Loneliness May Drive Addictive Behavior in College Students, but Stress and Flexibility Hold the Key

Loneliness has quietly become one of the most pervasive health complaints on university campuses, and a new study suggests it may do far more than darken a student’s mood. Researchers in Türkiye report that lonely college students are significantly more likely to engage in addictive behaviors, and that the pathway connecting social isolation to addiction appears to run directly through the experience of stress. The findings, published in the journal Current Psychology, also reveal something encouraging: a psychological trait called psychological flexibility can act as a buffer, dramatically weakening the lonely road toward addictive habits.

The study was conducted by Kıvanç Uzun of Uşak University, Gökmen Arslan of Burdur Mehmet Akif Ersoy University and the University of Melbourne, and Umut Aydoğdu of Burdur Mehmet Akif Ersoy University. Drawing on the psychological literature on emerging adulthood, the researchers set out to answer a question that has grown urgent for campus mental health services worldwide: why do some lonely students turn to substances and compulsive behaviors while others seem to cope without harm, and can the difference be measured and targeted?

To find out, the team recruited 419 students from a state university in Türkiye. The sample was 56.10 percent female and 43.90 percent male, with participants ranging in age from 20 to 28 years and an average age of 25.96 years with a standard deviation of 2.38. Because the research used a cross-sectional design, in which all measurements were collected at a single point in time, the authors are careful not to claim that loneliness causes addiction outright. Nevertheless, the statistical architecture of the study, which combined mediation and moderation analyses, allowed them to map the relationships among the variables with unusual precision.

The theoretical scaffolding for the work draws on several influential frameworks. Agnew’s general strain theory proposes that negative emotional states such as strain and distress push individuals toward deviant or compensatory behaviors, including substance use, as a way of managing painful feelings. The self-medication hypothesis, articulated by Khantzian, similarly holds that people use addictive substances to regulate overwhelming emotions. Loneliness research by Cacioppo and colleagues has shown that perceived social isolation is not merely an unpleasant feeling but a biological warning signal that reshapes cognition, heightens vigilance to social threat, and dysregulates stress physiology. In parallel, Sinha’s work on chronic stress has documented how sustained stress exposure sensitizes the brain’s antireward circuitry, increasing vulnerability to addiction.

Within this framework, the researchers proposed a mediated moderation model. First, they hypothesized that loneliness would positively predict addictive behavior. Second, they proposed that perceived stress, measured with established instruments rooted in the classic Cohen, Kamarck and Mermelstein conception of stress as the degree to which life situations are appraised as unpredictable, uncontrollable and overloaded, would partially mediate this link. The logic is straightforward: loneliness generates distress, and distress generates a motivational pressure to escape, and addictive behaviors offer the fastest available escape route. Third, and most novel, the team predicted that psychological flexibility would moderate the entire process, protecting lonely students from translating their pain into addictive action.

Psychological flexibility, the central construct of acceptance and commitment therapy as developed by Hayes and colleagues, refers to the ability to contact the present moment fully and, depending on what the situation affords, either persist with or change one’s behavior in the service of chosen values. Psychologically flexible individuals can feel lonely, notice the feeling, accept its presence without being dominated by it, and still choose actions aligned with their long-term goals. Psychologically inflexible individuals, by contrast, tend to fuse with painful thoughts, experientially avoid discomfort, and reach for immediate relief, a pattern that Hayes and colleagues have long associated with behavioral disorders including addiction.

The results confirmed the central predictions. Loneliness significantly and positively predicted addictive behavior among the students. Perceived stress played a partial mediating role in this relationship, meaning that part of the effect of loneliness on addictive behavior traveled through elevated stress. In other words, lonely students felt more stressed, and the stress itself contributed to the pull of addictive behaviors. The persistence of a direct effect alongside the indirect pathway suggests that stress is an important mechanism but not the only one; loneliness may also erode the social controls, belongingness and meaning structures that normally restrain compulsive behavior, a possibility consistent with Hirschi’s social control perspectives and recent cross-national evidence from Savolainen and colleagues linking perceived loneliness to youth addictive behaviors.

The moderation findings were the study’s most striking contribution. When students reported low psychological flexibility, the relationship between loneliness and addictive behavior was strong. When students reported high psychological flexibility, the relationship was not statistically significant. Put plainly, psychological flexibility appeared to neutralize the lonely-to-addictive pathway entirely at high levels. The finding echoes prior work by Kelly and colleagues, who studied US veterans during the COVID-19 pandemic and found that psychological flexibility influenced how loneliness related to substance use and functioning, and a growing body of intervention research showing that acceptance and commitment therapy can improve psychological flexibility in populations affected by substance dependence.

For campus health professionals, the practical implications are twofold. Preventive programs that merely address substance availability or peer pressure may miss the deeper engine of the problem, which is the emotional cascade from isolation through stress to escape-seeking behavior. Counseling services could instead screen for loneliness and perceived stress as early warning signals, and incorporate skills-based training in acceptance, mindfulness, values clarification and cognitive defusion, the active ingredients of flexibility-focused interventions. The authors emphasize that their results provide an important basis for preventive and supportive interventions targeting addictive behavior among college students, and prior meta-analytic work on risk and protective factors for addiction supports the notion that strengthening individual regulatory capacity can meaningfully reduce risk.

The study also carries broader cultural and developmental significance. Emerging adults aged roughly eighteen to twenty-nine occupy a uniquely volatile developmental window, as Arnett’s theory of emerging adulthood describes, in which identity exploration, residential mobility and unstable social networks converge. In collectivist cultural contexts such as Türkiye, where family connectedness and relatedness carry particular weight as Kağıtçıbaşı’s cross-cultural work has shown, separation from familiar support networks may generate especially acute loneliness. At the same time, the digital environments in which contemporary students socialize have been repeatedly linked to problematic use patterns, from smartphone and internet addiction to social media compulsion, with studies such as those by Bian and Leung and by Li and colleagues demonstrating that loneliness fuels these behaviors through boredom and diminished self-control.

As with any cross-sectional study, important limitations temper the conclusions. The single-time-point design cannot establish the direction of causality; it is plausible that addictive behaviors also deepen loneliness by damaging relationships, a bidirectional dynamic suggested by Mendelian randomization research on loneliness, smoking and alcohol use by Wootton and colleagues. The sample, drawn from one state university in a single country, limits generalizability, and self-report measures remain vulnerable to bias. Still, the size of the sample, the theoretical grounding of the model and the consistency of the findings with prior international evidence lend the work considerable weight.

What makes the research resonate beyond academia is its quietly hopeful message. Loneliness, stress and addiction form a chain that many students know intimately, but the chain has a weak link, and that link is trainable. Psychological flexibility is not a fixed trait reserved for the resilient few; it is a set of learnable skills that can be cultivated through therapy, structured programs and everyday practice. If lonely students can learn to hold their pain without being ruled by it, the study suggests, the path from isolation to addiction may be blocked before it begins. In an era when universities worldwide report rising mental health concerns, that insight may prove one of the most valuable tools campuses can offer.

Subject of Research: The roles of perceived stress and psychological flexibility in the relationship between loneliness and addictive behavior among college students.

Article Title: I feel lonely and stressed: psychological flexibility and addictive behavior in college students

Article References: Uzun, K., Arslan, G., & Aydoğdu, U. (2026). I feel lonely and stressed: psychological flexibility and addictive behavior in college students. Current Psychology, 45(18), Article 1525. https://doi.org/10.1007/s12144-026-10078-9

Image Credits: AI Generated

DOI: 10.1007/s12144-026-10078-9

Keywords: loneliness, addictive behavior, psychological flexibility, perceived stress, college students, mental health, moderated mediation, emerging adulthood, substance use, acceptance and commitment therapy, prevention, Current Psychology