If we apply these principles to the question of the jury, the task is clear. We want to find out whether the coöperation, the discussion, and the repeated voting of a number of persons is helping or hindering them in the effort to judge correctly upon a complex situation. We must therefore artificially create a situation which brings into action the judgment, the discussion, and the vote; but if we are loyal to the idea of experimenting, we must keep the experiment free from all those features of a real jury deliberation that have nothing to do with the mental action itself. Moreover, it is evident that the situations to be judged must allow a definite knowledge as to the objective truth. The experimenter must know which verdict of his voters corresponds to the real facts. Secondly, the situation must be difficult, in order that a real doubt may prevail. If all the voters were on one side from the start, no discussion would be needed. Thirdly, it must be a rather complex situation, in order that the judgment may be influenced by a number of motives. Only in this case will it be possible for the discussion to point out factors which the other party may have overlooked, thus giving a chance for changes of mind. All these demands must be fulfilled if the experiment is really to picture the jury function. But it would be utterly superfluous, and would make the exact measurement impossible, if the material on which the judgment is to be based were of the same kind of which the evidence in the court-room is composed. The trial by jury in an actual criminal case may involve many picturesque and interesting details, but the mental act of judging is no different when the most trivial objects are chosen.