How well can we sense each other’s loneliness?

Experts have likened loneliness to a disease that changes the brain. Sadly, these changes often affect people in ways that further isolates them – for example, lonely people are more sensitive to negative facial expressions. If we're to break this cycle and provide friendship to the lonely, a starting point is to recognise that a person is feeling isolated. A new study in Journal of Research in Personality tests whether and how well we can do this.

Maike Luhmann and her colleagues asked 463 young adults in Germany (average age 18) to complete a short questionnaire about their feelings of loneliness – for example, one item was "How often do you feel left out?". Each participant's romantic partner (if they had one), one of their parents, and their best friend also filled out the same questionnaire about the participant. The researchers simply looked to see how well the different assessments correlated. For comparison, the same procedure was performed for self- and other-ratings of life satisfaction.

There was a strong correlation between the participants' own self-rated loneliness scores and those given by their friends, partners and relatives (in statistical terms, the correlation between own ratings and best friend ratings was 0.37; for partners it was 0.66; and for parents it was 0.43). The researchers described this convergence as "substantial" and of similar size to that found for life satisfaction.

As the correlation scores show, the participants' partners tended to be better at judging their loneliness than did their friends and parents.  In fact, there was no significant difference in statistical terms between participants' ratings of their own loneliness and the ratings given to them by their partners. In contrast, parents and friends tended to underestimate the participants' loneliness.

Interestingly, parents' and friends' ratings of the participants' loneliness were not correlated with each other, but both were correlated with the partner ratings. The researchers said this suggests: "parents utilise information that is also available to the partners but not to the friends. Similarly friends utilise information that is also available to the partners, but not to the parents. In sum, partners have access to more information than both parents and friends."

This is one of the first ever studies to explore whether we can judge each other's loneliness and there is still much that we still don't know. Apart from the fact that these findings might not generalise to other cultures or older age groups, we also don't have any detail on how the friends, parents and partners were making their judgments – for example, were they using circumstantial cues (such as the relevant participant being on their own a lot) or behavioural cues, such as the participant acting withdrawn or hostile?

While more research is clearly needed, Luhmann and her team said their findings already have practical implications – for example, they suggest that asking friends and family about a person's loneliness is likely to be useful and reliable in cases in which someone is unable or unwilling to report their own feelings, such as might be true of young children or older people in care.

Further reading

Luhmann, M., Bohn, J., Holtmann, J., Koch, T., & Eid, M. (2016). I'm lonely, can't you tell? Convergent validity of self- and informant ratings of loneliness Journal of Research in Personality, 61, 50-60 DOI: 10.1016/j.jrp.2016.02.002