I have another possible explanation, which I think deserves a far greater “probability mass″: images make scientific articles seem more plausible for (some of) the same reasons they make advertising or magazine articles seem more plausible—i.e., precognitive reasons which may have little to do with the articles’ content being scientific. McCabe and Castel don’t control for this, but it is somewhat supported by their comparison of their study with Weisberg’s:
The simple addition of cognitive neuroscience explanations may affect people’s conscious deliberation about the quality of scientific explanations, whereas the brain images may influence a less consciously controlled aspect of ratings in the current experiments.
“-Scientific content, -scientific images” includes most advertising, which is pretty obviously made more convincing through images. For an example of “+scientific content, -scientific images”, think of the many articles in (say) New Scientist that are made more pleasant (and quite possibly more convincing) by more-or-less purely aesthetic graphics.
I can also think of some “less consciously controlled” reasons that are science-specific. Images of brain scans lend a kind of “hard science” sheen to the articles’ claims—in much the same way that CGI molecules spinning around hair follicles add to shampoo advertising’s claims of sheen (“-scientific content, +scientific images”). McCabe & Castel again:
This sort of visual evidence of physical systems at work, which is typical of ‘‘harder’’ sciences like physics and chemistry, is not typically apparent in studies of cognition, where the evidence for cognitive processes is indirect, by nature. Indeed, it is important to note that while brain images give the appearance of direct measurement of the physical substrate of cognitive processes, techniques like fMRI measure changes in relative oxygenation of blood in regions of the brain, which is also indirect. Of course, it is unlikely that this subtlety is appreciated by lay readers.
In other words, images of brain scans create the impression that underlying physical mechanisms are better understood than they actually are. This is also an issue in pop science reporting:
[...] many cognitive neuroscientists have expressed frustration at what they see as the oversimplification of their data, and have suggested that efforts be made to influence media coverage of brain imaging research to include discussion of the limitations of fMRI, in order to reduce the misrepresentation of these data.
So how does this study pertain to physicalism? As I see it, this study underscores the ease with which intelligent people—including physicalists—can be fooled into thinking that scientific studies explain more than they do by the use of overly-concrete, hard-science-flavored imagery (and language). It shows how easy it is to jump from an image of a presumed physical substrate for some phenomenon to the belief that we better-understand that phenomenon. In other words, it shows how the impression of reductionism can function as a curiosity-stopper.
As I understand it, that is a common criticism of reductionism in practice.
Also, this is why I’m uncomfortable with the overuse of overly-precise terms from maths and science—like referring to one’s own “probability mass” on Less Wrong, or the Churchlands bemoaning their “seratonin levels” rather than saying they feel horrible (see here, p. 69). Sometimes an unwarranted science-y aesthetic can mislead.
Luke—your typology of ends reminds me of something I was reading recently by Jonathan Edwards. I know this is not an atheology post, and the Edwards work isn’t particularly empirical, but I thought it might be an interesting antecedent besides.