Because states are abstract entities, they often require embodiment for mass publics and elites to understand them. This embodiment often occurs as personification, where the state is associated with the most salient figure in the political system, but embodiment can also occur through political institutions and social groups. Surprisingly, there is virtually no systematic empirical work on the political and psychological consequences of state personification, or other forms of embodiment. In this experiment, we investigate how various ways of embodying the state influence attitude formation processes. Drawing on the on-line/memory-based processing and entitativity literatures, we hypothesize that personification of the state should facilitate on-line processing and stronger attitudes, whereas embodying the state as a parliamentary institution should produce weaker attitudes that are formed in a memory-based fashion. The results support these hypotheses. Embodiment as a social group produced inconsistent results. This study provides the first systematic evidence that the widespread practice of personification of the state has robust and potentially far-reaching attitudinal consequences that have meaningful implications for strategic interaction, perception and learning, and attitude change in the international realm.