The paper is concerned with the computational evaluation of a new branching heuristic for evolved DPLL Satisfiability solvers. Such heuristic, like several other recent ones, is based on the conflicts obtained during the solution of an instance. A score is associated to each literal. When a conflict occurs, some scores are incremented with different values. The branching variable is then selected by choosing the maximum score. This branching heuristic is introduced in two evolved DPLL solvers: ZChaff and Simo. Experiments on several benchmark series, both satisfiable and unsatisfiable, demonstrate advantages of the proposed heuristic.