The risk of radicalization: A neuroscientific perspective on key
psychological traits and cognitive processes
Literature thesis
MSc Brain and Cognitive Sciences
Institute for Interdisciplinary Sciences
Universiteit van Amsterdam
Supervisor: dhr. prof. dr. J.H. (Jan Henk) Kamphuis
Co-assessor: dhr. dr. A.R. (Allard) Feddes
D. de Bruin
10594884
19-12-2019
Table of Contents
Abstract 2
1 Introduction 2
1.1 Defining radicalization and terrorism 3
1.2 Explaining radicalization and terrorism 3
1.2.1 Terrorism as a psychopathology 3
1.2.2 A predominant social psychological model of radicalization 4
1.2.3 Individual traits to explain radicalization 4
2 Radicalization as a social process 6
2.1 The neural underpinnings of social reward processing 6
2.2 The impact of not feeling connected: exclusion and discrimination 7
2.3 Social connections: ingroup versus outgroup 8
2.3.1 Connectedness with the ingroup 8
2.3.2 Distance to the outgroup 8
2.4 Being social and radicalization 9
3 Psychological traits associated with radicalization 11
3.1 Justice sensitivity 11
3.1.1 Personality traits associated with justice sensitivity 11
3.1.2 Neural correlates of justice sensitivity 12
3.1.3 Justice sensitivity and radicalization 12
3.2 Self-esteem and the quest for significance 13
3.2.1 Personality traits associated with self-esteem and the quest for significance 13 3.2.2 Neural correlates of self-esteem and the quest for significance 14 3.2.3 Self-esteem, the quest for significance and radicalization 15
3.3 Psychological traits and radicalization 16
4 Cognitive traits associated with radicalization 17
4.1 Neural correlates of cognitive flexibility and dogmatism 17
4.2 Neural correlates of cognitive complexity and need for structure 17
4.3 Neural correlates of cognitive capacity and cognitive control 18
4.4 Cognitive traits and radicalization 19
5 Discussion and conclusion 21
The risk of radicalization: A neuroscientific perspective on key
psychological traits and cognitive processes
Abstract
As radicalization is a complex process with large interindividual differences in the factors driving it, studying the underlying processes has proved to be challenging. Scholars agree that there is not one “terrorist profile”, but research instead suggests that there are traits that can be associated with an increased risk of radicalization. This literature review therefore focusses on the psychological and cognitive traits linked to radicalization, and their neural underpinnings. Radicalization is often seen as a social process, with the need for social belonging being a strong motivator to join a radicalized group. Feelings of exclusion can result in more antisocial behavior, more extremist viewpoints, and more violence towards outgroups. Higher sensitivity to social group dynamics can thus be linked to radicalization. When looking at psychological traits, justice sensitivity and the need for significance are often associated with radicalization. Differences in cognitive traits might also explain why some people are more prone to radicalize. The straightforward, black-and-white nature of extremist ideologies can be more attractive for individuals with higher cognitive inflexibility and less cognitive complexity. The beforementioned psychological and cognitive traits can all be linked to individual differences in neural processing, including altered reward processing, cognitive control and dopamine sensitivity. These findings suggest that neural and neurobiological correlates might play a role in the process of radicalization. Despite radicalization being a heterogenous process with many different factors affecting it, this review thus suggests that psychological and cognitive traits and their underlying neural mechanisms can help explain why some individuals radicalize, and others do not.
1 Introduction
Why do some people become radicalized, whereas others do not? With an increase in terroristically motivated incidents in the European Union over the past years (Europol, 2019), understanding radicalization and terrorism is a priority for governments worldwide (UNESCO, 2017). For years, scholars of different disciplines have been trying to construct a “terrorist profile” (Russell & Miller, 1983), but it has become clear that there is no typical profile of the radicalized (Horgan, 2008; Horgan, 2014). Terrorism and radicalization are complex societal problems, influenced by multiple factors and different for every individual (Victoroff, 2005). Despite this variability in routes to radicalization, the quest for common factors leading to radicalization has suggested that there are similarities between individuals that radicalize (e.g. Russell & Miller, 1983; Borum, 2003; Precht, 2007; Silber & Bhatt, 2007; McCauley & Moskalenko, 2008; Borum, 2011; King & Taylor, 2011; Doosje et al., 2016; van den Bos, 2019). The majority of research has focused on the demographic, economic and social factors that might lead to radicalization, whereas less focus has been on individual psychological processes and traits that might be associated with a higher risk of radicalization and violent behavior (King & Taylor, 2008). It thus remains unclear if there are certain psychological processes, personality traits or cognitive traits that make individuals more prone to radicalize. The neuroscience behind radicalization and terrorism is even less clear (Decety et al., 2018).
Understanding why some people radicalize whereas others do not can be very helpful in countering radicalization and in deradicalizing individuals with extreme viewpoints. If we know what makes individuals more likely to adopt extremist viewpoints or to join a terrorist organization, this knowledge can be used to develop more effective intervention strategies.
The aim of this literature review is to look into psychological and cognitive traits associated with a higher risk of radicalization and their neural underpinnings. In so doing, this review addresses the question why some people radicalize, whereas other do not from a psychological and neuroscientific point of view. First, radicalization and terrorism will be defined and a few prominent psychological theories on
radicalization will be briefly discussed. This review will then look at radicalization from three perspectives: a) radicalization as a social process, b) psychological traits associated with radicalization and c) cognitive traits associated with radicalization. The neural underpinnings of these processes and personality traits linked to them will be discussed.
1.1 Defining radicalization and terrorism
Finding a consensus on the definition of terrorism has turned out to be difficult (Cooper, 2001), with different scholars using different definitions (Schmid, 2011). In this review, the definition of terrorism as proposed by Borum (2004) is used: acts of violence intentionally perpetrated on civilian
non-combatants with the goal of furthering some ideological, religious or political objective. Terrorism
thus involves intentionally using violence. This violence is directed towards outgroup members, with the aim of protecting one’s own group (the ingroup) (McCauley & Moskalenko, 2008). As mentioned before, terrorism can be the result of ideological, religious or political beliefs. These ideologies often manifest themselves in terrorist groups.
Terrorism is often preceded by a process referred to as radicalization. Radicalization can be defined as
the process of developing extremist ideologies and beliefs (Borum, 2012). McCauley & Moskalenko
(2008) describe radicalization as a change in beliefs, feelings, and behaviors in directions that
increasingly justify intergroup violence and demand sacrifice in defense of the ingroup. Some scholars
make a differentiation between cognitive radicalization and behavioral radicalization. Cognitive radicalization refers to the development of extremist beliefs, whereas behavioral radicalization refers to extremist behavior (Neumann, 2013). Therefore, terrorism could be seen as a form of behavioral radicalization. Whereas in most cases behavioral extremism follows from cognitive extremism, this is not always the case (Guhl, 2018; Nilsson, 2018). Likewise, cognitive radicalization does not always lead to behavioral radicalization (Guhl, 2018). Understanding which factors influence the development of cognitive radicalization into behavioral radicalization and the use of violence is one of the top priorities in the field of terrorism research (UNESCO, 2017).
1.2 Explaining radicalization and terrorism
Throughout the years, terrorism has been studied from different theoretical perspectives and disciplines (Victoroff, 2005; Trip et al., 2019). Radicalization and the development of extremist beliefs and behavior are thought to be influenced by multiple factors, including demographics, economic status, social groups, cultural and political beliefs and psychological processes (Hudson, 1999; Gottschalk & Gottschalk, 2004; Victoroff, 2005; Silke, 2008). These factors can be studied on different levels: micro (individual), meso (group) and macro (societal) level (Doosje et al., 2016). Due to the heterogenous and multilevel nature of radicalization and the large interindividual variability between individuals who radicalize, studying it has been proven to be difficult (King & Taylor, 2011). Despite these challenges, there are studies that have tried to identify common factors in the process of radicalization. The focus of this review will be on the psychological processes and individual traits underlying these processes. First, two of the most influential psychological theories on radicalization will be briefly discussed.
1.2.1 Terrorism as a psychopathology
Theories of terrorism vary in the degree to which they use psychological traits or processes to explain why people radicalize. One widely studied theory explains terrorism and radicalization as the result of psychopathologies (Victoroff, 2005). Around the 1970s, it was proposed that terrorism was the result of psychopathy (Victoroff, 2005). Terrorists were suggested to be paranoid, crazy, or otherwise mentally ill. The range of psychopathological disorders suggested to be associated with terrorism has since then broadened. Traits like pathological hatred, authoritarianism (Gottschalk & Gottschalk, 2004), narcissism (Pearlstein, 1991), and other psychological disorders like schizophrenia and borderline personality disorder have since been linked to terrorism (Weenink, 2015).
More recently, researchers have started to make a distinction between different types of terrorists (Schulten et al., 2019). The most common form of terrorism is group terrorism, in which individuals perceive problems in society which motivates them to join a terrorist group (Gill, 2012; Doosje et al., 2016). Different from the group terrorists are the lone actors or lone wolves. These terrorists operate independently from an organization and carry out their terroristic attacks individually (Spaaij, 2010). When looking at these types of terrorists separately, research suggests that psychopathology does not explain group terrorism (Silke, 1998; Victoroff, 2005; Borum, 2014; Gill & Corner, 2017; Schulten
et al., 2019), but that psychopathological disorders can be linked to lone wolf terrorism (Spaaij, 2010;
Schulten et al., 2019). In line with this, research by Corner & Gill (2015) suggests that the odds of lone actor terrorists having a mental illness are 13.5 times higher than the odds of a group actor having a mental illness. Compared to the general population, the prevalence of psychopathological disorders has been found to be higher in lone actor terrorists. More specifically, schizophrenia, delusional disorders and autism spectrum disorders were found to be more prevalent within this group of terrorists (Schulten et al., 2019). These findings thus suggest that psychopathology can play a role in the radicalization of lone wolf terrorists. However, as was mentioned before, the most common form of terrorism is group terrorism. Due to the lack of scientific evidence supporting an association between this type of terrorism and psychopathology, the psychopathological theories of terrorism have lost popularity over the years (Gill & Corner, 2017; Piccinni et al., 2018).
1.2.2 A predominant social psychological model of radicalization
As the majority of scholars now agree that terrorism is not a psychopathology and that radicalization can happen to “normal” individuals (Silke, 1998; Victoroff, 2005; Doosje et al., 2016; Gill & Corner, 2017; Piccinni et al., 2018), theories now try to explain which factors drive these individuals to terrorism. There are many different psychological models that try to explain the process of radicalization. Common across models is the idea that radicalization is a diffuse process that takes time and that involves different factors and dynamics (Neumann, 2013). Understanding why people radicalize could also help to deradicalize these individuals (Kruglanski et al., 2014; Doosje et al., 2016).
One of the most influential psychological models of radicalization and terrorism is the staircase model proposed by Moghaddam (2005). This model tries to explain why some people radicalize towards terrorism by using a six-floor staircase as a metaphor. Individuals who progress from the ground floor to the first floor perceive injustice and relative deprivation and are seeking ways to achieve greater justice. Perceiving injustice and relative deprivation has been suggested as an important factor in the process of radicalization by many scholars (Doosje et al., 2013; Borum, 2014; van den Bos, 2019). Individuals on the first floor progress to the second floor if they feel like they cannot do anything about the injustice, which leads to feelings of aggression towards outgroups or authorities. On the third floor, these individuals start to justify violence to reach their goal and start engaging more with terrorist organizations. Actual recruitment to terrorist organizations takes place on the fourth floor. On the fifth and final floor, individuals are equipped and prepared to commit terrorist acts
(Moghaddam, 2005).
Like the model of Moghaddam, other models describe similar radicalization processes in which some, but not all individuals progress to more extreme forms of radicalization and terrorism (Borum, 2003; Precht, 2007; Silber & Bhatt, 2007; McCauley & Moskalenko, 2008; Borum, 2011; King & Taylor, 2011; Doosje et al., 2016). The progression to terrorism through radicalization can thus be seen as a process involving different steps, factors and dynamics.
1.2.3 Individual traits to explain radicalization
Most scholar now agree that there is no “terrorist profile” (Horgan, 2008; Horgan, 2014) and that most individuals that radicalize are remarkably normal (Silke, 1998; Victoroff, 2005; Doosje et al., 2016; Gill & Corner, 2017). However, the majority of theories and models on radicalization do agree that there are certain common processes or factors that are associated with radicalization (Horgan, 2008). Therefore, the aim of this review is to look at psychological and cognitive traits that have been
associated with radicalization, and that might explain why some individuals are more prone to radicalize than others.
It is important to first define what is meant by psychological and cognitive traits. Psychological traits are used to describe individual differences in psychological processes. Personality traits are qualities or characteristics that define an individual, his or her actions and attitude. Cognitive traits are used to describe individual differences in cognitive processes, which refers to mental processes rather than behavior or attitude. These different types of traits are all intertwined and the distinction might not always be clear, but it does provide some structure in studying the process of radicalization. All traits are present in every individual, but individuals differ in the strength of these traits. Every individual can be placed somewhere on the spectrum for every trait (Harkness & Lilienfeld, 1997). In this review, the most common distinction is low versus high levels of certain traits.
There are many social psychological theories on the process of radicalization (Victoroff, 2005, Trip et
al., 2019), but research focusing specifically on psychological and cognitive traits linked to
radicalization is scarce (King & Taylor, 2008). The in this review discussed traits were selected after reviewing the existing literature on psychological and cognitive theories of radicalization. As radicalization is often suggested to be a social process (Precht, 2007; Silber & Bhatt, 2007; McCauley & Moskalenko, 2008; Schmid, 2013), this review will first look at individual differences in social processing. Based on social psychological theories, both justice sensitivity (Borum, 2003; Eidelson & Eidelson, 2003; Moghaddam, 2005; Precht, 2007; van Zomeren et al., 2008; Davis & Cragin, 2009; Borum, 2011; King & Taylor, 2011; Doosje et al., 2013; Doosje et al., 2016; van den Bos, 2019) and the quest for significance (Hogg, 2007; Kruglanski & Orehek, 2011; Hogg & Adelman, 2013; Hogg et al., 2013; Kruglanski et al., 2013; Dugas & Kruglanski, 2014; Kruglanski et al., 2014; Doosje et al., 2016; Lyons-Padilla et al., 2016; Jasko et al., 2016; Kruglanski et al., 2018; van den Bos, 2019) are important factors in the process of radicalization, and therefore this review will focus on these two psychological traits. There are some studies that have linked differences in cognitive processing to radicalization (Victoroff, 2005; Wiktorowicz, 2005), including differences in cognitive flexibility (Victoroff, 2005; Borum, 2014; van den Bos, 2019; Zmigrod et al., 2019), cognitive complexity (Neuberg & Newson, 1993; Victoroff, 2005; Borum, 2014; Soliman et al., 2016; van den Bos, 2019) and cognitive capacity (Victoroff, 2005). Based on the earlier suggested association between these cognitive traits and radicalization, these traits will be discussed more in-depth in this review.
Interestingly, the trait approach can also be linked to the psychopathological theories on terrorism and radicalization. Psychopathologies are diagnosed based on the presence of certain traits, and it is thus possible that certain psychological traits are more strongly associated with terroristic behavior rather than the complete psychopathology. This stresses the importance of looking at traits rather than at psychopathologies and could explain the mixed findings on the link between psychopathologies and radicalization (Weenink, 2015; GIll & Corner, 2017; Piccinni et al., 2018; Schulten et al., 2019).
Many researchers have been trying to understand the neurobiology and neural correlates underlying psychological and cognitive traits. Cloninger’s (1987) three-dimensional personality model of temperament links specific neurotransmitters to a comprehensive model of personality dimensions. Depue & Lenzenweger (2001) proposed a model of personality buildup of four higher order traits. This model also suggests that these traits are linked to specific neurotransmitters (Paris, 2005). These models thus suggest that neurobiological processes underlie individual differences in traits. Similarly, researchers have linked differences in neural activity to individual differences in traits (Brühl et al., 2011; Kennis et al., 2013). Therefore, this review will also focus on the neurobiology and the neural correlates underlying different psychological and cognitive traits that have been linked to radicalization.
2 Radicalization as a social process
One of the most important psychological traits for humans is being social. Humans are social species and being social has always had benefits throughout evolution (Seeman, 1996; Gintis et al., 2003). Some even suggest that being social is the main driver of the evolution of the human brain (Dunbar, 1998). Social connectedness is highly related to happiness (Haller & Hadler, 2006; Fowler & Christakis, 2008) and humans have a strong need to be socially connected to others (Seppala et al., 2013). Social ties are also suggested to play an important role in the process of radicalization (Precht, 2007; Silber & Bhatt, 2007; McCauley & Moskalenko, 2008; Schmid, 2013). As the majority of radicalization and terrorism happens in group context, this review will first focus on radicalization as a social process. Different social processes and the putative neural and biological correlates underlying them will be reviewed. Then, the link between being social and radicalization will be discussed.
2.1 The neural underpinnings of social reward processing
Because of the strong need to be socially connected, humans are also very sensitive to social rewards or punishments (Izuma et al., 2008; Lebreton et al., 2009). Social rewards like social acceptance, a positive social appraisal and social interactions elicit activity in the brain reward system, which includes the striatum, the lateral and medial prefrontal cortex and the amygdala (Rilling et al., 2002; McClure et al., 2004; Izuma et al., 2008; Ebstein et al., 2010; Bhanji & Delgado, 2014).
Individual differences in social reward dependence, which is defined as the favorable disposition towards social relationships and attachments, have been linked to structural differences in the brain. By using voxel-based morphometry (VBM) on structural magnetic resonance imaging (MRI) scans, individual differences in gray matter volume can be investigated. Increased gray matter density has been associated with higher involvement and better functioning of brain regions (Maguire et al., 2000). Lebreton et al. (2009) found that a higher reward dependence was associated with increased gray matter density in the orbitofrontal cortex, basal ganglia and temporal lobes. The training of social skills by targeted socio-cognitive training has been found to increase gray matter thickness in the temporal cortices (Valk et al., 2017). These findings thus suggest that structural differences in brain areas related to reward processing are linked to individual differences in sensitivity to social rewards.
In addition to the association between functional and anatomical differences and social reward sensitivity, social rewards have also been linked to certain neurotransmitters and hormones. Dopamine is important for the value appraisal and prediction of rewards (Schultz, 2002). This neurotransmitter has been extensively linked to reward and pleasure and plays an important role in drug addiction and hedonic wanting (Rilling et al., 2002; Berridge, 2007; Blum et al., 2009). Hypersensitivity of the dopamine system has been linked to antisocial traits like aggression (Seo & Patrick, 2008; Buckholtz et al., 2010) and to multiple disorders, including psychopathy (Buckholtz et
al., 2010), schizophrenia (Brisch et al., 2014), borderline personality disorder (Friedel, 2004) and
social anxiety disorder (Stein et al., 2002).
Two other neuropeptides that have been extensively linked to social reward are oxytocin and vasopressin (Carter, 2014). These molecules selectively influence each other, and together they are crucial for social bonding. Vasopressin is mostly associated with survival, adaptation and defensive behaviors. Oxytocin, in popular science also known as the “hormone of love”, heavily attributes to the high levels of social sensitivity in humans and plays an important role in all social interactions. It is also important for the feeling of emotional safety (Carter, 2014). Individual differences in levels of oxytocin have been linked to differences in social behaviors (Carter, 2014) and oxytocin has been found to selectively improve empathic accuracy in individuals with less social proficiency (Bartz et al., 2010).
Taken together, these findings thus suggest that individuals can differ in their sensitivity to social reward. Individual differences in social proficiency have been linked to differences in functional, structural and hormonal levels, which have also been linked to certain personality traits and disorders.
2.2 The impact of not feeling connected: exclusion and discrimination
Social interactions and connections are thus highly rewarding for individuals, but this sensitivity to social rewards also makes people more sensitive to social punishment or exclusion (Izuma et al., 2008; Lebreton et al., 2009). Social rejection can result in actual physical pain (MacDonald & Leary, 2015), social exclusion and loneliness negatively impact physical health (Hawkley & Cacioppo, 2010; Jaremka et al. 2013; Leigh-Hunt et al., 2017), and can lead to higher negative affect, social withdrawal and a lack of trust (Ernst & Cacioppo, 1999; Cacioppo et al., 2006). Social exclusion has also been associated with psychological disorders like clinical depression, borderline personality and schizophrenia (Ernst & Cacioppo, 1999; Cacioppo et al., 2014; Leigh-Hunt et al., 2017).
Several neurobiological mechanisms have been linked to perceived exclusion and loneliness. On an hormonal level, differences in sensitivity to dopamine and oxytocin have been associated with loneliness. Lonely individuals have been found to be less sensitive to the effects of oxytocin and research suggests that diminished oxytocinergic signaling contributes to the association between perceived social isolation and health (Norman et al., 2012). Differences in oxytocin receptor genes have also been linked to differences in state loneliness (van Roekel et al., 2013).
Social exclusion has also been linked to altered dopamine sensitivity. The social defeat hypothesis, proposed by Selten & Cantor-Graae (2005), states that long-term exposure to social exclusion may lead to sensitization of the mesolimbic dopamine system. This can in turn lead to increased risk of psychological disorders like schizophrenia (Selten & Cantor-Graae, 2005; Selten et al., 2013). One specific form of social exclusion, discrimination, has been studied more extensively. The incidence of schizophrenia and psychotic disorders has been found to be higher among ethnic minorities (Veling
et al., 2007; Selten et al., 2013; Vinkers et al., 2013). Research on the prevalence of these disorders
suggests that perceived discrimination and social exclusion increase the risk of these psychological disorders. It remains unclear if this increased risk is the result of dopaminergic sensitization caused by social exclusion, but these findings do suggest that social exclusion and dopaminergic sensitivity are somehow related.
In addition to altered dopamine sensitivity, the threat of social exclusion, described as stereotype threat, has also been linked to altered cognitive performance (Schmader et al., 2008). Stereotype threat has been found to induce a stress response, negative feelings and negative emotions which can then lead to a lower working memory efficiency and lower performance on social and cognitive tasks (Schmader et al., 2008). Social exclusion thus not only affects an individual’s emotional state, but also one’s cognitive processing abilities.
Individual differences in loneliness sensitivity have also been linked to functional and structural differences in brain areas associated with social cognition. Higher perceived loneliness has been associated with a decrease in gray matter volume in the posterior temporal sulcus (Kanai et al., 2012) and decreased regional white matter density in multiple areas including the anterior insula, the posterior temporoparietal junction, the posterior superior temporal sulcus and the dorsomedial prefrontal cortex (Nakagawa et al., 2015). Functional differences in response to social stimuli have also been found. Research using functional magnetic resonance imaging (fMRI) to measure neural reactivity has found that lonely individuals show less activation in response to social stimuli in the ventral striatum. This brain area has been previously linked to reward processing. Moreover, lonely individuals show greater activation of the visual cortex in response to stimuli showing the distress of others, which suggests that these individuals pay more attention to the distress of other people (Cacioppo et al., 2009).
Research thus suggests perceived loneliness and exclusion are associated with functional and structural brain differences, differential reactivity to hormones and neurotransmitters, increased risks of psychological disorders and more negative affect and lower trust towards others.
2.3 Social connections: ingroup versus outgroup
The need for social connectedness is apparent. Being socially connected might lead to benefits for oneself, but research has shown that more socially connected individuals are also more likely to become distant from outgroup members (Waytz & Epley, 2012). Two processes that might result from the need for social connectedness and being more social will be discussed: connectedness with the ingroup and distance to the outgroup.
2.3.1 Connectedness with the ingroup
As has been mentioned before, feelings of social connectedness are highly related to happiness (Haller & Hadler, 2006; Fowler & Christakis, 2008). On a neurobiological level, social inclusion activates the brain reward system (Rilling et al., 2002; McClure et al., 2004; Izuma et al., 2008; Ebstein
et al., 2010; Bhanji & Delgado, 2014) and is associated with increased sensitivity to neurotransmitters
and neuropeptides like oxytocin, the “hormone of love” (Rilling et al., 2002; Berridge, 2007; Blum et
al., 2009; Carter, 2014). Connectedness with a group can also lead to changes in beliefs and behavior.
Within a group, shared principles and beliefs are very important, as they function as behavioral imperatives (Ellemers & van den Bos, 2012). Group members are thus likely to take over the moral principles of their group (Kundu & Cummins, 2013) and change their moral behavior accordingly (van den Bos, 2019). As individuals in a group are likely to share similar ideas and beliefs, group membership can lead to polarization. Polarization is the process of shifting to more extreme beliefs (Sunstein, 2012; McCauley & Moskalenko, 2008; van den Bos, 2019). Individuals with stronger attitudes are less open to other ideas and often resist changing their beliefs, even if counterevidence is provided (Kaplan et al., 2016). This resistance has been linked to neural reactivity in the dorsomedial prefrontal cortex and the orbitofrontal cortex, areas that are part of the brain’s default mode network (DMN). The DMN is involved in the processing of deeply held beliefs (Kaplan et al., 2016). In addition to resistance to counterevidence, individuals are biased to look for and interpret information in line with their own beliefs, which is known as the “confirmation bias”. Information not in line with their beliefs will be more easily discarded whereas information in line with prior beliefs is easier accepted (Sunstein, 2012; Taber & Lodge, 2016; van den Bos, 2019).
As a result of this positive endorsement of a group's beliefs and values, group members can develop a belief of moral superiority (Ellemers & van den Bos, 2012). This position provides group members of a positive image that they can be proud of (Tajfel & Turner, 1979). This moral superiority also helps individuals to justify ingroup behavior, even if it involves breaking the rules (Leidner et al., 2010; Iyer
et al., 2012; van den Bos, 2019). Ingroup superiority can also lead to moral disengagement, which is
the process of distancing oneself from the consequences of one’s actions and constructing a version of reality in which their actions are not reprehensible (Bandura, 2002; Leidner et al., 2010; van den Bos, 2019). Higher moral disengagement has been linked to traits like aggression (Barchia & Bussey, 2011; Kokkinos et al., 2016) and narcissism (Jones et al., 2017).
Ingroup connectedness can thus have positive effects for the individual and plays an important role in happiness, but it can also lead to moral superiority, group polarization, resistance to other ideas and moral disengagement.
2.3.2 Distance to the outgroup
As has been mentioned before, being social can have substantial benefits for an individual, but it might also lead to a more distant attitude towards outgroups (Waytz & Epley, 2012). The before mentioned belief of moral superiority has been found to reduce willingness to help outgroup members (Szabó et al., 2017). In addition, Waytz & Epley (2012) showed that increased connectedness to the ingroup can enable more dehumanization of outgroup members. This in turn mediated the willingness to harm outgroup members. These findings are in line with earlier research by Milgram (1963). Milgram studied a concept known as ‘social obedience’, which is a form of social influence. He found that under the right circumstances and with enough social pressure, the majority of individuals are able to administer what they believe are lethal shocks to other individuals (Milgram,
1963). These findings have since then been replicated by multiple other studies (Doliński et al., 2017). Obedience has been linked to certain personality traits, including agreeableness and conscientiousness (Bègue et al., 2015; Decety et al., 2018). To our knowledge, the neural correlates of obedience remain unclear. A study on social norm compliance under the threat of punishment by Spitzer et al. (2007) suggests that norm compliance is positively correlated with activation in the lateral orbitofrontal cortex and the right dorsolateral prefrontal cortex. Individual differences in activation in these areas were associated with behavioral differences in norm compliance. Activity in the dorsolateral prefrontal cortex has been linked to cognitive control. The lateral orbitofrontal cortex has been suggested to play a role in the evaluation of punishments (Spitzer et al., 2007). It remains unclear if neural reactivity in these areas is also associated with obedience in a social context and
without the threat of punishment.
A study by Cikara et al. (2011) found that neural reactivity in the brain reward systems is also affected by group membership. They studied neural reactivity using functional magnetic resonance imaging (fMRI) in baseball fans in response to baseball game outcomes. Subjectively negative outcomes, defined as the failure of the favored team or success of the rival team, activated the anterior cingulate cortex and the insula. Subjectively positive outcomes, which were success of the favored team or failure of the rival team (also against a third team), activated the ventral striatum, a brain area linked to reward processing. These findings suggest that failure of rivals (the outgroup) can be rewarding for individuals of a different group. In addition, activity in the ventral striatum correlated positively with the self-reported likelihood of using aggression against fans of the rival team. This suggests that willingness to harm outgroup members is also associated with neural reward processing (Cikara et al., 2011). Similarly, Hein et al. (2010) found differential activation associated with helping in- and outgroup members. Costly helping (experiencing pain) ingroup members was associated with increased activity in the anterior insula. The insula has been linked to the processing of empathy, and neural reactivity of the insula was found to be associated with self-reported empathic concern. Not helping the outgroup members was associated with higher activation in the nucleus accumbens. While the above-mentioned findings on an increased distance to the outgroup as a result of ingroup connectedness have been replicated by multiple studies, there are individual differences in the behavioral and neural response towards excluders. Will et al. (2015) found that in response to social exclusion in a virtual ball-tossing game, some participants monetarily punished their excluders, whereas others refrained from punishment. Punishment of excluders was linked to increased activation in the pre-supplementary motor area and the anterior insula. Refraining from punishment was associated with activation in the temporoparietal junction, dorsomedial prefrontal cortex, dorsal anterior cingulate cortex and the ventrolateral and dorsolateral prefrontal cortex. These areas have been linked to cognitive control. These findings thus suggest that individual differences in response to social exclusion might elicit differential neural activation.
Taken together, social connectedness can thus lead to less willingness to help outgroup members, more willingness to punish them and more dehumanization of the outgroup. This does not apply to all individuals, as individuals differentially respond to social exclusion. How individuals respond to social exclusion is suggested to be linked to cognitive control.
2.4 Being social and radicalization
Radicalization is often proposed to be a social process (Precht, 2007; Silber & Bhatt, 2007; McCauley & Moskalenko, 2008; Schmid, 2013). Based on the abovementioned research, individual differences in sensitivity to social rejection, social reward and social inclusion can all be linked to processes that have also been associated with radicalization. Social rejection and feelings of exclusion have been found to inflict physical pain (MacDonald & Leary, 2015) and have been linked to higher negative affect, social withdrawal (Ernst & Cacioppo, 1999; Cacioppo et al., 2006) and multiple psychological disorders (Ernst & Cacioppo, 1999; Cacioppo et al., 2014; Leigh-Hunt et al., 2017). Differences in sensitivity to social reward have been linked to structural (Lebreton et al., 2009; Valk et al., 2017), functional (Rilling et al., 2002; McClure et al., 2004; Izuma et al., 2008; Ebstein et al., 2010; Bhanji &
Delgado, 2014) and neurobiological (Bartz et al., 2010; Carter, 2014) differences in the brain. The neurotransmitter dopamine plays an important role in social processing, and hypersensitivity of the dopamine system has been linked to both antisocial behaviors (Seo & Patrick, 2008; Buckholtz et al., 2010) and psychological disorders (Stein et al., 2002; Friedel, 2004; Buckholtz et al., 2010; Brisch et
al., 2014). In addition, feelings of social exclusion have also been associated with alterations in the
brain. Extended social exclusion has been linked to increased dopamine sensitivity (Selten & Cantor-Graae, 2005), which in turn has been associated with an increased risk of psychological and psychotic disorders (Selten & Cantor-Graae, 2005; Veling et al., 2007; Selten et al., 2013; Vinkers et al., 2013). Individual differences in loneliness sensitivity have too been linked to structural (Kanai et al., 2012; Nakagawa et al., 2015), functional (Cacioppo et al., 2009) and neurobiological alterations. Loneliness has been suggested to negatively impact oxytocinergic signaling (Norman et al., 2012; van Roekel et
al., 2013), which contributes to the increase in antisocial behavior (Bartz et al., 2010; Carter, 2014).
These findings suggest that social sensitivity might make individuals more prone to feeling excluded, which can lead to a more antisocial attitude and which has often been suggested to be a key motivation to join a radicalized group (Orlina & Desjardins, 2012; Doosje et al., 2016; van den Bos, 2019).
Joining a radicalized group can increase the feelings of connectedness, but it might also lead to more antisocial behavior towards outgroup members (Waytz & Epley, 2012). Group members are likely to take over the moral principles of their group (Kundu & Cummins, 2013; van den Bos, 2019), and this like-mindedness can lead to polarization towards more extreme beliefs (Sunstein, 2012; McCauley & Moskalenko, 2008; van den Bos, 2019). Stronger attitudes lead to more resistance to other ideas (Sunstein, 2012; Kaplan et al., 2016; Taber & Lodge, 2016; van den Bos, 2019). In addition, these shared beliefs can lead to a feeling of moral superiority (Ellemers & van den Bos, 2012) which can in turn lead to moral disengagement (Bandura, 2002; Leidner et al., 2010; van den Bos, 2019) and justification of illegal behavior by group members (Leidner et al., 2010; Iyer et al., 2012; van den Bos, 2019). Higher moral disengagement has been linked to traits like aggression (Barchia & Bussey, 2011; Kokkinos et al., 2016) and narcissism (Jones et al., 2017). Furthermore, higher connectedness to the ingroup and moral superiority can lead to more distance towards the outgroup. It can lead to less willingness to help outgroup members (Szabó et al., 2017), dehumanization (Waytz & Epley, 2012) and higher willingness to harm outgroup members (Milgram, 1963; Doliński et al., 2017). Group membership has been shown to affect neural reactivity in the brain reward system and empathy related brain regions (Hein et al., 2010; Cikara et al., 2011; Will et al., 2015). Research findings on individual differences in social sensitivity and on the effects of group membership can thus elucidate why some individuals might be more susceptible to radicalized groups and extreme ideologies.
3 Psychological traits associated with radicalization
In addition to looking at radicalization as a social process, this review focuses on individual psychological traits that have been linked to radicalization in0 prior studies. This review will focus on two psychological traits that have been extensively suggested to play a role in the process of radicalization: justice sensitivity and self-esteem and the quest for significance. Personality traits associated with higher and lower levels of these psychological traits and the neural correlates underlying them will be reviewed. Finally, the link to radicalization will be discussed.
3.1 Justice sensitivity
Perceived injustice has been proposed to play an important role in the process of radicalization (Borum, 2003; Eidelson & Eidelson, 2003; Moghaddam, 2005; Precht, 2007; van Zomeren et al., 2008; Davis & Cragin, 2009; Borum, 2011; King & Taylor, 2011; Doosje et al., 2013; Doosje et al., 2016; van den Bos, 2019). Research has shown that individuals differ in their sensitivity to perceived injustice (Huseman et al., 1987; Schmitt, Neumann, & Montada, 1995; Schmitt, 1996). This personality traits described as justice sensitivity entails the readiness to perceive injustice and the strength of the response to injustice (Huseman et al., 1987; Lotz et al., 2011). Individuals high in justice sensitivity perceive injustice more frequently, react more negative to it and are more willing to redress justice than individuals low in justice sensitivity (Schmitt, Neumann, & Montada, 1995; Mohiyeddini & Schmitt, 1997; Baumert et al., 2010; Bondü et al., 2017).
Injustice can be perceived from different perspectives; victims experience injustice themselves, while observers notice an unfair treatment of others (Mikula, 1994; Fetchenhauer & Huang, 2004). Research suggests that distinct processes and different motivational factors underlie these different perspectives (Yoder & Decety, 2019). Individuals who see themselves as victims of injustice are more focused on restoring justice for themselves (Gollwitzer et al., 2005; Mohiyeddini & Schmitt, 1997) and were found to be more willing to exploit others for their own gain (Fetchenhauer & Huang, 2004). They were also less willing to engage in altering the disadvantages of others compared to individuals who observe injustice (Gollwitzer et al., 2005; Gollwitzer et al., 2009). High victim justice sensitivity has thus been linked to more antisocial behavior, whereas high sensitivity to observing injustice has been linked to more prosocial behavior (Gollwitzer et al., 2005; Gollwitzer et al., 2009; Lotz et al., 2011). These different motivations are suggested to influence moral judgments and judgments of justice differently; individuals with a prosocial motivation are more likely to condemn moral transgression, whereas antisocially motivated individuals who want to avoid exploitation are more likely to condone those same behaviors (Yoder & Decety, 2019).
3.1.1 Personality traits associated with justice sensitivity
Research on the association between certain personality traits and justice sensitivity has found similar results. High victim sensitivity has been found to be more highly correlated with self-related, anti-social concerns like paranoia, vengeance, suspiciousness and narcissism (Schmitt et al., 2005). High justice sensitivity has also been linked to emotional lability (Kaplánová, 2018). Similarly, Lovas & Wolt (2002) suggest that individuals that are more aggressive, egocentric and introspective are more sensitive to injustice. High justice sensitivity can thus lead to a more hostile attitude. More other-related concerns like social responsibility, empathy and agreeableness were found to be more highly correlated with high observer sensitivity and more prosocial behavior (Schmitt et al., 2005). Empathy has been studied in more detail, as it is believed to be an important factor in the perceiving of justice (Decety & Cowell, 2015). A distinction can be made between cognitive empathy, which is the ability to understand what somebody else is feeling, and emotional or affective empathy, which reflects the capacity to share or feel others’ emotional states (Decety & Yoder, 2016). Research has linked higher cognitive empathy to higher justice sensitivity, but no association between emotional empathy and justice sensitivity towards others was found (Decety & Yoder, 2016).
3.1.2 Neural correlates of justice sensitivity
More recently, neuroscientists have started to study the neurobiological mechanisms underlying these individual differences in justice sensitivity. Structural differences in gray matter thickness have been linked to differences in justice sensitivity (Eres et al., 2015; Baumgartner et al., 2016). Structural magnetic resonance imaging studies suggest that a larger gray matter volume and thickness of the insula are associated with higher justice sensitivity (Baumgartner et al., 2016) and with more affective empathy (Banissy et al., 2012; Eres et al., 2015). The insula plays an important role in social affect, empathy and interoception (Uddin et al., 2017) and is therefore important in social decision making (Droutman et al., 2015; Eres et al., 2015; Baumgartner et al., 2016). Greater gray matter density of the dorsal medial prefrontal cortex has been linked to cognitive empathy (Eres et al., 2015). Cognitive empathy relies heavily on the capacity to attribute the mental states of others to oneself (known as the “Theory of Mind”), and the dorsal medial prefrontal cortex has been suggested to play in important role in this process (Decety, 2011).
Functional neuroimaging studies have found similar results. A meta-analysis of 40 functional magnetic resonance imaging (fMRI) studies by Fan et al. (2011) suggests that the insula is involved in affective or emotional empathy, whereas the midcingulate cortex (adjacent to the dorsal medial prefrontal cortex) was more frequently involved in cognitive empathy.
In addition to the beforementioned structural differences, Yoder & Decety (2014) have found functional differences associated with interindividual variability in justice sensitivity. This study suggests that justice sensitivity modulates the neural response to morally laden actions. This change in neural activity was found in higher-order computational nodes in the temporoparietal junction (TPJ) and the dorsolateral and dorsomedial prefrontal cortex. Activity in these regions has been associated with maintaining goals and guiding moral decision-making and was found to predict praise and blame ratings. These findings thus suggest that justice sensitivity modulates activity in response to moral judgments of others in regions associated with goal representation and moral decision making (Yoder & Decety, 2014).
As was mentioned before, various processes and motivations underlying justice sensitivity regarding others and oneself have been posited (Yoder & Decety, 2019). Research on the neural correlates of fairness judgments including oneself or others suggests that there are only partially overlapping neural processes underlying these judgments. The prefrontal medial cortex (Corradi-Dell’Acqua et al., 2013) and the middle cingulate cortex (Civai et al., 2012) have been linked to judgments of unfairness for oneself, whereas the anterior insula has been linked to unfairness judgment for both oneself and others (Civai et al., 2012; Corradi-Dell’Acqua et al., 2013; Yoder & Decety, 2019). A study by Wu & Tian (2017) looked at differences between different types of justice sensitivity by studying regional homogeneity in resting-state fMRI. Regional homogeneity entails the degree of temporal synchronization in neural activity between regions and can be used to correlate regional activation with individual differences in justice sensitivity. Their results suggest that victim sensitivity is associated with activation in the paracentral lobule, whereas observer sensitivity is correlated with activation in the temporal pole. This again suggests that different neural processes underlie these types of justice sensitivity.
Taken together, findings on the neural correlates of justice sensitivity suggest that functional and anatomical differences in the insula and the prefrontal cortex are associated with individual differences in justice sensitivity. However, justice sensitivity can be primarily driven by prosocial or antisocial motivation, and these have shown associations with distinctive underlying neural correlates. Making a distinction between justice sensitivity regarding oneself and others is therefore important.
3.1.3 Justice sensitivity and radicalization
As has been mentioned before, the majority of models of radicalization incorporate perceived injustice as an important factor in the process of radicalization (Borum, 2003; Moghaddam, 2005; Precht, 2007; Borum, 2011; Doosje et al., 2016; van den Bos, 2019). These models propose that as a
result of perceived injustice for the ingroup, people get more prone to radicalize. These models thus suggest that these individuals see themselves as victims of injustice.
Individuals with higher justice sensitivity (Schmitt, Neumann, & Montada, 1995; Mohiyeddini & Schmitt, 1997; Baumert et al., 2010; Bondü et al., 2017) who see themselves as a victim of the injustice are likely to develop a more antisocial viewpoint towards solving the injustice (Mohiyeddini & Schmitt, 1997; Fetchenhauer & Huang, 2004 Gollwitzer et al., 2005; Gollwitzer et al., 2009; Lotz et
al., 2011; Yoder & Decety, 2019). This high victim sensitivity has been linked to other antisocial traits
like narcissism, aggression and egocentrism (Lovas & Wolt, 2002; Schmitt et al., 2005; Kaplánová, 2018). These traits might lead to a hostile attitude towards outgroups (McCauley & Moskalenko, 2008). Research has shown that anger and hostility mediate the relationship between fairness perceptions and retaliation (Barclay et al., 2005). Individual differences in justice sensitivity can thus make individuals more likely to act upon perceived injustice in an antisocial way. High justice sensitivity has also been linked to disorders like borderline and depression (Bondü et al., 2017; Lis et
al., 2018). These findings thus suggest that justice sensitivity can be seen as an important factor in
the process of radicalization.
3.2 Self-esteem and the quest for significance
An often-proposed motivational factor in the process of radicalization is the quest for significance (Kruglanski et al., 2013). The quest for significance has been posited as a fundamental human motivation and is defined as the need to be recognize and to matter (Maslow, 1943). It has also been described as self-love (Kruglanski & Orehek, 2011). Feelings of insignificance can have multiple causes, for instance poor socioeconomic status or personal loss or failure (Doosje et al., 2016). Competence and meaning something are socially defined, which makes the quest for significance an attainment of what an culture says is worth attaining (Kruglanski et al., 2013).
Closely related to the quest for significance is self-esteem (Kruglanski et al., 2013), described as one’s attitude towards oneself. Feelings of insignificance are associated with lower self-esteem and more uncertainty about one’s own worth, whereas feelings of significance can be linked to higher self-esteem. As feelings of self-esteem and significance are so important, individuals are likely to be motivated to decrease their feelings of uncertainty about themselves (Hogg, 2011).
Self-esteem has also been suggested to play an important role in how people deal with situations in their daily lives. Individuals low in self-esteem are more affected by both positive and negative experiences (Greenier et al., 1999). Self-esteem has also been linked to how individuals deal with their fear of death (Greenberg et al., 1997). The Terror Management Theory proposes that death anxiety drives people to more strongly defend their worldviews and maintain their self-esteem. Self-esteem thus functions as a buffer in dealing with the awareness of one’s own vulnerability and mortality. The authors propose that because of this important function, self-esteem is a basic human need (Greenberg et al., 1997).
One way of increasing self-esteem and feelings of significance is by joining a social group (Hogg, 2007, 2011). This process of using group identification to decrease uncertainty was first described by the uncertainty-identity theory of Hogg (2000). Groups have internalized values and beliefs on how to behave, which can be used to prescribe one’s own attitudes and behavior. Individuals also judge themselves and others conform these group norms. In this way, group identification can reduce self-related uncertainty (Hogg, 2007). Feelings of uncertainty and low self-esteem can thus be a motivational factor to seek praise and recognition by joining a social group (Borum, 2014).
3.2.1 Personality traits associated with self-esteem and the quest for significance
Feelings of insignificance can thus be linked to more uncertainty and lower self-esteem and accomplishing a feeling of significance is believed to be a strong human motivation (Maslow, 1943). Lower levels of self-esteem have been linked to more defensive extremism (McGregor & Jordan, 2007) and more dogmatic thinking (Montgomery, 1972; Goldsmith & Goldsmith, 1982; Ottati et al., 2015). High self-esteem has been associated with better personal resilience, but individuals high in self-esteem have also been found to respond more antisocial after threat (Park & Crocker, 2005;
McGregor et al., 2009). Extreme high self-esteem is often believed to be linked to more antisocial, dark personality traits (Stenason, 2014).
One well studied personality trait in relation to pathological self-esteem is narcissism. Research on esteem makes the distinction between two types of esteem: implicit, unconscious self-esteem and the explicit, more conscious evaluations of oneself (Gregg & Sedikides, 2010). Narcissism is suggested to be linked to high levels of explicit self-esteem, but low levels of implicit self-esteem. This theory suggests that grandiosity is used to mask one’s low implicit self-esteem (Zeigler-Hill, 2006; Bosson et al., 2008). Findings on the link between self-esteem and this dark personality traits are mixed. High levels of explicit self-esteem have often been linked to narcissism (Tracy et al., 2009; Gregg & Sedikides, 2010; Stenason, 2014), but there are also studies that did not find this link (Vater
et al., 2013; Di Pierro et al., 2016). Similarly, both high (Stenason, 2014) and low (Sakellaropoulo &
Baldwin, 2007; Gregg & Sedikides, 2010) implicit self-esteem have been associated with narcissism. Next to narcissism, Machiavellianism, defined as highly manipulative and interpersonal coldness, was found to be associated with lower implicit self-esteem. Low explicit self-esteem in general has been linked to low attachment to society and externalizing behavior (Stenason, 2014).
Self-esteem has also been linked to other personality traits. Authoritarianism is suggested to be associated with low self-esteem and high insecurity (Larsen & Schwendiman, 1969). Low self-esteem has also been linked to depression (Orth & Robins, 2013) and more sensitivity to social exclusion (Onoda et al., 2010). On the other side of the spectrum, personality traits like emotional stability, extraversion and openness to experience have been linked to higher levels of self-esteem (Robins et
al., 2001). Findings on the relation between self-esteem and aggression are mixed (Ostrowsky, 2010).
Both high and low self-esteem have been linked to aggression and antisocial behavior (Donnellan et
al., 2005; Bushman et al., 2009). These findings thus suggest that high self-esteem can lead to better
resilience and more prosocial behavior, but that it is also associated with more antisocial traits.
3.2.2 Neural correlates of self-esteem and the quest for significance
Individual differences in self-esteem have been linked to both structural and functional differences in the brain. Research on white matter integrity, which is a measure of structural connectivity, suggests that individual variability in white matter structural integrity between the medial prefrontal cortex and the striatum is linked to individual differences in self-esteem (Chavez & Heatherton, 2015). The medial prefrontal cortex is believed to be important in self-representation, whereas the ventral striatum is involved in reward processing. Connectivity of these areas may contribute to self-esteem by integrating self-representation with feelings of reward (Chavez & Heatherton, 2015).
Studies on volumetric structural differences have related individual differences in self-esteem to hippocampal volume (Kubarych et al., 2012) and gray matter volume in the anterior cingulate cortex, the right lateral prefrontal cortex and the hypothalamus (Agroskin et al., 2014). Self-esteem was positively correlated with brain volume in these areas. These areas have all been associated with emotional self-regulation in response to stress. It is suggested that higher self-esteem makes individuals less vulnerable to stress (Kubarych et al., 2012; Agroskin et al., 2014).
Studies on the functional underpinnings of self-esteem have identified multiple cortical midline structures involved in the processing of self-worth, including areas in the medial prefrontal cortex and the cingulate cortex (Northoff et al., 2006). More specifically, the ventromedial prefrontal cortex has been suggested to be involved in the assignment of personal value to self-related information (D’Argembeau, 2013), but has also been linked to being evaluated by others (Will et al., 2017). Higher functional connectivity between the ventromedial prefrontal cortex and the hippocampus has been linked to higher self-esteem (Pan et al., 2016). The anterior insula and hippocampus are involved in emotion regulation and research has found increased activation in these areas during self-evaluation (Doerig et al., 2014). Activation of the anterior cingulate cortex has been linked to self-criticism (Longe et al., 2010). These findings thus suggest that self-esteem is build up from different constructs represented by activation in multiple frontal areas.
Next to individual differences in self-esteem, changes in self-esteem within individuals have also been linked to changes in neural activity. Greater activity in the dorsal anterior cingulate cortex (Yang et al.,
2012) and the anterior insula was associated with lower subjective ratings of self-esteem in response to feedback (Eisenberger et al., 2011). Changes in activity in the ventral striatum, the anterior cingulate cortex and the ventromedial prefrontal cortex have also been linked to changes in self-esteem within individuals (Will et al., 2017). These findings thus suggest that activity in these areas is associated with individual differences in self-esteem, but that these areas also modulate feelings of self-esteem within an individual.
Individual differences in self-esteem have also been found to moderate the reaction to social feedback. Research suggests that individuals low in self-esteem are more sensitive to social rejection and social feedback. Individuals with low self-esteem showed more neural reactivity in the anterior cingulate cortex and the anterior insula in response to social rejection compared to individuals high in self-esteem (Kashdan et al., 2014). Similar results have been found regarding social feedback. Individuals with low self-esteem showed enhanced activity in the anterior cingulate cortex and the medial prefrontal cortex in response to positive versus negative social feedback, whereas this increased activation was not found in individuals high in self-esteem (Somerville et al., 2010). In addition, differences in self-esteem have been linked to differences in social pain resulting from exclusion (Onoda et al., 2010). Individuals with lower trait self-esteem subjectively rated social inclusion as more painful, and these individuals also show greater activation of the dorsal anterior cingulate cortex. Activity in this region has previously been linked to social pain and social rejection. Next to that, these individuals showed increased positive connectivity between the dorsal anterior cingulate cortex and the prefrontal cortex compared to individuals with high self-esteem. This connectivity might contribute to the more negative self-representation in low self-esteem (Onoda et
al., 2010).
3.2.3 Self-esteem, the quest for significance and radicalization
As feelings of self-worth and significance are so important (Maslow, 1943), individuals will be very motivated to change their feeling of insignificance. This quest for significance has been extensively suggested to be a factor in the process of radicalization (Hogg, 2007; Kruglanski & Orehek, 2011; Hogg & Adelman, 2013; Hogg et al., 2013; Kruglanski et al., 2013; Dugas & Kruglanski, 2014; Kruglanski et al., 2014; Doosje et al., 2016; Lyons-Padilla et al., 2016; Jasko et al., 2016; Kruglanski et
al., 2018; van den Bos, 2019). According to the quest for significance model of radicalization, the
quest for significance is the crucial motivational component of radicalization, next to the ideological component and the group component (Kruglanski et al., 2014). As has been discussed before, radicalization can also be seen as a social process and joining a social group can increase one’s feeling of significance. Having feelings of insignificance and low self-esteem can thus make it more appealing to join a radicalized group (Doosje et al., 2016), which will offer these individuals a sense of inclusion and purpose (Lyons-Padilla et al., 2016). This can lead to increased self-esteem (Hogg, 2013). In addition to the motivational power of the quest for significance, uncertainty and low self-esteem can also drive people towards more extreme, totalistic groups (Hogg, 2007; Hogg et al., 2013). The need to defend one’s world view is also moderated by self-esteem (Greenberg et al., 1997).
Next to the motivational power of the quest for significance, there are also other traits associated with self-esteem that have been linked to radicalization. Low levels of self-esteem have been associated with more defensive extremism (McGregor & Jordan, 2007), dogmatic thinking (Montgomery, 1972; Goldsmith & Goldsmith, 1982; Ottati et al., 2015), higher sensitivity towards social exclusion (Onoda et al., 2010) and low attachment to society (Stenason, 2014). On the other side of the spectrum, high levels of self-esteem have been linked to better resilience and more prosocial behavior (Robins et al., 2001), but have also been associated with antisocial behavior, narcissism (Tracy et al., 2009; Gregg & Sedikides, 2010; Stenason, 2014) and aggression (Bushman et
al., 2009).
As mentioned before, findings on the neural underpinnings of self-esteem suggests that individuals low in self-esteem are more sensitive to social rejection (Kashdan et al., 2014) and experience more social pain after exclusion (Onoda et al., 2010). This higher sensitivity could lead to a stronger motivation to obtain feelings of significance and inclusion. This could in turn lead to a higher risk of
radicalization. Individual differences in neural reactivity and self-esteem could thus possibly underlie radicalization.
Taken together, the above-mentioned findings suggest that the quest for significance and self-esteem play an important role in the process of radicalization.
3.3 Psychological traits and radicalization
Both feelings of injustice and the quest for significance are nominated by many scholars as important factors in the process of radicalization (Borum, 2003; Moghaddam, 2005; Precht, 2007; Borum, 2011; Hogg et al., 2013; Doosje et al., 2016; Lyons-Padilla et al., 2016; Kruglanski et al., 2018; van den Bos, 2019). Based on the above-mentioned findings on the association between these psychological traits and radicalization, research does suggest that individual differences in these traits might make individuals more prone to radicalization. Radicalization is a complex problem influenced by different factors and different for every individual, but these findings do suggest that there are traits that can be associated with a higher risk of radicalization. This review does not try to argue that radicalization can be explained by solely looking at psychological traits, but psychological traits might partly explain why some individuals radicalize, whereas other do not.
4 Cognitive traits associated with radicalization
Another category of theories that try to explain radicalization are the cognitive theories (Victoroff, 2005). These theories suggest that radicalization is merely driven by cognitive processes. One often mentioned cognitive factor in the process of radicalization is the “cognitive opening”, first proposed by Wiktorowicz (2005). This is the moment an individual becomes receptive to the possibility of new ideas and worldviews, and thus to a radicalized ideology. It still remains unclear which cognitive processes underlie this cognitive opening (Trip et al., 2019). In the following, three cognitive traits that have been associated with radicalization by previous studies will be discussed to try to explain why some individuals are more vulnerable to radicalization: cognitive flexibility and dogmatism, cognitive complexity and need for structure and cognitive capacity and cognitive control.
4.1 Neural correlates of cognitive flexibility and dogmatism
A cognitive trait that has extensively been associated with a higher risk of radicalization is cognitive inflexibility or rigid thinking (Victoroff, 2005; Borum, 2014; van den Bos, 2019; Zmigrod et al., 2019). As mentioned before, individuals with strong beliefs are often resistant to changing their beliefs (Kaplan et al., 2016), which has been linked to rigidity and inflexibility of the thinking process in general (Rokeach, 1948). Research focusing on the link between cognitive inflexibility and extremist attitudes suggests that cognitive inflexibility significantly predicted ideological attachment and the endorsement of violence to protect the ingroup. Cognitive rigidity was found to be an antecedent of ideological attachment and self-sacrifice to save the ingroup (Zmigrod et al., 2019).
A process closely related to cognitive inflexibility and belief resistance is dogmatic thinking. Dogmatism is the tendency to hold on strongly to one’s own beliefs and to reject and consider as inferior other opinions (Borum, 2014; van Prooijen & Krouwel, 2016; Friedman & Jack, 2018). Dogmatic intolerance has been linked to more extremist opinions (van Prooijen & Krouwel, 2016) and has been suggested to be negatively related to analytical reasoning and perspective taking (Friedman & Jack, 2018). Interestingly, Friedman & Jack (2018) have found that dogmatism has different effects in religious and non-religious individuals. Within religious individuals, higher levels of dogmatism were associated with increased prosocial intentions, whereas higher levels of dogmatism were negatively related with empathic concern in non-religious individuals. This suggests that one’s belief system and how dogmatism effects an individual’s behavior are related.
Neuroimaging studies have associated cognitive flexibility with activation in the prefrontal cortex (Leber et al., 2008; Armbruster et al., 2012; Dajani & Uddin, 2015), an area that also plays an important role in cognitive control. Dopamine receptor signaling has also been linked to cognitive flexibility, with cognitive flexibility depending on dopamine receptor signaling (van Holstein et al., 2011). Cognitive inflexibility has also been linked to multiple psychiatric disorder that are thought to be affected by dysfunctions in the dopamine system, including schizophrenia and delusional disorders (Moritz & Woodward, 2005; Garety et al., 2015; Waltz, 2017). Individual differences in cognitive flexibility thus seem to be associated with cognitive control and dopaminergic signaling.
4.2 Neural correlates of cognitive complexity and need for structure
The association between cognitive complexity and more extreme beliefs has also been studied. Cognitive complexity can be defined as how able an individual is in understanding and combining different ideas, options and constructs. Cognitive complexity has been linked to better social perspective taking (Hale & Delia, 1976) and less extreme evaluation of the in- and outgroup (Ben-Ari
et al., 1992). Focusing on radicalization specifically, Soliman et al. (2016) have found that cognitive
complexity is a good predictor of radicalization.
Related to cognitive complexity is one’s need for structure (Neuberg & Newson, 1993; Victoroff, 2005). Individuals with a higher personal need for structure are more likely to organize information in less complex ways. These individuals are also more likely to stereotype others (Neuberg & Newson, 1993; Schaller et al., 1995) and are less forgiving (Neuberg & Newson, 1993; Eaton et al., 2006). This need for structure can also result in more rigid thinking (van den Bos, 2019). The absolutist nature of
