Facing a (security) threat, what is the best thing to do? The overall research questions driving this study are about how security professionals assess, reason, and decide about security risks, and where their justification is founded on. The presented results are explorative, based on trying primarily to understand individual professional judgment and secondary, if possible, to explain the reasons behind this judgement. This work will not answer all the questions, it is, however, a valuable start to understand the difficult task the professionals in this domain are facing: preparing for, and thus predicting possible security threats....@en

Professionals working in both the physical and cybersecurity domain need to assess and evaluate security risks. As information on risks in general and security risks in particular is often imperfect and intractable, these professionals are facing a challenge in judging both likelihood and consequences, but how much do their existing psychological biases play a role in these judgments? In this paper, we present new empirical evidence on the perception of the information position and confidence levels of security professionals, the influence of detailed information and the conjunction fallacy, and the level of noise in security assessments. This paper adds to the literature by examining, for the first time, risk assessments by professionals in realistic, real life, security cases. The results show clear indications for overconfidence, comparative ignorance, influence of the conjunction fallacy, and influence of individual experience on security decision making in the professional security domain. The observed phenomena might have far reaching effects on security risk management in organizations and society.@en

Paper ID – 0348@en

Risk assessments in the (cyber) security domain are often, if not always, based on subjective expert judgement. For the first time, to the best of our knowledge, the individual preferences of professionals from the security domain are studied. In on online survey they are asked to mention, rate and rank their preferences when assessing a security risk. The survey setup allows to differentiate between easily accessible or “on top of mind” attributes and guided or stimulated attributes. The security professionals are also challenged to both non-compensatory and compensatory decision making on the relevance of the attributes. The results of this explorative study indicate a clear difference and shift in the individual perceived relevance of attributes in these different settings. Another remarkable finding of this study is the predominant focus on impact attributes by the respondents and the less significant position of likelihood or probability. The majority of professionals seem to ignore likelihood in their security risk assessment. This might be due to so called probability neglect as introduced by other scholars. the security in organisations and society is depending on the assessment and judgement of these professionals, understanding their preferences and the influence of cognitive biases is paramount. This study contributes to this body of knowledge and might raise attention to this important topic in both the academic and professional security domain.@en

Security professionals play a decisive role in security risk decision making, with important implications for security in organisations and society. Because of this subjective input in security understanding possible biases in this process is paramount. In this paper, well known biases as observed and described in prospect theory are studied in individual security risk decision making by security professionals. To this end, we distributed a questionnaire among security professionals including both original dilemmas from prospect theory and dilemmas adapted to the context of incident prevention. It was hypothesised that security professionals dealing with risks and decision making under risk on an almost daily basis would or should be less vulnerable to decision biases involving risks, in particular when framed in terms of incident prevention. The results show that security professionals are vulnerable to decision biases at the same scale as lay people, but some biases are weaker when decision problems are framed in terms of security as opposed to monetary gains and losses. Of the individual characteristics defining experience, only the general education level observably affects vulnerability for biases in security decision making in this study. A higher general education level leads to a significantly higher vulnerability to decision biases. By highlighting the vulnerability of security professionals to decision biases, this study contributes essential awareness and knowledge for improved decision making, for example by different representation of probabilities and uncertainty.@en