After three months trapped in a cell no bigger than a parking space, William White lost the ability to speak, then eat. He lived under perpetual fluorescent light, unable to interact with others, see the sun, shave, or exercise. He soon fell into a dissociative state and collapsed. Prison authorities moved White out of solitary confinement, but his suffering persisted in filthy conditions under violent guards. At one point, White “became nearly completely unresponsive, cut his stomach, and began writing on the walls with blood.” 1 A return to normal life post-incarceration was impossible. White described suffering from visions during the day and insomnia, delirium, and “waking nightmares” at night. He claimed that his newly unstable mental state ended his marriage and forced him to close his business. When White was reincarcerated a few years later, a clinical psychologist diagnosed him with post-traumatic stress disorder (PTSD) and “specifically related the … diagnosis to trauma [White] had [earlier] experienced while incarcerated.” When White sought restitution for his treatment in 2024, however, the 10th Circuit Court of Appeals unanimously dismissed him, writing that he suffered “[no] more than de minimis physical injury.” 2
White’s predicament is sadly common. United States law classifies PTSD as a purely emotional injury and thus less worthy of compensation than a material harm to the body. Because PTSD and other psychological harms are not clearly visible to observers and resist quantifiable measurement, courts consider them less substantial or even self-conjured.
New technologies now challenge these assumptions, as brain scans reveal the structural and functional brain changes associated with PTSD. We can now consistently see damage in the cortex of victims with millimeter resolution.Reference Goense 3 In this paper, I argue that foundational neuroimaging research should therefore change PTSD’s classification from not just an emotional but also a physical injury, making it easier for plaintiffs to win compensation.
While there is growing support to erode the legal barrier between physical and emotional tort damages, little scholarship has yet been dedicated to the particular case of PTSD. Its distinct neurological markers and prevalence among diverse segments of society make it a promising beachhead for the movement. Establishing PTSD as a physical injury could set a precedent for other conditions like chronic pain, bringing tort law further in line with neuroscience advances.
In Part I, I offer background on tort law and its varying treatment of emotional and physical harm claims. In Part II, I describe how neuroimaging works, what it measures about PTSD, and how it is beginning to change legal perspectives on the disorder. In Part III, I highlight the importance of neuroimaging for marginalized groups, who not only are at a higher risk of PTSD, but also have traditionally faced greater skepticism and indifference to their claims of emotional harm. In Part IV, I assess the technology’s current limitations and make the case that aggregate, but not personalized, neuroimaging should pass state and federal evidentiary standards. Finally, in Part V, I offer recommendations for how neuroimaging should be regulated and incorporated into tort jurisprudence.
Psychiatric illness has long been overlooked and underappreciated by society. “It’s all in your mind,” the cliched dismissal goes. But the mind is no longer such a mysterious, inaccessible place, and PTSD is no longer an invisible injury. If the law ascribes to the principle that seeing is believing, then shouldn’t the suffering of White and other victims now garner more consideration?
I. The Problem of Invisible Injuries
Emotional and physical injuries are primarily redressed through tort law. To understand the unique challenges that PTSD victims face in winning compensation, let us first consider the underlying logic of civil liability.
A. Why Tort Law Disfavors Emotional Injury
Tort law serves two distinct purposes that inform a court’s decisions: corrective justice for the individual and economic deterrence for society.Reference Grey 4 Corrective justice aims to remedy the harm of one actor, often through financial compensation to the victim that is proportional to both the extent of their suffering and the severity of the wrong.Reference Grey 5 Economic deterrence uses the victim and tortfeasor as a means to reinforce normative behaviors in the general public. The awarded compensation reflects a tortious act’s cost to society, and it aims to disincentivize potential offenders by forcing them to internalize the risk of harming others. 6 In both models, a jury must estimate the severity of harm inflicted. Legal precedent in the United States distinguishes emotional from physical harms, with the latter proving far easier to recover from in court.
This bifurcation of mind and body, known as substance dualism, derives primarily from the challenge of proving the existence and severity of emotional harms. A plaintiff with broken ribs can show x-rays of the snapped bone to substantiate a physical injury, but a plaintiff claiming emotional injury has traditionally relied on self-report — perhaps bolstered by psychological examination. Courts subject emotional harm to greater scrutiny because they feel limited in objectively quantifying its scope and would prefer to focus on less suppositional physical harms. 7
Accompanying the evidentiary challenges is the risk of malingering litigants who falsify or exaggerate emotional injuries to bolster their damage awards. Although the Restatement (Second) of Torts acknowledges that “the mere fact that a person can offer no physical evidence of pain … does not mean that his reported symptoms are disingenuous,” courts nevertheless can deny recovery “because of the fear of fictitious or trivial claims, [and] distrust of the proof offered.”Reference Cassin 8
Finally, substance dualism reflects an underlying belief that psychological harms are intrinsically less severe than physical ones. As the Supreme Court ruled in the asbestos litigation case Metro-North Commuter Railroad Co. v. Buckley, it would be unethical to “[permit] immediate large-scale recoveries for widespread emotional distress caused by fear of future disease” if that “[diminishes] the likelihood of recovery by those who later suffer from the disease.” 9
B. How Tort Law Treats Emotional and Physical Injury
Now that we understand the traditional reasons for favoring physical over emotional injury claims, we can explore how this preference manifests in legal doctrine.
1. Physical Injury
A plaintiff seeking compensation for physical harm must still prove the existence of an injury and the defendant’s culpability. But they benefit from the eggshell skull rule, first established in Vosburg v. Putney, which holds that a wrongdoer is liable for all physical injuries resulting from their offense, regardless of whether an average person would suffer to the same degree. 10 For example, if you trip an old woman with osteoporosis and she fractures her hip, the victim’s low bone density does not mitigate your responsibility. The burden of a plaintiff’s ailing body rests on the defendant. But as we shall see, the burden of a plaintiff’s ailing mind rests largely on the plaintiff.
2. Intentional Infliction of Emotional Distress
According to the legal scholar Betsy Grey, emotional injury began as a “parasitic harm to personal injury or property damage claims.” If emotional harm was substantiated, a victim recovered damages “as attached to some other tort.” 11 Only in the mid-20th century did emotional injury begin to gain recognition as a stand-alone claim with tracks for intentional or negligent infliction of emotional distress (subsequently referred to as IIED and NIED, respectively). According to the Restatement (Third) of Torts, if “an actor, who by extreme and outrageous conduct, intentionally or recklessly causes severe emotional harm to another,” they are liable for this harm under IIED. 12 A plaintiff must not only show that the actor purposefully caused the harm, but that their conduct went “beyond the bounds of human decency such that it would be regarded as intolerable in a civilized community.” 13 The emotional distress is typically categorized as severe only if “a person of ordinary sensitivities in the same circumstances” could not bear it. 14 This is a challenging standard for plaintiffs to prove. If you trip an old woman with a phobia of falling, and she develops PTSD but no concomitant physical injuries, you will likely not be liable. She must internalize the burden of her frail psyche and demonstrate that a woman of similar age without her anxiety disorder would suffer equally debilitating emotional trauma. 15
3. Negligent Infliction of Emotional Distress
NIED claims face an even tougher evidentiary onus. In addition to the reasonableness test just mentioned, many states require plaintiffs to show they were in immediate risk of physical harm (a “zone of danger”) that triggered their emotional distress. 16 This restriction tethers NIED recovery to physical injury, and thus implicitly reinforces the pre-eminence of the latter in tort jurisprudence. Normative concerns of unchecked liability exposure for NIED are assuaged by proof that victims almost suffered physical harm. As one commentator argued, “the coercive machinery of the state should only be used to prohibit especially negligent inflictions of distress,” and the previously discussed provisions are convenient heuristics for severity.Reference Kolber 17
4. PTSD as Emotional Injury
Efficiency, however, comes at the cost of accuracy. The IIED and NIED standards disadvantage plaintiffs with largely invisible injuries such as PTSD, who may not have consistent physical symptoms and yet still suffer enormously. Before delving further into the law’s treatment of PTSD, it is important to contextualize the causes, symptoms, and demographic risk factors for the disorder.
The DSM-5Reference Halpin 18 lists “exposure to actual or threatened death, serious injury, or sexual violence” as potential inducers of PTSD, which can manifest in a wide range of behaviors across four categories: intrusion symptoms, avoidance of negative stimuli, cognitive deterioration, and dysregulated arousal. 19 Patients often experience sleep disturbances, hypervigilance, and involuntary flashbacks. They might struggle to remember key details about the traumatic event due to dissociative amnesia, become estranged from friends and family, and engage in reckless behaviors and negative self-narratives. To receive an official PTSD diagnosis, the DSM-5 requires that these symptoms “persist for more than one month … interfere with regular functioning, and [cannot] be attributable to medication, substance use, or other illness.” 20 It’s estimated that approximately seven percent of people in the United States will develop PTSD at some point in their lives. 21 The condition is often chronic, debilitating, and financially ruinous — up to forty percent of patients are unemployed.Reference Sweeton 22 Women are roughly twice as likely to develop PTSD than men.Reference Hinojosa 23 Black Americans have the highest lifetime occurrence of PTSD among racial groups, and all minorities are less likely to seek out or complete psychotherapy treatment than White Americans, due in part to limited mental health resources in areas of poverty.Reference Roberts 24 Rates of PTSD among veterans and the general prison population typically range from ten to thirty percent. 25 Those formerly held in solitary confinement exhibit even greater prevalence.Reference Hagan 26
Despite the alterations in sleep and behavior that PTSD can engender, most courts do not consider the disorder a bodily injury. 27 For example, the Kentucky Court of Appeals ruled in the 1920s that a child who witnessed a trolley car kill her father was not eligible for emotional damages. Although PTSD did not become an official diagnosis until 1980, four physicians diagnosed her with its precursor — “traumatic neurosis” — and testified that the incident likely “brought about the condition from which she is now suffering.” Nevertheless, the court ruled that “fright caused by negligence is not of itself a cause of action.” 28 The trolley did not strike the daughter, so she was not eligible for recompense. Such stringent barriers have continued to thwart plaintiffs with PTSD over the last century.
C. The Challenges of Proving One’s Suffering
The challenges don’t stop there. If a jury does find a victim credible, they must then determine a compensating amount for the offender to pay. This is an imperfect art at best and an arbitrary guess at worst. In both physical and emotional tort cases, the price accounts for “pain and suffering” — a broad category of noneconomic damages encompassing harms such as bodily discomfort, anxiety, insomnia, and grief.Reference Geistfeld 29 There is no objective way to determine the severity of pain and suffering damages, nor to translate them into appropriate monetary awards. Pain is personal after all — difficult to describe and for others to comprehend.
As a result, courts leave juries with vague guidance. The Seventh Circuit Court of Appeals admits that “there is no exact standard for setting the damages to be awarded on account of pain and suffering. You are to determine an amount that will fairly compensate Plaintiff for the injury he has sustained.” 30 Indeed, “the only real measuring stick, if it can be so described, is your collective enlightened conscience.”Reference Eades 31
The frameworks that juries typically rely on to fill this void of instruction are severely flawed. For instance, in the “per diem” method, jurors determine the cost of pain and suffering damages for a day, and then multiply that amount by the number of days a plaintiff was injured and will likely continue to be. 32 But the first calculation is still subjective, and the quantitative second step merely engenders a false confidence in the final value. To make things more difficult, juries are not told of past decisions in similar cases.Reference Avraham 33 While this prevents biasing their analysis of a plaintiff’s unique circumstances, it also leads to variation in awards for comparable injuries. 34 Pain and suffering compensation is therefore highly inconsistent as well as highly subjective. This is true regardless of whether an emotional injury is attached to a physical one.
Marginalized groups experience the greatest repercussions from these problems. They not only tend to underreport their pain, but also face “thick skin bias” — the misconception that those with disadvantaged lives suffer less from negative experiences.Reference Cheek and Shafir 35 Studies show that as legal standards become more ambiguous, jurors weigh thick skin bias and other extralegal factors (such as race, gender, and physical appearance) more heavily in their decision making.Reference Kaplan and Miller 36 Thus, even if a minority plaintiff with PTSD substantiates an emotional injury claim, they might not obtain fair restitution.
Neuroimaging has the potential to remedy this. The technologies described in the next section illustrate PTSD’s physical damage to the brain, making it easier to quantify the severity of patients’ symptoms. This empirical data can reduce the influence of jurors’ prejudices (conscious or not) about the plaintiff’s demographics and the legitimacy of psychological distress. The effects of PTSD on brain structure and function suggest that substance dualism is an antediluvian distinction. As one commentator phrased it, “physical experiences have mental and emotional components; mental and emotional conditions have physical correlates.” 37 The two are inextricably linked. As invisible injuries become visible, PTSD victims should no longer face inordinate scrutiny and receive insufficient redress for their condition.
II. Neuroscience and PTSD Tort Law
A. How Neuroimaging Works
Neuroscience has percolated through a wide range of legal fields in the last several years, from criminal sentencing to juvenile justice and tort law. Defendants have used brain scans with varying success to ask for sentence reductions, prove mental incompetence, or suggest a lack of mens rea.Reference Greely and Farahany 38 In the case Graham v. Florida, the Supreme Court banned minors from life imprisonment without parole for non-homicidal crimes and cited research on adolescent prefrontal cortex development. 39 In the later case Brown v. Entertainment Merchants Association, the court considered neuroscience findings that violent video games induce antisocial behaviors. 40 Given the significant intersections of law and neuroscience, it is surprising that until the late 20th century, we had few techniques to map the brain’s composition, let alone visualize its activity in real time. Now there are dozens.
The primary methods to ascertain brain structure are magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI). With the former, a cylindrical MRI machine creates a magnetic field around a patient that aligns the atoms in their body. 41 Radio waves emitted by the machine disrupt the atoms’ positions. When the signal pauses, the atoms “send back” radio waves that the MRI can compile into a detailed image of tissue. DTI uses MRI to measure specifically the structural composition of white matter tracts in the brain. 42 White matter is brain tissue coated with myelin — an insulating substance that improves the conduction of electrical signals. By measuring the diffusion of water molecules along white matter fibers, DTI creates detailed maps of the brain’s connections.
Functional magnetic resonance imaging (fMRI) builds on MRI technology to measure the brain’s response to stimuli. An active brain region requires a greater blood supply to support its metabolic needs. Once a hemoglobin molecule releases oxygen in this area, it slightly disrupts the induced magnetic field. This creates a blood oxygen level-dependent (BOLD) signal, which an fMRI machine detects to generate a composite map of blood flow in the brain.Reference Pustilnik 43 A positron emission tomography (PET) scan also uses blood flow to indirectly measure neural activity, although by a distinct mechanism. The machine tracks the movement of an injected radioactive compound through vasculature. 44 As a brain region activates, the tracer moves towards it. fMRI has largely supplanted PET due to its higher resolution and lower costs, although PET retains useful medical applications in tumor imaging. 45 Single-photon emission computed tomography (SPECT) uses a different radioactive tracer to PET that emits gamma radiation instead of positrons. 46 This measurement of brain metabolism is cheaper and more accessible, however it sacrifices some spatial resolution. Electroencephalography (EEG) and magnetoencephalography (MEG) are other popular techniques. EEG electrodes on a patient’s scalp detect the electrical activity of synchronously firing neurons, while MEG measures their corresponding magnetic field.Reference Yen 47 Both have exquisite time sensitivity (detecting unique brain waves separated by milliseconds), but less positional accuracy than fMRI. 48
Research on PTSD’s neurological correlates has proliferated over the last few decades as these technologies have grown more precise and accessible. While emotional suffering is intrinsically subjective, brain scans now quantify its harms in unprecedented ways.
B. Neuroimaging Detects the Physicality of PTSD
So what exactly has neuroimaging elucidated about PTSD’s mark on the brain? One of the most consistent findings is overactivation of the amygdala.Reference Koenigs and Grafman 49 This subcortical region normally triggers our fight or flight response to acute trauma, but overactivation likely contributes to the chronic anxiety, flashbacks, and intrusive memories characteristic of PTSD. 50 This effect is compounded by reduced gray matter volume in the prefrontal cortex and anterior cingulate cortex (ACC), which play essential roles in executive functioning and emotional regulation. 51 In other words, if your stress response is a race car, then your amygdala is the gas and your prefrontal cortex is the brake. PTSD floors the former and short circuits the latter.
Other characteristics of the disorder include decreased bilateral volume and underactivation of the hippocampus, which can impair sleep, spatial navigation, and learning.Reference Kitayama 52 MRI and DTI studies also find reduced white matter integrity of midbrain structures like the corpus callosum and uncinate fasciculus — both of which are essential for signal transduction to the emotional regulatory center known as the limbic system. 53
These collective changes form a brain fingerprint that can identify victims of PTSD.
In one blinded study, a lab used MEG to analyze the brains of 74 PTSD-affected war veterans and 250 healthy controls.Reference Bottalico and Bruni 54 The researchers’ positive predictive value of PTSD subjects was 97.3%, and their negative predictive value of control subjects was 87.6%, yielding an overall classification accuracy of 92.4% (p value < 0.001). 55 MEG signal intensity (relative to the healthy controls) also corresponded with the severity of the veterans’ PTSD symptoms, suggesting that an even finer level of detection is possible. 56 Brain scans could potentially determine if victims suffer from PTSD and by how much. Scientists have replicated these findings with other neuroimaging tools, albeit with varying precision. For instance, a machine-learning analysis of MRI scans obtained from 51 individuals with PTSD and 76 trauma-exposed healthy controls distinguished them with 70.6% accuracy.Reference Zilcha-Mano 57 Control subjects had stronger connectivity within the Executive Control Network (ECN) — composed of cell clusters in the prefrontal cortex. There was a strong negative correlation between ECN connectivity and PTSD symptom severity (p < 0.001).Reference Shin 58
PTSD’s neural biomarkers not only distinguish victims from healthy controls, but also from those with associated neurological disorders. For example, as many as 52% of PTSD cases are comorbid with major depressive disorder (MDD). 59 An MRI analysis correctly identified 52 subjects with PTSD and MDD from 51 subjects with PTSD alone with 76.7% accuracy. 60 The comorbid group showed more pronounced abnormalities in brain regions associated with the reward system, such as the basal ganglia network, which aligns with the anhedonic symptoms of MDD. There is also significant diagnostic overlap between PTSD and traumatic brain injury (TBI), which results from blunt force trauma to the head.Reference Amen 61 There have been over half a million diagnosed cases of TBI in military personnel since 2000, 62 and between 33% and 42% additionally suffer from PTSD. The disorders share many symptoms such as anxiety, impulsivity, memory impairment, and insomnia. Given the unique treatments for TBI and PTSD, there is an important diagnostic need to reliably distinguish them, as well as identify cases of co-occurrence. 63 A 2015 SPECT study of 104 subjects with PTSD, 104 with TBI, 73 with both conditions, and 116 controls (with no neurological disorders) was able to distinguish the groups with 100% accuracy from resting-state and on-task scans. 64 Taken together, these studies suggest that PTSD induces consistent changes in brain physiology that are distinct from other psychological disorders sharing behavioral symptoms.
These data are promising, but the current research does have some constraints. Small sample sizes of most fMRI studies undermine confidence in the brain patterns they find. Scientists may claim statistical significance, but they cannot fully account for wide variations in brain physiology with just a few subjects. The problem stems in part from expensive equipment and a dearth of funding. As the legal scholar Henry Greely observes, “one hour-long scan of one subject, when overhead and analysis costs are included, will often cost a researcher more than $1,000.” And while “genetics projects can get multi-million dollar grants; fMRI research has to make do with much smaller budgets, and hence, much smaller numbers.”Reference Greely 65 Studies aren’t just undersized; they also contradict each other frequently. For instance, there is still disagreement over whether PTSD increases, decreases, or does not alter connectivity between the amygdala and Anterior Cingulate Cortex, as well as the functions of specific nodes in the prefrontal cortex and hippocampus. 66
As we shall consider in more detail in Part IV, these scientific uncertainties are one of many barriers to using brain imaging on individual plaintiffs. There is significant heterogeneity in neurological responses to PTSD, and we don’t yet have an adequate grasp of these nuances to reliably diagnose or characterize the disorder from one person’s scans. But aggregate research does indicate how the disorder affects brains on average. We know what circuits regulate fear and memory, and we can measure their atrophy or hyperactivity in research subjects. As the next sections will discuss, these scientific revelations are beginning to undermine the traditional classification of PTSD as a solely emotional injury in tort law.
C. Old and New Perspectives on PTSD Injury
State and federal courts at all levels have long upheld the interpretation that bodily injury must primarily be “corporeal” in nature. 67 If physical harm is caused by emotional harm, it does not count. For example, in Terrafranca v. Virgin Atlantic Airways Ltd, a plaintiff argued that a mid-flight bomb threat triggered PTSD, which in turn gave her anorexia. 68 The court ruled that her eating disorder was a “mere manifestation of fear or anxiety,” and not the “direct, concrete bodily injury” required for recovery. 69 Virgin Atlantic would be liable for her weight loss if she caught a GI infection from its food. But since her condition stemmed from mental anguish rather than acute physical harm, it was exempt.
Courts typically treat PTSD as bodily injury only when caused directly by physical harm. Even the co-occurrence of the two injuries is not enough. The Eighth Circuit ruled in one 2002 case that a plaintiff’s minor cuts and bruises from a plane crash were unrelated to her PTSD, and therefore could not trigger recovery for the disorder as a physical injury. 70 “Damages for mental injury must proximately flow from physical injuries caused by the accident,” the court insisted. 71
Such intricate cause-and-effect standards have thwarted many other PTSD victims, but research now substantiates that PTSD’s psychological symptoms do arise, at least in part, from physical damage to neuroanatomy. In the novel case Allen v. Bloomfield Hills School District, the plaintiff introduced PET scans that, according to expert testimony, showed pronounced “decreases in frontal and subcortical activity consistent with [PTSD].” 72 The trial court ruled that Allen did not suffer bodily injury. The Michigan Court of Appeals reversed the decision given that he showed “objective medical evidence that a mental or emotional trauma can indeed result [from] physical changes to the brain.” Since the brain is an organ of the body, “brain injury is a bodily injury.” 73
So far, no other plaintiff has successfully used their own brain scans to substantiate physical injury from PTSD. Apart from the scientific limitations previously mentioned, many plaintiffs also face barriers to access or affording the technology. But even without brain scans, the substantial neuroscience research on PTSD may still educate a jury and shape a case’s outcome. For example, in Pekin Insurance Co. v. Hugh, the Supreme Court of Iowa ruled that a plaintiff’s PTSD was bodily injury because “every emotional disturbance has a physical aspect.” 74 The robust research on the brain changes underlying PTSD’s behavioral effects made the distinction between “physical [and] psychological…artificial and arbitrary.” 75 In the following section, we will explore how similar contextualizing evidence may be especially beneficial for three populations: victims of domestic abuse, racial trauma, and solitary confinement. By expanding PTSD’s general classification to physical injury from the confines of IIED and NIED, aggregate neuroimaging research gives these plaintiffs a better chance of winning compensation.
III. The Benefits of PTSD Neuroimaging for Marginalized Populations
Anyone can suffer trauma. But the quantifiable evidence of neuroimaging is especially important for plaintiffs from marginalized groups, whose PTSD is more likely to be unacknowledged or underappreciated due to its ubiquity.
A. Domestic Abuse
One in four American women suffer from intimate partner violence (IPV), and the risk is even greater for women of color, those who identify as LGBTQ+, or who live below the poverty line.Reference Lipsky 76 Over half of IPV victims can incur PTSD and lasting neurological damage in addition to the physical trauma of battery.Reference Golding 77 Even if a woman’s PTSD symptoms diminish over time, her cortisol levels can remain elevated, increasing her susceptibility to immune dysfunction, heart disease, and all-cause mortality.Reference Meszaros 78
Neuroimaging illustrates the consequences of such a heightened stress response on the brain. Domestic abuse victims with PTSD have a hyperactive amygdala and reduced white matter integrity in the corpus callosum.Reference Likitlersuang 79 However, they do not show reliable decreases in hippocampal volume or frontal gray matter — an intriguing departure from most PTSD victims. This might be due in part to the injury’s chronic nature. A perpetrator is often a spouse or shares children with a victim, making it challenging to fully sever an abusive relationship. Unlike veterans or holocaust survivors, whose trauma has definitively ended, victims of IPV may consequently participate in a study while their maltreatment is ongoing, or in tentative remission. 80
It is important to emphasize that these studies assessed the average effects of domestic violence over many brains. Current technology cannot reliably detect PTSD in an individual victim of IPV. But given the steady improvements in neuroimaging, the future looks promising. The preliminary findings of IPV’s unique brain fingerprint offer a tantalizing possibility that brain imaging soon may be able to not only substantiate PTSD’s physical effects, but also verify its origin. If researchers are correct that PTSD from combat and domestic violence cause distinct neurological injuries, then a plaintiff might theoretically be able to use their brain scans to support a causal link between a specific trauma and their disorder (although they would likely need additional evidence). 81
In the meantime, aggregate experimental evidence still benefits IPV victims. They can cite studies on the brain’s negative responses to domestic violence, providing contextualizing evidence for juries that consider IPV cases. For example, People v. Torres considered brain imaging studies on battered woman syndrome — a subcategory of PTSD. 82 Expert witness Dr. Julie Blackman testified on the growing scientific consensus that prolonged abuse may engender neurological changes and irrational behavior. In Blackman’s opinion, the defendant Lydia Torres suffered from battered woman syndrome, and it likely contributed to her decision to shoot her abusive husband. While neuroimaging of PTSD in domestic abuse contexts is still in its infancy, it is already beginning to illuminate the consequences of IPV. The jury for People v. Torres learned that the brain may bruise as easily as the body and often takes much longer to heal.
B. Racial Trauma
Some commentators propose that neuroimaging could also benefit tort claims of race-based trauma. IIED was a popular strategy in the wake of the Civil Rights Movement for people of color to win compensation for racial violence.Reference Mansoor 83 But while the common manifestations of racism have evolved over the last few decades (from acute violence to pervasive microaggressions, for example), IIED has not evolved in tandem. 84 Many courts do not consider insidious forms of racial harassment to be the “extreme and outrageous conduct” necessary for emotional injury claims. 85 For instance, in Graham v. Guilderland Cent. School District, an African American student sought restitution after a teacher called her a racial slur. The court ruled that “the use of religious, ethnic or racial aspersions to denigrate a person … is not sufficiently egregious conduct to sustain a claim of [IIED].” 86 A singular instance of name calling did not exceed the expectation that all members of society are subject to some degree of insult. 87
Although expressions of racial harassment have changed, IIED’s antiquated nature fails to acknowledge that the health consequences have not. Both overt and ambiguous racist incidents increase victims’ risk of heart disease, hypertension, diabetes, and asthma.Reference Richman 88 Studies have found that the accumulation of microaggressions can contribute to severe behavioral symptoms of PTSD (e.g. flashbacks, hypervigilance, insomnia, and substance abuse) akin to those from macroaggressions.Reference Nadal 89
There are also significant changes to the brain. For example, the degree of circuit connectivity between the amygdala and thalamus in a group of Black adults was directly proportional to their self-reported exposure to discrimination.Reference Webb 90 DTI scans of 116 Black women also found that their self-reported racial discrimination was inversely related to the integrity of white matter tracts in the corpus callosum and longitudinal fasciculus (even after controlling for lifetime trauma exposure).Reference Fani 91 These brain changes are typical for PTSD victims and have long-term implications — increasing victims’ risk of neurodegenerative disorders and future mental health problems.
Victims have yet to cite these studies in tort litigation, but considering the analogous use of neuroscience evidence in IPV cases, it seems reasonable that the scientific literature could inform a court that absent blatant violence, racial invectives often still have grave physical effects. Neuroimaging research may thus complement the shared experiences of victims of chronic microaggressions and encourage jurors to take plaintiffs’ claims more seriously. It will be more difficult, however, for plaintiffs to prove that microaggressions from a particular defendant are responsible, rather than the lifelong accumulation of harm from multiple parties. Even if the plaintiff in Graham v. Guilderland Cent. School District had a brain scan with classic PTSD signatures, it would not be sufficient evidence to place culpability solely on her teacher. Although her attorney could cite the last wrongdoer doctrine, which places culpability on the perpetrator whose act was the proximate cause of harm, this argument is tenuous. While the immediate pain from a gunshot wound can be traced to the individual who pulled the trigger a second before, it may be more difficult to trace the beginning of one’s suffering from racial trauma, and thus identify which insult served as the causal bullet. There are also First Amendment issues worth considering. Perpetrators of verbal harassment have successfully avoided IIED liability by citing free speech protections. 92 It is unclear whether a hypothetical future device that could reliably detect brain damage after a single microaggression would be enough to erode this defense. The value our society places on upholding the First Amendment may outweigh, in the jury’s opinion, the plaintiff’s buttressed claim for restitution. Given these challenges, it appears the most realistic applications for neuroimaging in racial trauma torts will be confined to contextualizing evidence. Aggregate research may do little to help prove the culpability of one person.
C. Solitary Confinement
One of the marginalized groups likely to benefit the most from neuroimaging research is prisoners. The 1996 Prison Litigation Reform Act (PLRA) prevents federal inmates from winning compensation for emotional injury without evidence of associated “physical injury or the commission of a sexual act.”Reference Filler and Greenfield 93 While intended to limit frivolous lawsuits, PLRA has in practice choked off meritorious ones as well. Five years after it passed, the number of federal inmate cases had dropped by 60 percent. 94
PLRA is especially catastrophic for prisoners trying to protest conditions of solitary confinement. Despite extensive research on the psychopathological effects (ranging from hallucinations and panic attacks to anxiety, depression, and PTSD), it has been nearly impossible for prisoners to win compensation due to their apparent lack of physical injury.Reference Hagan 95 Brain imaging now correlates these behavioral symptoms with deleterious changes in brain structure and function similar to other victims of PTSD. Extreme isolation decreases “cortical volume …[and] myelin production,” while altering “activity in the reward system and the amygdala.”Reference Coppola 96 EEG measurements have found “brain injury-like wave alterations … linked with hyperresponsivity [and] a complete breakdown of the identity of the isolated individual.” 97 Not only does extreme isolation generate neurobiological deficits, but it also reinforces existing ones. 98 This is especially alarming since 44% of the general inmate population already suffers from mental illness. 99 Instead of rehabilitating these convicts, prisons that employ solitary confinement are likely exacerbating their symptoms.
In the groundbreaking case Ashker v. Governor of California, prisoners used brain imaging studies to challenge their treatment at the Pelican Bay State Prison. 100 Over a thousand inmates had been kept isolated for 23 hours a day in cells without windows, sometimes for decades. They spoke with relatives through an intercom; physical contact was never allowed.Reference Lobel and Akil 101 The plaintiffs’ counsel cited the abundant literature on social isolation’s cognitive effects and extrapolated these findings to the prisoners’ situation. 102 This foundational scientific evidence bolstered the argument that their extreme isolation constituted physical injury. The accompanying testimony of several neuroscientists and psychiatrists helped the plaintiffs pressure California into a settlement that overhauled its approach to solitary confinement and ended indefinite isolation at the Pelican Bay facility. 103 This success has been replicated in cases across the country. 104
The neuroscience evidence was persuasive because it corroborated common-sense logic, rather than attempting to make a dispositive conclusion for the court. As Justice Kennedy wrote in another case, “the human toll wrought by extended terms of isolation has long been understood and questioned by writers and commentators…. [R]esearch confirms what this Court suggested over a century ago. Years on end of near total isolation exact a terrible price.” 105
Prisoners in solitary confinement, abused women, and racial minorities are just a few examples of marginalized groups that could benefit from brain scanning. The technology will likely be less useful for certain victims, such as those suffering from racial trauma. Although its utility in court for individuals may vary, neuroimaging significantly benefits the population of minority victims by refuting the myth that they have “thicker skins” that make them impervious to psychosomatic harm. Society has habitually responded to the injuries of marginalized people with silence, indifference, and victim-blaming. But the behavioral symptoms of PTSD now have clear physical substrates in the brain, regardless of one’s social status. By offering credence to the claims of people from all backgrounds, brain scanning rebalances courts’ understanding of psychological harm.
IV. Evidentiary Limitations on Individual Neuroimaging
There is an understandable buzz around neuroimaging’s potential to redefine legal treatment of PTSD and other mental disorders. But we should temper such excitement with levelheaded analysis of the technology’s current deficiencies in accuracy, and the risk that juries will overemphasize its findings in their decision making. The cases we’ve considered largely relied on aggregate neuroimaging research to educate courts on PTSD’s physical underpinnings. In this section, I describe why such foundational research may pass state and federal admissibility standards in certain contexts. I go on to argue that individual neuroimaging of plaintiffs should not yet do the same. Courts should exclude personal brain scans until they achieve a greater level of precision and reliability. While neuroimaging techniques have improved rapidly and elucidated physical underpinnings of psychological injury, it will remain challenging for lawyers to overcome evidentiary limitations on their use in many cases.
A. Evidentiary Standards
State and federal courts can evaluate the admissibility of scientific evidence in different ways. If a federal judge deems the specialized knowledge relevant to a case, they then consider if it will “help the [jury] … determine a fact in the issue.” 106 According to standards developed in Daubert v. Merrill Dow Pharmaceuticals, the evidence must also be “the product of reliable principles and methods.” 107 There is little guidance on what constitutes “reliable” beyond the ability to produce replicable data, but the majority opinion lists a few factors that judges can further opt to consider, such as “whether the theory or technique … has been subjected to peer review” and “its known or potential error rate.” 108 While some states follow the federal guidelines, others apply a standard first articulated in Frye v. United States. 109 For scientific evidence to be deemed admissible, it must first gain “general acceptance in the particular field in which it belongs.” 110 This process can take decades to achieve, if it happens at all. 111 The judge must delegate the evaluation of a novel technique to the consensus of experts.
B. Admissibility of Aggregate Evidence
Despite these distinctions, both state and federal courts have the same overarching goal: to only admit evidence that is reliable, relevant, and helpful. It is possible for aggregate neuroimaging research to pass all three qualifications. Although the technology is nascent, the peer-review process gives some confidence that published studies are generally reliable. The information may be relevant and helpful to jurors by educating them on PTSD’s essential biology and physical impact on victims.Reference Herringa 112 The probative value of population neuroimaging studies — and thus their likelihood of admission at trial — will increase as the demographics of study participants more closely match those of a plaintiff. For example, a study on racial trauma in Hispanic communities may be deemed relevant if the plaintiff in a racial trauma lawsuit is also Hispanic. 113
Aggregate research may further serve to correct jurors’ implicit biases by revealing unique elements of PTSD unknown to the general public. Seventy percent of people globally will experience a potentially traumatic event in their life, but only 5.6% go on to develop diagnosable PTSD. 114 Fact finders who have suffered trauma often attempt to understand a plaintiff’s conditions based on their own experience. 115 This may mislead them, because painful memories are not akin to the intrusive symptoms of PTSD.Reference Clapp 116 Neuroimaging studies can serve an important role alongside expert witnesses to educate courts about the true qualities of PTSD and prevent mistaken inferences from personal knowledge. Each jury that compensates a PTSD plaintiff for physical as well as emotional injury will contribute to a new precedent in tort jurisprudence. The quantitative evidence of neuroimaging research may thus incrementally nudge the law toward a more just treatment of the disorder as a physical harm.
C. Probative Issues of Individual Neuroimaging
Compared to foundational population studies, personalized PTSD neuroimaging faces far greater challenges in reliability and relevance that ought to compromise its use in litigation. To begin, it is incredibly rare for plaintiffs to have comparable brain scans from before and after a traumatic event. Absent baseline data, it is difficult to conclude that the trauma caused any structural or functional abnormalities observed.Reference Jones 117 Perhaps the plaintiff’s brain was always that way. And even if a plaintiff does have a baseline, the relevance of a post-trauma scan declines the longer it is taken after the incident. Brains naturally change over time in response to environmental and genetic factors, so there is no guarantee that an abnormality detected six years after trauma was also there a few months after it, or will still be there six years, or months, from now.Reference Greely and Wagner 118 There is little ability to determine brain composition and activity beyond the present. To complicate this “time machine” problem further, psychotherapy can ameliorate some of the brain effects of PTSD — increasing hippocampal volume and inferior frontal junction inhibition of the amygdala. 119 Plaintiffs who take scans after treatment may no longer exhibit PTSD’s brain morphology. While time does not erase the right to restitution for their period of illness, it might erase the evidence in their brain.
Scans vary over time and also between subjects. Neuroimaging research characterizes PTSD’s effects over large populations, but these averages don’t represent all individuals. 120 A plaintiff may have severe PTSD without the neurotypical signature. Put another way, brain changes can corroborate a victim’s suffering, but their absence cannot disprove it.
There are also fundamental issues with the technology that impair confident analysis of individual scans. fMRI does not directly measure brain activity, and one’s age or health might influence their blood oxygen levels. In addition, Greely and Illes point out that postmenopausal “women using hormone replacement therapy may have enhanced cerebrovascular reactivity, thereby increasing the speed and size of their blood flow responses” compared to healthy age-based controls.Reference Greely and Illes 121 Anti-hypertensive medication and neurological conditions such as Alzheimer’s disease, stroke, and glioma engender the opposite effect. It is challenging to control for all these overlapping and confounding factors. Despite incredible advances over the past few decades, neuroimaging is still an imperfect tool. When used on individual plaintiffs, it does not seem sufficiently reliable to pass either the Daubert or Frye standards.
D. Prejudicial Issues of Individual Neuroimaging
Not everyone agrees. The late legal scholar Frederick Schauer admitted that individual fMRI data is substandard, but he still advocated for its use given the fallibility of current fact-finding methods.Reference Schauer 122 Emotional injury litigation relies heavily on witness testimony, and witnesses are not faithful reporters of the truth. “Omissions, exaggerations, shadings, fudgings, slantings, bendings, and hedgings are an omnipresent feature of modern litigation,” Schauer observed. “So too is flat-out lying.” 123 In his view, the oath and cross examination are insufficient counter-measures. Indeed, studies suggest that the prevalence of malingering PTSD plaintiffs in disability and related forensic assessment contexts is approximately 15 ± 15%.Reference Young 124 Traditional methods to unearth fabrications, such as the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5) and the Trauma Symptom Inventory (TSI), are far from perfect. CAPS-5, a comprehensive interview that includes questions on the frequency, intensity, and subjective distress of symptoms as well as their effects on social functioning, is widely considered the gold-standard in malingering assessments. Yet it still relies on an estimation from the evaluator on the overall validity of the interview — a highly subjective calculation for which the manual provides minimal guidance. 125 fMRI neuroimaging of individuals may be “bad science,” but according to Schauer, it is “good evidence” compared to many of the current methods employed and would improve the overall accuracy of a trial. 126
Schauer made a strong case, but I believe prejudicial concerns about the technology outweigh its marginal benefit to factual accuracy. According to Federal Rule 403, even if a judge deems scientific evidence reliable and relevant, they may still exclude it if “its probative value is substantially outweighed by a danger of … confusing the issues [or] misleading the jury.” 127 Fact finders may trust neuroimaging more than is warranted because they are impressed by the technology’s complexity and captivated by the colorful outputs. It may be bad science, but it will likely be perceived as excellent science.Reference Brown and Murphy 128 For example, when mock jurors were presented with fMRI scans that exhibited a high-risk defendant’s brain damage in addition to expert testimony from a psychologist describing him as “psychotic,” they were 22 times less likely to sentence him to death compared to jurors who just heard from the psychologist.Reference Aono 129 Although emerging research tempers some of these findings, there is undoubtedly a risk that neuroimaging bias could distort jurors’ reasoning as much as a deceitful plaintiff.Reference Schweitzer and Saks 130 Neuroimaging’s imperfections should not discourage its development nor eliminate its use in applicable cases as a source of aggregate contextualizing evidence. But it would be prudent to ban the use of individual neuroimaging for PTSD plaintiffs until the technology better aligns with society’s lofty expectations for its ability.
V. Recommendations
A. FDA Oversight of Clinical Trials
If there is a government moratorium on PTSD neuroimaging, then it is fitting that a government agency should set the quantitative standards for efficacy and certify whether they are met. The FDA seems best situated to take on this role given the need for expertise in clinical trial design and medical devices. However, since this application of neuroimaging is not “intended for use in the … cure, mitigation, or prevention of disease,” it falls beyond the normal scope of FDA oversight. 131 Congress would likely need to pass legislation expanding the agency’s jurisdiction.
To prove the efficacy of individual PTSD neuroimaging, FDA trials should be conducted on a large number of subjects with varied age, race, socioeconomic standing, and trauma history. Brain signatures vary between trauma types, so evidence that fMRI can detect individuals’ PTSD from domestic abuse does not guarantee its accuracy for race-based trauma. Separate studies should be conducted on victims from each trauma, as well as those who have experienced both.
Finally, and perhaps most importantly, who will pay for this? There is currently little corporate interest in fMRI applications for emotional injury torts (although this may change as the technology improves). 132 Government funding seems a reasonable alternative since accurate detection of trauma’s neurological substrates can advance research in the public’s interest. In short, rigorous clinical trials will ensure that the technology is capable of its task. We ought to validate neuroimaging’s soundness before it ostensibly validates a plaintiff’s suffering.
B. Individual Neuroimaging Should Not Be Required
It is only a matter of time until neuroimaging exceeds government standards. But even when fMRI is exquisitely accurate, the legal system should establish that plaintiffs do not need such evidence to substantiate their PTSD. The seductive allure of brain scans might otherwise create a “CSI effect,” where juries and judges expect a popular form of scientific evidence and make negative assumptions about a claimant in its absence. 133
This has occurred in other legal contexts for neuroimaging. One study by Nita Farahany of criminal cases between 2005 and 2012 found that those citing neurobiological evidence achieved more favorable outcomes.Reference Farahany 134 A complementary finding is that defendants appealing their case on the basis of ineffective counsel — known as the Strickland Standard — are more successful when their previous lawyer failed to obtain neurobiological evidence. 135 The burden of proof for the Strickland Standard is high. According to one observer, “a lawyer with a pulse will be deemed effective.”Reference Miller 136 They need not be awake, sober, or prepared. Nevertheless, in a review of 553 criminal cases citing neuroscientific evidence, 27.65% of Strickland claims were successful.Reference Denno 137 This is astounding. For comparison, there was only a 4% success rate for Strickland claims in California during the same period.Reference Benner 138 It appears that judges already view neuroimaging as essential when there is a reasonable probability that brain abnormalities affected criminal defendants. The absence of brain scans might risk wrongful conviction or excessive sentencing. Thus, lawyers who do not obtain this evidence are not doing their jobs.
In the case of PTSD tort claims, a CSI effect would harm litigants unable to acquire scans due to cost or accessibility. Ake v. Oklahoma established the government’s obligation to pay for expert examination (including neuroimaging) of an indigent criminal defendant if a court deems it crucial to their case. 139 Indigent tort plaintiffs have no analogous opportunity for state funding and must instead rely on pro bono services or contingency fee arrangements, where attorneys cover upfront costs in return for a portion of awarded damages. This can restrict access to necessary tests, as law firms may hesitate to advance funds without a high likelihood of recovery. 140 The marginalized groups who could benefit the most from neuroimaging therefore could also suffer the most from a tort system that expects brain scans without helping to pay for them. The CSI effect might also harm those whose PTSD does not have a neurotypical fingerprint. 141 Despite other significant evidence, a jury could be overly biased by a false-negative brain scan.
To avoid these issues, courts should encourage a multidisciplinary approach to substantiating PTSD. A good model for this is traumatic brain injury (TBI), which is legally classified as a bodily injury yet shares multiple symptoms with PTSD that make it challenging for clinicians to distinguish them.Reference Wortzel and Arciniegas 142 TBI plaintiffs do not need to show structural brain damage to win compensation (and in fact, only 5–8% of traumatic brain injuries are detectable by neuroimaging). 143 Instead, plaintiffs draw on psychological testing, neurological examinations, and behavioral symptoms to substantiate their injuries and win compensation. 144 Courts do not need to see the brain damage to know it is there.
A similar approach should apply for PTSD. Since there is robust research that PTSD causes physical harm to the brain, it does not make sense to require every plaintiff to prove this fact again on themselves. Individual neuroimaging will no doubt be a wonderful tool, but it should remain one of many that a plaintiff can use. 145
Conclusion
While there has been social progress in acknowledging the severity of psychiatric illnesses, a pejorative skepticism of PTSD claimants remains in tort jurisprudence that hurts their ability to win compensation. Neuroimaging is not a panacea for legal biases due to struggles with technical accuracy and its own propensity to distort jurors’ evaluation of evidence. But while the technology cannot yet produce definitive measurements for individuals, it still serves valuable purposes. Aggregate research may educate jurors about PTSD’s neurobiology and lend crucial support to the claims of minority plaintiffs. It no longer makes sense to define the disorder as an exclusively emotional injury when each of its symptoms correlates to clear damages in the brain.
Neuroimaging not only challenges the tort classification of PTSD, but the entire principle of substance dualism. If we now consider PTSD to be both an emotional and physical injury, then what about other psychiatric disorders or chronic pain? The utility and authority of the dividing line weaken the more we cross it, and we will inevitably do so again and again. Just as the definition of fetal viability keeps changing as prenatal technology improves, so too will the meaning of physical injury as neuroimaging improves. It is difficult to maintain legal boundaries that do not keep pace with scientific advancement.
There is still considerable hesitation to cast aside substance dualism. The Restatement (Third) of Torts indicates that “some degree of emotional harm is part of the price of living in a complex society,” and we have a certain obligation to self-soothe without legal recourse. 146 Prioritizing physical tort claims reinforces the social expectation of emotional resiliency and limits spurious IIED and NIED torts.
But there’s a crucial flaw in this standard: an injury’s visibility does not necessarily equate to its severity. A broken arm will heal in a few months, but a broken mind can be permanent. William White suffered more from solitary confinement than most Americans can imagine. The pain of months-long isolation deprived him of reason, language, and eventually identity. He had no control over his fate; no ability to self-soothe. And yet the lack of an obvious wound sabotaged his tort claim. His suffering was largely invisible, so the law did not deem it compensable. If the goal of tort law is to remedy the most grievous harms, then substance dualism fails.
What’s the alternative? Dismantling the physical-mental divide may help plaintiffs like White, but might the pendulum of justice swing too far in the other direction — allowing disgruntled neighbors to go to court, fMRI scans in hand, and win recovery for petty insults hurled over the hedge?
I don’t believe the fall of substance dualism will engender such a society of victimhood. The power of the jury ensures that there will always be a dividing line between serious and spurious claims. Neuroimaging may substantiate a plaintiff’s suffering, but it cannot determine if that suffering warrants compensation. No matter how advanced technology becomes, the decision will rest on juries’ evolving interpretation of community values, expectations, and priorities. What is the definition of an unreasonable harm? What steps should victims take to enhance their own recovery? Only we can answer these questions. Science may guide the law, but it does not control it.
Acknowledgements
The author would like to thank Henry Greely for his exceptional teaching and guidance on this project.
