A new paper in the Journal of Evolutionary Biology helps debunk the peppered moth icon, and cites Discovery Institute’s Jonathan Wells in the process.
You wouldn’t know it from the summary on PhysOrg, or from the abstract of the paper in the Journal of Evolutionary Biology, but some South Korean scientists have joined in the cause to dismantle a celebrated “Icon of Evolution,” the myth of the peppered moth. That’s right; in the first paragraph, Kang, Moon, Li and Jablonski set right to work, pulling the rug out from Kettlewell (1955) and all his copycats who made the peppered moth Exhibit A for the power of natural selection:

Evolution of adaptations through natural selection is the central theory in biology (Darwin,1859; Fisher,1930), and crypticity of moths (morphological phenotype) has been the icon of morphological adaptations to avoid predation throughout the history of evolutionary biology (Thayer,1909; Cott,1940; Kettlewell,1955a,b; Endler,1984).

That word “icon” was no accident. The authors cited Jonathan Wells’s book Icons of Evolution (2000) twice:

The importance of previously unexplored aspect [sic] of behaviour in moth crypticity has became apparent in recent discussions (Majerus,1998; Wells,2000) of Kettlewell’s classic research (Kettlewell,1955a,b), as well as in popular attempts to criticize the evolution as a scientific discipline (Wells, 2000; Hooper, 2003).

The stage is set for a classic showdown: either Wells was right to criticize this icon, or we’re going to show those ID people that Darwin’s famous icon can withstand the criticism. We note in passing, though, that Majerus (the late staunch evolutionist who devoted his final years to proving the validity of peppered moths as evidence for evolution), and Wells were mentioned together in “recent discussions” about the importance of the behavioral aspect of the story.
In a series of clever experiments, the researchers used native South Korean moths that employ camouflage when resting on their native habitat, pine trees. But instead of pinning the moths to the trees, or placing moth images on bark images, they thought of a ridiculously simple method: “Rather than arranging moths in positions that are believed to imitate their naturally achieved resting spot and orientation, we allowed the moths to do this for us.” Why didn’t anyone think of this before?
Actually, Kettlewell apparently did: his original experimental paradigm was, “a moth positions itself — a predator attempts to detect it.” Somehow, neither he nor later investigators got around to it. Instead, they interfered with nature: the experimental paradigm morphed into, “an experimenter arranges the prey — a predator attempts to detect it.” The South Korean team recognized the bias in this plan. As if quoting Wells, they said (overlooking the Koreanized English):

Researchers used specimens pinned on a tree bark (or photographs of thereof), or images copy-pasted onto a tree bark image, in body orientations chosen by the experimenter based on the knowledge on how the moths position their bodies… This paradigm ignores that moths may choose landing spots on a bark differently than humans do. This paradigm also ignores that the moths after landing actively search for a “suitable” resting spot and that the body orientation in nature functions as adaptation for camouflage only when matched with the natural behavior to seek the resting spot.

So, Kang et al. decided to go back to the original plan and let the moths do what they do naturally without investigators interfering. They let the moths choose the trees to land on, and they photographed what they did after they landed. It changed the picture entirely.
Into the forest they went, releasing their captive moths (two native species), and watching them with Canon cameras in hand. They photographed the initial landing spot and orientation, then watched them come to a final resting spot and orientation — often some centimeters from where they initially landed. Calibrating their photographs carefully with controls, they brought these photographs into the lab, cropped them to remove extraneous cues, and showed them to human subjects as “experimental predators.”
Using software that flashed the images on a screen, they asked the human predators to hunt for the moths in the photographs within a reasonable time limit. A major discovery followed. The moths were much harder to spot after they moved from their initial landing spot to the final resting spot. You can see this for yourself in some of the photographs they published. The moths moved around, some of them rotating, till they became almost invisible. You can also watch the moth behavior in movie clips from the Supporting Information page. Sure enough; those clever insects demonstrated an uncanny ability to blend perfectly into the bark by moving around. If you had photographed only their initial landing spot (or had glued them to the tree trunk), you wouldn’t know this.
It’s astonishing (and deplorable) that during the 57 years that the peppered moth has been promoted as a stellar case of evolution in action, nobody performed this experiment until now: “… no direct test of adaptive role of this behavior has been conducted,” the team says on page 2, and again on page 5, “This is the first study that directly measures the detection of moths by visual predators before and after a moth performed the body positioning behavior.”
Yet the behavior of the moth is key to the story. In the real world, as opposed to the staged peppered-moth photos we have all seen in textbooks, the moths would not have stood out like sore thumbs on the tree trunks. They would have wandered about for a place to blend in; most likely, they would not have landed in such conspicuous spots in the first place (Wells, p. 148). Letting the moths do what comes naturally is what the experimenters should have done. The old peppered moth experiments, consequently, have been invalidated (again).

In summary, by using one of the iconic examples of evolution, we showed how a morphological adaptation cannot be fully understood without taking into account a full behavioural phenotype responsible in natural situations for increasing the adaptive function of the morphological trait.

To be sure, Kang et al. do not deny that natural selection might be at work. They propose that morphological and behavioral adaptations might “co-evolve” to produce the moths’ highly effective camouflage. Some might criticize their methods on various grounds.
Possible criticisms include: (1) “Human predators” are a poor substitute for the moths’ natural bird predators. The authors acknowledge this, but explain that both birds and humans rely on visual cues to first locate the moths; besides, previous investigators also used humans to test camouflage effectiveness. (2) UV light, to which some birds are sensitive, might affect the visibility of the moths on the trees. The authors took this into account, but from their measurements of the UV spectrum of tree trunks and moths, they believe it would be a minor factor, at most, for UV-sensing birds. (3) They only studied two Korean moth species, and not peppered moths. The authors realize this; they were not making a global generalization. They were just cautioning investigators to be aware that behavioral phenotypes must be considered in all future studies of adaptive camouflage; and, thereby, they invalidated all previous work that had ignored this factor.
But at the very least, they yielded some highly significant ground. For one, they could not ratify the work of previous investigators working for six decades. For another, they revealed a key part of the story that previous investigators had all overlooked. And finally, the paper shows that criticisms of Darwinism by ID writers like Dr. Wells are having an impact — even in the Journal of Evolutionary Biology.