Analysing forensic processes: Taking time into account

A great deal of forensic psychology concerns sequences of behaviours or events. In this paper, we review some recent efforts to examine forensic issues as sequences, discuss some of the contemporary methodologies involved, and highlight some of the lessons that emerge from this research. Specifically, we show: (i) how research on public violence has benefited from studying incidents as patterns of cues and responses among perpetrators and bystanders; (ii) how regularities in the histories of those who undertake suicide terrorism may be identified by mapping their life events on a graphical timeline; and (iii) how sequence-based correlation coefficients make it possible to test detailed theories about the ways perpetrators respond to the various influence attempts of police negotiators. We conclude by encouraging forensic psychologists to conceptualise their own areas of investigation as a sequence of events rather than a collection of variables. 1 To be counted as analysing process, an article had to consider data from at least three time points using a methodology that respected the temporal relations between the three points. Our decision not to include articles using two time points (e.g. the before–after designs prevalent in recidivism research) was made because such designs speak effectively to what happens but say less about the process by which it happens. A list of the papers identified as examining process is available at: http://www.lancs.ac.uk/staff/taylorpj/ifp. Please get in touch if you think we have overlooked an article relevant to process. 2 In case of doubt, the number of articles published in the British Journal of Social Psychology (BJSP) is significantly greater than those published in the top 10 forensic journals, as measured using a one-way t-test comparing BJSP’s count to the distribution of number of articles published in the top 10 forensic journals, t(9) = –7.87, p < .01. Paul J. Taylor et al. 46 Issues in Forensic Psychology No. 8 6-month and 12-month follow-up would be counted as measuring process. So would a study using time-series analysis to examine the unfolding violent interaction between husband and wife in conflict. In total, we found just 16 papers that examined process as set out by our criteria. This equates to just over two articles a year, and less than 1 per cent of all papers published by these journals over the seven years. By way of comparison, the British Journal of Social Psychology alone published seven articles on process within the same period. The absence of process in forensic publications makes it difficult to draw any conclusion except that forensic psychology has not focused on process nearly as much as it should have (for a similar conclusion in sociology see Abbott, 1990). However, this state of affairs is arguably not without good reason. There continues to be a paucity of available texts on sequence methodologies and their application (particularly compared to other methods such as general linear modelling or qualitative analysis), which makes it difficult for researchers to identify and use available methods. There is also an overriding tendency for researchers to organise their studies in terms of variables rather than events. A variable-centric approach is familiar and, when used appropriately, provides a rigorous way of testing theory. But it limits drastically what researchers can say about the importance of history and change to their phenomenon of interest. For example, the reason for an officer’s decision during a murder enquiry can only be understood in the context of previous decisions, evidence seen, and the order in which these decisions and evidence transpired. It is only by using sequence methods that we can begin to understand this decision because it is only sequence methods that examine the decision within the context of an unfolding set of behaviours and events. In sum, methods that examine temporality open up ways of answering some of the questions that forensic psychologists really want to answer. In the remainder of this paper we seek to provide a flavour of what studying process can offer forensic psychology, and how it has already begun to reshape practice. We do so by examining three areas of research. Our point of departure is recent studies of violence in the ‘night-time economy’. These studies demonstrate what can be learned about group violence simply by conceptualising an incident as a sequence of actions and reactions, rather than as a collection of occurring variables. We then introduce a graphical approach to understanding process by examining a sequence of events that occur over a much larger time scale – that of an individual’s life. We review how examining the life histories of female suicide terrorists can provide important insights into the critical events, turning points, and processes of radicalisation that underlie their involvement. Finally, we review research on hostage negotiation, conceptualised as a sequence of cues and responses, and how sequence methods have made it possible to test hypotheses about differences in the effectiveness of influence strategies across cultures. Answers to these hypotheses have informed negotiation strategy. One doesn’t have to look far One of the most striking things to emerge from studying cases as sequences of events is that it is not necessary to use sophisticated analyses to derive important findings. Often, simply deconstructing an event into a sequence of behaviours can be sufficient to illuminate what has previously been overlooked. The recent work on violence in the night-time economy by Mark Levine and his colleagues (Levine et al., 2007) is a compelling example of this. Violence and antisocial behaviour in public places continues to rise in the UK, against a background of decline in most crime figures (Nicholas et al., 2007). While alcohol consumption undoubtedly plays a part in this phenomenon, it cannot account for the variety of behaviour observed among individuals who engage in night-time economy activities. The behaviour of the violent few is modified both by Analysing forensic processes: Taking time into account Issues in Forensic Psychology No. 8 47 the social context in which drinking takes place, and by the cultural traditions that either inhibit or facilitate aggression in these contexts. Understanding when and why violent episodes erupt therefore offers much to crime reduction initiatives. To derive an initial understanding of night-time violence, Levine et al. (2007) collected 42 CCTV clips of violent episodes that did not include the intervention of the police or private security forces (e.g. bouncers). They then coded the observable behaviour of those involved in the incidents as either escalating (e.g. punch, kick) or deescalating acts (e.g. hold back), and noted both who acted, and to whom their behaviour was directed. So, for example, a coded behaviour might be a protagonist acting aggressively towards the target (i.e. protagonist–escalate–target), or a bystander acting to de-escalate another bystander (i.e. bystander–de-escalating–bystander). They then arranged these actor–action–target codes into time order. By analysing the resulting sequences of acts, Levine et al. were able identify certain features that lead to severe violence rather than de-escalation, and to tell a compelling story about the patterns of behaviour that occur in night-time economy violence. The first step in examining the nighttime violence data was to look at the contingencies among the actor–action–target codes. In its simplest form, a contingency is a count of the number of occasions that one event (e.g. protagonist–escalate–target) is followed by a second event (e.g. bystander–de-escalating–bystander) over the sequence being examined. Table 1 gives an example of contingencies in the night-time violence data. The left panel of Table 1 shows part of a sequence of behaviours and the right panel the counted contingencies. To derive the counts shown in the right panel, move from top to bottom of the sequence in the left panel and, on each occasion, note down the behaviour that immediately precedes the behaviour reached. The results when tallied and put in tabular form should match the table shown in the right-hand side of Table 1. As can be seen from Table 1, some behaviours occur Behaviour Sequence Contingency Table Behavioural Response Protagonist Target Bystander