“When heuristics don’t yield the results we expect, you’d think we would eventually realize that something’s wrong. Even if we don’t locate the biases, we should be able to see the discrepancy between what we wanted and what we got, right? Well, not necessarily. As it turns out, we have biases that support our biases! If we’re partial to one option—perhaps because it’s more memorable, or framed to minimize loss, or seemingly consistent with a promising pattern—we tend to search for information that will justify choosing that option. On the one hand, it’s sensible to make choices that we can defend with data and a list of reasons. On the other hand, if we’re not careful, we’re likely to conduct an imbalanced analysis, falling prey to a cluster of errors collectively known as “confirmation biases.”