Loneliness hurts.

Reynolds points to an article by Cacioppo et. al., who contribute to a special issue of Cortex magazine on Neuro-cognitive mechanisms of social interactions. From her review:

For early humans, being alone was no way to live. Those on the tribe’s periphery faced increased risks of starvation, predation and early death. And so humans (like other communal creatures) evolved what seem to be specific biological reactions to social threats. A social animal that feels itself to be isolated from its kind begins to behave nervously and experiences unhealthy physiological responses. The body produces more stress-­related biochemicals, leading to inflammation and a reduced ability to fight viral infections. These adaptations might help explain why many chronically lonely people have an overabundance of stress-­related cells and weakened immune systems. But how they see the world — how loneliness affects their thinking — may be just as consequential to their health.

The abstract of the Cacioppo et. al. article, which finds that lonely people become inadvertently hypervigilant to social threats, which makes the loneliness worse:

Being on the social perimeter is not only sad, it is dangerous. Our evolutionary model of the effects of perceived social isolation (loneliness) on the brain as well as a growing body of behavioral research suggests that loneliness promotes short-term self-preservation, including an increased implicit vigilance for social, in contrast to nonsocial, threats. However, this hypothesis has not been tested previously in a neuroimaging study. We therefore used high density EEG and a social Stroop interference task to test the hypothesis that implicit attention to negative social, in contrast to nonsocial, Words in the Stroop task differs between individuals high versus low in loneliness and to investigate the brain dynamics of implicit processing for negative social (vs nonsocial) stimuli in lonely individuals, compared to nonlonely individuals (N = 70). The present study provides the first evidence that negative social stimuli are differentiated from negative nonsocial stimuli more quickly in the lonely than nonlonely brains. Given the timing of this differentiation in the brain and the fact that participants were performing a Stroop task, these results also suggest that these differences reflect implicit rather than explicit attentional differences between lonely and nonlonely individuals. Source estimates were performed for purposes of hypothesis generation regarding underlying neural mechanisms, and the results implicated the neural circuits reminiscent of orienting and executive control aspects of attention as contributing to these differences. Together, the results are in accord with the evolutionary model of loneliness.