Ideally, confession alone should not be deemed a sufficient ground for conviction, it should be merely something that might make further investigation and, ultimately, establishing "the whole truth" easier. (Looks like the latter is not the goal of the Western judicial system, hence the wide acceptance of the "plea bargain" practice which effectively encourages false confessions.)

Don't chew all day = have one set of problems

Chew all day = another set of problems

We can't win, can we?

A recent piece taking on the topic from a different framing: http://joyousandswift.org/hyperstimulus/

It doesn't deal in addiction but overwhelming stimuli, whether or not they're repeated. Animals can be fooled by unrealistically strong experiences - you can trick a goose into 'hatching' a volleyball - and it appears that humans have somewhat similar responses.

The insights in the two articles aren't vastly different, but there's one intriguing thought: there are actual selection pressures against this. When animals are fooled by superstimuli they're things that are rare (brood parasitism) or nonexistent (volleyballs) - if the extreme stimulus was common they would refine their processing. Some human cases may be hard to avoid (e.g. dopamine receptors) but some might be simpler. It's starting to look like overriding these responses is a major skill for everyday human functioning.

On the topic of creating more friendly urban areas for birds, it has been suggested that making the sides of very tall buildings more amenable to raptor nesting (falcon, hawk) that not only would it benefit those species it could help manage pigeon and starling populations.

That said, having had a hawk nesting site near my house there are a lot of feather piles to deal with :-)