Scenario 5: improvised explosive device material is found in an apartment. The only witness to activity there is an alcoholic in withdrawal.
Problem: How can law enforcement facilitate good reporting from individual citizens, including those who might usually provide unreliable information (e.g., distressed homeless persons, people who were inebriated when they saw the event)?
Strategies: Memories often exhibit what is called "state dependency." That is, if an event is perceived while someone is under the influence of a drug (e.g., alcohol), then it is remembered better if the recall occurs under the same state (that is, while drinking). A similar effect sometimes has been found to operate for moods--that is, memories also are recalled better if the person is in the same mood they were in when they acquired the memory. This poses a potential problem for effectively interviewing people with histories of the abuse of alcohol or other drugs.
Problem: Can we identify the likely characteristics of suicide bombers in the Middle East at present, ask how many of the same characteristics (especially, support structures) might or do exist in the United States, and use these characteristics to identify people who might be considered at risk? Similarly, can we use what is known about the men who engaged in the attacks on the United States on 9/11, to offer a likely description of such a terrorist?
Strategies: Data mining techniques might be useful here. The data mining technique is useful when someone is confronted with a huge collection of individual items (e.g., telephone call narratives, case histories, medical histories, indices of aberrant behavior) and needs these items be prioritized according to which should be pursued thoroughly. It also allows the user to find patterns of behavior that are almost impossible for the human observer to perceive. Importantly, then, this is not just a technique that is faster than a human. It is a technique that allows a human to view aspects of a large data set that they otherwise would not be able to see, and to discover interesting associations among data in the database. (10) (Some examples of this technique are in Appendix 5, "Data Mining Methodology.")
Problem: How can law enforcement create standard operating procedures that could be provided to local and regional offices to use when an event like that described in the scenario occurs?
Strategies: Most real-life decisions involve multiple decision stages--that is, a sequence of actions that are taken over time. Each action results in some consequence, which then affects the next action. During the past 30 years, decision researchers have learned a great deal about the basic principles of multistage decisions, which are represented in terms of a decision tree. (An example of a decision tree is provided in Appendix 4, Decision Tree Methodology.) The creation of a decision tree requires the collaboration of a technical advisor with those who are the recipients of the incoming information (and will use the decision tree). However, once such a tree is established, using it requires relatively little technical skill, so that it can be broadly applied.
It is important to note that decision trees, like data mining techniques, serve only to report to the user, who then makes a final decision regarding action and follow-up. It is important also to recognize that the use of these techniques can be effective while maintaining much of the privacy of the person or persons who supplied the information, thereby not violating civil liberties of the populace at large and protecting important sources of information.
Implications for practice, training and research
This scenario is an instance where it appears as if the social sciences have the necessary and useful tools, and law enforcement and intelligence have the data. Research is needed on how to:
* Provide social scientists with the data without compromising security, and
* Implement these research tools within the local and national cultures of law enforcement and intelligence agencies.
Social scientists in Israel have collected data pertinent to the Palestinian suicide bombers. (11) Research in the United States is needed to understand whether similar conditions exists in this country, and whether such conditions might develop in the future. Research also is needed as to whether the 19 attackers of September 11 shared characteristics with the Palestinian suicide bombers.
(10) Data mining comes from analyses of self-organizing map (SOM) networks, one of the most important network architectures developed during the 1980s. The main function of SOM networks is to map the input data from an n-dimensional space to a lower dimensional (usually one or two-dimensional) plot while maintaining the original topological relations. Therefore, it can be viewed as an analog of factor analysis.
(11) McCauley, C. (in press). Understanding the 9/11 perpetrators: crazy, lost in hate, or martyred? In N. Matuszak (Ed.), History behind the headlines, Vol. 5. Farmington Place, MI: Gale Publishing Group.
|Printer friendly Cite/link Email Feedback|
|Publication:||Countering Terrorism: Integration of Practice and Theory|
|Date:||Feb 28, 2002|
|Previous Article:||Scenario 4: three persons are arrested near a nuclear power plant; two refuse to talk and the third (who recently traveled to Indonesia, Yemen and...|
|Next Article:||Scenario 6 (a-c): how can law enforcement build effective ties to local Muslim communities, and what can these communities do to mobilize their...|