Metacognition in text comprehension. Does disfluency lead to more accurate judgments, better control, and higher performance?

• Applicant: Dr. Christoph Mengelkamp
• Subject Area: Developmental and Educational Psychology
• Term: from 2014 to 2015
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 268189691

Metacognition is defined as thinking about one´s own thinking (Flavell, 1979) and plays an important role in text comprehension. Metacognitive monitoring and control are two important components of metacognition that affect each other. Monitoring is defined as "subjective assessment of one´s own cognitive processes and knowledge, whereas control refers to the processes that regulate cognitive processes and behavior" (Koriat, Maayan, & Nussinson, 2006, p. 38; Mengelkamp & Bannert, 2012). Many theories of metacognition emphasize the importance of monitoring and control (e.g. Nelson & Narens, 1990; Zimmerman, 1990). Thereby often a sequential structure is implied (Koriat, 2012): monitoring affects control and control affects performance. However, monitoring and control are interactive (Efklides, Kourkoulou, Mitsiou & Ziliaskopoulou, 2006, Mengelkamp & Bannert, 2010; Koriat, 2012): monitoring affects control (monitoring-based control), but also is affected by control (control-based monitoring). Yet, there are only few studies that investigate the interaction of monitoring, control and performance. But understanding this interaction is important to deduce interventions to foster performance (de Bruin & van Gog, 2012). Therefore the aim of the planed study is to investigate the interplay between monitoring, control and performance. Disfluency seems to be one way to affect control and monitoring. Support comes from the cue utilization approach (Koriat, 1997) in which fluency or respectively ease of processing is mentioned as a cue for judgments. If students learn with a disfluent text that is hard to read, they may judge it as difficult to learn (ease of learning judgment). Further this text may require more effort and deeper processing (control) which can be measured using eye-tracking (e.g. longer fixations, more fixations and shorter saccades, see Rayner, Reichle, Stroud, Williams & Pollatsek, 2006). Because of higher effort for disfluent texts students may predict lower performance (judgment of learning). Studies (e.g. Dunlosky & Metcalfe, 2009, Pieger, Mengelkamp & Bannert, subm.) show that students are often overconfident when predicting their own performance and therefore disfluency should reduce this overconfidence and lead to more accurate control-based monitoring. When rereading the text this more accurate monitoring should lead to adequate monitoring-based control which can be measured by eye-tracking. As a result the performance in a knowledge test should increase. The planned experimental study investigates these questions using different font types to manipulate the fluency.

• DFG Programme: Research Grants