The present article, of quantitative cut, aims at testing –via linear logistic test model (LLTM)– a set of cognitive operations (i.e., rules) influencing the item difficulty of a fluid intelligence test on different samples of students. In Study 1, high school students (n = 1751) were randomly assigned to study and validation samples. The former sample served to test a proposed set of rules as variables affecting item difficulty and the latter aided to bring evidences of validity of these rules. In Study 2, university students (n = 162) were recruited to determine whether the influence of these rules on the level of difficulty of the items could be generalized to this new group. Study 1 brings evidence of the viability of a set of cognitive operations underlying the process of solving the items while Study 2 suggests individual differences in respondents’ solution strategies. The same strategy of analysis could be applied to the construction of other tests and may help educators, researchers and decision-makers to improve their pursue of relying on the most refined instruments.