Current research on persuasion is dominated by two dual-process theories: the elaboration likelihood model and the heuristic-systematic model. According to them, persuasion can be accomplished via two distinct routes. Lately, there has appeared an alternative approach - the unimodel, which views persuasion phenomena in terms of a single process. The aim of the experiment was an attempt to verify both models. The experimental design included all the variables recognized as important from the perspective of both approaches (involvement, source expertise, source information length, argument quality, argument length, evidence appearance order). The research concerned attitudes toward the European Parliament election and was carried out in two stages. In the first stage, attitudes toward the elections were measured. In the target experiment, the key manipulations were made and the attitudes and the involvement were measured. The analyses showed, that low involvement subjects were influenced by argument strength, when the arguments were easy to elaborate. The results provide a support for the unimodel's notions.