Cause and reason

An ‘explanation’ may be defined as ‘a thing which explains, makes clear, or accounts for something’ (Oxford English Dictionary 2019). This broad definition encompasses the more specific types of explanation below. The distinction between cause and reason has been extensively debated. According to Buss (Reference Buss1978) ‘causes are that which bring about change’ and ‘reasons are that for which change was brought about’. Thus, an insulting comment can be the cause of an act of aggression, for which the reason was to retaliate. Buss's definitions are accepted for the framework presented in this article, but although Buss's categorisation implies mutual exclusivity, it needs to be recognised that the distinction can be very fine. For example, the desire to obtain someone else's material possession to fund a drug habit may be both a reason for, and a cause of, a street robbery. However, whereas the desire may be sufficient in itself as a reason, it will form only part of the causal explanation. If the offender was intoxicated, then the disinhibiting effects of the intoxicant may be another part of the causal explanation. Applying Dennett's typology of explanation, reasons answer ‘What for?’ questions (Dennett Reference Dennett2017). According to Dennett, causes, on the other hand, answer ‘How come?’ questions and are best described in terms of a process narrative. A process narrative is particularly useful when developing an explanatory formulation.

Intention, motivation and function

Motivation and intention can have narrower meanings than reason. Obtaining a material possession that belonged to someone else as an explanation for an act of aggression in the course of a street robbery is a particular sort of reason for which the term ‘intention’ is used. Intention may be defined as referring to the thing consciously intended, ‘that which is intended or purposed’ (Oxford English Dictionary 2019). Not all reasons that represent the actor's mind are intentions. For example, explaining reactive violence in terms of an explosive outburst of anger relies on a mind-based explanation which is not an intention. The cause is the anger and a plausible reason is to express that anger, but this may not have been the conscious objective.

The ordinary use of motivation does not distinguish it from reason, but in the hands of psychiatrists and psychologists the term is used to describe a change in physiological or psychological conditions that behaviour is aimed at achieving (Peters Reference Peters1960). It can be seen as a driving force for the action. Motivation used in this way emphasises an internal objective that can be distinguished from intention, which tends to refer to an observable objective (e.g. behaviour or environmental event). Applying this type of distinction would lead one to conclude that the intention of the street robbery was to obtain the item, whereas the motivation was to satisfy a desire to achieve a state of mind associated with drug use, which will become possible by selling the stolen item. Bearing in mind this distinction in the assessment of violent acts facilitates the identification of the mental processes. This can be further illustrated by considering the scenario of an intoxicated pub patron impulsively lashing out at another customer who had made a mocking comment. The aggressor reported that his intention was to shut the other man up. Further exploration uncovered that he was motivated by wanting both to reduce his sense of humiliation and to avoid a loss of perceived standing among his peers, who were witness to the interaction. In another example the intention of an aggressive rapist was to commit a rape, whereas he was motivated by a wish to enhance his anticipated sexual arousal with the feelings he would experience when directly witnessing signs that his victim was fearful of him.

As with motivation, the term function is used to describe an internal state-based objective of the act. Whereas motivation is more commonly used to describe proximal processes, function tends to draw on wider psychological processes and involves interpretive assumptions (Hofmann Reference Hofmann2017). For example, in the reactive aggression pub scenario mentioned above, the motivation was to reduce feelings of humiliation and avoid perceived loss of status. The function was to counter a worsening of pervasive underlying self-devaluative feelings by providing the aggressor with positive feelings associated with expectations about the way his peers will view him.

Rationalisation and justification

The term rationalisation is often used to describe a potentially logical, but not necessarily valid, reason for an action offered by the actor after the act has occurred. Justification is a particular type of rationalisation that introduces a moral element into the explanation, i.e. relying on a judgement that the reason for the action was at some level a good or a just one. In these senses, rationalisations and justifications offered by the subject of the assessment may interfere with the assessor's attempts to reach an informed and balanced explanation, and they are therefore processes of which the clinician needs to be aware. Although minimisation (or, in its extreme form, denial) may go along with justification, it is not a type of explanation in itself. Rather, it is the assessor's judgement about the discrepancy between the perpetrator's account and the account suggested by independent observers.

Trigger

Perpetrator accounts of aggressive incidents often begin with a description of something that provokes a change in the individual's mind in a way that leads to the aggression. This may be an environmental event (e.g. an action by another person) and, depending on the model or context, is referred to as a stimulus, cue, trigger, aversive event or provocation. From a subjective standpoint the aggressive episode actually starts with the attention to the event and then the attribution of meaning to it (discussed below). However, the objective properties of environmental events need to be taken into account (Finkel Reference Finkel and Hall2018). Although the aggressor may refer to a single event, there is often a series of proximal events that have a similar effect on them.

Internal events may trigger an aggressive state of mind. In the case of post-traumatic stress disorder associated with violence, a re-experiencing phenomenon (e.g. memory or flashback) can lead to a change in subjective experience so that aggression is more likely (Silva Reference Silva, Derecho and Leong2001). Command auditory hallucinations encouraging violence or persecutory delusions associated with hostility or feelings of imminent threat are other examples of internal triggers (Lamsma Reference Lamsma and Harte2015). Whether the identified trigger is an environmental event, a memory or a psychotic experience, it is the change in the perpetrator's affect state that is critical in influencing the likelihood of an aggressive response (discussed further below). In some cases, the affect change can occur spontaneously. For example, affective instability in borderline personality disorder involves a sudden and dramatic switch to a negative affect state and in some individuals this state is associated with aggressive urges.

Attention

Where an aggressive state of mind is triggered by an event, the subjective phase starts with attention to the event. Attention to potential triggers varies both between individuals (Chan Reference Chan, Raine and Lee2010) and within the same individual depending on other contextual factors (Banks Reference Banks, Scott and Weems2018). Studies of the role of attention in the mental processes leading to aggression tend to focus on events that the aggressor reacts against, i.e. threat-related stimuli (Salum Reference Salum, Mogg and Bradley2016). However, some situations activate attention in a way that leads to aggression because they represent opportunities. For example, a lone female in a secluded area may attract the attention of an individual with sexually sadistic urges, or a shop empty of customers may be of interest to a person seeking an opportunity to commit an armed robbery. These situations may also be deliberately sought out with a view to carrying out the aggressive act. In these scenarios, there may have been an existing preparedness for the use of violence, but the resistance offered by a victim may provoke an angry aggressive reaction. The resistance serves as a trigger for reactive violence for which there had been instrumental preparation, illustrating the overlap between reactive and instrumental constructs.

Meaning

An event has the potential to become a trigger on account of the meaning ascribed to it. Such events may include the actions of others, such as the adoption of a threatening posture or the utterance of a verbal insult. However, some aggression-prone individuals may attribute threatening meaning to postures or comments that most would not consider threatening, particularly if there is some ambiguity of intent. A tendency to attribute hostile intent to the actions of another is associated with the likelihood of aggression (Dodge Reference Dodge, Malone and Lansford2015; Martinelli Reference Martinelli, Ackerman and Bernhard2018).

As well as threat, other types of meaning that may be attributed to actions so as to trigger violence include challenge, rejection, abandonment, criticism, subjugation, disrespect, humiliation and injustice. The tendency to attribute malign intent is a feature of paranoia. Beliefs about the malign intent of others in paranoid psychosis do not always require a trigger event. In paranoid personality disorder, there is usually an occurrence that has led to the attribution of paranoid meaning. The tendency to persistently bear grudges may lead to considerable delay between the identified event and the violent act. More fleeting paranoid reactions to the actions of others are a feature of borderline personality disorder.

It is possible for an aggressive state of mind to develop in response to an action without the attribution of a particular intent to the victim. As already noted, an affect change without a clear environmental trigger may lead to aggression. Further, a potential aggressor may experience frustration and anger in response to an incident that is unrelated to the victim but then the anger is expressed in the form of aggression towards the victim.

Quantitative studies of the psychological correlates of aggressive acts understandably rely on variables representing aspects of the aggressor's mind. A framework based on these variables does not do justice to the interactive nature of an aggressive encounter. This can be illustrated by the case of a patient who appeared hypervigilant to signs of malign intent and complained that people he met were more reserved or unfriendly with him from the outset than they were with others. Observations revealed that he approached new interactions in a way that the other party in the interaction found abrupt and aloof. When a detailed subjective account was taken, it became clear (to the assessor and the patient) that his default stance was one that was likely to be interpreted as defensive, which in turn would provoke others to amend their interactional style in a way that he found unfriendly. It was recognised that there had been a positive feedback loop in which he had relied on his experience of how people interacted with him as evidence to support his view that it was safer to assume that the world was unjust and that he should be suspicious of others.

Preparedness

Neurobiological studies (e.g. Repple Reference Repple, Pawliczek and Voss2017) have found evidence for one pattern of neural activation that appears to represent a state of preparedness for aggression and another distinct pattern that is thought to represent a top-down regulatory process. The state of preparedness is considered to be equivalent to the subjective experience of an impulse or urge. The regulatory processes can deactivate the impulse, which may be experienced as an urge being resisted.

The likelihood that an aggressive impulse will arise in response to an event ascribed with meaning will depend on a number of factors. Social information processing models of aggression refer to scripts, which have been defined as ‘templates that provide a description of events and mental rules to guide behaviour in particular situations’ (Gilbert Reference Gilbert and Daffern2017). Aggression is more likely if aggressive scripts are readily available. This depends in part on immediate factors. Attributing hostility, the presence of a weapon or exposure to media violence increases the availability of aggressive scripts (Allen Reference Allen, Anderson and Bushman2018; Huesmann Reference Huesmann2018). Wider factors also play a part. A tendency to generate aggressive scripts may have developed in the context of adverse early life experiences and/or antisocial peer-group influences (Huesmann Reference Huesmann and Bushman2016). A general tendency to ruminate can facilitate script rehearsal, as can certain types of psychopathology. Autism spectrum disorder is characterised by a restricted and repetitive pattern of thinking, which in some rare cases may include thinking with a violent content. The bizarre fantasies and preoccupations that are a diagnostic feature of schizotypal personality disorder, and psychotic symptoms in psychotic disorders, are sometimes repetitively replayed in the patient's mind. If the content is violent, there is a potential for aggressive scripts being readily available. By definition, sexual sadistic disorder is manifest in fantasies, urges and behaviour related to the suffering of others.

The process of script availability is conceptually linked to the notion of normative beliefs (Yang Reference Yang and Mulvey2012; Huesmann Reference Huesmann2018). A normative belief that aggression is an acceptable means of resolving conflict will increase the likelihood of an aggressive action impulse in response to a relevant stimulus. The shaping of normative beliefs occurs particularly in the context of significant relationships in the developmental period (Botelho Reference Botelho and Goncalves2016) and through peer influences. The latter influences are powerful in membership of violent gangs, in which the potential for violence is both promoted and a requirement for continued membership (Wood Reference Wood2014). In-group versus out-group biases, which can facilitate aggressive impulses to members of an out-group (Bohm Reference Bohm, Rusch and Gurerk2016), are exaggerated in gang culture (Vasquez Reference Vasquez, Wenborne and Peers2015). However, they can also be activated in other contexts, such as in the presence of racial biases (Mekawi Reference Mekawi, Bresin and Hunter2016). Normative beliefs, in turn, are related to the notion of sense of identity. Clinical experience suggests that some individuals whose sense of identity and agency relies heavily on their potential for aggression (e.g. to counter general feelings of self-devaluation and lack of agency secondary to the effects of previous adversity) are more likely to generate aggressive responses in aversive environmental and/or affect conditions.

The choice of action in a particular situation will depend on not just the type, but also the range of options available. A maladaptive response is more likely if adaptive options are not easily generated (Blair Reference Blair2016). Thus, the chances of an aggressive impulse being activated is greater if an aggression-prone individual has poor social problem-solving difficulties, since alternative prosocial scripts are less readily available.

In scenarios in which violence is associated with auditory hallucinations, the association is sometimes explained by the content of the voices appearing, from a subjective perspective, to encourage aggression (e.g. command hallucinations), thus provoking an aggressive action impulse.

Evaluation and inhibition/activation

The likelihood of one particular action impulse being chosen above the others depends on an evaluation of the anticipated consequences of the action (Lerner Reference Lerner, Li and Valdesolo2015). Negatively evaluated action impulses will tend to be inhibited. For most action impulses, including many instances of violence, inhibition occurs with very little or no deliberate reflection. Dampening of inhibitory processes that counter action impulses leads to impulsivity. Psychiatric conditions associated with a general attenuation of these processes include attention-deficit hyperactivity disorder (ADHD) and executive dysfunction. If an individual with such a condition is liable to aggressive action impulses then, as with other action impulses, there is an increased risk of translation into action owing to the non-specific effect of reduced action impulse inhibition (Blader Reference Blader and Daviss2018). Violence in the context of acute psychosis is sometimes linked to the patient's experience of the control of their actions being overridden, often by external forces (Link Reference Link, Stueve, Monahan and Steadman1994).

There are inhibitory processes that are more specific to violence. The normal developmental trajectory involves a peak prevalence of aggression at around 2–3 years of age (Tremblay Reference Tremblay, Nagin and Séguin2004). A key process accounting for the subsequent reduction in the frequency of aggression is the development of the capacity to implicitly adopt the anticipated perspective of the victim of a future aggressive action (Decety Reference Decety and Cowell2018). As a consequence, aggressive action impulses are normatively inhibited by the aversive feelings associated with adopting the anticipated perspective of a suffering victim of violence.

This type of inhibitory effect is attenuated if there is an impairment of the general capacity to readily perspective-take, such as in autism spectrum disorder (Chen Reference Chen, Martinez and Cheng2018). In psychopathy, there is a more specific disturbance in relation to certain emotional states of others, such as fearfulness and disgust (Igoumenou Reference Igoumenou, Harmer and Yang2017). If this disturbance is enduring, as in psychopathy, then more deliberate planning involving explicit focus on the effects of aggression is not inhibited. The related, but more dynamic, process of moral disengagement has been used to describe how distancing oneself from accepted moral standards can attenuate processes that would otherwise inhibit aggression (Moore Reference Moore2015). Moral disengagement consequent on extremist ideological indoctrination has been identified in acts of terrorist violence (Bandura Reference Bandura1999). A temporary reduction in emotional perspective-taking associated with dissociation under stress is part of the explanation for some aggressive outbursts by people with borderline personality disorder (Falk Reference Falk, Herpertz and Bertsch2015). As with ADHD, it is not that individuals meeting the criteria for these conditions necessarily experience greater levels of aggressive action impulses. Rather, in the event of such impulses, the individual's capacity to inhibit the impulse may be attenuated.

Emotional processes

The role of emotional processes throughout this sequence has been increasingly recognised (e.g. Lemerise Reference Lemerise and Arsenio2000). In the moment, the goal of an action is the achievement of a preferred emotional state (Lerner Reference Lerner, Li and Valdesolo2015). This may be more obvious in the case of violence that is classified as reactive, where the action can be seen as a means to shift from an aversive state to one that is anticipated to be more favourable. Emotional dysregulation is associated with an increased risk of aggression. The relevance of the affect state to violence is not just a feature of its valence (e.g. negative or aversive) or stability, but also its type. There is a particular association between anger and aggression (Wyckoff Reference Wyckoff2016). The way attention is allocated and the way stimuli are understood is dependent on the individual's emotional state at the time. For instance, anger and irritability increase the likelihood of attention bias to threat-related cues (Salum Reference Salum, Mogg and Bradley2016) and of the misinterpretation of cues (Li Reference Li, Sun and Yee Ho2016). Other negative emotions, such as fear, are associated with aggression in some circumstances (Roberton Reference Roberton, Daffern and Bucks2012).

Aversive conditions that are associated with an increased likelihood of aggression include pain, stress and heat (Groves Reference Groves and Anderson2018). However, an aversive state is not always a necessary baseline condition. The role of excitement or arousal-seeking has been linked with aggressive behaviour in antisocial and dissocial personality disorders (Howard Reference Howard2011). Instrumental aggression can also be explained as a way of reaching a predicted preferred affect state (e.g. from a baseline state to the anticipated affect state associated with illegally acquiring an item of value; or the use of domestic violence in the context of a wider coercive dynamic to reach an affect state associated with dominance and control). The intensity of the affect associated with the meaning ascribed to a stimulus can be influenced by contextual factors. Thus, the experience of being challenged or humiliated is likely to be heightened by the presence and reaction of onlookers (Mann Reference Mann, Feddes and Leiser2017). This context may also lead to an aggressive action impulse being favoured since it is associated with an anticipated status boost and/or an avoidance of status loss in comparison with the anticipated outcomes of a non-aggressive response (Anderson Reference Anderson, Hildreth and Howland2015). This may in part be dependent on the individual's perception about the onlookers' beliefs and attitudes.

Substance use

Empirical studies have found a robust relationship between violence and the use of alcohol and drugs (Duke Reference Duke, Smith and Oberleitner2018). Much of the research into the mechanisms that may account for this association has focused on the effects of alcohol. The pharmacological effects that increase the likelihood of violence include (a) narrowing of attention (which for aggression-prone individuals may lead to a narrowing of attention on a stimulus perceived as provoking), (b) reduction in the capacity to generate a range of potential responses to the perceived provocation (resulting in a temporary state akin to poor problem-solving), (c) an intensification of the affect state (which may increase the likelihood of aggression if the affect state is anger or fear) and (d) reduction in the power of inhibitory influences (such as implicit perspective-taking or explicit consequences of antisocial behaviour) (Graham Reference Graham, West and Wells2000; Parrott Reference Parrott and Eckhardt2018; George Reference George, Grant, Coccaro and McCloskey2019). Social factors that may also contribute to the association between violence and substance use include the circumstances in which the substance is obtained (e.g. engagement in antisocial behaviour with antisocial peers) and consumed (e.g. where groups of people are in confined spaces and in-group/out-group dynamics are more prevalent) (Perdersen Reference Perdersen, Copes and Sandberg2016).

Sexual violence vignette

A is a 25-year-old male who, in the early hours of the morning, was walking home from a night out with his friends when he saw an adult female ahead of him walking alone. He caught up with the woman and physically assaulted and attempted to rape her. He fled when he heard calls from a window overlooking the alley where the offence took place. On assessment, he admitted to a history of voyeurism, but he did not have any previous convictions. He disclosed that since his teenage years his level of sexual arousal had been enhanced by images of him engaging in aggressive sexual acts with non-consenting post-pubescent females. He reported that he had not planned the offence and the point at which he definitely knew he was going to commit the offence was shortly after seeing the victim. However, he said that earlier that evening he had fallen out with his friends and he recognised that, when frustrated for any reason, he was more likely to seek out aggressive sexual pornography. A's subjectivity is evaluated in Box 3.

Psychotic violence vignette

B is a 23-year-old man who was transferred to hospital from police custody under section 2 of the Mental Health Act 1983. He had been taken into custody after being arrested for an assault on another man at a railway station. During the in-patient assessment, B was deemed to be suffering from a psychotic illness. Several months before his admission, he had developed a belief that he had been placed under surveillance in his flat by neighbours whom he thought were working for the government. Later his beliefs became more generalised in that he thought the surveillance was continuing wherever he went and that he was about to be abducted and tortured. He also experienced voices warning him to be careful about people as he passed them in the street. Sometimes the voices would encourage him to attack his persecutors, although he did not want to. On the day of his arrest, he had become so overwhelmed by his experiences that he decided to leave the area and travel by train to London. As he came out of a toilet cubicle at the railway station, he saw a man standing in front of him. He was convinced that the man touched his ear as if activating a hidden microphone and the voice told him that the man was going to prevent him leaving. B struck the victim's face and ran out of the toilets. Box 4 outlines an evaluation of B's subjectivity.

Violent robbery vignette

C committed a robbery of a newsagents when he was 19 years old. Seeing that the shop was empty, he went in and walked straight up to the counter. He opened his jacket to show that he was holding a knife and demanded that the cashier give him the money from the till. She had started placing money on the counter when a young man entered the shop. C grabbed the money and made for the door but the man, who had not seen the knife, grabbed him. C lashed out with the knife, causing the man to sustain a cut to his neck and a stab wound to his thigh. When interviewed, C stated that he had a heroin and cocaine addiction and he had planned to spend the proceeds of the robbery on drugs. He had taken cocaine before the offence. C's subjectivity is evaluated in Box 5.