In the judgment task, on the other hand, children and adults focused on water level only and did
not discriminate the two diameters. Hence, the judgment task differed significantly from the manual
tilting task in terms of the knowledge children and adults showed about the effect of glass diameter on
the tilting angle. The fact that participants did not show any knowledge about the effect of diameter in
the judgment task but correctly tilted the thin glasses farther in the manual tilting task suggests that
the manual tilting allowed them to access some action-based knowledge they could not retrieve in an
abstract judgment task. This dissociation also implies that the manual tilting task was in fact solved by
using an imagery strategy and was not based on abstract formal knowledge (cf. Schwartz, 1999; Schwartz
& Black, 1999). Furthermore, the large effect size of water level in the judgment task and the
highly similar consistencies rule out the possibility that the absence of an effect of diameter was
due to noisier data or a mere lack of power. They rather show that participants were able to handle
the judgment task quite well.