A test of a neo-Piagetian model of the Water Level Task.

A neo-Piagetian model (Pascual-Leone & Morra, 1991) of the Water-level task (WLT) predicts that WLT performance depends on three factors: physical knowledge, field independence, and M capacity. The latter is an attentional capacity construct defined in neo-Piagetian theories. The model predicts that a minimum M capacity is necessary for accurate performance on WLT items. It also predicts that M capacity mediates the relationship between field independence and physical knowledge on one side and WLT on the other. Tests of field independence and M capacity, a paper-and-pencil version of the WLT, and a set of questions on the liquid horizontality principle were administered to 337 participants, aged 5 to 13 years. Six predictions derived from the model were tested; all of them were supported by the data, with only one exception, namely, on items with sideways and capsized bottles the M capacity required for accurate performance was one unit lower than predicted. It is concluded that the results strongly support the model under consideration, but suggest an amendment to one of its details. Gender differences were also examined; a small but significant gender difference in the WLT remained when field independence, physical knowledge, and M capacity were controlled for.