On 2015 Nov 11, Lydia Maniatis commented:

Feldman seems unaware of a fundamental principle of perception. His error (can you spot it?) is contained in the following statements:

“What we can say, often, is whether estimates from distinct perceptual systems “agree.” When I reach out and grasp the banana, I may find that it feels the same size as it looked. This is extremely useful, because it means that the perceptual system is coherent: estimates drawn from different sorts of evidence agree.”

In fact, there is a fundamental asymmetry between our visual sense and our body senses. The body senses defer to vision – what vision says, goes. With respect to haptics, this is described as “visual capture.” In the relevant experiment, a subject observes a straight rod through a distorting lens that makes it appear curved. It feels curved.

In the tilted room, our eyes tell us we're tilted, even though our body senses tell us (or should tell us) that we're upright. We feel tilted.

In the rubber hand illusion, our eyes tell us that the rubber hand is being tickled in the same rhythm as we are experiencing tickling on our real hand. The experience is percieved as originating in the hand that we see, not the hand that is actually doing the feeling.

Thus, when this false argument is used to promote the “Bayesian inference” brand, it is moot:

“Not incidentally, this is the central principle underlying Bayesian inference, which Dennis Lindley (2006) paraphrased as (his emphasis) “BE COHERENT.”

In the case of perception, the “central principle' seems to be resting on a false assumption.

Coherence is also a central principle of scientific argument. It is impossible, in this brief report, for a reader to evaluate Feldman's remaining claims for “B'ian inference.” For a critical analysis of very sloppy arguments, please see my PubPeer comment on Feldman's (2015) “Bayesian models of perceptual organization.”

On 2015 Nov 11, Lydia Maniatis commented:

Feldman seems unaware of a fundamental principle of perception. His error (can you spot it?) is contained in the following statements:

“What we can say, often, is whether estimates from distinct perceptual systems “agree.” When I reach out and grasp the banana, I may find that it feels the same size as it looked. This is extremely useful, because it means that the perceptual system is coherent: estimates drawn from different sorts of evidence agree.”

In fact, there is a fundamental asymmetry between our visual sense and our body senses. The body senses defer to vision – what vision says, goes. With respect to haptics, this is described as “visual capture.” In the relevant experiment, a subject observes a straight rod through a distorting lens that makes it appear curved. It feels curved.

In the tilted room, our eyes tell us we're tilted, even though our body senses tell us (or should tell us) that we're upright. We feel tilted.

In the rubber hand illusion, our eyes tell us that the rubber hand is being tickled in the same rhythm as we are experiencing tickling on our real hand. The experience is percieved as originating in the hand that we see, not the hand that is actually doing the feeling.

Thus, when this false argument is used to promote the “Bayesian inference” brand, it is moot:

“Not incidentally, this is the central principle underlying Bayesian inference, which Dennis Lindley (2006) paraphrased as (his emphasis) “BE COHERENT.”

In the case of perception, the “central principle' seems to be resting on a false assumption.

Coherence is also a central principle of scientific argument. It is impossible, in this brief report, for a reader to evaluate Feldman's remaining claims for “B'ian inference.” For a critical analysis of very sloppy arguments, please see my PubPeer comment on Feldman's (2015) “Bayesian models of perceptual organization.”