Education-line Home Page

Metacognitive Knowledge as a Factor Modulating Self-Regulation in Multimedia Learning

Alessandro Antonietti

Department of Psychology - Cognitive Psychology Laboratory
Catholic University of the Sacred Heart
Largo Gemelli 1, 20123 Milano (Italy)

Paper presented at the European Conference on Educational Research, University of Geneva, 13-15 September 2006


Learners develop personal ideas about the educational materials that they are required to use. In this perspective, learners should be aware of what computer-supported tools activate in their minds. So far disparate studies have been conducted to investigate students’ beliefs about technologically supported learning. Thus, a more systematic approach seemed to be needed. We have tried to identify a homogeneous level of beliefs—the metacognitive knowledge—and a precise issue—the psychological effects of computer-supported educational tools. Metacognitive knowledge about such an issue was analysed by means of a questionnaire which included items about the motivational and emotional aspects of learning, the behaviour to have during the learning process, the mental abilities and the style of thinking required, the cognitive benefits. Results of various studies carried out by administering such a questionnaire showed that students ranked the psychological effects of the computer-supported tools in a relative different order according to the kind of tool and attributed distinctive effects to each tool. Gender and expertise played a minor role in modulating undergraduates’ beliefs. Implications for instruction are discussed.

In recent years metacognition has been proposed as a promising perspective in the field of education (Hacker & Graesser, 1998; Metcalfe & Shimamura, 1994) since it suggests a pedagogical approach aimed at inducing students to self-regulate their learning in order to become autonomous and critical knowledge constructors (Boekaerts, Pintrich & Zeidner, 2000). Usually metacognition is defined as the awareness, the knowledge and the control of cognitive objects and cognitive processes. Historically, the notion of learners thinking about their own thinking dates back to at least Plato and Aristotle (Brown, 1987), but the first attempts to define and classify the domain of metacognition was made by Flavell (1979) who proposed a model of metacognition whose key concept is "metacognitive knowledge", which refers to the part of personal knowledge about how the mind works when engaged in perceiving, comprehending, memorising and re-elaborating notions.

According to Flavell (1981), metacognitive knowledge consists of sets of beliefs about personal attributes, task features and strategies. All these aspects are relevant not only to learning in general, but also to the specific forms of learning which are mediated by technological devices (Veenman, Elshout & Busato, 1994; Veenman, Prins & Elshout, 2002). In fact, a student who is advised to use a multimedia presentation to understand a topic should consider if his/her individual skills and thinking styles (personal attributes) will allow him/her to benefit from the simultaneous presentation of words and pictures. Moreover, the task itself might be an issue of metacognitive knowledge. For instance, people should be aware that browsing a Website to look for data or practical notions is different from browsing it to find topics on which to write an essay, since the process of knowledge construction in the Web is rather peculiar. Finally, beliefs about the relevant strategies to be applied can play an important role in using an instructional instrument. For example, specific conversational scripts and collaborative working patterns should be known by a person whose goal is to participate in an electronic forum about cultural topics.

Metacognitive knowledge can influence learning outcomes since beliefs about mental processes involved in learning induce students to allocate cognitive resources and to choose thinking strategies which are consistent with those beliefs themselves (Schraw & Moshman, 1995). If beliefs are correct, a cognitive functioning relevant to the task to be carried out will be activated, alternatively the individual might be induced to implement mental operations which are not optimal (Huffaker & Calvert, 2003-2004). This stresses the need to investigate the metacognitive beliefs that students develop regarding what happens in the human mind when learning comes through a computer-supported device. A variety of such tools are now available in educational settings. What they have in common, besides the fact that they are run by a computer, is that they mediate the access to knowledge by providing learners with symbolic materials (texts, pictures, three-dimensional representations and so on) which can be managed to obtain a better understanding of the content to be learned (Lajoie & Derry, 1993).

A disparate set of studies have provided us with some notions about students' metacognitive knowledge concerning educational tools and their implications. University students are convinced that computers are necessary in school life, that they can enhance academic performance and that the Web helps to find information, makes communication easier, and has potential as a learning tool (Liaw, 2002). It has been proven that students can detect and report differences between traditional and technology-based tools for testing (Richardson et al., 2002; Stricker, Wilder & Rock, 2004) and for training (Antonietti, Rasi, Imperio & Sacco, 2000; Crosier, Cobb & Wilson, 2000). Learners' opinions appear to be rather sophisticated: For instance, as far as the learning process involved is concerned, students can identify differences between CD-Roms and Internet-based courses (Workman, 2004). Students have precise ideas about the strengths and weaknesses of e-learning (Weiner, 2003) and computer-supported instructional tools (Colley, 2003), conceptualise the benefits which are likely to be expected (Mouza & Bell, 2001; Wilson & Whitelock, 1998), and can choose between two types of operating systems based on their perceptions of the specific features (Messier, 2003).

Undergraduates are also aware of the specific goals of the online courses in which they are enrolled (Bennett & Scholes, 2001) and are aware of what motivates them in a Web-based course (Gao & Lehman, 2003). Undergraduates can judge the quality of online courses by referring to some psychological aspects (Hackman & Walker, 1995; Hassett, Ingram, Hassett & Marino, 2003; Koohang, 2004; Small & Lohrasbi, 2003) and can evaluate the satisfaction provided by the course attended, by offering judgments about the quality of the instructors, of the technology, of the course materials, of the course management, and of the interactivity (Bolliger & Martindale, 2004). Furthermore, students can determine the satisfaction and the quality of the learning results according to the instructional condition assigned (Ocker & Yaverbaum, 2001). Moreover, students can assess the nature, the reliability, and the strategies to access information on the Web (Dinet, Marquet & Nissen, 2003) and can relate learning outcomes to the level of interactivity experienced during the course (Gao, 2003; Jiang & Ting, 2000). Undergraduates are also aware of the processes activated by new e-learning tools. For instance, students engaged in instructional activities carried out within a virtual learning environment have come to realise that in such an environment the place, the time and the pace of work is more flexible; They have also learnt to appreciate the opportunity of reviewing the materials as many times as deemed necessary as well as the chance of being responsible for their own learning (Richardson, 2001). Students especially have appreciated an Internet-supported learning environment because it included real-life problems in which previously acquired knowledge can be applied (Chuang & Tsai, 2005).

All these findings concern what students believe about the various psychological aspects (motivation, pleasure, satisfaction, cognitive approach, goal-setting, decision making, and so on) concerning the computer-supported tools used. However, a review of the literature in this field shows that heterogeneous issues have been analysed, making it difficult to draw a systematic view of students' beliefs. Thus, we decided to design well-focused investigations aimed at considering different aspects of the metacognitive representation simultaneously by restricting our attention to one type of student belief only, namely, the psychological effects of computer-supported tools employed for educational aims. An effect was considered as psychological if it matched two criteria: (i) It referred to a mental state, process, strategy, function or skill; (ii) It referred to individual, but not social, abilities, preferences or outcomes. Such a topic, scarcely considered so far, concerns what computers introduce specifically, in terms of what happens in the learner’s mind, during the learning process and thus it seemed worth investigating both as a phenomenon in itself (that is, as a piece of metacognitive knowledge which accompanies computer use) and as a factor which can influences e-learning.

The studies mentioned above showed that learners are sensitive to the characteristics of the instructional tools they use and share well-defined opinions about them. They can identify the opportunities which are offered in terms of enjoyment and involvement, availability of training material, flexible control over the process, quality of the outcomes, and so on. However, previous studies have only considered a particular implementation of a given kind of tool (usually an implementation that students have directly experienced) and only one kind of tool at a time—for instance, Web-based courses (Yazon, Mayer-Smith & Redfield, 2002), e-mail use (Wissick, Dubay, Helman & Cates, 1995; Yu & Yu, 2002), navigating the Internet (Gal-Ezer & Lupo 2002), journeys into virtual reality (Spicer & Stratford 2001), multimedia encyclopaedias (Wishart, 2000), or multimedia presentations (Perry & Perry 1998). Furthermore, in those investigations students were asked to compare computer-supported tools to traditional instruments only. The studies that we carried out differ from the previous ones since (i) they investigated a homogeneous level of subjective reactions to computer-supported instructional tools, that is, their metacognitive representation; (ii) They were focused on a specific topic, namely, the psychological effects of these tools; (iii) They do not consider particular implementations of one kind of tool, but categories of tools.

We realise that nowadays undergraduates have access, in several contexts, to a variety of computer-supported tools and to different implementations of the same kind of tool. Therefore, we can assume that they can generalise their opinions to categories of tools. Furthermore, we argue that conceiving differences among tools is important both to select the most relevant tool in consideration of the goal to be reached and to use the chosen tool in the proper way. For instance, a science teacher might suggest his students choose either a virtual environment to carry out laboratory experiments or a multimedia program concerning the same concepts to be learned. He might think that the virtual environment could help students to gain a first-level, intuitive understanding of the simulated phenomena since it allows them to try to modify the values of the variables involved and become aware of the corresponding effects, whereas the multimedia tool could favour learning thanks to the visual representation of the whole set of the relationships implied in the phenomena and to the oral explanation that is contemporaneously given: The simultaneous presentation of visual and auditory information will foster a better retention of the notions provided and facilitate their application to novel situations. It is important that students are aware of the distinct advantages provided by the two kinds of tools in order to select the tool that matches their preferences, cognitive styles, study habits, time constraints, and so on, and that is more suitable to the type of learning that they want to experience. Another example: A geography teacher asks her students to employ a hypertext concerning African countries, which are the topic of her next lesson. Students should be aware of both the risks they might encounter working through hypertexts (for instance, their minds might be loaded by an excessive amount of information, together with a parallel, unproductive increase of attentional effort) and of the possible benefits (being facilitated in finding links and analogies among topics, in comparing different points of view, in supporting motivation through curiosity, and so forth).

The findings of a set of studies (see references), which we carried out by mean of a questionnaire focused on the metarepresentation of the psychological effects of multimedia tools, can be summarised as follows.

The questionnaire was used for the first time by making reference to Virtual Reality (VR) tools (Antonietti, Rasi, Imperio & Sacco, 2000). Students' opinions about the opportunities and the implications of VR in instruction were investigated by administering the questionnaire to humanities and engineering undergraduates. The questionnaire asked to rate a series of statements concerning motivation and emotion, skills, cognitive styles, benefits and learning outcomes associated with the use of VR in education. The representation which emerged was internally consistent and articulated into specific dimensions. It was not affected by gender, by the previous use of VR software, and by the knowledge of the main topics concerning the introduction of IT in instruction. Also the direct participation in a training session based on an immersive VR experience did not influence such a representation, which was partially modulated by the kind of course attended by students.

Subsequently the questionnaire was administered by making reference to multimedia tools. In Antonietti and Giorgetti's (2004) investigation the questionnaire was administered to 50 (Study 1) and to 170 (Study 2) boys and girls attending different university courses. Respondents identified a large number of not-trivial instructional opportunities of multimedia and showed a system of well-defined and internally-articulated beliefs about psychological implications of multimedia tools. Undergraduates’ conceptions were not consistently affected by gender differences, by the frequency of computer use, by computer expertise, by previous experience with multimedia educational software, and by cognitive style. Afterwards the questionnaire was administered to 493 boys and girls enrolled in university courses in Italian universities (Giorgetti and Antonietti, 2005). Their conceptions were not consistently affected by gender differences, by competence in computer using, and by the faculty and the year of course attended.

Opinions about the psychological correlates of multimedia computer-supported instructional tools were analyzed also by asking teachers to express their opinions (Antonietti & Giorgeti, 2006). The questionnaire was distributed to 272 teachers working in kindergarten, primary, and secondary schools. Gender and previous experience with multimedia, as well as the disciplinary fields taught by secondary school teachers, were taken into account. No significant gender effect was found. Differences of school level, disciplinary field, and direct experience with multimedia tools affected a part of the teachers’ representation.

We wondered whether students have different metacognitive beliefs in relation to different kinds of computer-supported educational tools. Thus, Antonietti and Balduzzi (2004) presented the items of the questionnaire three times; each time they made reference to a different kind (multimedia, hypertext, and interactive) of tools. The questionnaire was filled out by 87 undergraduates attending different faculties. Results showed that students tended to attribute similar effects to the distinct kinds of computer-supported tools. In a second study (Antonietti, Colombo & Lozotsev, submitted) items were presented five times, by making reference, respectively, to online courses, hypertexts, Web forums, multimedia presentations and virtual simulations. The questionnaire was filled out by 99 undergraduates attending engineering courses. Results showed that students ranked the psychological effects of the computer-supported tools in a relative different order according to the kind of tool and attributed distinctive effects to each tool. Gender and expertise played a minor role in modulating undergraduates’ beliefs. Differences between the two studies, where similar lists of statements were employed, might be due to differences in the samples of students who filled out the questionnaires: Engineering undergraduates, thanks to their course of study and/or to the higher exposure to technological devices, may be more sensitive to the differences existing among different kinds of tools. Alternatively, it is possible that the differences depended on the issues that students were asked to evaluate. In Antonietti and Balduzzi’s (2004) investigation, undergraduates were requested to answer by considering different dimensions—that is, properties defined at an abstract or functional level—of the computer-supported tools, whereas in the present study they were requested to consider different types of tools, defined on the basis of concrete, physical, perceptual and/or operational properties. Presumably differentiating "things" such as forums, three-dimensional simulations, and so on is easier than differentiating concepts such as interactivity, and so forth.

Factor analyses carried out in each study yielded partially overlapping outcomes. They showed that facilitating effects, creativity, analytical thinking and negative effects emerged as factors in all the studies. The metacognitive representation of the psychological effects of computer-supported educational tools was relatively unaffected by gender and by the level of computer expertise. Gender differences in computer attitudes have often been reported, in recent years too (Lee, 2003; McIlroy, Bunting, Tierney & Gordon, 2001; Oosterwegel, Littleton & Light, 2004; Shaw & Marlow, 2000; Volman & Van Eck, 2001). It might be that gender affects only the overall attitude towards computers but not—as in the present study—metacognitive knowledge about specific issues related to the role of information technologies in instruction. This conclusion is supported by more recent studies still where gender differences failed to emerge if participants' opinions concerned specific issues related to computer use (Chuang & Tsai, 2005; Liu, 2004). The same argument can be used to explain the lack of solid effects owing to expertise, as reported by Garland and Noyes (2004).

More recently, we focussed on a specific theory. To find a way to make the multimedia learning always effective, Richard Mayer devised a cognitive framework which led to hypothesise a general principle: students should learn better from words and pictures than from words alone. Mayer also conjectured that naives theories that people develop concerning multimedia are opposite to his assumptions. Is this hypothesis concrete? Actually Mayer never tested his conjectures about the common sense theory, focusing his researches on the factual validity of his principles. Hence, studying naïve conceptions can be seen as a new empirical research field concerning multimedia learning. In a study (Colombo & Antonietti, 2006), university students were presented with Mayer’s research protocols and asked to guess the research outcomes. Their ratings were compared with those actually achieved by Mayer. We found that Mayer’s expectations (naïve theories conflicting with cognitive multimedia principles) were not confirmed.

The implications of the findings are threefold, since they have something to say to learners, trainers, and software developers. Nowadays students are faced with a variety of technological educational devices. In some cases they have the opportunity to choose among them. In these circumstances, in order to select the tool which is more adequate, they must be aware of what a certain kind of instrument involves in terms of the mental work required. Obviously, the capacity to identify the specific features of each kind of tool does not assure that the best choice among available tools is made; The metacognitive competence concerning the distinctions among tools is a necessary, but not sufficient, condition which precedes the selection of the relevant technological device. But also when a single kind of tool is available, this awareness is important—if tools are to be used in a way which should be consistent with the features of the tools themselves, with the learner’s personal characteristics, and with the task demands and the goals to be achieved. The results of the studies raise the question of whether well-focused interventions are needed to lead students to reject inadequate metacognitive conceptions and become aware of the properties that such tools have and of which they might not otherwise be aware. For instance, it appeared that undergraduates failed to differentiate the tiring effects deriving from various categories of computer-supported educational tools. The amount and type of effort required by hypertext navigation is usually different from that needed by the participation in a Web-forum. The lack of awareness about these aspects might lead students to approach new instructional devices by assuming the same, a-specific attitude and, consequently, to neglect possible causes of failure or ignore interesting possibilities offered by the technology.

Educators have their own beliefs about multimedia tools in learning. If we assume a wide ecological perspective about the introduction of technology in education, we are invited to pay attention, within a systemic framework, not only to the tools and to their features, but also to the personal representation of those tools which all the involved actors develop (Zhao & Franck, 2003). Hence, not only students' metacognitive knowledge about computer-supported instructional instruments, but also educators' beliefs should be taken into account. For instance, when teachers work in a team, it should be checked that they have the same opinions about the psychological aspects of the technological devices that they plan to employ as the other members of the team in order to be sure that a coherent view is shared by them and, consequently, the same pedagogical guidelines are followed by them . Trainers are also asked to pay attention to what students think about the computer-supported devices. Such attention is important in order to verify whether what is relevant in the educators’ opinion is also relevant in the learners' opinion.

Finally, software developers should know what potential users think of the kind of instructional tools that they are producing, in order to reflect about possible lack of correspondences between the functions with which they equip the tools and learners’ expectations (Norman, 2002). It is not unlikely that software designers might include in a tool opportunities—such as the possibility to shift from one representational format to another, to draw automatically a schematic overview of the structure of the contents to be learned, to follow an approach based on autonomous discoveries, and so on—that will not be recognised by users since they do not perceive a particular tool as providing those opportunities or since they fail to attribute an important role within the learning process to those opportunities.


Antonietti A., A framework to conceptualise technologicy enhanced environments. In R. Carneiro, K. Steffens & J. Underwood (Eds.), Self-regulated learning in technology enhanced learning environments, Staker, Aachen 2005, pp. 59-63.

Antonietti A., Balduzzi E., Can students identify different psychological implications in multimedia, hypertext, and interactive educational tools? In Cantoni L. & McLoughlin C. (Eds.), Proceedings of ED-MEDIA 2004 - World Conference on Educational Multimedia, Hypermedia & Telecommunications, Lugano (Switzerland), 2004, pp. 3490-3497.

Antonietti A., Calcaterra A., Colombo B., Giorgetti M., Attorno al computer. Strumenti e ricerche su nuove tecnologie e apprendimento, Carocci, Roma 2003.

Antonietti A., Cantoia M., Imparare con il computer. Come costruire contesti di apprendimento per il software, Erickson, Trento 2001.

Antonietti, A., Catellani, P., Ciceri, R., Gilli, G., I corsi on line all'Università: aspetti cognitivi e psico-sociali. In Scurati C. (Ed.), E-learning/Università. Esperienze, analisi, proposte, Vita e Pensiero, Milano 2004, pp. 243-283.

Antonietti A., Colombo B., The effects of computer-supported learning tools: A bi-circulare bi-directional framework, submitted.

Antonietti A., Colombo B., Lozotsev Y., Undergraduates metacognitive knowledge about the psychological effects of different kinds of computer-supported instructional tools, submitted.

Antonietti A., Giorgetti M., Students’ conceptions about learning from multimedia. In Niegemann H., Brünken R. & Leutner D. (Eds.), Instructional design for multimedia learning, Waxmann, Münster-New York 2004, pp. 249-265.

Antonietti A., Giorgetti M., Teachers’ beliefs about psychological aspects of learning through multimedia, Computers in Human Behavior, 22, 2006, 267-282.

Antonietti A., Rasi C., Imperio E., Sacco M., The representation of virtual reality in education, Education and Information Technologies, 5, 2000, 317-327.

Colombo B., Antonietti A., Are the cognitive principles underlying multimedia learning counterintuitive? A study of undergraduates' folk conceptions. In Clarebout G. & Elen J. (Eds.), Avoiding semplicity, confronting complexity. Advances in studying and designing powerful (computer-based) learning environments, Sense Publishers, Rotterdam 2006, pp. 67-76.

Giorgetti M., Antonietti A., Come gli studenti si rappresentano il ruolo della multimedialità nella formazione, Rassegna di Psicologia, 22 (1), 2005, 141-159.

This document was added to the Education-Line database on 06 October 2006