Share this article on:

Guilty by Suspicion? Criminal Behavior in Frontotemporal Lobar Degeneration

Diehl-Schmid, Janine MD*; Perneczky, Robert MD*; Koch, Julia MD*; Nedopil, Norbert MD, PhD; Kurz, Alexander MD*

Cognitive and Behavioral Neurology: June 2013 - Volume 26 - Issue 2 - p 73–77
doi: 10.1097/WNN.0b013e31829cff11
Original Studies

Objective: Our aim was to compare the frequency of criminal conduct in patients with behavioral variant frontotemporal dementia (bvFTD), semantic dementia (SD), and Alzheimer disease.

Background: A few small-scale studies of antisocial and criminal behavior in patients with frontotemporal lobar degeneration have focused on the clinical subtype bvFTD. It is not yet known whether antisocial behavior affects patients with other clinical subtypes of frontotemporal lobar degeneration, like SD, and patients with other dementing disorders, like Alzheimer disease.

Methods: We used a standardized caregiver interview to assess criminal behavior in 83 outpatients: 32 with bvFTD, 18 with SD, and 33 with Alzheimer disease.

Results: We found criminal behavior (theft, willful damage to property, housebreaking, assault, or indecent behavior) in 54% of the patients with bvFTD and 56% of those with SD, but only 12% of those with Alzheimer disease.

Conclusions: Just over half of our patients with bvFTD or SD had committed crimes. When middle-aged or older patients commit minor crimes, frontotemporal lobar degeneration should be considered as a possible cause. If an affected person faces criminal charges, the court might take incapability or diminished responsibility into account in reaching a verdict.

*Department of Psychiatry, Technische Universität München, Munich, Germany

Department for Forensic Psychiatry, Ludwig-Maximilians-Universität München, Munich, Germany

The authors declare no conflicts of interest.

Reprints: Janine Diehl-Schmid, MD, Klinik und Poliklinik für Psychiatrie und Psychotherapie der TU München, Ismaninger Str. 22, Munich 81675, Germany (e-mail:

Received August 8, 2012

Accepted December 27, 2012

Reader Benefit: After reading this article, readers will be aware that (minor) criminal behavior is a frequent symptom in frontotemporal lobar degeneration.

AD=Alzheimer disease; bvFTD=behavioral variant frontotemporal dementia; FTLD=frontotemporal lobar degeneration; SD=semantic dementia.

Prefrontal cortex dysfunction has long been linked to social misconduct. Case reports have described patients with “acquired sociopathy” after prefrontal craniocerebral injury (Blair and Cipolotti, 2000; Meyers et al, 1992). In war veterans, a consistent relationship has been found between orbitofrontal lesions and later antisocial behavior (Blumer and Benson, 1975). Various imaging studies support a connection between the frontal lobes and antisocial behavior.

Key areas found to be structurally or functionally impaired in antisocial populations include dorsal and ventral regions of the prefrontal cortex, as well as the amygdala, hippocampus, angular gyrus, anterior cingulate, and temporal cortex (Raine and Yang, 2006). The prefrontal cortex has been shown to be vital for social cognitive and moral judgment processing (Forbes and Grafman, 2010), and this is the most crucial brain structure found to be compromised in violent and antisocial populations (Yang and Raine, 2009). Given the importance of the frontal lobes for appropriate social behavior, it is not surprising that we find antisocial behavior in patients with degenerative disorders of the frontal lobes.

Frontotemporal lobar degeneration (FTLD) causes progressive dementing conditions with prominent behavioral abnormalities. The most common clinical phenotype of FTLD is behavioral variant frontotemporal dementia (bvFTD). Cerebral dysfunction predominantly in the frontal lobes generates a syndrome characterized by early decline in social behavior and personal conduct, early loss of insight, and emotional blunting (Neary et al, 1998). Patients with bvFTD often transgress social norms through sociopathic behavior, a loss of empathy or appreciation of the feelings of others, and a loss of insight into their own behavior and its consequences (Mendez et al, 2008). Structural and functional imaging studies have shown that patients with bvFTD have early changes in a medial paralimbic network that includes the anterior cingulate, orbitofrontal, and frontoinsular cortices (Rabinovici and Miller, 2010). As the disease progresses, the neurodegenerative process spreads throughout the frontal cortex, except for the motor cortex, and into temporal and parietal areas (Diehl et al, 2007).

Another clinical variant of FTLD is semantic dementia (SD). SD is a disorder of language, semantics, and recognition of visual percepts, caused by predominant anterior temporal pathology (Neary et al, 1998). Although semantic deficits dominate the clinical picture, most patients with SD also undergo behavioral alterations such as loss of empathy, increasing irritability, disinhibition, selfishness, food fads, and perseverative behaviors (Snowden et al, 2001). Only rarely, however, have transgressions of social norms been described in SD (Diehl et al, 2006). Regions of brain atrophy in SD are the medial and lateral portions of the bilateral anterior temporal lobes, including the amygdalae and anterior hippocampi, the anterior fusiform gyri, the posterior portion of the insula, the caudate nuclei, and the ventromedial frontal cortex (Gorno-Tempini et al, 2004).

Case reports (Mendez, 2010) and small studies have described antisocial and even criminal behavior in patients with FTLD. Our own group has shown that antisocial behavior is much more prevalent in patients with FTLD than in those with Alzheimer disease (AD) (Diehl et al, 2006). Miller et al (1997) found antisocial behaviors including assault, indecent exposure, shoplifting, and hit-and-run driving in 10 of 22 patients with bvFTD, but only 1 of 22 patients with AD had a history of socially disruptive conduct. Mendez et al (2005) reported sociopathic behavior in 16 of 28 patients with bvFTD. Patient numbers were small in all studies, and both US studies, Mendez et al (2005) and Miller et al (1997), excluded patients with SD.

Because previous reports have had limited numbers, we aimed to study a large group of patients to compare the frequency of criminal conduct among those with bvFTD, SD, and AD.

Back to Top | Article Outline


Assessment of Criminal Behavior

For the purposes of our study, we defined criminal behavior as conduct ranging from indecent behavior to misdemeanors to serious crimes.

Based on our clinical experience with patients with FTLD, we designed a structured caregiver interview to ascertain specific criminal behaviors: theft, willful damage to property, breaking and entering, assault, indecent behavior, illegal use of weapons, and any other criminal behavior. We interviewed patients’ caregivers, asking them to rate the frequency of the patients’ criminal behavior since the onset of their dementia: rarely (less than once a year), occasionally (about once a year), frequently (several times a year but not as often as every month), or many times (at least once a month). We also asked caregivers for their opinion about intentionality, whether the behavior was accidental (eg, forgetting to pay in a supermarket) or willful. Further, we asked the proxies if, to their knowledge, the patients had ever shown indecent or criminal behavior before their disease began. The interviewers were not blind to the patients’ diagnoses.

The interview has been published in the German language (Diehl et al, 2006).

Informed consent according to the Declaration of Helsinki was obtained from or on behalf of every patient. The study protocol was approved by the local ethics committee.

Back to Top | Article Outline


This study was an extension of some previous work done by our group (Diehl et al, 2006). For this project we randomly chose patients from a large outpatient population that had been diagnosed with FTLD at a center for cognitive disorders. We excluded patients whose symptoms had started less than a year earlier. Of 100 caregivers whom we asked to participate, 83 agreed and completed their interviews: 80 spouses, 1 sister, 1 father, and 1 son.

Our final sample had 50 patients with FTLD (of whom 32 had bvFTD and 18 had SD) and 33 patients with AD. BvFTD and SD had been diagnosed according to the revised Lund-Manchester criteria (Neary et al, 1998), and AD according to the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) (World Health Organization, 2010). All diagnoses had been based on a thorough neurologic and psychiatric examination, informant interview, neuropsychological evaluation, standard blood tests, structural neuroimaging (computed tomography or magnetic resonance imaging), and fluorodeoxyglucose (18F-FDG) positron emission tomography.

Back to Top | Article Outline

Statistical Methods

For statistical analyses, we used the Statistical Package for the Social Sciences, v18.0 (SPSS Inc, Chicago, IL). Data on patient characteristics are given as mean±standard deviation. We used the χ2 test or, if appropriate, the Fisher exact test to compare frequency data. We used an independent 2-sample t test to compare means. We set statistical significance at P<0.05.

Back to Top | Article Outline



Table 1 lists characteristics of our 50 patients with FTLD and our 33 patients with AD. The only significant difference between the groups was age: The patients with FTLD had their first symptoms at a mean age of 56.6 years versus 66.8 years for the patients with AD (P=0.001).



To break down the characteristics of the 2 subgroups with FTLD: Of the 32 patients with bvFTD, 59% were men and 41% were women; the bvFTD subgroup had a mean education of 13.0±3.5 years, mean age at disease onset of 55.6±9.4 years, mean age at interview of 58.7±10.2 years, and mean duration of disease from onset until the interview of 5.7±3.6 years. Of the 18 patients with SD, 61% were men and 39% were women; the SD subgroup had a mean education of 12.5±2.8 years, mean age at disease onset of 58.8±6.7 years, mean age at interview of 64.6±4.8 years, and mean duration of disease from onset until the interview of 7.3±5.9 years. Sex, education, age at onset, and duration of the disease did not differ significantly between the bvFTD and SD subgroups. The only real difference was that at the time of the interview, the patients with SD were significantly older than those with bvFTD (P=0.027).

Back to Top | Article Outline

Criminal Behavior

None of the patients had shown criminal behavior before the onset of their disease. The interviews revealed a significant difference in frequency of criminal behavior once they were ill. Caregivers described criminal behavior in just 4 (12%) of the 33 patients with AD, but in 27 (54%) of the 50 patients with FTLD (Table 2).



Within the FTLD group, 17 (53%) of the patients with bvFTD and 10 (56%) of those with SD showed criminal behavior. None of the differences between the subgroups reached statistical significance (Table 3).



Eleven (35%) of the patients with bvFTD and 7 (39%) of those with SD had stolen someone else’s property, most often food, cosmetics, or money. One patient had stolen expensive clothing from a boutique. Most of the patients stole regularly, once a week to once a month.

Five (16%) of the patients with bvFTD and 5 (28%) with SD had willfully damaged someone else’s property: cars, bicycles, a computer, a sewing machine, or an expensive tray.

Eight (25%) of the patients with bvFTD, 1 (6%) with SD, and 1 (3%) with AD had entered someone else’s home without permission. Most often, they had entered the homes of relatives (ex-wife, daughter), acquaintances, or neighbors, by damaging windows or doors.

Eight (25%) of the patients with bvFTD, 2 (11%) with SD, and 2 (6%) with AD had physically assaulted spouses, relatives, friends, or strangers. Ten of the 12 patients had regularly threatened other persons. The other 2 patients had actually hurt someone.

Ten (31%) of the patients with bvFTD and 7 (39%) with SD had shown indecent behavior. Eight patients had exhibited themselves in public and 9 had touched strangers in an indecent or sexual way. Half of the patients had engaged in indecent behavior only once or twice, but the other half did so frequently.

None of the patients’ proxies reported use of weapons or any other criminal behavior.

While the patients with and without criminal behavior did not differ significantly by sex, education, age at disease onset, or age at interview, we found a significant difference in duration of the disease from onset until the interview. The patients with criminal behavior had twice the disease duration of the patients without criminal behavior (P=0.009) (Table 4).



Back to Top | Article Outline


We found criminal behavior in 54% of patients with FTLD (bvFTD and SD) versus 12% of patients with AD. Men and women had similar rates of criminal behavior. Most of the patients’ offenses were minor, mainly involving property, and usually the material damage was small.

Because none of the patients had shown criminal behavior before the onset of their FTLD, we presume that their behavior was caused by the neurodegenerative disease. Disease duration from symptom onset until the caregiver interview was significantly shorter in patients without than with criminal behavior. The finding that antisocial or criminal behavior increased with advancing disease is consistent with the progression of the neurodegenerative process in the relevant neuroanatomic structures.

Our percentages of patients with bvFTD showing criminal behavior agree with existing case reports and studies (Mendez, 2010). Miller et al (1997) found almost identical percentages. Beyond this, we were able to show that antisocial behavior was as common in patients with SD as in those with bvFTD.

It is unclear why patients with FTLD commit crimes. We were able to ask 7 patients about the reasons for their behavior. One patient, who had scratched numerous cars with his umbrella, stated that he was angry because his driving license had been revoked. Another patient, who had regularly stolen food from shops, answered that he was not willing to pay so much money for the food. None of the 7 patients showed remorse or shame. Most did not see any need to explain themselves.

Their attitude is in line with observations that patients with FTLD do not feel concerned enough about their actions to be deterred from violating social norms. A number of studies have suggested that patients with FTLD understand social rules and conventions, as well as the consequences of their behavior (Lough and Hodges, 2002; Lough et al, 2006; Mendez, 2010; Mendez et al, 2005). However, further studies are needed to investigate whether patients are really able to process this knowledge properly in complex social situations.

The criminal behavior and the lack of concern, guilt, or shame are most likely caused by the neurodegenerative process disrupting a neuromoral network localized in the frontal and anterior temporal lobes (Mendez, 2009). Moll et al (2005) have argued that moral phenomena emerge from the integration of contextual social knowledge, represented as event knowledge in the prefrontal cortex; social semantic knowledge, stored in the anterior and posterior temporal lobes; and motivational and basic emotional states, which depend on corticolimbic circuits.

Mendez (2010) has suggested that the neuromoral network that is damaged in bvFTD is the ventromedial prefrontal cortex, with its rich interconnections with limbic structures (especially the orbitofrontal cortex and the amygdalae). Lesions in the ventromedial prefrontal cortex cause patients with bvFTD not to feel the strong automatic negative “gut reactions” to moral violations that prevent other people from committing morally impermissible acts. Patients’ orbitofrontal dysfunction leads to disinhibition and compulsion, and lesions of their amygdalae cause a loss of empathy (Mendez, 2010).

The fact that the anterior temporal cortex and the amygdalae are typically affected in patients with SD (Chan et al, 2001) explains their lack of empathy as well as their loss of social semantic knowledge, leading to their impaired social cognition. In addition, structural imaging studies of patients with SD have shown atrophy of the ventromedial prefrontal cortex. Because SD affects key structures of social cognition, it is not surprising that a significant number of our patients with SD had a history of criminal behavior.

Only a minority of our patients with AD had shown antisocial behavior. The neurodegenerative process in AD starts in the medial temporal lobes and the fusiform gyrus, and spreads into the posterior temporal lobes and parietal lobes. As the disease progresses, neurodegeneration reaches the frontal lobes (Rabinovici et al, 2008; Whitwell, 2010). However, even patients with early AD have prominent atrophy of the amygdalae (Poulin et al, 2011), supporting findings that AD impairs social cognition and theory of mind (Freedman et al, 2013). Therefore, it is unclear why patients with AD do not tend toward criminal behavior. A possible explanation is that the neurodegenerative process spreads into the frontal lobes only in the later stages of AD, when cognitive deficits like disorientation, memory impairment, and executive deficits prevent patients from committing crimes.

The results of our study have neuropsychiatric and forensic implications. When previously law-abiding middle-aged or older patients suddenly start committing minor crimes, particularly theft or shoplifting, a neurodegenerative brain disorder should be considered and a thorough psychiatric and neurologic examination arranged. If the patient has a diagnosis of FTLD, the patient’s family needs to be informed that the crimes are a symptom of the disease rather than the patient’s fault.

Only if a patient has received the correct diagnosis can incapability (incapacity) or diminished responsibility be claimed when a crime is committed. The German Criminal Code (German Criminal Code in the version promulgated on November 13, 1998) makes allowances for incapability (Section 20) and diminished responsibility (Section 21). Here is the definition for Section 20, with an interpolated paraphrase from Section 21:

Any person who at the time of the commission of the offence is incapable [or has a substantially diminished capacity] of appreciating the unlawfulness of their actions or of acting in accordance with any such appreciation due to a pathological mental disorder, a profound consciousness disorder, debility or any other serious mental abnormality, shall be deemed to act without guilt.

The German legal concept of “incapability of acting in accordance with the appreciation of the unlawfulness of an offence” applies perfectly to patients with FTLD, who know right from wrong but are unable to act in accordance with that knowledge because they have lost empathy and emotional moral judgment, and they cannot suppress an impulse to act in an unacceptable manner (Lough et al, 2006).

In the United States, the M’Naghten rule (United States Code, Title 18, Section 17) (Legal Information Institute, 2013) defines legal incapability much more strictly. It requires that the defendant “was unable to appreciate the nature and quality of the wrongfulness of his acts.” According to the M’Naghten rule, patients with FTLD would not qualify to plead “not guilty by reason of insanity.”

Our study has some limitations. First, the frequency of antisocial and criminal behavior might have been overestimated. We recruited our patients from a cognitive disorder center affiliated with a university department of psychiatry. We cannot rule out the possibility that such a center accepts only patients who have severe behavioral disturbances, leaving people with milder dementia to be managed by regular neurology departments.

Second, our study could not use the findings from the patients’ neuropsychological testing or neuroimaging. Our center had confirmed the diagnosis of FTLD in each of our patients before we arranged the caregiver interview. At the time of the interview, some patients had already been showing criminal behavior for years. Some other patients, who had been diagnosed with FTLD elsewhere, had already stopped their criminal behavior before coming to our center. Thus, the “peak” of most patients’ criminal behavior did not coincide with the time of their clinical evaluation, neuropsychological testing, and imaging.

Third, our patients with criminal behavior had suffered significantly longer from their disease than the patients with no criminal behavior. Patients who had not engaged in criminal behavior at the time of the interview may have started committing crimes afterward.

Finally, it is possible that more of our patients had been motivated to commit crimes, but they could not act on their urges because of a lack of opportunities or internal drive.

The search to correlate criminal conduct with neuropsychological and neuroimaging data in patients with neurodegenerative diseases is a major challenge that requires an elaborate prospective experimental design. The findings may shed further light on the neural substrates of antisocial and criminal behavior.

Back to Top | Article Outline


Blair R, Cipolotti L.Impaired social response reversal: a case of acquired sociopathy.Brain.2000;123:1122–1141.
Blumer D, Benson DFBenson DF, Blumer D.Personality changes with frontal and temporal lobe lesions.Psychiatric Aspects of Neurologic Disease.1975.New York:Grune & Stratton;151–170.
Chan D, Fox NC, Scahill RI, et al..Patterns of temporal lobe atrophy in semantic dementia and Alzheimer’s disease.Ann Neurol.2001;49:433–442.
Diehl J, Ernst J, Krapp S, et al..Misdemeanor in frontotemporal dementia.Fortschr Neurol Psychiatr.2006;74:203–210.
Diehl J, Grimmer T, Drzezga A, et al..Decline of cerebral glucose metabolism in frontotemporal dementia: a longitudinal 18F-FDG-PET-study.Neurobiol Aging.2007;28:42–50.
Forbes C, Grafman J.The role of the human prefrontal cortex in social cognition and moral judgement.Annu Rev Neurosci.2010;33:299–324.
Freedman M, Binns M, Black S, et al..Theory of mind and recognition of facial emotion in dementia: challenge to current concepts.Alzheimer Dis Assoc Disord.2013;27:56–61.
German Criminal Code in the version promulgated on November 13, 1998Federal Law Gazette [Bundesgesetzblatt] I, p. 3322, last amended by Article 3 of the Law of October 2, 2009, Federal Law Gazette I, p. 3214.
Gorno-Tempini ML, Dronkers NF, Rankin KP, et al..Cognition and anatomy in three variants of primary progressive aphasia.Ann Neurol.2004;55:335–346.
Legal Information Institute2013United States Code. Title 18, Crimes and Criminal Procedure. Section 17, Insanity defense. Ithaca, NY: Cornell University Law School. Available at: Accessed May 21, 2013.
Lough S, Hodges J.Measuring and modifying abnormal social cognition in frontal variant frontotemporal dementia.J Psychosom Res.2002;53:639–646.
Lough S, Kipps C, Treise C, et al..Social reasoning, emotion and empathy in frontotemporal dementia.Neuropsychologia.2006;44:950–958.
Mendez MF.The neurobiology of moral behavior: review and neuropsychiatric implications.CNS Spectr.2009;14:608–620.
Mendez MF.The unique predisposition to criminal violations in frontotemporal dementia.J Am Acad Psychiatry Law.2010;38:318–323.
Mendez MF, Anderson E, Shapira JS.An investigation of moral judgement in frontotemporal dementia.Cogn Behav Neurol.2005;18:193–197.
Mendez MF, Lauterbach E, Sampson S.An evidence-based review of the psychopathology of frontotemporal dementia: a report of the ANPA Committee on Research.J Neuropsychiatry Clin Neurosci.2008;20:130–149.
Meyers CA, Berman SA, Scheibel RS, et al..Case report: acquired antisocial personality disorder with unilateral left orbital frontal lobe damage.J Psychiatry Neurosci.1992;17:121–125.
Miller B, Darby A, Benson D, et al..Aggressive, socially disruptive and antisocial behaviour associated with frontotemporal dementia.Br J Psychiatry.1997;170:150–155.
Moll J, Zahn R, de Oliveira-Souza R, et al..Opinion: the neural basis of human moral cognition.Nat Rev Neurosci.2005;6:799–809.
Neary D, Snowden J, Gustafson L, et al..Frontotemporal lobar degeneration: a consensus on clinical diagnostic criteria.Neurology.1998;51:1546–1554.
Poulin SP, Dautoff R, Morris JC, et al..Amygdala atrophy is prominent in early Alzheimer’s disease and relates to symptom severity.Psychiatry Res.2011;194:7–13.
Rabinovici G, Seeley W, Kim E, et al..Distinct MRI atrophy patterns in autopsy-proven Alzheimer’s disease and frontotemporal lobar degeneration.Am J Alzheimers Dis Other Demen.2008;22:474–488.
Rabinovici GD, Miller BL.Frontotemporal lobar degeneration: epidemiology, pathophysiology, diagnosis and management.CNS Drugs.2010;24:375–398.
Raine A, Yang Y.Neural foundations to moral reasoning and antisocial behavior.Soc Cogn Affect Neurosci.2006;1:203–213.
Snowden J, Bathgate D, Varma A, et al..Distinct behavioural profiles in frontotemporal dementia and semantic dementia.J Neurol Neurosurg Psychiatry.2001;70:323–332.
Whitwell J.Progression of atrophy in Alzheimer’s disease and related disorders.Neurotox Res.2010;18:339–346.
.International Statistical Classification of Diseases and Related Health Problems, 10th Revision [ICD-10].2010.Geneva:World Health Organization.
Yang Y, Raine A.Prefrontal structural and functional brain imaging findings in antisocial, violent, and psychopathic individuals: a meta-analysis.Mol Psychiatry.2009;16:227–236.

frontotemporal dementia; frontotemporal lobar degeneration; semantic dementia; criminal behavior; misdemeanor

© 2013 by Lippincott Williams & Wilkins.