A mathematical model of the acute respiratory distress syndrome (ARDS) lung, incorporating simulated gravitational superimposed pressure and alveolar opening and closing pressures, was used to study the mean tidal pressure-volume (PV) slope ("effective compliance") during incremental and decremental positive end-expiratory pressure (PEEP) trials with constant tidal volume (VT) "ventilation." During incremental PEEP, the PEEP giving maximum mean tidal PV slope did not coincide with "open lung PEEP" (minimum PEEP preventing end expiratory collapse of 97.5% of alveoli inflated at end-inspiration), and it varied greatly with varying VT and "lung mechanics." Incremental PEEP with a low VT tests recruitment by the peak pressure, not prevention of collapse by PEEP. During decremental PEEP with a low VT, maximum mean tidal PV slope occurred with PEEP 2-3.5 cm H2O below open-lung PEEP, unless closing pressure was high. High VT, high "specific compliance," and high opening pressures caused slightly greater underestimation of open-lung PEEP. Maximum mean tidal PV slope was always higher (e.g., 93.7 versus 16.69 ml/cm H2O), and the variation in PV slope with PEEP was greater, during decremental PEEP. The maximum PV slope during a decremental PEEP trial with a low VT may be a useful method to determine open-lung PEEP in ARDS, and should be studied clinically.