Abstract
Sense of humor is potentially relevant to social functioning in dementias, but has been little studied in these diseases. We designed a semi-structured informant questionnaire to assess humor behavior and preferences in patients with behavioral variant frontotemporal dementia (bvFTD;
Keywords
Introduction
Humor is a ubiquitous and highly valued social attribute and clinical experience suggests abnormalities of humor may be prominent in neurodegenerative diseases [notably, the frontotemporal lobar degenerations (FTLD)] that impair social and emotional signal decoding [1–3]. The syndromes of behavioral variant frontotemporal dementia (bvFTD) and semantic dementia (SD) are associated with impaired understanding of cartoons and sarcasm [1, 3] and this may develop premorbidly [4]. Although humor abnormalities are not generally regarded as a cardinal feature of Alzheimer’s disease (AD), emerging evidence suggests that humor might be affected alongside other aspects of social cognition in AD [5]. However, information concerning humor expression and awareness in neurodegenerative diseases remains limited.
Here we addressed this issue in patients representing canonical syndromes of FTLD [bvFTD, SD, and progressive nonfluent aphasia (PNFA)] and AD. We designed a semi-structured questionnaire to assess humor behavior and preferences, both in the current phase of established disease and retrospectively prior to clinical onset, in comparison to healthy older individuals.
Materials AND Methods
Participants
Forty-eight patients fulfilling current consensus criteria [6–8] for bvFTD (
Syndromic diagnoses were further corroborated in all cases by neuropsychological assessment, brain imaging (CT in three patients, MRI in the remainder), and/or cerebrospinal fluid examination. Participants had a comprehensive general neuropsychological assessment including standard measures of executive (Wechsler Abbreviated Scale of Intelligence, WASI) [9], social cognitive (The Assessment of Social Inference Test, TASIT) [10], semantic (British Picture Vocabulary Scale, BPVS) [11], and visual perceptual functions (see Table 1). All patients had neuroimaging findings compatible with their clinical syndromic diagnosis (corresponding to a diagnosis of ‘probable bvFTD’) in the bvFTD group [6]. Cerebrospinal fluid tau and Aβ1-42 assays in 20 patients (six bvFTD, six PNFA, eight AD) supported the clinical diagnosis in all cases, based on local laboratory reference ranges (normal ranges; total tau <320, Aβ1-42 220–2000, tau/Aβ1-42 ratio >0.8 predictive of AD). Of the genetic cases, four also had cerebrospinal fluid data available, confirming a non-AD-likeprofile.
The study was approved by the local institutional ethics committee and all participants gave written informed consent following Declaration of Helsinki guidelines.
Humor questionnaire
In order to assess patients’ sense of humor in daily life, we designed a semi-structured questionnaire comprising seven items (Table 2). Questionnaires were completed for each patient by a normal informant who had known them well for at least 15 years. The questionnaire recorded perceived changes in the patient’s sense of humor over the course of the illness and an item adapted from the Cambridge Behavioural Inventory (CBI) was used to quantify any tendency to express humor in scenarios that others would not generally find funny (rated 0–4; 0 = never, 1 = a few times per month, 2 = a few times per week, 3 = daily, 4 = constantly). In addition, the questionnaire recorded patients’ total daily life comedy exposure in broadcast and print media (estimated hours per week) and their liking for comedy (on a 10-point Likert scale), both currently and 15 years previously. This interval was chosen arbitrarily, but was designed to capture any alterations in humor preferences before the onset of typical clinical symptoms, while minimizing potential confounding effects from normal cognitive aging, informant knowledge, and recall bias. Patients with disease duration longer than 15 years were accordingly not included in the study. The questionnaire assessed three broad comedy genres or categories; farcical or slapstick (e.g.
Statistical analyses
Demographic characteristics, neuropsychological and behavioral rating data were compared between groups. Data on participant gender, country of origin and altered sense of humor (present/absent) were analyzed using two-tailed Fisher’s exact tests. Kruskal Wallis tests were used to compare other demographic characteristics, comedy exposure and liking for particular comedy genres between groups. Relations between humor preference ratings and gender were assessed in the healthy control cohort using the Wilcoxon rank-sum test. Spearman’s tests were used to assess correlations of humor measures with general disease measures (symptom duration, Mini-Mental State Examination (MMSE) score) and nonverbal executive performance (WASI Matrices score) in the combined patient cohort, semantic performance (BPVS score) in the SD group and social cognitive performance (TASIT scores) in the bvFTD group. For all tests,
Results
Participant groups did not differ significantly in mean age (
Humor questionnaire data are summarized in Table 3 and representative informant comments are in Table 4. Three patients with bvFTD— one with a pathogenic C9orf72 mutation, one with a MAPT mutation, and one with no identified mutation on screening— were not entered into the study because estimated symptom duration was >15 years in these cases. In each case, the patient’s caregiver described alterations in their sense of humor similar to other patients with bvFTD. Questionnaire informants were mainly the patients’ primary caregivers (in most cases, a cohabiting spouse) or a sibling or child who had been in long-term regular contact (at least monthly) with the patient (Table 3). Most participants had grown up in the United Kingdom; a few had spent part of their childhoods abroad in countries affiliated with Britain (Table 3). One patient with bvFTD was excluded owing to the fact he was a French national (and had therefore experienced a comedy milieu not shared by the rest of the cohort). Participant groups did not differ significantly according to country of origin (
Altered sense of humor was reported significantly more frequently in bvFTD (
The CBI measure of increased tendency to show humor was significantly correlated with executive impairment (WASI Matrices) for the combined patient cohort (rho = –0.36,
Informant comments (Table 4) revealed a number of instances in which patients were reported to show frankly inappropriate humor responses such as laughter over others’ misadventure (e.g. watching news stories about natural disasters, witnessing a spouse injure herself) or impersonal stimuli (e.g. a car badly parked, a barking dog). In a
Questionnaire data on liking for particular comedy genres in individual patients are presented in Fig. 1. These data show that the majority of patients with bvFTD and SD showed reduced liking for comedy, while most patients with PNFA showed no change in liking for comedy across genres following the onset of their illness. However, a few patients in each group showed increased liking for comedy; this occurred most frequently for slapstick comedy and in patients with bvFTD and PNFA (20% of patients in each of these groups).
Discussion
Here we have shown that canonical dementia syndromes commonly produce an altered sense of humor and this alteration differs qualitatively and quantitatively between dementia syndromes. In this series, altered humor was universal in bvFTD and SD, but occurred in a substantial minority of patients with PNFA and AD. Increased fatuity and relative predilection for childlike or slapstick humor and less pleasure in other comedy genres were features of all dementia syndromes, while frankly inappropriate humor in response to unpleasant or impersonal stimuli was a hallmark of bvFTD and SD. Moreover, selectively altered humor responsiveness was reported to have occurred well before the onset of more typical symptoms in association with both FTLD and AD: this was manifest as less pleasure in satirical comedy premorbidly. Development of abnormal humor expression correlated with executive impairment across syndromes and with clinical disease duration in AD, but not FTLD syndromes; supporting the clinical impression that sense of humor is often impoverished early in FTLD, but relatively preserved initially in AD.
The most striking alterations of humor responsiveness here occurred in FTLD syndromes characterized by impaired interpersonal functioning and for comedy genres (satirical, absurdist) most reliant on social cognition processes. This corroborates recent cognitive profiling of these syndromes using a novel paradigm assessing patients’ perception of humor in nonverbal cartoons [12]: both bvFTD and SD were associated with impaired detection of humorous intent in cartoon scenarios requiring psychological insight. Whereas the appreciation of slapstick humor typically entails detection of surface and physical incongruities, appreciation of satirical and absurd comedic scenarios requires a model of our place in the world with an understanding of social norms and often, others’ beliefs and intentions (‘theory of mind’ [1]). Transgression of those norms or mental states can then be perceived as surprising and solving the puzzle this poses, as ultimately pleasurable (when we ‘get’ the joke [13]). In patients with bvFTD and SD, this puzzle-solving behavioral algorithm appears to be not simply defective, but promiscuous: such patients are apt to assign humorous value in highly inappropriate contexts.
This interpretation would align an altered sense of humor with other forms of aberrant reward processing in FTLD [14], further demonstrating that abnormal valuation can extend from primary biological reinforcers such as food and sex to complex, abstract sensory stimuli [15] and potentially reflecting shared mechanisms of abnormal reward decoding in striatal and mesolimbic brain networks. The lack of correlation here between humor measures and a standard measure of social cognitive function (TASIT) in the bvFTD group might therefore appear initially somewhat counterintuitive; while arguments to a negative finding must be cautious particularly in the face of small case numbers, this might reflect the modularity of social cognitive subcomponents and suggests that substrates for humor decoding may be at least partly separable from other social cognition processes [12]. It is also noteworthy that only a minority of patients with bvFTD were reported as showing enhanced liking for slapstick comedy (Fig. 1) despite a clear tendency to increased fatuity and inability to suppress humor responses. This might indicate that humor behaviors in these patients become ‘mirthless’ (dissociated from subjective pleasure) or alternatively, that the behavioral correlates of such pleasure are harder for normal informants to decode. Further in this regard, our finding that sense of humor is commonly altered in AD is somewhat surprising and should motivate further study. The cognitive basis of this alteration may differ fundamentally in AD and FTLD: in particular, humor alterations in AD might reflect over-identification with the plight of others, as a manifestation of eroded emotional boundaries [5].
This study has several limitations that should guide future work. Clinical group sizes here were relatively small and patients were assessed using third-person reports while control data were based on self-report, both potentially subject to recall bias. There is a need to validate the questionnaire we propose in future work and in participants from other ethnic and cultural backgrounds. Humor is heavily influenced by social and cultural context and assessment of humor appreciation will likely require tailoring to these factors. Humor behavior and its potential as a disease biomarker should be studied prospectively, ideally from the presymptomatic phase of genetically-mediated dementias and with direct autonomic, structural and functional neuroanatomical correlation to capture subjective alterations in the experience of humor and mirth. Ideally, humor should also be assessed in the setting of neurological disease that spares cognitive function, in order to disambiguate cognitive from nonspecific chronic disease effects. The nature of humor alterations in neurodegenerative diseases requires further clarification: humor abnormalities have probably been under-recognized in AD and their relation to poorly understood phenomena such as abnormal laughter in PNFA [16] remain to be clarified. In particular, there is a need to investigate in greater detail the relations between humor alterations and other components of social cognition in these diseases. More broadly, the present findings have implications for the social functioning and quality of life of patients and those who care for them and this should be explored explicitly. We hope that our findings will stimulate interest in humor as an engaging, ecologically relevant and informative index of social functioning in neurodegenerative disease.
Footnotes
ACKNOWLEDGMENTS
We are grateful to all patients and healthy volunteers for their participation.
The Dementia Research Centre is supported by Alzheimer’s Research UK, the Brain Research Trust and the Wolfson Foundation. This work was funded by the Wellcome Trust, the UK Medical Research Council and the NIHR Queen Square Dementia Biomedical Research Unit. HLG holds an Alzheimer Research UK PhD Fellowship. JDR is funded by a National Institute for Health Research Rare Disease Translational Research Collaboration Fellowship. JDW is supported by a Wellcome Trust Senior Clinical Fellowship (Grant No 091673/Z/10/Z).
