The players types are increasingly

demanding especially in relation to the

experience that the game provides. In shooter

games, although the bots often have the same

level of skill and follow specific rules throughout

the game, not presenting their own strategies.

This fact can mean a fragility in the game design

process, since players expect even more realism

in shooter games. In this context, the creation

of bots showing believable behaviors can be

decisive for the success of a shooter game

next to its target audience. This work presents

a proposal of revisiting the Bartle’s typology

in order to adapt it to bots behavior. Then,

considering a shooter battle arena scenario, this

work assumed that realistic and efficient bots for

shooter games can be developed based on four

behavior types: killer, achiever, explorer, and

beginner. Each bot behavior was defined through

a finite state machine based on the verification

of the environment and decision making. From

the experiment’s results analyses, it is possible

to conclude that as important as the behavior

of the bots, is the interaction of this with the

environment, which plays a fundamental role

in the question of player experience and the

believability related to immersion into the battle

scenario.