Fig. 1
From: Nevertheless, partisanship persisted: fake news warnings help briefly, but bias returns with time
Visual explanation of the sleeper effect for three hypothetical groups of people. The top group receives no information so they never believe it, the second only receives the information so they believe it (with some fading), and the last receives the information plus a cue that tells them not to trust the original information. We would expect that after the cue, if trusted, people would look like the top group and continue not believing the information, but instead their belief rises over time, getting closer to the group that never received the discounting cue