Maartje Raijmakers - How to test children’s logical reasoning abilities?

Strategic games require reasoning about other people’s and one’s own beliefs or intentions. Although they have clear commonalities with psychological tests of theory of mind, they are not clearly related to theory of mind tests for children between 9 and 10 years of age (Flobbe, Verbrugge, Hendriks & Krämer, 2008). There are multiple reasons why logical reasoning tasks for children might measure other abilities than theory of mind alone. First, limited executive functions might seriously hamper advanced recursive reasoning. To test this hypothesis, we studied children’s (5–12 years of age) individual differences in how they played a strategic game by analyzing the strategies that they applied in a zero, first, and second-order reasoning task with statistical latent variable techniques to detect strategies that were not necessarily well-defined beforehand. Second, instruction might not be an optimal way of preparing children for the test. We tested children’s (N = 98) abilities for deductive logical reasoning in an adaptive training environment were children receive intensive adaptive training (rekentuin.nl or MathsGarden.com; Gierasimczuk, Raijmakers & van der Maas, 2013). As yet, children (N > 30.000) advance from low level to higher levels without getting any instruction. In an experimental study with a pretest, posttest, and delayed posttest design (N = 98, Age 8 – 12 years) we tested two additional interventions: self-explanation and instruction in contrast with just solving items. This study shows the importance of dedicated training to reach optimal performance in children’s logical reasoning.