Abstract
The insanity defense remains one of the most difficult evaluations for a forensic examiner because it requires a retrospective reconstruction of an individual's mental state and extracting that mental state to specific psycholegal criteria (e.g., capacity to understand wrongfulness) at the time of an alleged crime. Some scholars have suggested that neuroimaging can improve the objectivity of insanity evaluations by highlighting brain deficits in the defendant that led to a compromised ability to understand wrongfulness or an inability to comport their behavior to the requirements of the law. This brief review highlights six principal limitations that, to date, have not been appropriately addressed by proponents of using neuroimaging in insanity evaluations. Researchers are encouraged to address these specific limitations to improve its application to insanity defense evaluations. We posit that until this work is completed, neuroimaging provides limited value to insanity evaluations.
1. The insanity defense through the years
The insanity defense has longstanding roots in philosophy and the legal system. Early conceptualizations of insanity were rooted in Aristotelian ethics. In Nicomachean Ethics (350 B.C.E./1902). Aristotle wrote of individuals who demonstrated “ignorance of the circumstances of the act and of the things affected by it, in this case the act is pitied and forgiven, because he who acts in ignorance of any of these circumstances is an involuntary agent” (pp. 1110b-1111a). Aristotle's (350 B.C.E./1902) conceptualization of the circumstances including the act itself mirrors actus reus (literally translated “guilty act”) and mens rea (“guilty mind”). In a similar vein, Plato wrote of persons committing crimes, “In a state of madness or when affected by disease … And if this be made evident to the judges elected to try the cause, on the appeal of the criminal or his advocate, and he be judged to have been in this state when he committed the offense, he shall simply pay for the hurt which he may have done to another” (Plato, 360 B.C.E./2013, Book IX).
There were several early legal formulations of insanity including The Wild Beast Test, The M'Naghten Test, and the Model Penal Code/American Law Institute. The latter two tests heavily focus on an inability to know/appreciate right from wrong due to a mental disease or defect (Cornell Law School, 2022)(“M'Naghten Rule,” 2022; Model Penal Code § 4.01 (1)). In the United States, seventeen states have adopted the M'Naghten standard, and fourteen states use the Model Penal Code/American Law Institute (ALI). Other states rely on some combination and four states do not have an insanity defense. The ALI wrongfulness component reads as follows, “A person is not responsible for criminal conduct if at the time of such conduct as a result of mental disease or defect he lacks substantial capacity either to appreciate the criminality [wrongfulness] of his conduct.” The ALI adds a volitional prong, referring to an individual who is not responsible if they lack sufficient capacity “to conform his conduct to the requirements of the law” (ALI, 1962).
Based on historical antecedents, one would assume that the insanity defense is strongly engrained and accepted as legitimate in American jurisprudence. However, extant data indicates significant distrust of insanity and mental illness with the public having strong negative views against the insanity defense (Borum & Fulero, 1999; Goossens et al., 2021; Hans, 1986). Research with perspective jurors (e.g., Cutler et al., 1992; Perlin, 1994; Poulson et al., 1997; Skeem & Golding, 2001) has consistently found that the insanity defense is widely unpopular. This unpopularity is rooted in the ideas that unscrupulous experts find non-insane individuals insane, a belief that the insanity defense is overused, and that the insanity defense is a legal loophole where individuals who commit crimes do not receive their “just desserts” (see Morse, 1985). These views remain steadfast even though the actual number of individuals pleading and being found not responsible for their criminal behavior due to mental illness is extremely low (Packer, 2009).
2. Challenges associated with insanity defense evaluations
Conducting insanity or criminal responsibility evaluations is an inherently difficult task for a forensic examiner. Insanity evaluations require that the clinician retrospectively evaluate a defendant's mental state at and around the time they committed the reported offense. Research on the reliability and validity of insanity reports reveals that there are significant shortfalls in these evaluations. For instance, research has indicated that insanity defense reports vary by evaluator, as agreement between evaluators on whether an individual meets the legal criteria for insanity has ranged from 55% to 76% (Gowensmith et al., 2012, 2013; Phillips et al., 1988), and this level of agreement is lower than other forensic opinions offered by evaluators (e.g., competency to proceed to trial). The reported level of disagreement belies the idea that forensic evaluators can reliably evaluate insanity. Several potential issues underscore the less than adequate reliability for insanity evaluations. These reasons include adversarial allegiance (i.e., the evaluator is more likely to side with who has retained them), lack of forensic training in conducting criminal responsibility evaluations, or using inadequate data to make decisions (Acklin et al., 2015; Ferguson & Ogloff, 2011; McAuliff & Arter, 2016).
The level of disagreement underscores concerns that insanity evaluations lack objectivity. The studies showing poor consistency for insanity opinions have been used to support the hypothesis that neuroimaging can provide a more thorough and objective approach to insanity evaluations. It has been suggested that imaging can reduce cognitive biases, especially backward reasoning (Scarpazza et al., 2018). This is best exemplified by observing a behavioral disturbance and considering brain-based explanations. This paper considers claims that neuroimaging imaging offers a pathway to improved reliability in insanity evaluations.
The development and application of neuroscience to psycholegal questions, such as competency to stand trial, criminal responsibility (i.e., insanity), and mitigation has become commonplace (Perlin & Lynch, 2016). The application of neuroimaging to court is commonly referred to as neurolaw, which is defined by Petoft (2015) as “an interdisciplinary field which links the brain to law, facilitates the pathway to better understanding of human behavior in order to regulate it accurately through incorporating neuroscience achievements in legal studies” (p. 53). Attempts to view, comprehend, and explain behavior by indirectly evaluating the brain has a lengthy history. Yet, the application of neuroscience to legal matters is not without peril.
In discussing the limitations with relying on neuroscience to answer psycholegal questions, Beecher-Monas and Garcia-Rill (2017) authored a comprehensive law review that unpacked ideas of brain imaging and brain science to inform questions of criminal responsibility. Later, these authors reframed the same review in a book that aptly described the fit of neuroscience in legal areas as “square peg, round hole” (Beecher-Monas & Garcia-Rill, 2020). Their thesis recognized that the promised revolution in conducting criminal responsibility evaluations using advanced technologies, like neuroimaging, has not yet come to fruition (Bigenwald & Chambon, 2019; Vitacco et al., 2020). Fundamental limitations in the use of neuroscience have been identified across publications. These limitations have persisted even considering increased access to imaging technology by lawyers and the individuals they represent.
3. Expansion of neurolaw and psycholegal applications
Neuroimaging in the forms of magnetic resonance imaging (MRI), functional MRI (fMRI), and positron emission tomography (PET) scans have greatly influenced the field of neurolaw as they have been employed to argue that brain deficits should mitigate punishment for criminal behavior (Aono et al., 2019). Before we address the problems associated with neuroimaging and the insanity defense, we must acknowledge that neuroimaging has several critical and relevant uses in many aspects of mental health evaluation, including civil and criminal trials. Ultimately, neuroimaging may improve decision-making for the trier of fact in specific instances, like mitigation (Denno, 2015). Consider the relevance of imaging in how the United States Supreme Court has conceptualized adolescent culpability in the “Miller trilogy” (Drinan, 2015, 2021). The Miller trilogy is a series of cases (Graham v. Florida, 2010; Miller v. Alabama, 2012; Roper v. Simmons, 2005) decided by the United States Supreme Court that fundamentally changed the landscape of juvenile sentencing. These three Supreme Court decisions, which incorporated neuroscientific evidence related to developmental immaturity, outlawed the most severe sentences for offenders who commit crimes as juveniles, including the death penalty and life sentences for non-homicides (Albert et al., 2013; Steinberg & Scott, 2003). The Court recognized that the presence of underdeveloped brains increased the chance to engage in violence and lowered, but did not eliminate, culpability. These specific applications are also useful when considering the role mental health can play in mitigation and how a trier of fact may utilize imaging data when determining a sentence in a criminal trial (Denno, 2015).
The rise of neurolaw has meant that more attorneys are willing to engage the use of neuroscientists to consider brain-based explanations for complicated behavior. In a comprehensive analysis, Denno (2015) analyzed 800 criminal cases from 1992 to 2012 and provided several case examples regarding the integration of neuroscience in court. It was summarized that “neuroscience is typically introduced for a well-established legal purpose … and precise information when determining a defendant's fate” (p. 494). Often, brain evidence of frontal lobe dysfunction is introduced to demonstrate that a defendant manifested difficulty with self-control, making the resulting act of violence less controlled by the defendant. It has been argued that evidence of frontal lobe dysfunction decreased legal culpability (Batts, 2009; Farahany, 2016; Perlin & Lynch, 2016) and especially useful when attempting to mitigate capital punishment. However, mitigation is quite different from insanity, especially when considering legal blameworthiness and the punishment associated with the outcome.
On the surface, the use of neuroimaging and neuroscience in cases that involve mental health is natural because mental health disorders are brain-based disorders associated with anatomical and functional abnormalities (Borgwardt et al., 2007). Consider schizophrenia as a primary example. Schizophrenia has been associated with enlarged lateral ventricles, loss of frontal lobe matter, manifestations in the corpus callosum, and grey matter disturbance in the temporal gyri, to name a few (Gur & Gur, 2010; Zhou et al., 2021). Bipolar disorder is another illustrative example. Bipolar disorder has been associated with increased brain activation in the prefrontal and anterior cingulate cortices and subcortical limbic structures (Clark & Sahakian, 2008). It has been suggested that if brain structures are implicated in the etiology and pathogenesis of mental disorders, that alone demonstrates the utility of brain imaging in legal proceedings. This premise is flawed.
The mere association between a brain-based finding and clinical diagnoses is insufficient to denote its validity in forensic evaluations. To underscore this point, a Consensus Report of the American Psychiatric Association Work Group on Neuroimaging Markers of Psychiatric Disorder published in 2012, reviewed the state of the science regarding the utility of neuroimaging in clinical work. This report recommended against using imaging data for clinical diagnoses (First et al., 2012). In a more recent paper, First et al. (2018) reiterated their earlier position and concluded, “While there is reason to be hopeful that these methods will eventually yield generalizable and replicable results that will permit their application in clinical practice, at this time, the majority of neuroimaging's utility remains as a research tool” (p. 2). In a staunch rebuttal, Henderson et al. (2020) focused on several strengths of neuroimaging by noting, “… functional neuroimaging can offer clues and information about psychiatric disorders and their comorbid conditions” (p. 7) and underscored that incremental steps are being taken to improve the use of neuroimaging for mental health diagnostics. The question for forensic examiners and courts is straightforward but not simple: Does the information obtained through neuroimaging provide incremental data that allows the trier of fact greater access to accurate legal findings? Simply put, the present paper questions the extent that neuroimaging adds to the accuracy of a forensic opinion of insanity beyond other pieces of evidence, like clinical interviewing and/or a review of collateral data?
The use and admission of imaging into the court for diagnostic purposes appears to have marched forward undeterred. Neuroimaging results are often viewed as a panacea to many psycholegal questions, like competency to proceed to trial (Perlin & Lynch, 2018; Philipsborn & Hamilton, 2021). Perlin and Lynch (2018) argued that using neuroimaging is consistent with competency evaluations completed through an approach of therapeutic jurisprudence and that using brain imaging “may help us more accurately determine whether some defendants are, in fact, competent or incompetent to stand trial” (p. 96). The primary limitation with this supposition is that competency to proceed is a functional test that requires evaluation of specific capabilities (e.g., ability to consult with one's attorney). It was not shown how the use of imaging brings a forensic evaluator closer to the goal of making an accurate determination. Many attorneys seem to believe that if they can show a brain issue on an imaging scan, one can infer a specific functional impairment. However, many of these associations have not been thoroughly vetted or scientifically validated. This leads to a problem dubbed “neurohype” which has been defined as “a broad class of neuroscientific claims that greatly outstrip available evidence” (Lilienfeld et al., 2018, p. 241).
4. Six issues plaguing the utility of neuroimaging in insanity evaluations
The idea that neuroimaging adds probative value to insanity evaluations has been discussed for years. The premise that data provided by neuroimaging and the multidisciplinary necessity of using these methodologies has been lauded, yet often the rationale behind using imaging in insanity evaluations has lacked empirical support. We discuss six predominant limitations that minimize the probative value of neuroimaging when used in insanity evaluations.
Imaging is prone to cognitive biases, showing a high level of subjective interpretation associated with its findings. Theoretical arguments have centered on improved objectivity as a primary reason for using neuroimaging in insanity evaluations. Consider how experts from the defense and prosecution have different takes on imaging findings and the implication for objectivity (United States v. Erickson, 2021). Research has shown potential reporting biases for imaging because of sex (David et al., 2018; Larrazabal et al., 2020) and race (Chekroud et al., 2014). It has also been suggested that so-called “denoising” techniques introduce bias into imaging (Kay, 2022). The notion that imaging provides objective brain data is not supported by extant literature (Chen et al., 2023; Kay, 2022), and with insanity examination the risk remains extremely high for the presence of adversarial allegiance.
Neuroimaging lacks evidence of incremental validity over traditional methods of evaluating insanity. Incremental validity is defined as “the degree to which the instrument provided measures that were more valid than alternative measures of the same variable” (Haynes & Lench, 2003). Incremental validity is critical to psychological and psychiatric science, and it ensures that methodologies being employed provide unique and important information to the decision-making process that is otherwise not available. Imaging has demonstrated incremental validity in other aspects of psychiatric diagnosis, including depression and anxiety (Ball et al., 2014). Even with some evidence of incremental validity, the consensus for basic diagnostics is that imaging is best used in conjunction with clinical data (Stoyanov et al., 2020). The absence of studies demonstrating the presence of incremental validity significantly limits the use of neuroimaging for insanity cases. On May 1, 2023, we conducted a literature search looking at published and unpublished research on the insanity defense and neuroimaging and were unable to locate a single study or data point that supports that insanity evaluations and opinions are improved with the addition of neuroimaging (see also Vitacco et al., 2020).
Issues with court admissibility have plagued imaging findings when used in insanity evaluations. Two Supreme Court decisions that guide admissibility of evidence continue to rule in all states. These decisions are codified in Frye v. United States (1923) and Daubert v. Merrell Dow Pharmaceuticals, Inc (1993). Frye (1947) is referred to as the “general acceptance” test, meaning that the procedure must be generally accepted in the field for its intended use in court. Daubert (1993) lays out four specific criteria that must be weighed for admissibility including general acceptability (see Frye, 1947), appropriate standards controlling procedure accuracy and control, known or potentially known error rate, and the testability of the procedure. Certainly, imaging and neuroimaging meet these criteria for diagnosing brain tumors as a plethora of scientific studies demonstrate imaging's sensitivity and specificity (e.g., Ghafourian et al., 2023). Yet, we could not find a single study that provides utility estimates of imaging for insanity evaluations. Lacking even basic statistics on error rates should hinder the admissibility of imaging in the complicated legal requirements of insanity evaluations. It should not be surprising that courts have approached the use of imaging in insanity cases and trials with skepticism (United States v. Erickson, 2021). However, based on case law, imaging will likely remain useful for mitigation, especially in capital cases (Lockett v. Ohio, 1978; Denno, 2015).
Brain plasticity and the problem of retrospective evaluation with technology that is specifically designed for a current perspective must not be overlooked. Insanity evaluations frequently occur after a significant amount of time has passed from the alleged incident. The retrospective nature of insanity evaluations fails to account for brain malleability. It is widely accepted that the brain is dynamic and can change based on circumstances and environment (Erickson, 2010). Brain plasticity is a barrier to neuroimaging accurately depicting the functionality of the brain in days, weeks, and months after an alleged offense (Poldrack, 2020). As such, imaging may be very useful to diagnosing brain tumors that are present (Baig et al., 2016), but have little utility for diagnosing specific mental health symptoms that wax and wane. Moreover, imaging provides little to no information on the specific symptom, its relation to behavior, or the nexus to the statutory requirements for the insanity defense.
Studies on neuroimaging have established no appropriate normative values concerning the evaluation of mental state at the time of the offense. To that end, we could not locate research studies demonstrating how “insanity” would look on a scan or if there what the differences would be for individuals with a mental illness not adjudicated insane compared with those adjudicated insane. Baskin et al. (2007) discussed the need for a higher level of clarity in imaging studies to improve their generalizability and acceptability. Fifteen years after that article, the field is continuing to search for clarity, and many promises linking imaging to behavior have remained unfulfilled. A recent double-blind study noted improvement in a brain-based response for alcohol use disorder (McCalley et al., 2022); however, similar results and data do not exist for psycholegal questions.
One of the key components of evaluating criminal responsibility is finding information that deals with the motivation that preceded the criminal behavior and the behaviors after the crime (Packer, 2009). This can be a complicated task. As an illustrative example, a forensic examiner may need to differentiate anger connected to real world events from anger based on delusional ideations (e.g., an innocent person is out for them). It is critical for the examiner to understand a planned act committed primarily for external gain (e.g., homicide in furtherance of a robbery) from an act that is done to satisfy a deific decree (e.g., command hallucination from God that someone is evil). Available data on imaging does not point to the ability to differentiate motivations. Of course, even individuals in the throes of psychosis can behave instrumentally. Imaging provides little to no data on disentangling complicated human motivations for criminal behavior.
5. Next steps in the use of neuroimaging in insanity evaluations
This manuscript, by focusing on the dearth of empirical evidence related to neuroimaging and insanity, also offers a research-based agenda for moving neuroimaging from theory to practice. There are several avenues of research that will propel the field forward. In the next paragraphs, we discuss the areas where research can be used to make imaging useful to insanity defense evaluations. We recognize that the retrospective nature of insanity evaluations cannot be changed by research, but the other significant limitations can be worked through with research agendas. The next paragraphs focus on how research can successfully address those limitations.
The first step involves developing normative values for insanity acquittees and imaging. This can only be accomplished through double-blind research where individuals who read brain scans are blind to the outcome and the individual evaluated. This type of research would provide an initial foundation for understanding neuroimaging results as they relate to use in insanity cases. This may be especially relevant with certain conditions. Redding (2006) suggested that neuroimaging may be especially profitable for understanding frontal lobe dysfunction relevant in losing the ability to conform one's conduct. To that end, Egbenya and Adjorlolo (2021) attempted to ascribe specific brain-based areas to the ability to understand right from wrong and loss of ability to conform one's behavior. However, they acknowledged potential difficulties in linking complex behavior with specific brain areas. As Baskin et al. (2007) argued, greater imaging clarity is needed to make this endeavor successful.
Until additional research is completed, linking brain regions with specific requirements of the law required for an insanity defense is too speculative and tentative for admissibility. In this case, the law will follow the sciences once specific research is conducted demonstrating its utility. Once research indicates sufficient reliability and validity using imaging in insanity evaluations, courts will likely become less skeptical of its inclusion. As of this writing, too many questions remain for courts, and they will likely remain dubious of many of the claims made by scientists and imaging findings.
The establishment of guidelines for the interpretation of imaging in forensic examinations is needed along with the realization that imaging may have a substantial role in some forensic issues. Consider mitigation, where findings may show diffuse damage that may be useful to attorneys attempting to argue that an individual has some lack of capacity and therefore should receive a lesser sentence or avoid capital punishment (Denno, 2015). However, the lack of programmatic research is glaring, and without the iterative nature of science, neuroimaging will continue to have minimal practical uses in insanity evaluations.
6. Conclusions
Proponents of using neuroimaging continue to trumpet its strength as an “objective” measure of mental illness and tout the use of technology to advance clinical judgment. However, the actual objectivity of neuroimaging findings relating to the insanity defense remains highly dubious and speculative (Vitacco et al., 2021). Consider this statement in an article by Rigoni et al. (2010) that underscores the relevance of neuroimaging in insanity cases; “Our claim is that the use of neuroscience and behavioral genetics do not change the rationale underlying the determination of criminal liability, which must be based on a causal link between the mental disorder and the crime. Rather, their use is crucial in providing objective data on the biological bases of a defendant's mental disorder” (p. 1). This highlights the problem of relying on imaging or other types of neurosciences to evaluate insanity cases as they methodologies are unable to find that “nexus.” As it currently stands, neuroscience, including imaging, has little to offer to improve upon clinical opinions for criminal responsibility. As noted in this manuscript, the empirical limitations of using neuroimaging should prohibit its use in insanity cases. The trier of fact should carefully consider whether imaging data support clinical diagnoses. Neuroimaging can fulfill its potential in the legal arena only after rigorous research is completed focusing on its reliability and validity in assisting in answering specific psycholegal questions.