Towards better computational models of the balance scale task: a reply to Shultz and Takane

In contrast to Shultz and Takane [Shultz, T.R., & Takane, Y. (2007). Rule following and rule use in the balance-scale task. Cognition, in press, doi:10.1016/j.cognition.2006.12.004.] we do not accept that the traditional Rule Assessment Method (RAM) of scoring responses on the balance scale task has advantages over latent class analysis (LCA): RAM is similar to a very restricted form of LCA. The apparent shortcomings of LCA are also less severe than they suggest. Via new simulations we show that LCA detects small classes reliably. We also counter their concerns regarding the torque difference effect and we underline the problems connectionist models have with correctly responding to balance items. Despite these differences in opinion we agree with Shultz and Takane on the possible avenues for future research.     

Re-thinking stages of cognitive development: an appraisal of connectionist models of the balance scale task

The present paper re-appraises connectionist attempts to explain how human cognitive development appears to progress through a series of sequential stages. Models of performance on the Piagetian balance scale task are the focus of attention. Limitations of these models are discussed and replications and extensions to the work are provided via the Cascade-Correlation algorithm. An application of multi-group latent class analysis for examining performance of the networks is described and these results reveal fundamental functional characteristics of the networks. Evidence is provided that strongly suggests that the networks are unable to acquire a mastery of torque and, although they do recover certain rules of operation that humans do, they also show a propensity to acquire rules never previously seen.     

Bridging the gap between theory and model: a reflection on the balance scale task

At first glance, the two lead articles in this issue share little except the balance scale task, yet we view them as complementary rather than unrelated or contradictory. Our Reflection focuses on the individual strengths of the two lead articles and, to a greater extent, the potential power of their combined perspectives. Our general approach is to allow psychological theory to suggest a model of performance that can both evaluate specific theoretical claims and reveal important features of the data that had been previously obscured using conventional statistical analyses. Our guiding principle is that model, theory, and data all should be connected.