This study examines children's comprehension of non-essential counting features (conventional rules). The objective of the study was to determine whether the presence or absence of cardinal values in pseudoerrors and the type of conventional rule violated affects children's performance. A detection task with pseudoerrors was presented through a computer game to 146 primary school children in grades 2 through 4. The same pseudoerrors were presented both with and without cardinal values; the pseudoerrors violated conventional rules of spatial adjacency, temporal adjacency, spatial–temporal adjacency, and left-to-right direction. Half of the participants within each age group were randomly assigned to an experimental condition that included pseudoerrors with a cardinal value, and the other half were assigned to a condition that included pseudoerrors without a cardinal value. The results show that when presented with a cardinal value, children more easily recognize the optional nature of non-essential counting features. Likewise, the type of conventional rule transgressed significantly affected the children's acceptance of pseudoerrors as valid counts. Participants penalized breaches of temporal and spatial–temporal adjacency to a greater degree than breaches of spatial adjacency and left-to-right direction.