He concluded: «Model conditioning need not be restricted to
calibration of parameters
against observations, but could also include more nebulous adjustment of parameters, for example, to fit expectations, maintain accepted conventions, or increase accord with other model results.