This　paper　aims　to　investigate　the　effects　of　human　behavior　and　contact　heterogeneity　on　the　spread　of　infectious　diseases.　For　this　purpose,　a　network-based　SIRS　epidemic　model　with　a　general　feedback　mechanism　is　proposed.　In　contrast　to　previous　models,　we　consider　the　different　fear　degrees　of　individuals　who　have　different　potential　number　of　contacts　with　others,　when　an　epidemic　prevails.　The　basic　reproductive　number　that　governs　the　global　dynamics　of　the　model　is　analytically　derived.　Accordingly,　the　permanence　of　the　disease　and　stability　conditions　of　the　equilibria　are　studied　in　detail.　It　is　shown　that　the　general　feedback　mechanism　cannot　change　the　basic　reproductive　number,　but　theoretical　and　numerical　results　indicate　that　it　plays　an　active　role　in　reducing　the　disease　damage.　The　obtained　results　generalize　and　improve　some　well-known　ones.