*Descriptive framework*

Modern information theory (triggered by the question: what if probabilities are unknown?):


Simplicity principle:


Algorithmic information theory (*mathematics*) [43–45]:


Structural information theory (*cognitive science*) [27–29]:


Precisals <sup>2</sup>−*C*, a.k.a. algorithmic probabilities, enable probabilistic formulation

A shared point, however, is that descriptive codes constitute reconstruction recipes for stimuli (just as computer algorithms are reconstruction recipes for the output they produce). Therefore, if a stimulus contains different contrast polarities, then these are necessarily also accounted for by descriptive codes of this stimulus. In Figure 1b, for instance, this implies that the contrast polarity changes in the triangles and diabolos make them more complex than the parallelograms without such changes are. Pinna and Conti knowingly ignored this and applied simplicity as if it is fundamentally blind to contrast polarity. Thereby, they missed the mark in their assessment of simplicity approaches.
