Metacognition project descriptions
A description of projects that are part of the special research area metacognition.
- The relation between intellectual and metacognitive skills from a developmental perspective
- High giftedness and metacognition
- Test anxiety and metacognitive skillfulness
- The assessment of metacognitive skills
Manita van der Stel,PhD
Since 1987, Veenman has obtained substantial evidence that intelligence and metacognitive skills, although correlated, have a unique contribution to learning outcomes. Moreover, both types of skills are known to increase with age. The main question of this research project is whether the development of metacognitive skills is advanced by intelligence or not. Secondly, it is investigated whether metacognitive skills are general or domain specific by nature. Both cross-sectional and longitudinal designs are being used with participants of 12 years old and up. Participants perform various learning tasks in different subject-matter domains, while thinking aloud. The transcribed protocols are analyzed on the quantity and quality of metacognitive skills. Results from the studies in this research project may have strong implications for the instruction of metacognitive skills in school. For instance, schooling may be an agent other than intelligence in the development of metacognitive skills. And if metacognitive skills are largely general by nature, then metacognitive skills should be taught throughout the school, and not separately within each discipline (see e.g. Pressley & Gaskin, 2006).
Supervisor: Marcel Veenman
Marcel Veenman, PhD
This research project (2009-2012) is funded by an ‘Onderwijsbewijs’ grant from the Dutch Ministry of Education (OC&W). In order to be diagnosed as ‘gifted’, highly intelligent (IQ>130) children also need to adequately apply their metacognitive skills (Veenman, 2008). Together with Marieke van Haaren from ICLON, Veenman investigates pre-academic secondary-school students from the Pre-University College who are selected on excellent school performance and motivation. In the first study, these students will be compared with randomly selected classmates on IQ (WAIS), metacognitive skills, and learning performance. Metacognitive skills will be assessed through a series of inductive learning tasks on the computer (cf. Veenman, Wilhelm, & Beishuizen, 2004). In the second study, half of the Pre-University students will receive metacognitive instructions during the learning task, while the other half serves as a control group, performing the learning task without instruction. Even excellent students are expected to benefit from such metacognitive instruction. In the last study, the diagnostic tools for metacognitive skillfulness will be further validated in pre-academic secondary education.
Marcel Veenman, PhD
Test anxiety is a problem well recognized by educators. Two types of test-anxious learners can be distinguished. Type-1 learners lack metacognitive skills, due to which they often experience failure and eventually develop test anxiety. This is referred to as an availability deficiency of metacognitive skills. Type-2 learners, conversely, experience task irrelevant thoughts as part of existing test anxiety. These irrelevant thoughts cause cognitive interference, preventing metacognitive skills from being executed. In fact, type-2 learners do have metacognitive skills at their disposal, but they suffer from a production deficiency of those skills. In this research project a successful method for distinguishing both types of test-anxious students has been developed. Providing test-anxious students with metacognitive cues during task performance improved both metacognitive and learning performance of type-2 learners, but not of type-1 learners (Veenman, Kerseboom, & Imthorn, 2000). This method, however, requires the assessment of metacognitive skillfulness through laborious techniques, such as the analysis of thinking-aloud protocols or systematical observation. Our aim is to look for easier to administer, yet valid, methods for distinguishing the two types of test-anxious learners. The practical importance is that both types need different treatments. Type-1 learners should receive extensive instruction of metacognitive skills from scratch on, whereas type-2 learners may benefit from a relaxation or desensitization training.
Marcel Veenman, PhD
A perennial issue in metacognition research pertains to the validity of methods for assessing metacognitive skills. A distinction can be made between off-line methods that are administered outside a task context, and on-line measures that are collected during actual task performance. Various off-line questionnaires (e.g. MSLQ, ILS) are being used worldwide because they are easy to administer to large groups. Although their internal consistency is well checked, little is known about their external and convergent validity. There is some evidence that learners do not actually do what they say they will do or have done (Veenman, 2005). On-line measures, such as observations and thinking-aloud protocols, do reflect the learner’s actual behavior and appear to be highly predictive of learning outcomes. However, these laborious methods can only be administered to individuals. In this research project various assessment methods are being compared in a multi-method design (Veenman, 2005). In such a design the convergent validity of different assessment methods can be established, as well as their external validity with respect to learning outcomes. Results will provide insight into what methods are valid for assessing metacognitive skills.