2006 Metacognitive Experiences- The Missing Link in the Self-Regulated Learning Process

ORIGINAL ARTICLE Metacognitive Experiences: The Missing Link in the Self-Regulated Learning Process A Rejoinder to Ainley and Patrick Anastasia Efklides Published online: 19 September 2006 # Springer Science + Business Media, Inc. 2006 Abstract The measurement of online self-regulation processes is a very important issue and in this rejoinder to Ainley and Patrick (this issue) I am arguing that including measures of metacognitive experiences, in conjunction with measures of other affective experiences, in various phases of task processing can increase the reliability and validity of online measures and our understanding of the self-regulation process. Furthermore, behavioral and performance measures as well as thinking aloud protocols can enrich not only the reliability and validity of our measures but also our awareness of the factors involved in the formation of the various facets of subjective experiences, be it affective or metacognitive. Keywords Measurement . Metacognitive experiences . Self-regulation Introduction The measurement of self-regulation process is a highly challenging issue and has attracted the attention of researchers on self-regulation per se (Kuhl & Fuhrmann, 1998) as well as researchers working on processes involved in self-regulated learning, such as metacognition (Efklides, 2002a; Schraw & Impara, 2000). The contribution of the article by Ainley and Patrick (this issue) lies in the discussion of measures that capture online changes of task- specific self-regulation processes. The authors focus on the measurement of interest through the “Between the Lines” software (see Ainley, Hidi, & Berndorff, 2002). As basis of their arguments they state the findings of one study that tracked the effects of students’ goal orientations on interest in predetermined points of task processing. These points, which correspond to different phases of self-regulated learning, are the following: (a) forethought, i.e., before starting the processing of the task; (b) on-task processes, i.e., during the processing of the task; (c) reaction and reflection, i.e., after the processing of the task. In what follows, I will argue that the micro-level approach adopted by Ainley and Patrick (this issue) is a very useful one, with considerable pros and cons. In their article the Educ Psychol Rev (2006) 18:287–291 DOI 10.1007/s10648-006-9021-4 A. Efklides (*) School of Psychology, Aristotle University of Thessaloniki, 541 24 Thessaloniki, Greece e-mail: efklides@psy.auth.gr authors deal explicitly with the advantages and disadvantages of the online, single-item self-report measures, such as the one they used, and, more specifically, with the problem of reliability and validity of such measures. I will extend their arguments and propose that single-item self-report online measures of affect can gain in rigor if measures of metacognitive experiences are also collected. The “Between the Lines” software has a lot of potential and inclusion of measures of metacognition can answer concerns over reliability and validity of single-item self-report measures. Even more importantly, this expanded approach can provide a deeper understanding of the self-regulation processes, which are not purely affective in nature. Metacognitive Experiences and Micro-Level Analysis The distinction between trait-like characteristics of the person, on the one hand, and online responses to the task at hand, on the other, is critical for our understanding of the formation of subjective experiences and their interaction with the features of the task as well as of the context. This was the strategy Ainley and Patrick followed in their study regarding the effects of achievement goal orientations on state interest, i.e., the task-specific interest. However, at the task-specific level interest is only one of the possible emotional or affective experiences that are present when people process tasks (Pekrun, Goetz, Titz, & Perry, 2002). Metacognitive experiences regarding task and processing features are also present (Efklides, 2001; Efklides & Petkaki, 2005). Metacognitive experiences are online feelings, judgments/estimates, and thoughts people are aware of during task processing (Efklides, 2001; Flavell, 1979). In a broader sense, the affective experiences that the person becomes aware of in connection with the task at hand—e.g., interest, liking, disappointment, etc.—are also part of the metacognitive experiences. These affective experiences co-occur with metacognitive feelings and metacognitive judgments/estimates, such as feeling of knowing, feeling of difficulty, feeling of familiarity, feeling of confidence, feeling of satisfaction, estimate of solution correctness, etc. Judgments regarding the similarity of tasks or control decisions, such as allocation of resources (e.g., effort) and thoughts on strategies to be used are also part of the online metacognitive experiences. Many of these metacognitive experiences are evident in spontaneous self-talk when solving a problem or in thinking aloud protocols. The importance of online subjective responses to the task lies in that they form the interface between the person and the task and provide the basis for control decisions during task processing, as well as after it. This is so, because they feed on one’s self-concept and motivation and, thus, influence future decisions regarding the engagement, or not, in similar tasks (Efklides & Tsiora, 2002). Therefore, metacognitive experiences are critical for self- regulated learning. It should be noted, however, that metacognitive experiences are not equivalent to “task perception” and “task analysis”, because the latter are based on cognitive analytic processes whereas metacognitive experiences— and, particularly, metacognitive feelings— are products of nonanalytic, nonconscious processes (Efklides, 2001). Thus, the very same task can be perceived, for example, as objectively difficult but subjectively easy if the person has the required knowledge and practice on similar tasks. Also, the teacher may evaluate a task— in terms of objective complexity or processing demands— as easy, but students experience difficulty when facing the task for the very simple reason that they are not familiar with the task. Furthermore, during the processing, the person’s subjective experiences may remain stable or change, because the person becomes aware of processing features, such as fluency 288 Educ Psychol Rev (2006) 18:287–291 or interruptions of processing. These are critical cues for the formation of metacognitive experiences (Efklides, 2001). Changes in processing features are reflected in the person’s subjective experiences and take the form of changing affect (i.e., from negative to positive mood), the form of emotions (i.e., from interest to boredom), the form of feelings (i.e., from feeling of difficulty to feeling of easiness), the form of judgments/estimates (i.e., “I do not know the answer”), the form of reactions to one’s self (i.e., “How stupid of me”), or thoughts on possible strategies in order to overcome interruptions of processing. The multifaceted, transient, and changing nature of subjective experiences necessitate, on the one hand, the use of multiple measures and, on the other, repeated measures of the same experience during the processing of a task. Ainley and Patrick (this issue) emphasize the temporal dimension of human experience which is captured by the micro-level (online) measures contrary to macro-level measures that address more stable and more general characteristics of the person.1 However, there is more to subjective experience than just temporal change. There is structure and interrelations between the various experiences, affective or metacognitive. As Efklides and her collaborators (see Efklides, 2002b) found, metacognitive experiences are organized both in terms of the phase in which they are measured, and in terms of the object, i.e., the processing feature they monitor. Thus, a basic distinction is between prospective metacognitive experiences and retrospective ones. The former refer to metacognitive experiences measured as soon as the person comes across a task and before starting working on it (i.e., at the forethought phase), whereas the latter refer to those measured after the completion of the task processing (i.e., at the reaction and reflection phase). Some of the metacognitive experiences are also present during the task processing (i.e., at the on-task phase). Metacognitive experiences that can be measured at the prospective phase are feeling of familiarity, feeling of knowing, feeling of difficulty, estimate of the predicted solution correctness or of the time needed to work on the task, as well as other affective responses such as interest and liking of the task. During task processing, metacognitive experiences that are present are feeling of difficulty, estimate of effort, estimate of time needed to complete the task, thoughts on strategies to be used. Also, present are affective responses, such as interest or other emotions. Finally, at the retrospective phase the metacognitive experiences that can be measured are the estimate of solution correctness, feeling of confidence, feeling of satisfaction, feeling of difficulty (Efklides, 2002a) and other affective responses, such as liking of the task, anger, etc. Why is the differentiation of metacognitive experiences along the phase of task processing so important? First, because different mechanisms seem to underlie the formation of metacognitive processes at each phase of processing (Nelson, 1996). Second, because the various measures at each phase provide the basis for determining the reliability and validity of the reported experience, be it metacognitive or affective, such as interest. For example, if state interest at the prospective phase is related to positive mood and heightened attention due to the person’s trait interest that is triggered by task features, then the response to the single item tapping interest should correlate with task-specific metacognitive experiences such as feeling of familiarity, since the task belongs to those that the person has previously worked with and has evaluated them as interesting. It should also correlate with affective responses such as the liking of the task. However, if the task is novel and the person’s interest is purely situational, then the relationship with feelings of 1 Actually, metacognitive knowledge is functioning at this general, macro-level, unlike metacognitive experiences that function at the micro-level. Educ Psychol Rev (2006) 18:287–291 289 familiarity and liking would be low because these two experiences presuppose previous encounter with the task (Iran-Nejad, 1987). In this way, one can make predictions about convergent and divergent validity depending on the specific task and its context. Furthermore, if interest facilitates engagement with a task, and this leads to increased processing fluency, then the measure of interest should be part of a scale that measures processing fluency. Such a scale would comprise items tapping feeling of familiarity as well as feeling of difficulty, estimate of effort, and estimate of solution correctness. Indeed, this was the case in some of the studies with my colleagues (Efklides, 2002a,b). Self-reported interest at the retrospective phase, however, is a judgment that is based on the affect generated by the task processing features and the processing outcome. Therefore, it would correlate with measures of feelings of satisfaction and confidence, particularly, with feeling of satisfaction, which is predicted by positive affect, that also underlies the experience of interest (Efklides & Petkaki, 2005). In this case, the measure of interest would be part of a scale that taps the evaluation of the outcome of task processing (Efklides, 2002b). Summing up, the argument I tried to put forth is that even though online measures of affect or metacognitive experiences are usually based on single items, this does not necessarily mean that we cannot test their reliability or validity. This can be done by including more measures of subjective experiences that are specific to the phase of processing. Behavioral and Thinking Aloud Measures Another way to answer concerns over reliability and validity of single self-report items is to use behavioral and performance measures as well. For example, using a microgenetic design and video recording of the whole process of problem solving (see Iiskala, Vauras, & Lehtinen, 2004) one can collect evidence on the direction of gaze, facial expressions, body movements, and verbal utterances and exchanges— if there are two or more participants collaborating. Also, evidence on the treatment of the task material (e.g., course of actions), use of technical strategies (e.g., underlining or taking notes, etc.), use of help from others, etc. and, finally, evidence on performance outcome. Another way of collecting online evidence on thinking processes is thinking aloud protocols. Behavioral and thinking aloud measures have their own limitations, because one cannot have evidence on cognitive or affective processes that do not have a manifest form. However, this kind of evidence can be associated with self-reports of metacognitive experiences or affect and, thus, increase the reliability of our measures, as well as our awareness of the limitations of each method used. Furthermore, inclusion of measures of effort expenditure— as well as inclusion of possible strategies in the processing of the task (e.g., quit or go on, change of strategy)— can lead to the testing of hypotheses about the relations of online affective and metacognitive factors with control decisions. Thus, it may be the case that interest or lack of it is critical for the decision to quit or go on, but other metacognitive experiences, such as feeling of difficulty, are critical for the regulation of effort or change of strategy, or even quitting the task even if the task is interesting. Thus, it can be concluded that the discussion of Ainley and Patrick (this issue) on the reliability and validity of online single item self-report measures is highly important, timely, and substantial. The more aware we become of the strengths and limitations of our research methods and measures, the more is the progress we can make in our understanding of the so complex, but so interesting, phenomena involved in online self-regulated learning and its implications for learning and instruction. 290 Educ Psychol Rev (2006) 18:287–291 References Ainley, M., Hidi, S., & Berndorff, D. (2002). Interest, learning and psychological processes that mediate their relationship. Journal of Educational Psychology, 94(3), 545 – 561. Efklides, A. (2001). Metacognitive experiences in problem solving: Metacognition, motivation, and self- regulation. In A. Efklides, J. Kuhl, & R. M. Sorrentino (Eds.), Trends and prospects in motivation research (pp. 297–323). Dordrecht, The Netherlands: Kluwer. Efklides, A. (2002a). The systemic nature of metacognitive experiences: Feelings, judgments, and their interrelations. In M. Izaute, P. Chambres, & P.-J. Marescaux (Eds.), Metacognition: Process, function, and use (pp. 19 – 34). Dordrecht, The Netherlands: Kluwer. Efklides, A. (2002b). Feelings as subjective evaluations of cognitive processing. How reliable are they? Psychology: The Journal of the Hellenic Psychological Society, 9, 163 – 184. Efklides, A., & Petkaki, C. (2005). Effects of mood on students’ metacognitive experiences. Learning and Instruction, 15, 415–431. Efklides, A., & Tsiora, A. (2002). Metacognitive experiences, self-concept, and self-regulation. Psychologia: An International Journal of Psychology in the Orient, 45, 221–235. Flavell, J. H. (1979). Metacognition and cognitive monitoring: A new area of cognitive developmental inquiry. American Psychologist, 34, 906 – 911. Iiskala, T., Vauras, M., & Lehtinen, E. (2004). Socially-shared metacognition in peer learning? Hellenic Journal of Psychology, 1, 147–178. Iran-Nejad (1987). Cognitive and affective causes of interest and liking. Journal of Educational Psychology, 79, 120–130. Kuhl, J., & Fuhrmann, A. (1998). Decomposing self-regulation and self-control: The volitional component inventory. In J. Heckhausen & C. S. Dweck (Eds.), Motivation and self-regulation across life span ( pp. 15–49). Cambridge, UK: Cambridge University Press. Nelson, T. O. (1996). Consciousness and metacognition. American Psychologist, 51, 102–116. Pekrun, R., Goetz, T., Titz, W., & Perry, R. (2002). Academic emotions in students’ self-regulated learning and achievement: A program of qualitative and quantitative research. Educational Psychologist, 37, 91–105. Schraw, G., & Impara, J. (Eds.) (2000). Issues in the measurement of metacognition. Lincoln, Nebraska: Buros Institute of Mental Measurements, University of Nebraska—Lincoln. Educ Psychol Rev (2006) 18:287–291 291 Metacognitive Experiences: The Missing Link in the Self-Regulated Learning Process Abstract Introduction Metacognitive Experiences and Micro-Level Analysis Behavioral and Thinking Aloud Measures References << /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /None /Binding /Left /CalGrayProfile (None) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (ISO Coated) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Error /CompatibilityLevel 1.3 /CompressObjects /Off /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJDFFile false /CreateJobTicket false /DefaultRenderingIntent /Perceptual /DetectBlends true /ColorConversionStrategy /sRGB /DoThumbnails true /EmbedAllFonts true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 524288 /LockDistillerParams true /MaxSubsetPct 100 /Optimize true /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveEPSInfo true /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts false /TransferFunctionInfo /Apply /UCRandBGInfo /Preserve /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true /AardvarkPSMT /AceBinghamSH /AddisonLibbySH /AGaramond-Italic /AGaramond-Regular /AkbarPlain /Albertus-Bold /AlbertusExtraBold-Regular /AlbertusMedium-Italic /AlbertusMedium-Regular /AlfonsoWhiteheadSH /Algerian /AllegroBT-Regular /AmarilloUSAF /AmazoneBT-Regular /AmeliaBT-Regular /AmerigoBT-BoldA /AmerTypewriterITCbyBT-Medium /AndaleMono /AndyMacarthurSH /Animals /AnneBoleynSH /Annifont /AntiqueOlive-Bold /AntiqueOliveCompact-Regular /AntiqueOlive-Italic /AntiqueOlive-Regular /AntonioMountbattenSH /ArabiaPSMT /AradLevelVI /ArchitecturePlain /Arial-Black /Arial-BlackItalic /Arial-BoldItalicMT /Arial-BoldMT /Arial-ItalicMT /ArialMT /ArialMTBlack-Regular /ArialNarrow /ArialNarrow-Bold /ArialNarrow-BoldItalic /ArialNarrow-Italic /ArialRoundedMTBold /ArialUnicodeLight /ArialUnicodeLight-Bold /ArialUnicodeLight-BoldItalic /ArialUnicodeLight-Italic /ArrowsAPlentySH /ArrusBT-Bold /ArrusBT-BoldItalic /ArrusBT-Italic /ArrusBT-Roman /Asiana /AssadSadatSH /AvalonPSMT /AvantGardeITCbyBT-Book /AvantGardeITCbyBT-BookOblique /AvantGardeITCbyBT-Demi /AvantGardeITCbyBT-DemiOblique /AvantGardeITCbyBT-Medium /AvantGardeITCbyBT-MediumOblique /BankGothicBT-Light /BankGothicBT-Medium /Baskerville-Bold /Baskerville-Normal /Baskerville-Normal-Italic /BaskOldFace /Bauhaus93 /Bavand /BazookaRegular /B