The evolution of cooperation is a fundamental problem in biology, especially for non-relatives, where indirect fitness benefits cannot counter within-group inequalities. Multilevel selection models show how cooperation can evolve if it generates a group-level advantage, even when cooperators are disadvantaged within their group. This allows the possibility of group selection, but few examples have been described in nature. Here we show that group selection can explain the evolution of cooperative nest founding in the harvester ant Pogonomyrmex californicus. Through most of this species’ range, colonies are founded by single queens, but in some populations nests are instead founded by cooperative groups of unrelated queens. In mixed groups of cooperative and single-founding queens, we found that aggressive individuals had a survival advantage within their nest, but foundress groups with such non-cooperators died out more often than those with only cooperative members. An agent-based model shows that the between-group advantage of the cooperative phenotype drives it to fixation, despite its within-group disadvantage, but only when population density is high enough to make between-group competition intense. Field data show higher nest density in a population where cooperative founding is common, consistent with greater density driving the evolution of cooperative foundation through group selection.

Detecting signs that someone is a member of a hostile outgroup can depend on very subtle cues. How do ecology-relevant motivational states affect such detections? This research investigated the detection of briefly-presented enemy (versus friend) insignias after participants were primed to be self-protective or revenge-minded. Despite being told to ignore the objectively nondiagnostic cues of ethnicity (Arab vs. Western/European), gender, and facial expressions of the targets, both priming manipulations enhanced biases to see Arab males as enemies. They also reduced the ability to detect ingroup enemies, even when these faces displayed angry expressions. These motivations had very different effects on accuracy, however, with self-protection enhancing overall accuracy and revenge-mindedness reducing it. These methods demonstrate the importance of considering how signal detection tasks that occur in motivationally-charged environments depart from results obtained in conventionally motivationally-inert laboratory settings.

Working memory (WM) theoretically affords the ability to privilege social threats and opportunities over other more mundane information, but few experiments have sought support for this contention. Using a functional logic, we predicted that threatening faces are likely to elicit encoding benefits in WM. Critically, however, threat depends on both the capacities and inclinations of the potential aggressor and the possible responses available to the perceiver. Two experiments demonstrate that participants more efficiently scan memory for angry facial expressions, but only when the faces also bear other cues that are heuristically associated with threat: masculinity in Study 1 and outgroup status in Study 2. Moreover, male participants showed robust speed and accuracy benefits, whereas female participants showed somewhat weaker effects, and only when threat was clearly expressed. Overall results indicate that working memory for faces depends on the accessibility of self-protective goals and on the functional relevance of other social attributes of the face.