Neural variability fingerprint predicts individuals' information security violation intentions

As the weakest links in information security defense are the individuals in an organizations, it is important to understand their information security behaviors. In the current study, we tested whether the neural variability pattern could predict an individual???s intention to engage in information security violations. Because cognitive neuroscience methods can provide a new perspective into psychological processes without common methodolog-ical biases or social desirability, we combined an adapted version of the information security paradigm (ISP) with functional magnetic resonance imaging (fMRI) technology. While completing an adapted ISP task, participants underwent an fMRI scan. We adopted a machine learning method to build a neural variability predictive model. Consistent with previous studies, we found that people were more likely to take actions under neutral condi-tions than in minor violation contexts and major violation contexts. Moreover, the neural variability predictive model, including nodes within the task control, default mode, visual, salience and attention networks, can pre-dict information security violation intentions. These results illustrate the predictive value of neural variability for information security violations and provide a new perspective for combining ISP with the fMRI technique to explore a neural predictive model of information security violation intention.