Hi 94501,
Example #1 above is the one true page for search and the one that all the canonicals reference.
If the pages are properly canonicalized then Example #1 will receive nearly all of the authority stemming from pages with this URL as the canonical tag.
I.e. Example #2 and #3 will pass authority to Example #1
Examples #2 and #3 shouldn't be in the index because the canonical points to url #1.
Setting a canonical tag doesn't guarantee that a page will not be indexed. To do that, you'd need to add a 'noindex' tag to the page.
Google chooses whether or not to index these pages and in many situations you want them indexed. For example: User searches for 'product X' and product x resides on the 3rd page of your category. Since Google has this page indexed (although the canonical points to the main page) it makes sense to show the page that contains the product the user was searching for.
Example #4 shouldn't be in the index, because it's just a source code that, again doesn't change the page and the canonical points to #1.
To make sure it is not indexed, you would need to add a 'noidex' tag and/or make sure the parameters are set in GWMT to ignore these pages.
But again, if the canonical is set properly then the authority passes to the main page and having this page indexed may not have negative impact.
Example #5 shouldn't be in the index because it's excluded in parameters as not affecting page content and the canonical is in place.
How long ago was the parameter setting applied in GWMT? Sometimes it takes a couple weeks to deindex pages that were already indexed by Google.