Definition
In-group bias is the tendency to favor members of one's own group over members of an out-group across many domains — trust, generosity, empathy, attention, the benefit of the doubt, and moral judgment. It is the behavioral consequence of the brain's us-versus-them sorting.
Sapolsky stresses that in-group bias often operates as warmth toward "us" rather than as overt hostility toward "them." Much of the bias is not hatred at all; it is the quiet, unremarked advantage that members of one's own group routinely receive.
Why it matters
How it works
Once the brain has sorted someone into the in-group, that person receives a cascade of advantages: more trust, more individuation, more empathy, more charitable interpretation of ambiguous behavior. The out-group receives the opposite by default — not necessarily malice, but the absence of these gifts.
This makes in-group bias the engine of group cohesion and also a quiet source of inequity. The same mechanism that lets a group cooperate and care for its own can leave outsiders systematically disadvantaged. Recognizing the favoritism — not just the hostility — is what makes the bias addressable.
The favoritism needs almost nothing to switch on. Henri Tajfel's "minimal group" studies of the 1970s assigned people to groups on trivial criteria — a preference for paintings by Klee versus Kandinsky, even a coin flip — and found that with no prior contact, no expected interaction, and full knowledge the grouping was arbitrary, participants still reliably allocated more to their own group. Cialdini's unity principle is the compliance application of the same machinery: being categorized as "one of us" produces the in-group advantage without the operator needing to be liked or trusted, which is why the honest question to ask of any "we" is whether it includes you with your actual interests or has been manufactured to borrow the cooperation effect.