The first argument makes no sense to me. The videos clearly show the facilitator rapidly typing away with Houben's finger even while he's looking away or has his eyes closed, which is by itself a very strong reason to be skeptical, especially in light of the past record of facilitated communication. The second argument is not only ad hominem, but further refuted by similar analysis by a neuroscientist. The last argument is a bit better, but wrongly assumes that the only alternative is that the doctor and family are lying. Facilitated communication isn't a matter of conscious fraud, it's a matter of self-deception of the facilitator (enhanced by the expectations and reactions of the family). Given the possibility of unconscious cuing of the facilitator by the doctor, as well as his own vested interest in a positive result, the test he described doing is still far from sufficient to overcome the evidence plainly displayed in the videos.
Unfortunately, there is a very strong incentive to believe on the part of the doctor, the facilitator, and the family. To find that the communications are coming from the facilitator would be emotionally devastating, and detrimental to the doctor's credibility. To test further is to risk a huge potential loss of what has apparently been gained, and I suspect it's unlikely that we'll see it happen.
But look at it from Houben's own perspective--further testing is absolutely in his own best interests. For if the facilitator is the one doing the communicating, not him, then he is being further exploited for the satisfaction of his doctor, facilitator, and family, not for his own benefit. He's not being treated respectfully or as an end, rather than as a means. If he is, in fact, minimally conscious as the brain scans suggest, then speaking on his behalf without his genuine input is doing him even greater harm.
If you reject the idea that an hour or so of Houben's time should be used to do a conclusive, double-blind test to see whether the communications are coming from him or from the facilitator, is it because you want to believe, rather than to know? There is clear possible harm to Hoeben from not doing such a test. There is no harm to Hoeben from such a test, though there's clearly the risk of painfully dissolving an illusion for the doctor, facilitator, and family. But Hoeben's interests should be placed above that risk.
(Previously on Houben, a post with many links and references.)
UPDATE (February 15, 2010): Houben has been put to the test, and it turns out the communications were, in fact, coming from the facilitator.
UPDATE (February 20, 2010): David Gorski at the Science-Based Medicine blog has a bit more from the Belgian Skeptics, who were involved in the test.