FOLK MODELS of HOME COMPUTER SECURITY
http://www.schneier.com/blog/archives/2011/03/folk_models_in.html
http://www.rickwash.com/papers/rwash-dissertation-final.pdf
http://www.rickwash.com/papers/rwash-homesec-soups10-final.pdf
by Rick Wash / at SOUPS (Symposium on Usable Privacy and Security)

Home computer systems are insecure because they are administered by untrained users. The rise of botnets has amplified this problem; attackers compromise these computers, aggregate them, and use the resulting network to attack third parties. Despite a large security industry that provides software and advice, home computer users remain vulnerable. I identify eight ‘folk models’ of security threats that are used by home computer users to decide what security software to use, and which expert security advice to follow: four conceptualizations of ‘viruses’ and other malware, and four conceptualizations of ‘hackers’ that break into computers. I illustrate how these models are used to justify ignoring expert security advice. Finally, I describe one reason why botnets are so difficult to eliminate: they cleverly take advantage of gaps in these models so that many home computer users do not take steps to protect against them.

1. INTRODUCTION
Home users are installing paid and free home security software at a rapidly increasing rate.{1} These systems include anti-virus software, anti-spyware software, personal firewall software, personal intrusion detection / prevention systems, computer login / password / fingerprint systems, and intrusion recovery software. Nonetheless, security intrusions and the costs they impose on other network users are also increasing. One possibility is that home users are starting to become well-informed about security risks, and that soon enough of them will protect their systems that the problem will resolve itself. However, given the “arms race” history in most other areas of networked security (with intruders becoming increasingly sophisticated and numerous over time), it is likely that the lack of user sophistication and non-compliance with recommended security system usage policies will continue to limit home computer security effectiveness. To design better security technologies, it helps to understand how users make security decisions, and to characterize the security problems that result from these decisions. To this end, I have conducted a qualitative study to understand users’ mental models [18, 11] of attackers and security technologies. Mental models describe how a user thinks about a problem; it is the model in the person’s mind of how things work. People use these models to make decisions about the effects of various actions [17]. In particular, I investigate the existence of folk models for home computer users. Folk models are mental models that are not necessarily accurate in the real world, thus leading to erroneous decision making, but are shared among similar members of a culture[11]. It is well-known that in technological contexts users often operate with incorrect folk models [1]. To understand the rationale for home users’ behavior, it is important to understand the decision model that people use. If technology is designed on the assumption that users have correct mental models of security threats and security systems, it will not induce the desired behavior when they are in fact making choices according to a different model. As an example, Kempton [19] studied folk models of thermostat technology in an attempt to understand the wasted energy that stems from poor choices in home heating. He found that his respondents possessed one of two mental models for how a thermostat works. Both models can lead to poor decisions, and both models can lead to correct decisions that the other model gets wrong. Kempton concludes that “Technical experts will evaluate folk theory from this perspective [correctness] – not by asking whether it fulfills the needs of the folk. But it is the latter criterion […] on which sound public policy must be based.” The same argument holds for technology design: whether the folk models are correct or not, technology should be designed to work well with the folk models actually employed by users.{2} For home computer security, I study two related research questions: 1) Potential threats : How do home computer users conceptualize the information security threats that they face? 2) Security responses : How do home computer users apply their mental models of security threats to make security-relevant decisions? Despite my focus on “home computer users,” many of these problems extend beyond the home; most of my analysis and understanding in this paper is likely to generalize to a whole class of users who are unsophisticated in their security decisions. This includes many university computers, computers in small business that lack IT support, and personal computers used for business purposes.

{1} Despite a worldwide recession, the computer security industry grew 18.6% in 2008, totaling over $13 billion according to a recent Gartner report [9]
{2} It may be that users can be re-educated to use more correct mental models, but generally it more difficult to re-educate


http://www.csoonline.com/documents/flash/ammap/ammap.swf