Instagram and loneliness: late-night posts, fewer likes are clues, study finds

How a young person uses Instagram can give moderately useful clues to how lonely they feel, according to a study by researchers at the University of California San Diego published on Friday, 6 November 2026, in the journal npj Mental Health Research. The team analysed the Instagram activity of 415 American college students, looking at their photos, the language they used and how they interacted with others. These signals predicted how lonely the students said they felt with fair accuracy, but they were weaker at spotting social isolation, meaning how few social contacts a person actually has. The researchers stressed that the study cannot show whether loneliness changes Instagram behaviour or the other way round.

Loneliness among young adults is a concern for health researchers, and it is often hidden: a student can be surrounded by classmates and still feel alone. This study suggests that the trail people leave online carries quiet signals of how they are feeling.

The researchers separated two kinds of loneliness. Emotional loneliness is the feeling of lacking a close, intimate bond, such as a best friend or partner. Social loneliness is the feeling of lacking a wider circle. The Instagram signals for each were different. Late-night posting, negative words and sadder expressions pointed to emotional loneliness, while fewer likes, less commenting and fewer photos with faces pointed to social loneliness. For readers, these are not diagnostic rules, but they are a useful mirror: patterns in your own feed may reflect how connected you feel.

The study also raises a privacy question that the authors themselves flag. If a model can estimate loneliness from public Instagram activity, the same method could be used by companies or others without people's consent.

For Indian students who use Instagram daily, the findings are a reminder that social media can reflect wellbeing as much as it shapes it. The study was done on American students, so its results may not translate directly to India, but the basic idea, that a late-night, low-interaction feed may be a sign worth noticing, is easy to apply to one's own habits.

Peepals calculation: an AUC of 0.77 means that, given one lonelier and one less lonely student, the model would rank them correctly about 77% of the time, against 50% for a coin toss; that is 27 percentage points better than chance (0.77 − 0.50). The late-night window the researchers flagged, 11 pm to 5 am, covers six hours, or a quarter of the day (6 / 24 = 0.25). Our calculation from the reported figures.

Short-term: The findings add to evidence that online behaviour mirrors wellbeing, and may prompt students to notice their own late-night, low-interaction patterns.

Long-term: With consent, similar methods could help universities and health services reach students who feel lonely; without safeguards, they could be misused to profile people.

Who is affected: College students and young adults who use Instagram, campus counselling services, researchers, and social media companies.

Questions and Answers