Tuesday, 26 July 2011

Brain scans could influence jurors more than other forms of evidence

It's surely just a matter of time until functional MRI brain scans are admitted in US and UK courts. Companies like No Lie MRI have appeared, and there have been at least two recent attempts by lawyers in the USA to submit fMRI-based brain imaging scans as trial evidence.

Functional MRI gauges fluctuating activity levels across the brain, with experts divided on the merits of using the technology as a high-tech lie detection measure (see earlier). The late David McCabe who died earlier this year, and his colleagues, have put that debate to one side. They asked: if fMRI evidence were to be allowed in courts, would it have a particularly influential effect on jurors' decisions? There's good reason to think it might. For example, a 2008 study by Deena Weisberg found that lay people and neuroscience students (but not neuroscience experts) were more satisfied by bad scientific explanations when they contained gratuitous mentions of neuroscience.

For the new study, 330 undergrads at Colorado State University read a vignette about a criminal trial in which a defendant was accused of killing his estranged wife and lover. Various points of evidence were mentioned and summaries of testimony and cross-examination were provided (the vignette amounted to two pages).

Crucially, a sub-set of the participants read a version in which fMRI evidence was cited: "... there was increased activation of frontal brain areas when Givens [the defendant] denied killing his wife and neighbour, as compared to when he truthfully answered questions." For comparison, other participants read a version that either included incriminating evidence from polygraph, from thermal imaging technology (which measures changes in facial skin temperature), or that contained no lie-detection technology.

The key finding was that participants who read the brain-imaging version were far more likely (76 per cent) to say they considered the defendant guilty, compared with participants who read the other versions (47 to 53 per cent). Moreover, the lie-detection evidence was more likely to be cited by participants in the fMRI condition as key to their decision, as compared with participants who read versions that didn't mention fMRI.

The participants were not entirely seduced by fMRI. Some of them were given a slightly different version of the fMRI vignette, in which the expert witness warned about the technology's unreliability. These participants came to a similar proportion of guilty verdicts as the participants who read the vignette versions that lacked fMRI evidence. So it seems the persuasive influence of fMRI evidence can be tempered easily enough if people are reminded of its limitations.

The researchers acknowledged the obvious weaknesses of their study: the use of students as mock jurors, the use of vignettes rather than a real trial, and so on. These caveats aside, they said their data show that fMRI evidence could be more influential than other types of evidence. "... [T]hough determining whether that indicates the evidence would lead to unfair prejudice, confusion of the issues, misleading the jury, or needless presentation of cumulative evidence is a complex issue," they said. "At the very least, it appears that juries should be informed of the limitations of fMRI evidence."

ResearchBlogging.orgMcCabe, D., Castel, A., and Rhodes, M. (2011). The Influence of fMRI Lie Detection Evidence on Juror Decision-Making. Behavioral Sciences and the Law DOI: 10.1002/bsl.993

Further reading: The brain on the stand, by Jeffrey Rosen, New York Times magazine.

This post was written by Christian Jarrett for the BPS Research Digest.


Nora Miller said...

I don't doubt that jurors will be swayed by brain images. They are very seductive. In her very enjoyable book, "Delusions of Gender", Cordelia Fine questions how much we can actually deduce from them to explain volitional behavior. Consider Craig Bennett's study that produced an fMRI image from a dead fish. It seems to me that study alone would suffice to counteract any use of such images as evidence in court.

As for determining the effect on jurors of the inclusion of such images, the wording used in the study just reported might have had as big an effect as the image: "... there was increased activation ...when ...the defendant denied killing his wife and neighbour, as compared to when he truthfully answered questions." (Emphasis mine.] Including "truthfully" where they did sets up the likely perception that the denial was known to be false, which might carry as much or more weight than the image itself. Was this exact wording was also used to introduce other technological evidence? Was a similarly biasing statement used for jurors who saw no technological evidence? Seems like a bit more work is needed here.

Christian Jarrett said...

Hi Nora - the text for the other lie-detection evidence was similar. Eg. for the polygraph: "there was increased physiological activity when Givens denied killing his wife and neighbour, as compared to when he truthfully answered questions."

Nora Miller said...

Thanks for the quick reply, Christian. Glad to see they used consistent language. Sounds like they got a meaningful result then. I guess we will have to rely on challenges to validity by opposing counsel to help offset the gee-whiz affect of the scan. I am regularly amazed by these studies that show how easily our views can be subtly swayed by the wording of a statement. Caveat auditor!

Richard said...

Who was it that described fMRI and the lovely pictures of parts of brains 'lighting up' as 'the new phrenology'?
The 'white coat' coupled with huge and expensive equipment plus fantastic computational jiggery pokery and super multicoloured images is impressive (anyone seen a bright blue brain recently?) but the interpretation, and the need for careful caution in seeking to assign any causality on the basis of correlational findings raises all sorts of ethical issues.
Another worry is the fact that anomalies are 'often' found in research scans (figures vary)- a. how are we to respond and b. all sorts of claims could be made for such 'abnormalities' in the context described.
Very interesting and thought provoking piece, as ever.

Carl Senior said...

I have to be honest but the application of brain imaging to judicial processes is very scary and deserves a lot of debate. Together with colleagues I have tried to detail the (very significant) limitations at the heart of this technology with should render their application moot in this context see e.g., http://eprints.aston.ac.uk/9892/. The McCabe and Weisberg studies are quite important as they do provide some evidence to suggest that this technology has a persuasive influence. However we even found that the chance to take part in a scan will see participants enduring a procedure that for some people can be considered quite uncomfortable and stressful see e.g., http://eprints.aston.ac.uk/4828/ also see http://eprints.aston.ac.uk/4835/ so its not just the ‘pretty images’ that has a persuasive effect but the technology itself. This is something that am very happy to note that the society has recognised and is considered in the revised code of ethics – all the best

Great article btw !

Joe said...

Not that surprising when you think that pictures of brain images can make news articles seem more convincing and along with the suggestion that pictures & reconstructions are more convincing than spoken testimony for jurors. Interesting though.

Post a Comment

Note: only a member of this blog may post a comment.