Users' brains scanned in bid to fix infosec

Mind map holds hope for better security design

Security system design and user education could benefit from neuralimaging that uses brain scans to determine the way personality types may react to threats, new research shows.

A team of five researchers from the University of Alabama conducted functional Magnetic Resonance Imaging (fMRI) scans of two dozen human subjects to disocver how they responded to security threats such as malware warnings and phishing email attempts.

The research field, normally the provence of medicial professionals, opened avenues for comprehensive study into the ways users react to other security defense measures including password memorisation which could result in more effective system design.

The academics determined areas of the brain that were active when users consumed malware warning messages and those which lit up when users considered the legitimacy of phishing emails.

"In the long-run, such studies may provide a neural signature for poor and good security decisions which can be used for predicting as well as correcting users' security behaviour," the team wrote in a paper (pdf).

"Future research may conduct subsequent evaluation with diverse participant samples, study the effect of warning fatigue or habituation, consider user-centered security domains other than phishing detection and malware warnings, and evaluate the effect of security training and education on users’ performance."

The work commenced by determining how users reacted to malware warnings and phishing scams. It found what some in the industry may have anticipated; users were alert malware warnings and overt phishing scams, and less so of surreptitious phish attempts. 

Despite this, 40 percent of users still took the phishing bait and 15 percent of users fell victim after bypassing malware warnings.

Those who clicked past warnings were likely to fall victim to phishing, the neuralimaging showed.

But unlike conventional thought, users did not blindly click past warnings like those thrown up by Google Chrome; they had done so after "actively" comprehending the threat message.

The research specifically found that users brain activity was heightened during all threats yet performance wavered depending on the users' knowledge of the type of threat and indeed whether a threat was obvious-enough to be detected.

The human brain activated the right frontal and left parietal regions -- the areas of the brain important to sensory processing -- when determining whether a website was legitimate. Those regions also lit up in previous tests when subjects were asked to determine the legitimacy of Rembrandt paintings.

It also found that individuals with "higher impulsive traits" may be less effective at evaluating threats than those who were not, using Barratt’s Impulsiveness Scale which was designed to assess the personality and behavioural construct of impulsiveness.

The distinction between impulsive and non-impulsive users they said warranted valuable study into whether individual trait characteristics should factor into user-centered security design.

"We see a clear path-forward for subsequent research using neuralimaging techniques to inform the design of user-centered security systems," the team of Ajaya Neupane; Nitesh Saxena; Keya Kuruvilla; Michael Georgescu, and Rajesh Kana wrote.

Join the CSO newsletter!

Error: Please check your email address.

Tags security threatsfMRI scansMangnetic Resonance Imagingmalware warningsGoogle Chromephishingthreats

More about Googleindeed

Show Comments

Featured Whitepapers

Editor's Recommendations

Solution Centres

Stories by Sam Bells

Latest Videos

More videos

Blog Posts