Policy learning is the mechanism through which actors involved in a policy subsystem revise their beliefs and preferences toward a policy over time – a crucial dynamic of stability or change of public policies. While the social dimension of this dynamic has been extensively researched, the individual psychology of policy learning remains a black box. Yet, this is a key missing link between policy learning and settings or practices that could model it. This paper addresses this research program by looking at two mental constructs susceptible to encourage policy actors to stick to their own point of view rather than to assimilate new policy information: egocentrism and self-esteem. The test of the hypotheses is based on regression analyses of a survey conducted in 2012 among 289 Belgian policy actors who had been involved, during the last two decades, in the European liberalization policy process of two network industries: the rail and electricity sectors. The findings are threefold. First, rational knowledge utilization remains a stronger cognitive dynamic of information processing than egocentrism and self-esteem. Second,still, egocentrism is not only a source of biased assimilation of policy information: it also directly induces a less positive alignment of policy actors’ preferences toward liberalization over time. Third, the results fail to confirm my theoretical expectations about the relation between self-esteem and policy learning. The theoretical and practical implications of these results are discussed.