People need to work to overcome their inherent biases in order to avoid falling for social engineering attacks, according to Heidi Mitchell at the Wall Street Journal.

“Criminals lure smart people into their traps by taking advantage of the unconscious, automatic processes that act as shortcuts to make our decision-making more efficient,” Mitchell explains. “These cognitive biases—arising from what’s often referred to as our ‘lizard brains’—can cause us to misinterpret information and make snap judgments that may be irrational or inaccurate.”

Professor Cleotilde “Coty” Gonzalez from Carnegie Mellon University told the Journal that criminals take advantage of human psychology to make their attacks more effective, explaining that “if something is presented as a loss, we are more willing to take a risk [to avoid it]; if it’s presented as a gain, we are OK with taking a safe option.”

As a result, people are more likely to fall for a scam that tells them they’re going to lose money, as opposed to one that offers to give them money.

Mitchell adds, “Or a scammer might send a message to your work email, claiming that there is a problem with an account at one of your corporate suppliers, and warning that your shipment—one that your boss is counting on—will be delayed unless you verify your account information in a link provided by them. The fake link leads to a fake website that looks like the real thing. By playing on your fear of losing access to your account, the scammer gets your credentials.”

Scammers also take advantage of authority bias and urgency bias to compel their victims to act. Authority bias can be seen in business email compromise (BEC) attacks, in which an attacker impersonates a person of authority within an organization and sends a request to a lower-level employee. Urgency bias is often tied into these attacks, and involves making the victim believe they must act quickly to fulfill a request.

New-school security awareness training can give your employees a healthy sense of suspicion so they can spot red flags associated with social engineering attacks.