The Intrepid Project, funded by the Swiss National Science Foundation, aims to develop a general understanding of how policy announcements
by state agencies are interpreted by journalists in ways that send signals, indicate intent, and otherwise provoke economic and political reactions.

​

To do this, journalistic accounts of policy announcements on central bank monetary policy and on foreign policy are hand-annotated to capture particular features. We then use machine learning techniques on semantic and syntactic properties of announcement texts that differ on issue domains (monetary vs. foreign policy), country studied (the United States, France, and Canada), and time period (when archival materials are available, from the late 1960s to 2018), matching those announcement texts to hand-coded journalistic texts. In this way, we aim to develop models of the announcement interpretation process.