Significance
Religious fundamentalism is a global and enduring phenomenon. Measuring religious fundamentalism following focal brain damage may lend insight into its neural basis. We use lesion network mapping, a technique that uses connectivity data to identify functional brain networks, to analyze two large, independent datasets of brain lesion patients. We found a network of brain regions that, when damaged, are linked to higher religious fundamentalism. This functional network was lateralized to the right hemisphere and overlaps with the locations of brain lesions associated with specific neuropsychiatric and behavioral conditions. Our findings shed light on neuroanatomy that may influence the emergence of religious fundamentalism, offering implications for understanding the relationship between brain networks and fundamentalist behavior.
Keywords: neurology, neuroscience, religion, human behavior, brain lesions
Abstract
Religious fundamentalism, characterized by rigid adherence to a set of beliefs putatively revealing inerrant truths, is ubiquitous across cultures and has a global impact on society. Understanding the psychological and neurobiological processes producing religious fundamentalism may inform a variety of scientific, sociological, and cultural questions. Research indicates that brain damage can alter religious fundamentalism. However, the precise brain regions involved with these changes remain unknown. Here, we analyzed brain lesions associated with varying levels of religious fundamentalism in two large datasets from independent laboratories. Lesions associated with greater fundamentalism were connected to a specific brain network with nodes in the right orbitofrontal, dorsolateral prefrontal, and inferior parietal lobe. This fundamentalism network was strongly right hemisphere lateralized and highly reproducible across the independent datasets (r = 0.82) with cross-validations between datasets. To explore the relationship of this network to lesions previously studied by our group, we tested for similarities to twenty-one lesion-associated conditions. Lesions associated with confabulation and criminal behavior showed a similar connectivity pattern as lesions associated with greater fundamentalism. Moreover, lesions associated with poststroke pain showed a similar connectivity pattern as lesions associated with lower fundamentalism. These findings are consistent with the current understanding of hemispheric specializations for reasoning and lend insight into previously observed epidemiological associations with fundamentalism, such as cognitive rigidity and outgroup hostility.
The basis of one’s religious beliefs and behaviors has been an enduring focus of psychological inquiry and, more recently, neuroscientific investigation. One facet of religiosity is religious fundamentalism: an adherence to religious doctrines believed to be inerrant, a devotion to religious practices considered immutable, and a perceived special relationship with a deity (1, 2). Fundamentalism is associated with authoritarianism (3, 4), cognitive rigidity (5–7), lower complexity of thought on religious issues (8, 9), a reduced likelihood of doubt (4, 9, 10) increased acceptance of misinformation (11, 12), delusion-like ideation (11), anti-intellectualism (13, 14), prejudicial attitudes (1, 15), and outgroup hostility (16). Fundamentalism may confer group-level advantages: including stronger ingroup commitment (17) and an increased sense of belonging and well-being (18).
Research examining the determinants of religious fundamentalism has often focused on environmental or socialization factors (19) such as socioeconomic and family affiliation variables (20, 21). However, neuroscientific and behavioral genetic research argues for an additional biological influence toward religious experiences and attitudes. Pharmacological, neuroimaging, and psychophysiological studies indicate that unique patterns of neural activity are correlated with religious states (22–25). Moreover, while general religiosity heritability estimates vary with age and metric (26), twin studies have shown that the specific construct of religious fundamentalism is strongly heritable (27–29). Thus, neurobiological factors likely predict a style of cognitive and emotional processing that tends to result in fundamentalist attitudes (10, 11). The cognitive and behavioral associations with fundamentalism may be clarified by the identification of brain regions or networks underlying religious fundamentalism (30).
Brain lesion studies allow for causal inferences into human behavior (31, 32), including religiosity (10, 33, 34). Prior research has shown that damage to the prefrontal cortex is associated with higher levels of religious fundamentalism (4, 5). However, this preliminary work was limited by sample size and a narrow focus on the prefrontal cortex. The prefrontal cortex is a large and highly connected region (35) that participates in a number of brain-wide networks (35, 36). Neurobiological factors that underlie a cognitive or emotional style of processing associated with religious fundamentalism are most likely a distributed brain network phenomenon (32). To date, no studies have investigated fundamentalism using a brain network approach. As a result, there remains a critical gap in the evidence relating to religious fundamentalism and the brain.
Here, we examined two large and independently collected human lesion datasets (N1 = 106; N2 = 84) to determine whether religious fundamentalism could be mapped to a functional network across the brain. We used an analytic method, lesion network mapping, which relies on functional connectivity data from healthy individuals to infer functional networks that may be disrupted by a lesion (33, 37). This technique allowed us to compare of our results with our extensive database of lesions associated with a variety of behavioral, neurological, and psychiatric conditions (N3 = 899).
Results
Lesions associated with religious fundamentalism occurred in multiple different brain regions across our two datasets (Fig. 1), suggesting that a brain network approach to mapping these lesion locations may be valuable. We combined both independent datasets and performed voxelwise statistical tests for lesion network associations with Religious Fundamentalism Scale (2) (RFS) scores (Ncombined = 190). Functional connectivity between each lesion location and the rest of the brain was computed using a publicly available normative connectome dataset from 1,000 healthy right-handed subjects (Fig. 2). Connections significantly associated with RFS scores were identified (Fig. 3, FWE Pcorrected < 0.05). Higher fundamentalism scores were associated with positive connectivity between lesion locations and right superior orbital frontal, right middle frontal, right inferior parietal, and right inferior temporal cortices, as well as the left cerebellum (Fig. 3B and Table 1). The fundamentalism network maxima clusters were strongly right lateralized in the cerebral cortex. Several of the maxima clusters did extend to adjacent white matter (Fig. 3B). We also observed left cerebellum maxima clusters, consistent with contralateral cerebro-cerebellar connections (38). Lower fundamentalism scores were associated with negative connectivity between lesion locations and the left paracentral lobule, right cerebellum, and bilateral middle temporal cortices (Fig. 3C and Table 1). Again, several of the clusters did extend to adjacent white matter (Fig. 3C). Post hoc analyses found 1) a similar topography of voxelwise associations when lesion size, age, chronicity, and sex were used as covariates (SI Appendix, Figs. S1 and S2), 2) no associations between lesion chronicity and fundamentalism, 3) no associations between proportion of gray versus white matter intersected by a lesion, and 4) no differences in fundamentalism associations with lesion connectivity between self-identified religious affiliations (SI Appendix, Fig. S3; SI Appendix).
Fig. 1.
Lesions associated with religious fundamentalism occur in many different brain regions beyond the prefrontal cortex. (A) Brain lesions of five patients with the highest Religious Fundamentalism Scale (RFS) scores for religious fundamentalism from dataset 1. (B) Brain lesions of five patients with the lowest RFS scores from dataset 1. (C) Brain lesions of five patients with the highest RFS scores from dataset 2. (D) Brain lesions of five patients with the lowest RFS scores from dataset 2. Note that lesions associated with high and low religious fundamentalism are highly heterogenous across brain regions and datasets.
Fig. 2.
Data-driven method for identifying a lesion network for religious fundamentalism. The network of brain regions functionally connected to each lesion was computed using resting-state functional connectivity data from a large database of healthy volunteers (N = 1,000). Lesions and lesion networks are shown for 4 of the 190 patients in the combined datasets (Ncombined = 190). Positively connected voxels are shown in warm colors, while negatively connected voxels are shown in cool colors. Connections associated with religious fundamentalism scores were then identified (Right). Warm colors in the group-level map indicate that functional connectivity with lesions is more likely associated with higher scores for religious fundamentalism. Conversely, cool colors in the group-level map indicate that functional connectivity with brain lesions is more likely associated with lower scores for religious fundamentalism.
Fig. 3.
Religious fundamentalism network derived from combined datasets (N = 190). (A) Whole-brain topography (unthresholded) for statistical associations between brain lesion connectivity and Religious Fundamentalism Scale (RFS) scores. Warm-colored regions indicate positive functional connectivity relative to lesions associated with higher religious fundamentalism. Cool-colored regions indicate negative functional connectivity relative to lesions associated with higher religious fundamentalism. (B) Voxels with statistically significant connectivity to brain lesions associated with higher religious fundamentalism based on continuous RFS scores across the datasets following family-wise error multiple comparison correction (FWE P < 0.05). (C) Voxels with statistically significant connectivity to brain lesions associated with lower religious fundamentalism based on continuous RFS scores across the datasets following family-wise error multiple comparison correction (FWE P < 0.05).
Table 1.
Locations of local maxima and local minima within our lesion-derived brain network for religious fundamentalism
| Cluster size (mm3) | Peak t | MNI coordinates | Peak structure |
|---|---|---|---|
| 13,787 | 5.1 | [18, 50, −11] | R Superior Orbital Frontal |
| 6,020 | 5.0 | [46, 14, 46] | R Middle Frontal |
| 3,200 | 4.8 | [50, −58, 56] | R Inferior Parietal |
| 2,414 | 4.8 | [4, 38, 40] | R Superior Medial Frontal |
| 2,273 | 4.6 | [−35, −76, −45] | L Crus 2, Cerebellum |
| 2,001 | 4.9 | [60, −28, −16] | R Inferior Temporal |
| 1,458 | 4.9 | [−12, −86, −26] | L Crus 2, Cerebellum |
| 768 | 4.7 | [−46, −76, −28] | L Crus 1, Cerebellum |
| 507 | 4.5 | [−20, 52, −10] | L Superior Orbital Frontal |
| 298 | 4.5 | [2, −30, 30] | R Middle Cingulum |
| 265 | 4.5 | [14, 6, 20] | R Caudate |
| 192 | 4.5 | [2, −12, 29] | R Middle Cingulum |
| 147 | 4.4 | [20, 23, 60] | R Superior Frontal |
| 135 | 4.4 | [26, 22, −11] | R Insula |
| 95 | 4.4 | [4, −8, 16] | R Fornix |
| 33,770 | −5.2 | [−10, −36, 76] | L Paracentral Lobule |
| 3,010 | −4.8 | [22, −50, −56] | R Cerebellum (8) |
| 2,494 | −4.7 | [−54, −68, 12] | L Middle Temporal |
| 2,162 | −4.8 | [56, −66, 4] | R Middle Temporal |
| 1,671 | −4.8 | [−26, −42, −56] | L Cerebellum (8) |
| 1,181 | −4.7 | [−42, −40, −24] | L Fusiform Gyrus |
| 409 | −4.4 | [30, −16, 62] | R Precentral Gyrus |
| 154 | −4.5 | [−12, −26, 0] | L Thalamus |
| 55 | −4.3 | [14, −19, 74] | R Precentral Gyrus |
All clusters achieved significance at a family-wise error (FWE) multiple comparison correction of P < 0.05.
Notably, religious fundamentalism whole-brain association maps derived independently from datasets 1 and 2 robustly replicated each other, demonstrating a strong spatial correlation (r = 0.82, P = 0.02; Fig. 4 A and B). Cross-validation analyses further indicated the reliability of the fundamentalism network results between the independently collected lesion datasets. Spatial correlations were calculated between the patient lesion networks in dataset 1 and the group-level religious fundamentalism map derived from dataset 2. These spatial correlations represent the similarity of an individual patient’s lesion network (dataset 1) to an independently derived religious fundamentalism map (dataset 2). These spatial correlations were significantly associated with the dataset 1 patients’ individual scores for religious fundamentalism (r = 0.28, P = 0.003). In the other direction, spatial correlations were calculated between the patient lesion networks in dataset 2 and the group-level religious fundamentalism map derived from dataset 1. These spatial correlations likewise represent the similarity of an individual patient’s lesion network (dataset 2) to an independently derived religious fundamentalism map (dataset 1). These spatial correlations were significantly associated with the dataset 2 patients’ individual scores for religious fundamentalism (r = 0.27, P = 0.01; Fig. 4 C and D; SI Appendix, Fig. S4). To further test for cross-validation, a “network damage score” for each lesion in dataset 1 was calculated by superimposing each patient lesion from dataset 1 onto the group-level fundamentalism network derived from dataset 2 and summing the overlap of the lesion locations from dataset 1 with the voxelwise values of the group-level fundamentalism map derived from dataset 2. Correlations were observed between religious fundamentalism scores in dataset 1 and the network damage score (r = 0.21, P = 0.03; SI Appendix, Fig. S5). This relationship was retained when controlling for lesion size (r = 0.32, P = 0.003; SI Appendix, Fig. S6). Similarly, a network damage score for each lesion in dataset 2 was calculated by superimposing each patient lesion from dataset 2 onto the group-level fundamentalism network derived from dataset 1 and summing the overlap of the lesion locations from dataset 1 with the voxelwise values of the group-level fundamentalism map derived from dataset 2. Correlations were also observed between religious fundamentalism scores in dataset 2 and the network damage score (r = 0.29, P = 0.008; SI Appendix, Fig. S5). Again, this relationship was retained when controlling for lesion size (r = 0.31, P = 0.001; SI Appendix, Fig. S6).
Fig. 4.
Cross-validation of lesion network mapping results across two independent datasets. (A) Finding: lesion network mapping in dataset 1 defined a whole-brain statistical map that relates lesion connectivity with Religious Fundamentalism Scale scores. Positive correlations between lesion connectivity and religious fundamentalism scores are displayed in warm colors, and negative correlations between lesion connectivity and religious fundamentalism scores are displayed in cool colors. (B) Replication: lesion network mapping of religious fundamentalism in dataset 2 demonstrated a strong spatial similarly to lesion network map for religious fundamentalism from dataset 1 (spatial correlation, r = 0.82, P = 0.02). (C) Cross-validations: Brain lesions from dataset 1 (white outlines) associated with high religious fundamentalism scores intersect the lesion network connections from dataset 2 that were also associated with high religious fundamentalism scores (shown in warm colors). (D) Conversely, brain lesion from dataset 1 (white outlines) associated with low religious fundamentalism scores intersect the lesion network connections from dataset 2 that were also associated with low religious fundamentalism scores (shown in cool colors; r = 0.21, P = 0.03). (E and F) The same cross-validation analysis was also significant when brain lesion locations from dataset 2 (white outlines) were compared to whole-brain map of lesion network correlations with religious fundamentalism derived from dataset 2 (r = 0.29, P = 0.008).
Finally, to test the utility of our lesion network mapping approach, we performed a data-driven analysis of lesion locations (rather than lesion networks) on our data. Voxel lesion-symptom mapping (VLSM) analyses did not yield significant results. The spatial correlation between the VLSM maps from the two datasets was not greater than would be predicted by chance (r = 0.069, P = 0.82; see SI Appendix, Supplementary Methods and SI Appendix, Fig. S7 for VLSM statistical approach and results).
Religious fundamentalism in healthy individuals has a number of epidemiological associations including cognitive rigidity, reduced likelihood to doubt, prejudice, and outgroup hostility. If neurobiological factors influence a cognitive and emotional processing style that tends to result in fundamentalist attitudes, then these factors may also support these epidemiological associations. To investigate this in more depth, we examined spatial similarities between our fundamentalism network and connectivity patterns for brain lesions associated with 21 behavioral, neurological, and psychiatric conditions (N3 = 899; see Fig. 5 for complete condition list). Strong spatial similarities were observed between our religious fundamentalism network and brain lesion connectivity associated with pathological confabulation (39) (spatial correlations: t24 = 8.3, P < 10−7, 95% CI, r = 0.32 to 0.53; Fig. 5). Confabulation is the generation of conspicuously false beliefs or memories without the intent to deceive (40) and is strongly associated with cognitive rigidity or perseveration during neuropsychological tasks (41, 42). Confabulation is also associated with failure to inhibit or doubt inaccurate memory elements during retrieval (10, 40). These aspects of confabulation are shared by healthy individuals high in fundamentalism, who also show cognitive rigidity and a reduced likelihood of doubt (4, 6, 7).
Fig. 5.
Lesions associated with behavioral, neurological, and psychiatric conditions intersect our religious fundamentalism network. (A) The average of voxel intensities within lesion locations associated with 21 different conditions (n = 899) are shown in a bar graph. Error bars reflect SE across different lesion locations within each lesion syndrome. (B and C) Lesions (white outlines) associated with confabulation (showing 4 of 25 cases) and criminal behavior (showing 4 of 17 cases) showed the strongest intersections with positive nodes of our religious fundamentalism network, similar to lesions associated with high religious fundamentalism. (D) Lesion locations associated with central poststroke pain (showing 4 of 23 cases) showed the strongest intersection with negative nodes of our fundamentalism circuit, similar to lesion locations associated with low religious fundamentalism.
In addition, we observed strong spatial similarities between the fundamentalism network and brain lesion connectivity associated with criminal behavior (spatial correlations: t16 = 5.5, P < 10−4, 95% CI, r = 0.28 to 0.63; Fig. 5). This is consistent with previously observed associations between high religious fundamentalism and increased hostility, aggression, and violence against outgroups (6, 16, 43).
Finally, strong negative spatial correlations were observed between our religious fundamentalism network and lesion connectivity associated with central poststroke pain (spatial correlations: t22 = −5.2, P < 10−4, 95% CI, r = −0.45 to −0.19; Fig. 5). This inverse association is consistent with prior work showing that religious stimuli may engage pain-inhibiting brain processes and induce analgesic effects (44). These data suggest that neurobiological factors in healthy individuals predict a style of cognitive and emotional processing that often results in religious fundamentalism.
Discussion
Brain lesions associated with religious fundamentalism occur in different brain regions but can be considered components of a single, connected brain network. This religious fundamentalism network was consistent and cross-validated in two large, etiologically diverse neurological patient samples collected from independent laboratories. It incorporated gray and white matter in the right superior orbital frontal, right middle frontal, right inferior parietal, right inferior temporal regions, and the left cerebellum. Our lesion network mapping findings were reproducible across independent datasets, and networks derived from one dataset could be used to predict religious fundamentalism scores in the opposite dataset. The cross-dataset replication and prediction depended on network mapping, as it was not present using lesion locations alone. Moreover, lesions to regions in the fundamentalism network were positively associated with confabulation and criminal behavior and negatively associated with pain.
Given the etiological differences between our patient samples for the two principal datasets, our robust replication of the fundamentalism brain network is remarkable. One sample included Vietnam war veterans with penetrating traumatic brain injury (45), and the other sample comprised neurological patients drawn from the Iowa Neurological Patient Registry (46) who sustained focal brain injury mainly from stroke or surgical resection for treatment of epilepsy or benign tumors. Traumatic brain injury, surgical resections, or stroke injury can lead to distinct sequela of symptoms following damage to the same region (47). Our reproduction and cross-validation across these etiologically diverse datasets increase our confidence that lesions within a well-defined network are associated with religious fundamentalism.
Our brain network for religious fundamentalism showed a right hemisphere lateralization. Hemispheric specialization has been associated with the emergence of modern human reasoning, moral cognition, self-awareness, creativity, divergent thinking, and the ability to reflect on one’s cognitions (48–50). Split-brain and other unilateral lesion evidence has indicated distinct roles of each cerebral hemisphere during reasoning. For example, it has been posited that while the left hemisphere draws inferences, hypotheses, and rationalizations from the evidence available to it, the right hemisphere stops the perseveration of incorrect ideas (51–53). Right hemisphere damage has been associated with confabulatory or delusional beliefs (54–56) with impairments in error monitoring or conflict detection. In the religious domain, right hemisphere damage may result in reduced recognition of belief conflict, less religious disbelief, and the persistence of more extreme religious beliefs.
It is important to interpret our brain network results cautiously and contextually. Many factors contribute to fundamentalism, including affective, cognitive, experiential, genetic, familial, institutional, developmental, and cultural variables. Given these multiple factors, some patients with lesions in our religious fundamentalism network may not display high fundamentalism. Although brain lesions to this network may increase the likelihood of religious fundamentalism, they should not be interpreted as an inevitable or sole cause of fundamentalism (4). Moreover, the reverse inference—viz., that individuals high in religious fundamentalism have brain lesions is unsubstantiated and unwarranted. Similarly, our lesion results do not imply that people with strong religious beliefs confabulate or that individuals high in religious fundamentalism commit crimes. Rather, our data may help us understand the style of cognitive or emotional processing that increase or decrease the probability of holding fundamentalism attitudes.
A limitation in the current work is that patients in our study are predominantly older Caucasian individuals from Christian backgrounds. This may limit generalizability to populations with more variabilities in age, ethnicity, culture, and religious tradition. In addition, our data did not assess fundamentalism prior to the lesion and, thus, fundamentalism changes comparing pre- and postlesion could not be directly measured. Future studies with diverse patient samples and both pre- and postlesion fundamentalism data are warranted to assess the generalizability and validity of these conclusions. Finally, it is possible that our lesion network mapping results may be improved by using a normative connectome that is better matched to the average age of our lesion patients. However, prior work from our group has shown that using age-matched or disease-matched connectomes produces negligible differences regarding network mapping (37, 57, 58). Nevertheless, future research examining this issue is warranted.
In sum, brain lesions associated with greater religious fundamentalism are characterized by a unique pattern of brain connectivity: a religious fundamentalism network. Specifically, lesion connectivity to the right orbitofrontal cortex, dorsolateral prefrontal cortex, and inferior parietal lobe corresponds to high fundamentalism attitudes. Our lateralized result is consistent with well-established hemispheric specializations in reasoning. This network spatially resembles brain lesion networks associated with pathological confabulation and criminal behavior. Overall, these results unify previously published reports on the neural substrates for religious fundamentalism and may prove useful for understanding neurobiological factors that influence the development of fundamentalism and its associated cognitive and emotional profile.
Materials and Methods
Subjects.
One hundred and ninety patients with brain damage from the W.F. Caveness Vietnam Head Injury Study Registry (VHIS) and the Iowa Neurological Patient Registry contributed to the two datasets of our study. Dataset 1 had 106 patients with penetrating traumatic brain injury who were male combat veterans from VHIS during phase 4 (40 to 45 y after injury) conducted at the National Institute of Neurological Disorders and Stroke (NINDS) in Bethesda, MD (45). Patients in dataset 1 self-identified religious affiliations included Protestant (39%), Roman Catholic (16%), other affiliation (12%), and 33% did not identify with a religious affiliation (5). Dataset 1 patients’ age ranged from 53 to 75 y (M = 58.3, SD = 3.0) at the time of testing. Lesion size in MNI space for dataset 1 ranged from 0.10 to 275.42 cubic centimeters (cc; M = 38.25, SD = 41.22). Dataset 2 had 84 patients from a demographically homogenous region (mainly rural Iowa) who experienced brain lesions from varied etiologies: stroke (51%), surgical resection for epilepsy or benign tumors (37%), focal contusion from traumatic brain injury (8%), and 4% had brain damage from a viral infection or a genetic condition. Patients in dataset 2 self-identified religious affiliations included Protestant (44%), Roman Catholic (25%), other affiliation (18%), and 13% did not identify with a religious affiliation. Dataset 2 patients’ age ranged from 23 to 80 y (M = 56.2, SD = 12.7) and 42 patients identified as male with 42 identifying as female. Lesion size in MNI space for dataset 2 ranged from 0.95 to 229.17 cc (M = 45.74, SD = 49.80). Chronicity in dataset 2 ranged from less than a year to 45 y (M = 10.9, SD = 8.7). All participants gave written informed content prior to data collection.
Religious Fundamentalism Assessment.
Religious fundamentalism was measured using the RFS (1, 2). The RFS is a widely used, well-validated scale with excellent internal consistency (Cronbach’s alpha > 0.90) (2), and is a standardized psychometric instrument assessing religious fundamentalism. Patients recorded responses to RFS statements on a Likert scale (Strongly agree to Strongly disagree). The RFS defines fundamentalism with four dimensions: 1) there is one set of religious teachings that contain the fundamental, inerrant truth about humanity and deity; 2) this truth is opposed to evil which must be actively fought; 3) this truth must be followed today according to the fundamental practices of the past; and 4) those who follow these fundamental teachings have a special relationship with the deity. Patients in dataset 1 were administered a randomized and balanced 10-item version of the 1992 RFS (1, 5) while patients in dataset 2 were administered the balanced 12-item 2004 version of the RFS (2, 4) (SI Appendix, Supplementary Methods). The balanced items of the scale rule out the possibility that liberal responding, per se, produces a high or low fundamentalism score.
Neuroanatomical Analyses.
Neuroimaging data in all patients were obtained in the chronic epoch, 3 or more months following lesion onset. Brain lesion neuroanatomical identification in dataset 1 patients was based on computerized tomography (CT) scans for each patient (5). The Analysis of Brain Lesion (ABLe) software version 2.8b (59) implemented in MEDx version 3.44 (Medical Numerics, Germantown, MD) was used to calculate lesion location and volume loss. Each CT scan was then spatially normalized to a CT template brain image in Montreal Neurological Institute (MNI-152) template space (resolution, 1 mm3) by applying an automated image registration algorithm (60) for registration accuracy. Brain lesion neuroanatomical identification in dataset 2 was based on either on CT or MRI scans for each patient. Each patient’s lesion was reconstructed in three dimensions using Brainvox, and the lesion contour was manually warped into a template brain creating a mask using the MAP-3 method (61). Lesion masks were warped and resampled from the template brain local standard space to MNI-152 template space using a symmetric normalization algorithm in ANTS (62). All lesions in both datasets were aligned to the same MNI-152 template space.
Lesion Network Mapping.
To maximize the statistical power in our analysis, we combined both independent datasets (N1 = 106; N2 = 84) and performed voxelwise statistical tests for lesion network associations with RFS scores (Ncombined = 190). We used lesion network mapping and previously validated methods to derive a brain network for religious fundamentalism in a data-driven fashion (33). First, resting-state functional connectivity between each lesion and the rest of the brain was computed using a publicly available normative connectome dataset from 1,000 healthy right-handed subjects (42.7% male subjects, ages 18 to 35 y, mean age 21.3 y). The connectome dataset was processed in accordance with the lesion network mapping procedure (63) including global signal regression, which results in a map of brain regions functionally connected to each lesion location referred to as a lesion network. In a voxelwise fashion for both independent datasets, a group-level map of brain associations with religious fundamentalism was derived. This was done using voxelwise permutation analysis of linear models (PALM) with RFS scores as a psychometric covariable (Fig. 2). We performed a family-wise error (FWE) multiple comparison correction on our results and tested for voxels that survived a conservative correction threshold of FWE P < 0.05. Coordinates for peak values in clusters surviving FWE correction were identified using MRIcroGL and the Automated Anatomical Labeling (AAL) atlas.
Cross-Validations Analyses.
Cross-validation testing was performed in two ways. First, spatial correlations were calculated between each individual patients’ brain lesion network maps compared with the group-level fundamentalism network map from the opposing dataset. Specifically, the group-level fundamentalism map derived from dataset 2 was spatially correlated with each individual patient’s lesion network map from dataset 1, and the group-level fundamentalism map derived from dataset 1 was spatially correlated with each individual patient’s lesion network map from dataset 2. This quantified the similarity of each patient’s lesion network to an independent group-level fundamentalism network derived from the opposing dataset. Each of these patient-to-group network similarity scores were then correlated with patients’ own RFS scores to test whether the similarity of an individual patients’ lesion network map to an independently derived fundamentalism network corresponded to patients’ psychometric scores for religious fundamentalism. Second, we calculated a network damage score for each patients’ brain lesion location relative to the opposite datasets’ group-level fundamentalism network. This was done by superimposing each patient’s brain lesion image onto the fundamentalism network map from the opposite dataset, then calculating the arithmetic sum of the statistical values for each voxel in the group-level fundamentalism map that was circumscribed by the patient’s lesion image (33). These network damage scores were then correlated with the patients’ RFS scores to test whether focal damage to a fundamentalism network map corresponded to patients’ psychometric scores for religious fundamentalism.
Analyses for Religious Fundamentalism Network and Lesions Associated with Behavioral, Neurological, and Psychiatric Conditions.
We compared the religious fundamentalism brain network to independent sets of previously published brain lesions and lesion networks associated with behavioral, neurological, and psychiatric conditions from a library of 899 lesions spanning 21 conditions. We used two methods of comparison in these analyses. First, we assessed the extent of the intersection between individual condition-associated lesions and the unthresholded fundamentalism network to compute a network damage score (33): Individual condition–associated lesions were superimposed on the combined dataset religious fundamentalism network, and the average of the t-value for voxels circumscribed within each lesion was calculated. A one-sample t test was then performed for network damage score values associated with brain lesion traces within a set of lesions sharing a common symptom. These t test results were then corrected for multiple comparisons across the 21 sets of condition-associated lesions. Second, we compared the spatial topography of the individual lesion network maps from each condition to the unthresholded religious fundamentalism network. Degree of spatial similarity to the fundamentalism network with condition-associated lesion networks was then calculated by performing a Pearson correlation between vectorized maps of our religious fundamentalism network and each of the lesion network maps for the 899 condition-associated lesions in our brain lesion library. A one-sample t test was then performed for spatial similarity values associated with lesion networks within a set of lesions sharing a common condition. These t test results were then corrected for multiple comparisons across the 21 sets of condition-associated lesions.
Supplementary Material
Appendix 01 (PDF)
Acknowledgments
We would like to note that our personal beliefs span a broad continuum from adherents of religious faiths through agnosticism to atheism. We approach the weighty subject matter of this research as earnest seekers of scientific data and encourage readers to receive our results in the spirit of open-minded empirical inquiry driven by scientific curiosity and without prejudice or malice to any group or faith. M.A.F. is supported by the Templeton World Charity Foundation, the Osher Center for Integrative Health at Harvard Medical School, and a generous donation from the Center for Emergence. E.W.A. is supported by a generous donation from Dr. Jerry Artz. I.K. is supported by the NIH (NINDS L30 NS134024). F.L.W.V.J.S. is supported by the American Epilepsy Society (846534) and the NIH (NINDS R01 NS127892). A.D.B. is supported by the NIH (NINDS R01 NS114405) and the Roy. J. Carver Trust. J.G. receives generous support from the John. F. Templeton Foundation, the Therapeutic Cognitive Neuroscience Fund (John Hopkins Medical School-Barry Gordon, M.D., Ph.D.) and the Rosenberg/Carlin Fund. M.D.F. is supported by grants from the NIH (R01MH113929, R21MH126271, R56AG069086, R21NS123813, R01NS127892, R01MH130666, UM1NS132358), Neuronetics, the Kaye Family Research Endowment, the Ellison / Baszucki Family Foundation, and the Manley Family. M.D.F. has intellectual property on the use of brain connectivity imaging to analyze lesions and guide brain stimulation, is a consultant for Magnus Medical, Soterix, Abbott, and Boston Scientific, and has received research funding from Neuronetics.
Author contributions
M.A.F., E.W.A., D.T., and M.D.F. designed research; M.A.F., E.W.A., A.D.B., J.M.N., F.L.W.V.J.S., S.S., J.R.B., and J.G. performed research; M.A.F. and M.D.F. contributed new reagents/analytic tools; M.A.F., E.W.A., I.K., and J.I.T. analyzed data; and M.A.F., E.W.A., I.K., D.T., A.D.B., J.M.N., F.L.W.V.J.S., S.S., J.S.A., J.A.N., J.R.B., J.G., and M.D.F. wrote the paper.
Competing interests
The authors declare no competing interest.
Footnotes
This article is a PNAS Direct Submission. L.Q.U. is a guest editor invited by the Editorial Board.
Data, Materials, and Software Availability
Supporting Information
References
- 1.Altemeyer B., Hunsberger B., Authoritarianism, religious fundamentalism, quest, and prejudice. Int. J. Psychol. Relig. 2, 113–133 (1992). [Google Scholar]
- 2.Altemeyer B., Hunsberger B., A revised religious fundamentalism scale: The short and sweet of it. Int. J. Psychol. Relig. 14, 47–54 (2004). [Google Scholar]
- 3.Altemeyer B., Hunsberger B., “Fundamentalism and authoritarianism” in Handbook of the Psychology of Religion and Spirituality, Paloutzian R. F., Park C. L., Eds. (Guilford Press, 2005), pp. 378–393. [Google Scholar]
- 4.Asp E. W., Ramchandran K., Tranel D., Authoritarianism, religious fundamentalism, and the human prefrontal cortex. Neuropsychology 26, 414–421 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zhong W., Cristofori I., Bulbulia J., Krueger F., Grafman J., Biological and cognitive underpinnings of religious fundamentalism. Neuropsychologia 100, 18–25 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Altemeyer B., The Authoritarian Specter (Harvard University Press, 1996). [Google Scholar]
- 7.Zmigrod L., Rentfrow P. J., Zmigrod S., Robbins T. W., Cognitive flexibility and religious disbelief. Psychol. Res. 83, 1749–1759 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Weeks M., Geisler S., Revisiting cognitive complexity of religious topics: Multiple complexity, religious fundamentalism, and a quest orientation. Psychol. Relig. Spiritual. 11, 433–441 (2019). [Google Scholar]
- 9.Hunsberger B., Alisat S., Pancer S. M., Pratt M., Religious fundamentalism and religious doubts: Content, connections, and complexity of thinking. Int. J. Psychol. Relig. 6, 201–220 (1996). [Google Scholar]
- 10.Asp E. W., Tranel D., “False Tagging Theory: Toward a unitary account of prefrontal cortex function” in Principles of Frontal Lobe Function, Stuss D. T., Knight R. T., Eds. (Oxford University Press, 2013), pp. 383–416. [Google Scholar]
- 11.Bronstein M. V., Pennycook G., Bear A., Rand D. G., Connon T. D., Belief in fake news is associated with delusionality, dogmatism, religious fundamentalism, and reduced analytic thinking. J. Appl. Res. Mem. Cogn. 8, 108–117 (2019). [Google Scholar]
- 12.Sobol M., Zajenkowski M., Jankowski K. S., Religious fundamentalism, delusions, and conspiracy beliefs related to the COVID-19 pandemic. Int. J. Environ. Res. Public Health 19, 1–6 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Merkley E., Loewen P. J., Anti-intellectualism and the mass public’s response to the COVID-19 pandemic. Nat. Hum. Behav. 5, 706–715 (2021). [DOI] [PubMed] [Google Scholar]
- 14.Oliver J. E., Rahn W. M., Rise of the Trumpenvolk populism in the 2016 election. Ann. Am. Acad. Polit. Soc. Sci. 667, 189–206 (2016). [Google Scholar]
- 15.Saroglou V., et al. , Fundamentalism as dogmatic belief, moral rigorism, and strong groupness across cultures: Dimensionality, underlying components, and related interreligious prejudice. Psychol. Relig. Spiritual. 14, 558–571 (2020). [Google Scholar]
- 16.Koopmans R., Religious fundamentalism and hostility against out-groups: A comparison of Muslims and Christians in Western Europe. J. Ethn. Migr. Stud. 41, 33–57 (2015). [Google Scholar]
- 17.Jackson L. M., Hunsberger B., An intergroup perspective on religion and prejudice. J. Sci. Study Relig. 38, 509–523 (1999). [Google Scholar]
- 18.Lueke N. A., Lueke A. K., Aghababaei N., Ferguson M. A., Bushman B. J., Fundamentalism and intrinsic religiosity as factors in well-being and social connectedness: An Iranian study. Psychol. Relig. Spiritual. 15, 398–406 (2021). [Google Scholar]
- 19.Beer J. M., Arnold R. D., Loehlin J. C., Genetic and environmental influences on MMPI factor scales: Joint model fitting to twin and adoption data. J. Pers. Soc. Psychol. 74, 818–827 (1998). [DOI] [PubMed] [Google Scholar]
- 20.Coreno T., Fundamentalism as a class culture. Sociol. Relig. 63, 335–360 (2002). [Google Scholar]
- 21.Wilson J., Sandomirsky S., Religious affiliation and the family. Sociol. Forum 6, 289–309 (1991). [Google Scholar]
- 22.Devinsky O., Lai G., Spirituality and religion in epilepsy. Epilepsy Behav. 12, 636–643 (2008). [DOI] [PubMed] [Google Scholar]
- 23.Griffiths R. R., Richards W. A., McCann U., Jesse R., Psilocybin can occasion mystical-type experiences having substantial and sustained personal meaning and spiritual significance. Psychopharmacology 187, 268–283 (2006). [DOI] [PubMed] [Google Scholar]
- 24.Azari N. P., et al. , Neural correlates of religious experience. Eur. J. Neurosci. 13, 1649–1652 (2001). [DOI] [PubMed] [Google Scholar]
- 25.Beauregard M., Paquette V., EEG activity in Carmelite nuns during a mystical experience. Neurosci. Lett. 444, 1–4 (2008). [DOI] [PubMed] [Google Scholar]
- 26.Button T. M., Stallings M. C., Rhee S. H., Corley R. P., Hewitt J. K., The etiology of stability and change in religious values and religious attendance. Behav. Genet. 41, 201–210 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Koenig L. B., McGue M., Krueger R. F., Bouchard T. J., Genetic and environmental influences on religiousness: Findings for retrospective and current religiousness ratings. J. Pers. 73, 471–488 (2005). [DOI] [PubMed] [Google Scholar]
- 28.Bouchard T. J., et al. , “Genetic influence on social attitudes: Another challenge to psychology from behavior genetics” in Behavior Genetic Principles: Perspectives in Development, Personality, and Psychopathology, DeLilla L. F., Ed. (American Psychological Association, 2004), pp. 89–104. [Google Scholar]
- 29.Ludeke S., Johnson W., Bouchard T. J., Obedience to traditional authority: A heritable factor underlying authoritarianism, conservatism and religiousness. Pers. Individ. Differ. 55, 375–380 (2013). [Google Scholar]
- 30.Pennycook G., Tranel D., Warner K., Asp E. W., “Beyond reasonable doubt: Cognitive and neuropsychological implications for religious disbelief” in Neurology and Religion, Coles A., Collicutt J., Eds. (Cambridge University Press, 2019), pp. 115–129. [Google Scholar]
- 31.Siddiqi S. H., Kording K. P., Parvizi J., Fox M. D., Causal mapping of human brain function. Nat. Rev. Neurosci. 23, 361–375 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Vaidya A. R., Pujara M. S., Petrides M., Murray E. A., Fellows L. K., Lesion studies in contemporary neuroscience. Trends Cogn. Sci. 23, 655–673 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Ferguson M. A., et al. , A neural circuit for spirituality and religiosity derived from patients with brain lesions. Biol. Psychiatry 91, 380–388 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Urgesi C., Aglioti S. M., Skrap M., Fabbro F., The spiritual brain: Selective cortical lesions modulate human self-transcendence. Neuron 65, 309–319 (2010). [DOI] [PubMed] [Google Scholar]
- 35.Petrides M., Pandya D. N., “Association pathways of the prefrontal cortex and functional observations” in Principles of Frontal Lobe Function, Stuss D. T., Knight R. T., Eds. (Oxford University Press, 2002), pp. 31–50. [Google Scholar]
- 36.Uddin L. Q., Yeo B. T., Spreng R. N., Towards a universal taxonomy of macro-scale functional human brain networks. Brain Topogr. 32, 926–942 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Boes A. D., et al. , Network localization of neurological symptoms from focal brain lesions. Brain 138, 3061–3075 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Palesi F., et al. , Contralateral cortico-ponto-cerebellar pathways reconstruction in humans in vivo: Implications for reciprocal cerebro-cerebellar structural connectivity in motor and non-motor areas. Sci. Rep. 7, 12841 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Bateman J. R., et al. , Network localization of spontaneous confabulation. J. Neuropsychiatry Clin. Neurosci. 36, 45–52 (2023). [DOI] [PubMed] [Google Scholar]
- 40.Moscovitch M., “Confabulation and the frontal systems: Strategic versus associative retrieval in neuropsychological theories of memory” in Varieties of Memory and Consciousness: Essays in Honour of Endel Tulving, Roediger H. L., Craik F., Eds. (Erlbaum, 1989), pp. 133–160. [Google Scholar]
- 41.Baddeley A., Wilson B., “Amnesia, autobiographical memory, and confabulation” in Autobiographical Memory, Rubin D., Ed. (Cambridge University Press, 1986), pp. 225–252. [Google Scholar]
- 42.Ciaramelli E., Ghetti S., Frattarelli M., Làdavas E., When true memory availability promotes false memory: Evidence from confabulating patients. Neuropsychologia 44, 1866–1877 (2006). [DOI] [PubMed] [Google Scholar]
- 43.Juergensmeyer M., Terror in the Mind of God: The Global Rise of Religious Violence (University of California Press, 2017). [Google Scholar]
- 44.Wiech K., et al. , An fMRI study measuring analgesia enhanced by religion as a belief system. Pain 139, 467–476 (2008). [DOI] [PubMed] [Google Scholar]
- 45.Raymont V., Salazar A. M., Krueger F., Grafman J., “Studying Injured Minds”—The Vietnam Head Injury Study and 40 years of brain injury research. Front. Neurol. 2, 1–13 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Tranel D., “Theories of clinical neuropsychology and brain-behavior relationships: Luria and beyond” in Textbook of Clinical Neuropsychology, Morgan J. E., Ricker J. H., Eds. (Taylor and Francis, 2007), pp. 27–39. [Google Scholar]
- 47.Damasio H., “The lesion method in cognitive neuroscience” in Handbook of Neuropsychology, Boller F., Grafman J., Rizzolatti G., Eds. (Elsevier, 2000), vol. 1, pp. 77–102. [Google Scholar]
- 48.Keenan J. P., Rubio J., Racioppi C., Johnson A., Barnacz A., The right hemisphere and the dark side of consciousness. Cortex 41, 695–704 (2005). [DOI] [PubMed] [Google Scholar]
- 49.Zwir I., et al. , Evolution of genetic networks for human creativity. Mol. Psychiatry 27, 354–376 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Filley C. M., Kletenik I., Churchland P. S., Morality and the brain: The right hemisphere and doing right. Cogn. Behav. Neurol. 33, 304–307 (2020). [DOI] [PubMed] [Google Scholar]
- 51.Turner B. O., Marinsek N., Ryhal E., Miller M. B., Hemispheric lateralization in reasoning. Ann. N. Y. Acad. Sci. 1359, 47–64 (2015). [DOI] [PubMed] [Google Scholar]
- 52.Gazzaniga M. S., “Brain modules and belief formation” in Self and Consciousness (Psychology Press, 2014), pp. 88–102. [Google Scholar]
- 53.Wolford G., Miller M. B., Gazzaniga M. S., The left hemisphere’s role in hypothesis formation. J. Neurosci. 20, 1–4 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Marcel A. J., Tegner R., Nimmo-Smith I., Anosognosia for plegia: Specificity, extension, partiality, and disunity of bodily unawareness. Cortex 40, 19–40 (2004). [DOI] [PubMed] [Google Scholar]
- 55.Feinberg T. E., Roane D. M., Kwan P. C., Schindler R. J., Haber L. D., Anosognosia and visuoverbal confabulation. Arch. Neurol. 51, 468–473 (1994). [DOI] [PubMed] [Google Scholar]
- 56.Vocat R., Staub F., Stroppini T., Vuilleumier P., Anosognosia for hemiplegia: A clinical–anatomical prospective study. Brain 133, 3578–3597 (2010). [DOI] [PubMed] [Google Scholar]
- 57.Darby R. R., Horn A., Cushman F., Fox M. D., Lesion network localization of criminal behavior. Proc. Natl. Acad. Sci. U. S. A. 115, 601–606 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Fox M. D., et al. , Resting-state networks link invasive and noninvasive brain stimulation across diverse psychiatric and neurological diseases. Proc. Natl. Acad. Sci. U.S.A. 111, E4367–E4375 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Makale M., et al. , Quantification of brain lesions using interactive automated software. Behav. Res. Methods Instrum. Comput. 34, 6–18 (2002). [DOI] [PubMed] [Google Scholar]
- 60.Woods R. P., Grafton S. T., Holmes C. J., Cherry S. R., Mazziotta J. C., Automated image registration: I. General methods and intrasubject, intramodality validation. J. Comput. Assist. Tomogr. 22, 139–152 (1998). [DOI] [PubMed] [Google Scholar]
- 61.Frank R. J., Damasio H., Grabowski T. J., Brainvox: An interactive, multimodal visualization and analysis system for neuroanatomical imaging. Neuroimage 5, 13–30 (1997). [DOI] [PubMed] [Google Scholar]
- 62.Avants B., Gee J. C., Geodesic estimation for large deformation anatomical shape averaging and interpolation. Math. Brain Imaging 23, S139–S150 (2004). [DOI] [PubMed] [Google Scholar]
- 63.Fox M. D., Mapping symptoms to brain networks with the human connectome. N. Engl. J. Med. 379, 2237–2245 (2018). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix 01 (PDF)