> I'm not a regular user of Chrome, so maybe that's how they all look Interestingly, that's an exact replica of the real Chrome Cleanup Tool that was/is in use

> I'm not a regular user of Chrome, so maybe that's how they all look Interestingly, that's an exact replica of the real Chrome Cleanup Tool that was/is in use when we ran the research (I'm a coauthor). The news article only reports overarching "did they heed the warning" stats, but in the published paper we actually didn't care whether they heeded it or not. We cared how much they deliberated over their decision before making a choice. We measured deliberation, or attention, via mouse cursor movement tracking. For example, straight-line trajectories towards the dismiss button indicate less cognitive attention than do less-than-straight-line trajectories (wavering, hover time, click latency, etc.). If someone deliberated for a long time, they might decide in the end that they thought it was spam. But at least they deliberated instead of making an automatic, hasty decision! Users were more likely to show signs of deliberation when the prompts were shown at low-interference times than they were at high-interference times.