[PS-1.4] Variance Distribution and Inclusion Relationship are Necessary but not Sufficient to Produce the Asymmetric Categorization Effect: Revisiting French et al. (2004)

Mermillod, M. 1, 2, 3 , Vermeulen, N. 4, 5 , Kaminsky, G. 6 , Gentaz, E. 7, 8 & Bonin, P. 9, 10, 3

1 Université Blaise Pascal, France
2 CNRS (UMR 6024)
3 Institut Universitaire de France
4 Psychology Department, Université catholique de Louvain (UCL), Belgium
5 Fund for Scientific Research (FRS-FNRS)
6 Toulouse University, Laboratoire CLLE-LTC, Maison de la Recherche, France
7 Grenoble Université, Université Pierre Mendes France, France
8 CNRS (UMR 5105)
9 Université de Bourgogne, France
10 CNRS (UMR 5022)

In different studies, Mareschal, French & Quinn (2001); Mareschal, Quinn, & French (2002) have shown that a simple connectionist autoencoder was able to simulate the asymmetric categorization effect reported by Quinn, Eimas & Rosenkrantz (1993) in 3- to 4-month-old infants. Moreover, French, Mareschal, Mermillod & Quinn (2004) have reported that a careful control of the variance distribution and inclusion relationship of the inputs is sufficient to reverse or even suppress the asymmetry. In the current paper, we show that variance distribution and the inclusion relationship are not sufficient to produce the asymmetry. We show that (i) correlation information reported at the level of the 10 features previously encoded for connectionist simulations was actually very weak and (ii) if we increase these correlations among the different features, we completely suppress the asymmetry categorization effect, even with the same distribution and inclusion relationship that were assumed to produce the asymmetry.