The Predictive Power of Neuroimaging

By Ethan Morris

This post was written as part of a class assignment from students who took a neuroethics course with Dr. Rommelfanger in Paris of Summer 2016. 

Ethan Morris is an undergraduate senior at Emory University, majoring in Neuroscience and Behavioral Biology with a minor in History. Ethan is a member of the Dilks Lab at Emory and is a legislator on the Emory University Student Government Association. Ethan is from Denver, Colorado and loves to ski.   

Background and Current Research

Neuroscience is a rapidly burgeoning field that is increasingly facing complex issues as scientists learn more about the human brain and by extension, about personal identity. One technology that has gained attention in the last two decades is brain imaging, a technique that uses various tools to evaluate the brain’s functional response to the world. Some of the more commonly used brain imaging devices are functional magnetic resonance imaging (fMRI) and positron emission tomography (PET), both of which measure blood flow (albeit by different mechanisms) through the brain. These blood flow results show which areas of the brain are metabolically active, and are thus activated by the task at hand. Using these devices, researchers can determine the activity of certain brain regions associated with certain types of sensory and perceptual processing, as well as cognitive function.

While used in clinical settings for neurological and psychiatric diagnoses, neuroimaging is also applied in a variety of research contexts to learn about the neural correlates of human behavior. One study examined fMRI activation levels in the amygdala, one of the brain’s centers for processing salient stimuli and emotion. The researchers found that white individuals displayed greater amygdala activation for unfamiliar black faces than familiar white faces, and moreover, there was a positive correlation between amygdala activation and unconscious racial bias (Phelps et al., 2000). Importantly, imaging cannot read human minds, but it is significant that brain-imaging patterns are being used currently to make important inferences about unconscious thoughts, even if they are not manifested behaviorally.

Image courtesy of WikiCommons

In another study, researchers found that prisoners with higher levels of psychopathy were more likely to have fewer connections between the parietal cortex and the anterior cingulate cortex (Philippi, 2015). The implication of this study is that it may be possible to identify psychopaths and potentially predict who is more likely to be rearrested based on brain connectivity. In a juvenile study, researchers used fMRI and found that certain patterns of functional connectivity between the premotor and prefrontal cortices were predictive of future impulsivity (Shannon et al., 2011). These studies demonstrate the current capability of neuroimaging to assess unconscious biases and perhaps predict future behavior, such as recidivism or impulsivity.

Ethical Considerations 

In order to inform policy, there are important ethical considerations regarding both current neuroimaging knowledge and future applications of this technology. Provided these studies are replicated and verified, neuroimaging might be used to infer unconscious attitudes and predict future behavior. Is it ethical to image the brains of prisoners to determine their likelihood of ending up back in jail? Even if certain images are correlated with rates of recidivism, it is still difficult to accurately predict future human behavior using neuroimaging. Brain images are transient portraits, which limits researchers’ abilities to extrapolate moment-to-moment brain states to label the brain and person (Fuchs, 2006). Additionally, brain imaging is susceptible to misinterpretation by researchers who do not fully understand the appropriate conclusions one can make with imaging. This could lead to dangerous conclusions from brain images about the entire identity of a person without meaningful evidence. Another limit of neuroimaging is the lack of causational data (e.g. brain activity X causing behavior Y). With neuroimaging, researchers are often only able to correlate brain images with certain functions or mental states (Miller, 2008). Knowing these limits, it does not seem possible right now to definitively predict future behavior. However, placed in eager hands, brain imaging could be used to predict recidivism, which may inevitably result in false positive results, placing prisoners at the mercy of their brain’s activity, perhaps without justification.

On a fundamental level, is it fair to judge a person for what their brain looks like? In the case of the correctional system, this may undermine its purported goal of assisting “offenders in becoming law-abiding citizens” (US Federal Bureau of Prisons). For example, consider a prisoner who appears completely rehabilitated, but whose brain images show a prefrontal cortex deficit associated with impulsivity and future recidivism. Would society deem it fair to place him under stricter parole than it would have without brain imaging? This can be reduced to whether brain images should be accounted for, even if what is observed does not manifest in behavior.

Another ethical concern is society’s widespread belief in free will. It is a commonplace belief that, as a human, one has an intrinsic ability to choose what one will do, no matter the environment or genetics that may predispose certain behaviors. Would society think it is ethical to judge a person for their neurobiology? Some may argue that it would contradict the belief that released prisoners have the ability to avoid committing another crime. Humans value the right to autonomy, or self-determination, so should parole boards meddle in the autonomy of others based on imaging conclusions about their risk for future behavior?

Image courtesy of Pixabay

This ethical issue is particularly pertinent for juvenile offenders. To what degree should the justice system implement brain imaging to predict recidivism or impulsivity if it has been shown the human brain does not finish developing until the mid-20s (Giedd, 1999)? Because studies have shown that adolescents gain white matter and lose impulsivity with age, it may not be ethical to use brain images to predict behavior if they are no longer accurate within a couple years (Casey, 2005). One final ethical consideration of neuroimaging is privacy. There is potential that in the future, scientists may be able to use brain imaging as an identity scanner. Scientists might be able to “’read personality features, psychiatric history, truthfulness and hidden deviations from a brain scan” (Fuchs, 2006). As Fuchs mentions, this application could get misappropriated quickly and invasively, as private companies and lawyers may misuse brain imaging to label and potentially defame people, all based on brain scans. The issue of consent also arises—how does someone lying in an fMRI scanner know what the person behind the operating computer is looking at?

Policy Recommendations 

With these technological limitations and ethical issues in mind, there are multiple policy recommendations to prevent violation of privacy and consent, false positives, and dangerous conclusions. On the issue of consent and privacy, the Department of Health and Human Services (HHS) should ensure that institutional review boards (IRBs) enforce limits on what researchers can image. These limitations should extend into the courtroom, where fMRI could be applied as superior or overriding evidence without sufficient basis. Researchers should only be able to image regions of the brain needed for their research and should be prohibited from using unrelated information that may be outside of participants’ consent/privacy and their research’s purview. Participant consent forms should contain explicit explanation regarding the technology, capabilities, and targeted brain areas so all parties are informed.

Whether through the US Security and Exchanges Commission (SEC) or the US Food and Drug Administration (FDA), the private sector should not have access to brain imaging in its current state. Due to the realistic limitations of imaging, false positives and unwarranted speculation about personal identity are likely to result from unregulated use of brain imaging and should be prevented to avoid personal judgments that may not have any tangible basis. Additionally, the possibility for this research to negatively influence public understanding of neuroscience dictates that powerful tools such as neuroimaging should not be introduced outside of research settings until the tool’s capabilities and limitations are fully understood.

The US Department of Justice should outlaw use of brain imaging in youth detention centers to avoid rampant false positive predictions, given the current knowledge about how decision-making improves with brain development. In addition, parole boards should not be allowed to use imaging to determine the chances an adult prisoner will commit another crime. If the justice system collectively decides brain images are paramount to demonstrated human behavior, brain imaging could theoretically antiquate and undermine the justice system’s efforts to improve actual human behavior. This must not be the case—the ultimate goal of prison should be to change behavior, not neurobiology.

 In the realm of research, review boards must vigorously review brain-imaging studies. It is simply too dangerous to publish conjectural conclusions about brain imaging because of the distinct possibility for misappropriation and for sensationalist media stories. The caveats and limitations of brain imaging (e.g. the lack of causational data) must be emphasized at the front line—the researchers—in an attempt to prevent false positives, and to prevent non-researchers from misapplying neuroimaging. Researchers and journals must be vigilant about disseminating correct results with appropriately stipulated interpretations to prevent misreporting. Furthermore, researchers must be held accountable by IRBs for their studies in an effort to prevent publications of flimsy associative data that may be misinterpreted. Additionally, there should be restrictions on researchers’ conflicts of interest regarding neuroimaging; for example, the US Department of Justice should be barred from funding neuroimaging in an effort to predict post-prison behavior until the technology proves reliable and causational.

Neuroimaging should continue to be used appropriately and realistically will be used in both research and clinical contexts going forward. Because of this, the HHS should implement a public education strategy to inform media members, lawyers and parole board members, as well as the public of the capabilities and limitations of brain imaging to prevent avoidable ethical problems and public fear of neuroimaging.

References 



Casey, B.J., A. Galvan, T.A. Hare. 2005. Changes in cerebral functional organization during cognitive development. Current Opinion in Neurobiology, 15: 239-244.



Federal Bureau of Prisons. About our agency: A foundation built on solid ground. Available here.



Fuchs, T. 2006. Ethical issues in neuroscience. Current Opinion in Psychiatry, 19: 600-607.



Giedd, J.N., J. Blumenthal, N.O. Jeffries, et al. 1999. Brain development during childhood and adolescence: a longitudinal MRI study. Nature Neuroscience, 2(10): 861-863.



Miller, G. 2008. Growing pains for fMRI. Science, 320(5882): 1412-1414.



Phelps, E.A., K.J. O’Connor, W.A. Cunningham, et al. 2000. Performance on indirect measures of race evaluation predicts amygdala activation. Journal of Cognitive Neuroscience, 12(5): 729-738. 



Philippi, C.L., M.S. Pujara, J.C. Motzkin, J. Newman, K.A. Kiehl, M. Koenigs. 2015. Altered resting-state functional connectivity in cortical networks in psychopathy. The Journal of Neuroscience, 35(15): 6068-6078.



Racine, E., O. Bar-Ilan, J. Illes. 2005. fMRI in the public eye. Nature Reviews Neuroscience, 6(2): 159-164. Shannon, B.J., M.E. Raichle, A.Z. Snyder, et al. 2011.



Premotor functional connectivity predicts impulsivity in juvenile offenders. PNAS, 108(27): 11241-11245. University of Washington. Brain imaging. Available here.

Want to cite this post?



Morris, E. (2016). The Predictive Power of Neuroimaging. The Neuroethics Blog. Retrieved on , from http://www.theneuroethicsblog.com/2016/10/the-predictive-power-of-neuroimaging.html

Neuroimaging in Predicting and Detecting Neurodegenerative Diseases and Mental Disorders

By Anayelly Medina

This post was written as part of a class assignment from students who took a neuroethics course with Dr. Rommelfanger in Paris of Summer 2016.



Anayelly is a Senior at Emory University majoring in Neuroscience and Behavioral Biology. 

If your doctor told you they could determine whether or not you would develop a neurodegenerative disease or mental disorder in the future through a brain scan, would you undergo the process? Detecting the predisposition to or possible development of disorders or diseases not only in adults but also in fetuses through genetic testing (i.e. preimplantation genetics) has been a topic of continued discussion and debate [2]. Furthermore, questions regarding the ethical implications of predictive genetic testing have been addressed by many over the past years [4,8]. However, more recently, neuroimaging and its possible use in detecting predispositions to neurodegenerative diseases as well as mental disorders has come to light. The ethical questions raised by the use of predictive neuroimaging technologies are similar to those posed by predictive genetic testing; nevertheless, given that the brain is the main structure analyzed and affected by these neurodegenerative and mental disorders, different questions (from those posed by predictive genetic testing) have also surfaced.

Computerized Axial Tomography (CAT), Positron Emission Tomography (PET) and radioactive tracers, Magnetic Resonance Imaging (MRI), and Functional Magnetic Resonance Imaging (fMRI) are all current neuroimaging technologies used in the field of neuroscience. While each of these technologies function differently, they ultimately all provide information on brain functioning or structure. Furthermore, these neuroscientific instruments have, in recent years, been used to explore the brain in order to determine predictive markers for neurodegenerative diseases and mental disorders, such as Parkinson’s disease, Schizophrenia, Huntington’s disease, and Alzheimer’s disease [1,9,11,12]. For example, Stoessl [11] explains how PET scans and radiotracers have shown evidence of abnormal dopamine dysfunction (a pathway known to be compromised in PD) in asymptomatic individuals from families with known inherited PD (although whether this dysfunction is an early measure of those who will develop PD or if it is associated with the inherited PD gene is unclear). In addition, Callicott et al. [1] provided evidence, through fMRI scans, that showed greater response in the right dorsolateral prefrontal cortex in cognitively intact siblings of patients with schizophrenia (this abnormal response was similar to that in patients with schizophrenia). Furthermore, Paulsen et al [9] used fMRI scans to show that striatal and white matter volumes in the brain could be used to predict diagnosis proximity (estimated years to diagnosis) of Huntington’s disease. Finally, the use of neuroimaging techniques to establish predictive markers of disease and mental disorders has been clearly seen in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) that was started in 2007 and is currently active [12]. The potential ability to predict whether or not an individual will develop a neurodegenerative disease or mental disorders seems like an initiative without faults. However, there are questions surrounding the ethics of such an ability that must be addressed.

Image courtesy of Flikr

The use of neuroimaging data to predict neurodegenerative diseases and mental disorders is an initiative that should continue to be pursued as it could help in the prevention or delay the disease or disorder by early intervention, but that research should also take into account the ethical implications of conducting and providing such information for and to the public. Some of the main ethical issues that have developed with the increased use of predictive neuroimaging include concerns surrounding intervention, privacy, and access. In terms of intervention, the main concern involves determining when to notify the patient—this would require having an established degree of probability, as well as prevalence of false positives, that would count as being sufficient to warrant patient knowledge of the neurodegenerative disease or mental disorder [3]. This would be further complicated when assessing brains of younger individuals given that their brains are still undergoing developmental changes, and the reliability of prognoses at such an early age has yet to be assessed. In addition to timing and accuracy, other issues involve the use of neuroimaging to predict diseases or mental disorders that have no cures or treatment as well as taking into account the impact that providing said information could have on the patient (such as the burden of knowledge [3] or the impact of stigma). Furthermore, the diseases and mental disorders that are being predicted using neuroimaging all affect the brain and its function, and thus possibly “also affect mental competence, mood, personality, and sense of self” [10]. In addition to intervention, privacy and discrimination are other issues at play as employers or insurers could determine whether or not a person is hired or what type of healthcare policy an individual receives based on predisposition/predictive neuroimaging tests [7]. Finally, the ethical concerns surrounding the access to neuroimaging technology must also be addressed. Neuroimaging scans are typically expensive, and its use in predicting the development of diseases and disorders may lead to more healthcare disparities; this could become a greater problem if the technology were to become commercialized and only accessible to those who are privileged [6;7].

In order to address these ethical issues, changes must be implemented at various stages of this predictive neuroimaging technology’s implementation. In addressing the issue of timing and accuracy in intervention, more studies looking at the correlation between various brain structures or functions and predisposition/prediction of neurodegenerative diseases and mental disorders must be conducted; specifically, studies in which the complementary approach of testing for genetic predisposition is controlled for would provide more conclusive and valid data supporting the associations made between brain scan findings and prediction of disease or mental disorder. Furthermore, more initiatives, like that of the Alzheimer’s Disease Neuroimaging Initiative (ADNI), should be created for other neurodegenerative diseases and mental disorders (with possible funding from the NIH, through the Functional Magnetic Resonance Imaging Core Facility (FMRIF)). Organizations like the ADNI hope to create a network of shared data, concerning biomarkers in the brain, in order to facilitate early detection of disease [12]. Moreover, initiatives like this could further facilitate the development and establishment of methods and protocols for predicting the onset of disease. In addition, regulation of the neuroimaging devices in predictive neuroimaging testing must also be implemented; as explained by Greely [5] the Food and Drug Administration (FDA) typically has jurisdiction over the use of drugs, biologicals, and medical devices, but if a test involves using a device (in this case a neuroimaging device) that has already been approved in the past, the FDA would not need to approve its use in a new test. Thus, an appeal to the FDA in order to reevaluate this decision should be pursued in order to establish safe and effective use of the technology. Furthermore, in order to protect the privacy of the patient, as well as protect against discrimination or unfair actions against the patient through the use of predictive neuroimaging data by insurers or employers (in terms of predictive data of neurocognitive disease or mental disorder development) protocols and regulations should be established by the U.S. Department of Human Health Services (HHS) and the U.S. Equal Employment Opportunity Commission. Finally, involvement of the U.S. DHHS in making this predictive neuroimaging accessible to those who cannot afford these services should also be established.

 References 



 1. Callicott, J. H., Egan, M. F., Mattay, V. S., Bertolino, A., Bone, A. D., Verchinksi, B.,

&Weinberger, D. R. 2003. Abnormal fMRI Response of the Dorsolateral Prefrontal Cortex in Cognitively Intact Siblings of Patients With Schizophrenia. American Journal of Psychiatry AJP 160(4): 709-719. doi:10.1176/appi.ajp.160.4.709.



 2. Farrimond, H. R., & Kelly, S. E. 2011. Public Viewpoints on New Non-invasive Prenatal Genetic Tests. Public Understanding of Science, 22(6): 730-744. doi:10.1177/0963662511424359



 3. Fuchs, T. 2006. Ethical Issues in Neuroscience. Current Opinion in Psychiatry 19(6): 600-607. doi:10.1097/01.yco.0000245752.75879.26.



 4. Fulda, K. G. 2006. Ethical Issues in Predictive Genetic Testing: A Public Health Perspective. Journal of Medical Ethics 32(3): 143-147. doi:10.1136/jme.2004.010272.



 5. Greely, H. 2004. Markula Center for Applied Ethics. The Neuroscience Revolution, Ethics, and the Law. Available here. (accessed June 19, 2016).



 6. Illes, J., & Racine, E. 2005. Imaging or Imagining? A Neuroethics Challenge Informed by Genetics. The American Journal of Bioethics 5(2): 5-18. doi:10.1080/15265160590923358 .



 7. Illes, J., Rosen, A., Greicius, M., & Racine, E., 2012. Ethics Analysis of Neuroimaging in Alzheimer’s Disease. Annals of the New York Academy of Sciences 1097: 278-295. doi:10.1196/annals.1379.030.



 8. Leah, DH., Williams J., Donahue MP., 2005. Ethical Issues in Genetic Testing. Journal of Midwifery & Women's Health 50(3): 234-240. doi:10.1016/j.jmwh.2004.12.016.



 9. Paulsen, J. S., Nopoulos, P. C., Aylward, E., Ross, C. A., Johnson, H., Magnotta, V. A., . . . Nance, M. 2010. Striatal and White Matter Predictors of Estimated Diagnosis for Huntington Disease. Brain Research Bulletin 82(3-4): 201-207. doi:10.1016/j.brainresbull.2010.04.003.



 10. Roskies, A. 2016. Neuroethics. The Stanford Encyclopedia of Philosophy. Available here. (accessed June 19, 2016).



 11. Stoessl, A. J. 2012. Neuroimaging in the Early Diagnosis of Neurodegenerative Disease. Translational Neurodegeneration 1(1), 5. doi:10.1186/2047-9158-1-5.



 12. Weiner, M. W., Veitch, D. P., Aisen, P. S., Beckett, L. A., Cairns, N. J., Cedarbaum, J., . . . Trojanowski, J. Q. 2015. 2014 Update of the Alzheimer's Disease Neuroimaging Initiative: A Review of Papers Published Since its Inception. Alzheimer's & Dementia 11(6). doi:10.1016/j.jalz.2014.11.001.

Want to cite this post?



Medina, A. (2016). Neuroimaging in Predicting and Detecting Neurodegenerative Diseases and Mental Disorders. The Neuroethics Blog. Retrieved on , from http://www.theneuroethicsblog.com/2016/10/neuroimaging-in-predicting-and.html