Home About RSS

Law and Neuroscience is more than fMRI

Much, though not all, of the work in Law and Neuroscience has focused on the uses of fMRI.  There are some good reasons for that focus.  Functional magnetic resonance imaging is our best current tool for seeing local regions of brain activity during particular thoughts or actions, as well as for seeing patterns of activity, increased or decreased.  But it is important for us to remember that fMRI is just one of many tools leading the ongoing revolution in neuroscience, with all of its likely consequences for human society and human law.

One reason to remember this is that fMRI has come under some prominent scientific fire lately – or, at least, the claimed results of some fMRI experiments have.  Nikos Logothetis led off with a long review in Nature last June:  What We Can Do and What We Cannot Do with fMRI, Nature 353:869-878 (2008).  More recently, an attack on size of the claimed correlations of many fMRI experiments has been circulating in preprint form:  Edward Vul, Christine Harris, Piotr Winkielman, and Harold Pashler, Voodoo Correlations in Social Neuroscience, Perspectives in Psychological Science (forthcoming, Sept. 2009).  The Vul piece may set a record for the most circulating comments before publication; it has already been the subject of a news article in Nature:  Alison Abbott, Brain Imaging Studies Under Fire (published on-line on Jan. 13, 2009).     Another article, by Sirota and Das, reports an experiment that showed substantially increased blood flow (and hence a BOLD signal) in the visual cortex of experimental subjects just before they were to be shown one in a series of visual stimuli, thus casting doubt, at least on some situations, on what the BOLD signal means. Yevgeniy B. Sirota and Aniruddha Das, Anticipatory Haemodynamic Signals in Sensory Cortex not Predicted by Local Neuronal Activity, Nature 457:475-79 (Jan. 22, 2009).  A brief blurb in the most recent issue of Nature Neuroscience discusses another, similar, study:  Neuroimaging:  Interpreting the Signal, Nature Neuroscience 10(3):166.

What does all this mean?  I’m a lawyer, not a neuroscientist – on one level, my answer has to be “beats me.”  (Now, what does it mean that so many of these studies are published in, or publicized in, Nature or its family of journals is another interesting question, though of a different sort, perhaps involving the sociology of science (or of Science and Nature) publishing.  I also don’t have a clue about that.)

More broadly, this seems to me part of the normal response to hot new technologies. The advocates for the technologies make broad, sometimes overbroad, claims.  The opponents point out questions, concerns, and inconsistencies, real or merely apparent.  When the technology is really hot, a kind of jealousy may also be in play, leading those who are working in the now-cooler technologies to relish any questions about the new kid in town.  (To this semi-outside observer, science doesn’t always look all that different from high school . . . though neither does academic law.)  My guess is that at least some of these papers will turn out to provide some useful cautions and corrections, while leaving the fundamental value of fMRI technology untouched, though how well that value will stand up in the face of time, and newer technologies, remains, of course, to be seen.  (The fact that neurosurgeons sometimes use fMRI to help them map out their surgeries gives me some confidence in its power.)

BUT regardless of the current state of opinion on fMRI technology, it is important to realize that Law and Neuroscience need not, should not,  and, in fact, does not derive from fMRI.  Functional MRI may be the most useful (best spatial resolution, least risk) technology for “mind-reading” applications of neuroscience – using neuroscience to try to detect deception, pain, or bias, as well as trying to detect “criminal predispositions”.

It is not even the only way to do the kind of functional imaging that can lead to these kinds of legally interesting (whether or not legally relevant or legally admissible) results.   Consider the use (more likely, in my view, misuse) of EEG technology in lie detection in both “Brain Fingerprinting” and in India’s “BEOS” technology.  And PET, SPECT, and even CT scans have been introduced in court as part of criminal cases, in excuse or mitigation or as part of sentencing.  fMRI is the hot neuroimaging technology and, in many ways, seems to be the best.  For those of us coming to this without a strong neuroscience background (including me), it may make sense to learn first about fMRI. But we should not forget that research is continuing with these other technologies and some of it may end up being quite important for law.

More importantly, Law and Neuroscience is much broader than just mind reading or responsibility – and thus than functional neuroimaging.  Legally relevant predictions might be done by fMRI, but they might also be done by structural scans (as well as by “neurogenetic” means or straight behavioral methods).  “Treatments” for criminal or other anti-social behavior will generally not involve fMRI but will involve drugs (to treat addiction or inappropriate sexual urges), vaccines (for drugs), surgery (for addiction or possibly other behaviors), deep brain stimulation (for various behaviors), or other methods (transcranial magnetic stimulation, for example).  (I’ve written on these issues in the Kansas Law Review 56:1103-38 (2008), for those interested.)   And the host of issues around “enhancement” are likely to involve technologies similar to those used in “treatment” – and not in fMRI or, more broadly, in functional neuroimaging.

Those of us who work in this area must be aware of, or at least open to being educated about, the weaknesses of fMRI, but we also must be aware of the strengths of other technologies for learning about, and changing, the human brain.  Law and Neuroscience is not just fMRI.

– Hank Greely

2 Responses to “Law and Neuroscience is more than fMRI”

  1. amyknight says:

    I’m a soon-to-be law student with a degree in cognitive science, and I’m really glad to hear someone calling for us to remember other technologies. EEG is my pet favorite, since that was the medium I worked with, and even when people do shocking things with it (last year’s conviction of the young Indian woman for murder comes to mind) it seems to be in the shadows of fMRI. EEG is simpler and cheaper, not to mention much more precise in its measurement of the timing of impulses, and I agree that it may be more likely to keep popping up as a proposed legal tool.

    But no matter which technologies grab our attention, doesn’t accepting brain-based evidence at all entail a solid belief in reductionism? Can we explain the complex behavior of individuals by looking at brain activity? Can we look at brain activity and extrapolate into actions and motives? The answer is either no, or not yet, depending on where you stand. And last time I checked (though I’m a few years out of my cognitive science program), there was still a debate going on about whether we’d ever be able to reduce human behavior and experience to the merely chemical. You could find an expert witness to tell you it could, and one to tell you it couldn’t.

    I’m wondering just how reliable or reproducible a neural signature would have to be in order for it to be legally useful. I’m sure this has been considered and talked about – I’d love to hear your thoughts. If the human brain can essentially be wired backwards (as it is in some left-handed people) and still function perfectly well, then can we really claim that the presence or absence of any given brain wave can prove something beyond a reasonable doubt? What do you think?

Leave a Reply