Yes, this is what they should use to indicate where the original content can be found, especially if they want that content to be indexed only for the main site/page that it can be found on.
In the case of
ign.com, I can see they have in their HTML source code:
<link rel="canonical" href="http://www.ign.com/" />
Without knowing the other URL's that you're referring to, I can only guess that they have this link in them also whenever duplicate content can be found. The alternative would be to specify in
robots.txt not to crawl these other review pages.
On the other hand, if they want the comments to also be indexed, and they are not dynamically generated or too complex for the search engine bots to index, they might just leave it to the search engine bot to decide. In this case, they would have pages competing against each other for the same content, with the potential of the main site/page losing out.
Here's a link from a Google Webmaster Tools answer for more information on this: About rel="canonical"