A comment on the generalization of the Marinatto-Weber quantum game scheme

Iqbal and Toor [Phys. Rev. A {\bf 65}, 022306 (2002)] and [Commun. Theor. Phys. {\bf 42}, 335 (2004)] generalized the Marinatto-Weber quantum scheme for $2 \times 2$ games in order to study bimatrix games of $3 \times 3$ dimension, in particular the Rock-Paper-Scissors game. In our paper we show that Iqbal and Toor's generalization exhibits certain undesirable property that can considerably influence the game result. To support our argumentation, in the further part of the paper we construct the protocol corresponding to the MW concept for any finite bimatrix game that is free from the fault.