This post by Stephen Downes presents an interesting link, but I don’t agree that it supports his thesis about human reasoning. Pattern recognition at this level is characteristic of many species and devices, and it’s a capability that is probably necessary for human reasoning whether or not such reasoning is based on rules and language.
In fact, it seems to me that if such a thing as distinctly human reasoning exists, then it shares many features with other more rudimentary forms of reasoning, but to say it is “based on” these is similar to saying it is based on biochemistry (or even physics if you want to go that far down into the foundations).
But if, as the base of human reasoning we are looking for a characteristic which distinguishes it from other forms of reasoning, then language is not enough since it appears that language at some level is practiced by other species. Some people suggest that an awareness of contingency is key, but anyone who has watched a cat learning to control a mouse may be inclined to disagree. But unless Chimp research proves otherwise it may be that the language of contingency is unique to us (and perhaps also that of propositional truth though I suspect that a chimp who can lie to keep a friend from finding a treat might eventually be trained to recognize and label the lie of another). Failing that it may just be that the concept of a distinctly human form of reasoning is a mere conceit based on quantity or scale rather than anything qualitatively different.