The relationship between outliers in datasets and availability bias
Squeaky wheels. When you have a bad headache, it's tough to think about how great your feet feel.
That's why I stopped watching the news about 50 years ago: the big news was the body count coming back home from Vietnam, followed by the body count of people murdered in various cities. And now it's the number of train derailments, car accidents, examples of human trafficking, etc., etc., etc. Which isn't to say that these things aren't bad and don't happen: they are and they do. But almost ~nothing~ good is ever reported. People seem to buy newspaper and watch TV that shows catastrophes. It's gotten where weather channels insist that people stay tuned for progress of a storm. As if watching the show--and, critically, the ads--will either stop the storm or give you some sort of life-saving warning about when to--what? Call a priest because you're about to die? From a rainstorm?
I'm not sure anyone has recently tried to put on good news and see if it sells. It would be a worthwhile experiment to try. Sure--throw in the bad news, too, but proportionately. Until someone does it, I'm more than happy not to watch commercial television. Haven't missed it for about half a century.