Rational choice accounts of political participation identify two major solutions to "the paradox of participation" — collective incentives and selective incentives. Prior findings regarding the viability of these solutions are seemingly inconclusive and contradictory. In our view, one important reason for this is that the applicability of these solutions varies across participatory modes. In this paper, we make a first attempt to develop a theoretical answer to the question of why this may be the case. We then test our predictions across four different modes of participation, using longitudinal data that eliminate or reduce the biases inherent in cross-sectional designs. Our results show different types of incentives to strike with distinctly variable force across different modes of participation. Most importantly, we find that whereas electoral modes of participation (voting and party activity) are affected by selective incentives only, the non-electoral modes (contacting and manifestations) are the consequence of both collective and selective incentives.