In　this　paper,　we　study　the　interplay　between　individual　behaviors　and　epidemic　spreading　in　a　dynamical　network.　We　distribute　agents　on　a　square-shaped　region　with　periodic　boundary　conditions.　Every　agent　is　regarded　as　a　node　of　the　network　and　a　wireless　link　is　established　between　two　agents　if　their　geographical　distance　is　less　than　a　certain　radius.　At　each　time,　every　agent　assesses　the　epidemic　situation　and　make　decisions　on　whether　it　should　stay　in　or　leave　its　current　place.　An　agent　will　leave　its　current　place　with　a　speed　if　the　number　of　infected　neighbors　reaches　or　exceeds　a　critical　value　E.　Owing　to　the　movement　of　agents,　the　network＇s　structure　is　dynamical.　Interestingly,　we　find　that　there　exists　an　optimal　value　of　E　leading　to　the　maximum　epidemic　threshold.　This　means　that　epidemic　spreading　can　be　effectively　controlled　by　risk-averse　migration.　Besides,　we　find　that　the　epidemic　threshold　increases　as　the　recovering　rate　increases,　decreases　as　the　contact　radius　increases,　and　is　maximized　by　an　optimal　moving　speed.　Our　findings　offer　a　deeper　understanding　of　epidemic　spreading　in　dynamical　networks.　©　2017　Elsevier　B.V.