The Dana Foundation‘s Cerebrum recently featured an article by Russel A. Poldrack, which addressed some common misconceptions about neuroimaging research. The piece starts off with this quote:
Colorful brain images may tempt researchers to make claims that outpace solid scientific data—and may tempt the public to believe those claims. In particular, although brain imaging has provided solid evidence of alterations in brain structures and functions associated with many psychiatric disorders, it can be used neither to diagnose such disorders nor to determine exactly how treatments work—at least not yet. Keeping some key ideas in mind can help us evaluate the next report of a brain-imaging “breakthrough.”
The issues and problems with the interpretation of brain images is no stranger to neuro-ethics discourse. For example, Illes et al., have provided conversation on the ethical, legal and social issues (ELSI) of advanced neuroimaging, in addition to other papers that explore epistemological and ethical issues that come with the current limitations of imaging technology (see: e.g., Racine, Bar-Illan & Illes, 2005; Illes 2007; Illes, Racine, & Kirschen 2006). Others, such as McCabe and Castel and Weisberg et al., have taken an empirical approach to understand the “temptation” of brain images by testing human participants directly. These latter studies demonstrated that presenting brain images or neuroscience information with research resulted in higher ratings of scientific reasoning, or as a more satisfying explanation – even if the information presented was “logically irrelevant” (as was the case in the Weisberg et al study).
The Dana article on “pipe dream” neuroimaging comes right on the heels of a couple other reports on some crucial methodological and statistical flaws in the data analysis process of brain imaging. The first, which received considerable online debate in blogs and elsewhere, was the paper “Voodoo Correlations in Social Neuroscience.” The lead author, Ed Vul, suggests that the analytic methods used in imaging indicate that many studies used faulty techniques to obtain their data. A second paper, currently in press for Nature Neuroscience, argues against “double-dipping” in systems neuroscience research. Double-dipping refers to “the use of the same dataset for selection and selective analysis,” and “will give distorted descriptive statistics and invalid statistical inference whenever the results statistics are not inherently independent of the selection criteria under the null hypothesis.” This, no doubt, gives more fuel to the fire for the extreme sceptics of neuroscience, medicine, and science in general. Who will guard the guardians of neuroscience?
What the Cerebrum article got me thinking of in particular, aside from what I briefly refer to above, and the – loosely – hermaneutical process of interpreting brain images, is how we as human beings relate to facts that we receive from science and medicine, and our relationship with technology; in particular technology that provides facts associated with the self. Joseph Dumit has explored this concept, and the persuasive power of brain images in his book Picturing Personhood. Using a neuro-anthropological perspective (pardon the neologism), Dumit examines the power that brain images have, as represented in the mass media, in altering understandings people have of their own bodies and brains – a term he calls the “objective self”. I have become particularly fascinated in the notion of objective self-fashioning, specifically as it relates to the research field of imaging genetics (or imaging genomics). In an up-coming paper I’m giving at the 20th Annual Canadian Bioethics Society Conference in Hamilton, ON, I argue that the power (and persuasiveness), increased sensitivity, and less statistical variability of the combined technologies (genetics + imaging) requires a heightened clinical sensitivity to the objective-self-fashioning process in the translation of knowledge derived from imaging genetics. This is not an argument for neuro- or genetic-exceptionalism. I welcome feedback on this claim as I develop my paper.
I will close with a quote from Hall and colleagues on addiction, the notion of disease, and neuroimaging and the “seductive allure” of neuroscience explanations:
A ‘disease’ that can be ‘seen’ in the many-hued splendour of a PET scan carries more conviction than one justified by the possibly exculpatory self-reports of addicts who claim that they are unable to control their drug use (p.867).
Neuroethics at the Core 
Hmm- brain imaging- intriguing but difficult stuff. I was always a bit skeptical when people started producing lovely coloured images representing arcane patterns of electrical discharge averaged through umpty filters and amplifiers and enhancers…There’s something there, but it takes more brain activity than I’ve been able to generate to understand half of it! I was interested in the imaging stuff that Stephen Kosslyn and mates had collected on discharges in particular pathways during input and output of info on different views of static and moving geometric objects. I was in work concerning processing of images of the body in people with anorexia nervosa as we had found indications they were avoiding correct naming of true size under pseudo signal detection paradigms. All a fuzz of weird stuff now!!