Traditional normative models of rational choice that require agents to find optimal choices among alternatives are computationally infeasible for real agents.

\(\neg K_i \phi \rightarrow K_i \neg K_i \phi\), which expresses that \(i\)’s failure to know \(\phi\) entails that he knows of this failure – are considered more controversial. , Hintikka 1962; Lenzen 1978; Fagin et al. 1995) is that the most defensible choices of logics of knowledge lie between the modal systems \(\textsf{S4}\) and \(\textsf{S5}\). [56] Note, however, that both of these results seem prima facie implausible relative to our everyday understanding of knowledge. For on the one han