Neuroskepticism - Neurotechnology in National Security and Defense - page 187

structure of the human brain—variations that occur even as between identical twins,
and the right and left hemispheres of the same brain (Weiss and Aldridge 2003).
So the colored area of activation or inhibition may not correlate precisely with the
area represented in the structural image.
This reinforces two points recently made by the neuroscientist and philosopher
Adina Roskies. First, as Roskies (2008) notes, “the conventions of the brain image
are representational translations of certain nonvisual properties related to neural
activity in the brain”—that is, the comparative magnetic properties of oxygenated
and deoxygenated hemoglobin; and second, “the choices that are made [in the con-
struction of a brain image] are not visible in or recoverable from the image itself.”
In my view, functional brain images might properly be considered the product of
“neuroscientific imaginaries” comprising the values, beliefs, and practices of neuro-
imaging communities. 3 This is not to say that functional neuroimages are complete
fabrications, entirely divorced from reality (in this case, brain functions). Rather,
brain images are the product of decisions about scanning parameters and the criteria
for statistical significance, in conjunction with acts of interpretation and represen-
tation that may tell us as much about the imagers as they do about those who are
imaged (although we must look beyond the images to learn much about either of
them).
This analysis is important because it highlights the chasm between the manner
in which brain images are constructed and the way lay viewers in particular com-
prehend them. Many of us appear to assume that visually arresting brain images
share the evidential characteristics and epistemic status of photographs (see Roskies
2008). Empirical support for this view has been provided by David McCabe and Alan
Castel, who have shown that readers attribute greater scientific value to articles sum-
marizing cognitive neuroscience research when those articles include brain images
than they do when the articles include no image, a bar graph, or a topographical map
of the brain. They found that this effect (albeit not large) was demonstrable regard-
less of whether the article included errors in reasoning (McCabe and Castel 2008).
One explanation for this may be, as Roskies contends, that neuroimages are “infer-
entially distant from brain activity, yet they appear not to be.” Put another way, she
argues, brain images are “seemingly revelatory.” In my view, this latter claim needs
to be probed a little further. To whom is what revealed, and by what means? The
answer to this question (one I next endeavor to provide) is vital to a nuanced under-
standing of some of the potential hazards of brain imaging in the arsenal of national
security neuroscience.
In spite of their ubiquity, brain images are essentially meaningless to the
uninitiated. When we look at these images in isolation, there is no “aha!” moment, no
epiphany. The images require the explanation of experts. However, at the same time,
the images also tend to reinforce the expert narrative. It is hard not to be impressed
by the expert (real or apparent) who can guide us through the images—who can bring
us to the point of revelation. And if the expert is compelling enough, it can be hard
to look at the image again without relying heavily on the expert's interpretive frame-
work. In this way, brain images and neuroscientific narratives rely on each other to
work their persuasive and pervasive magic. This point may be illustrated by a notable
recent analog regarding the use of a different kind of image in a national security