Strategies to enhance intelligence analysis

If you've ever watched a thriller about undercover agents, you probably have the impression that intelligence officers are models of objectivity, pragmatism and sharp, unbiased thinking. However, in reality even the most well-trained and highly honed intelligence brain is still a human brain. As such it is vulnerable to influences that may steer it towards ill-judged decisions – these are known as 'cognitive biases'. The RECOBIA project, which held its final conference in Brussels last week, sought to explore and assess these cognitive biases, investigate their affect on the practice of intelligence analysis and develop mitigation strategies to address their impact.

Speaking at the conference, RECOBIA project coordinator Frederik Schumann of CEIS defined cognitive biases as 'psychological distortions produced by the use of rules of thumb which fasten our cognitive processing of information'. One example of a cognitive bias would be our tendency to trust information from an 'older, white, male scientist' regardless of whether their expertise is relevant to solving our specific problem. Another is the 'familiarity bias' – our tendency to trust information from a person we have met before over information from someone we don't know. Frederik explained how, over the past three years, the nine RECOBIA partners have worked together to develop strategies that will ultimately help mitigate the impact of cognitive biases in intelligence to improve quality in the field.

A core step in the RECOBIA process was identifying seven Key Intelligence Tasks (KITs) carried out by those working in the field of intelligence on a daily basis. These KITs were then examined in detail in order to identify and document 28 cognitive biases that intelligence officers may be subject to in the course of their work. The RECOBIA team then brought these concepts to life with 40 specific situations that illustrate the impact and effects of cognitive biases.

The identified KITs, cognitive biases and specific situations formed the basis for the RECOBIA scenario which was presented at the conference. Using a fictional intelligence officer named Peter and an imagined border incident between 'Syldavia' and 'Borduria', the scenario demonstrates how cognitive biases play out in the course of an invented intelligence situation, and how RECOBIA mitigation strategies, also presented at the conference, can be applied to enhance the performance of intelligence analysts. As Frederik noted, 'Recognising the situations in which cognitive biases might occur and knowing and applying mitigation strategies is in the general interest of every intelligence officer'.

The RECOBIA team keenly emphasised the fact that all of the project outputs were developed in conjunction with end-users over the course of six workshops. In total, more than 100 end-users representing 21 national agencies and five European institutions gathered at one or multiple workshops where they interrogated, contributed to and ultimately validated the RECOBIA research.

The project will conclude on 31 January but the team is adamant that there is more to be done in this area. They believe that the next natural steps would include working with the project findings to design tools and training sessions with solutions tailored to specific needs of intelligence organisations. Additionally, the RECOBIA team notes, there is work to be conducted in the area of social cognitive biases.