アジェンダ・ゲーム : メディアの「螺旋」効果と限定合理性

抄録

Recently the problem of media's influence over society has been frequently argued, but never seems to be settled. Major contributions to this field such as the agenda- setting theory or the spiral of silence hypothesis have partly solved it, but have left much more to be solved. What makes this problem so opaque is the lack of a systematic analysis of the interaction between the media and the audience. H.A.Simon found that constituents' voting behavious are bounded by the "frames"by which they decide national issues. His concept of bounded rationality can be the cognitive foundation of the influence of agenda set by the media. It is an optimizing solution to the problem in which few values among many parameters are known exactly and it takes high information cost and algorithmic complexity to decide it by him/herself. A very simple game-theoretical model is introduced to formulate this cognitive view: we model the"spiral"effect as interdependence of two players' optimal strategies (agenda) in a"coordination game"played by the media and the audience (or a medium and another medium), in which one player's payoff is positive if and only if his/her strategy corresponds with another's. Examining its quasi-dynamic behaviour, we found that the more correlated media and audience's agenda, the stronger the spiral effect would be. An important reason for this interdependence is undecidedness of the coordination game: i.e.any cooperative solution is equivalent as long as one's strategy corresponds with another, so there is no criteria for deciding which of these"multiple equilibria"is more desirable than another. What makes such correspondence possible is not the players' rationality but their common knowledge that each other selects more"salient"agenda. This deductive conclusion of game theory coincides with the common wisdom of media theory, and might be a logical foundation of the spiral effect in setting agenda. These findings suggest how to asses the media's influence over the audience: it is not their proper"power"but their function to create such correspondence or momentum that makes them seem so influential. And this momentum will be accelerated by growing ignorance and undecidedness about national issues, because the information matrices we face are becoming more and more complex and multi-dimensional today. Since we will be more bounded and interdependent by sharing information, the media's seemingly strong power is only one symptom of the instability and precariousness of the coming"information society".