and motivation) provide access to the operational variables needed to study higher-order cognitive development, Azevedo suggests that computer-assisted instruction can be provided by tutoring agents, the functions of which would be modelled after the interventions supplied by human tutors. The purpose of such interventions is to facilitate "qualitative shifts in students' mental models" (p. 203), which represent accommodation to new information and the adaptive restructuring of unsophisticated understandings. To study the role of scaffolding, Azevedo and his associates recorded think-aloud protocols provided by students who studied complex and challenging science topics (physiology and ecology) using hypermedia tools. The research team compiled a list of thirty-three SRL sub-processes (based on the five areas listed above), and qualitatively coded the discourses transcribed from the think-aloud activities as students worked the problems. In addition, pre- and post-tests measured declarative knowledge as well as the quality of the students' mental models of the processes being studied. Their independent variable was the type of scaffolding provided by human tutors: no scaffolding, fixed scaffolding (subgoals related to academic content) and adaptive scaffolding (related both to content and to the processes of self-regulated learning). After a series of studies with students at different levels of education, it was reported that, "adaptive scaffolding by a human tutor who provides timely content and process-related scaffolding during learning tends to lead to significant qualitative mental model shifts for middle school, high school, and college students" (p. 204). Declarative knowledge gains were greater for college students who received either fixed or adaptive scaffolding, while younger students benefited from adaptive scaffolding (but not from the nonadaptive type). Think-aloud protocols indicated that students without scaffolding displayed the least self-regulatory behaviour; those in the fixed-scaffolding condition monitored their progress on the tasks, but those who were provided with adaptive scaffolding "engaged in an inordinate amount of help-seeking from the human tutor" (p. 204), and used self-regulatory strategies more than the other groups. Azevedo concluded, "In sum, the think-aloud data and discourse analyses tend to indicate that successful students regulate their learning by using significantly more metacognitive processes and strategies" (p. 205). Ultimately, Azevedo claims, "[I]t would make sense for a CBLE to emulate the regulatory behaviour of the human tutor... the system would ideally need to dynamically modify its scaffolding methods to foster the students' self-regulatory behavior during learning" (p. 205). He notes that the day has not yet arrived when electronic tutors can detect a learner's pedagogical needs with the same sensitivity as a human teacher, but (as we progress towards meeting this "technical challenge") we can develop computer-based tutoring systems that recommend goals and strategies which will facilitate the learning of self-regulatory processes, and thus support students in increasing their understandings of complex problems.