Research co-authored by a psychologist from Effat University's College of Humanities identifies the internal mechanism that converts online harassment into suicidal thought
Research co-authored by a psychologist from Effat University's College of Humanities identifies the internal mechanism that converts online harassment into suicidal thought — and argues that prevention programs need to start addressing it.
The research linking cyberbullying to suicide risk is extensive. Study after study, across different countries and age groups, has confirmed that young people involved in online bullying — as perpetrators and as victims — are significantly more likely to experience suicidal ideation than those who are not. Middle-school students who are victimized online are nearly twice as likely to report suicide attempts as their uninvolved peers. Adolescents involved in bullying perpetration face more than twice the risk of suicidal ideation in the following year compared to those not involved at all.
What this body of research has been slower to establish is the mechanism. Cyberbullying and suicidal ideation are correlated. But correlation is not an intervention strategy. To build prevention programs that actually work, researchers need to understand what is happening in the psychological space between online harassment and thoughts of self-harm — what internal processes are being activated, and whether those processes can be identified and treated before they escalate.
A study published in BMC Psychiatry in February 2024, co-authored by Souheil Hallit of the Psychology Department at Effat University's College of Humanities, makes a specific and testable argument about what that mechanism looks like — and the findings point toward a variable that most current prevention frameworks have not been considering.
The Argument
The study was conducted as part of the PEARLS project, a large binational cross-cultural initiative covering Lebanon and Tunisia. The Lebanese portion of the data involved 3,103 healthy community participants — people with no prior history of diagnosed mental illness or antipsychotic medication — surveyed between June and September 2022. The mean age was 21.73 years, 63.6% were female, and 18.8% reported suicidal ideation at the time of the survey.
The variable the researchers chose to investigate as a potential mediator was psychotic experiences — not full psychotic disorders, but subclinical symptoms that occur in otherwise healthy people. These include positive psychotic experiences, such as unusual perceptions and paranoid-style thinking, and negative psychotic experiences, such as emotional flatness and social withdrawal. Both types were measured separately and tested as mediators in the path from cyberbullying to suicidal ideation.
The mediation analysis confirmed the hypothesis across all four models tested — positive and negative psychotic experiences each partially mediating the association between both cyberbullying perpetration and cyberbullying victimization and suicidal ideation. The chain worked in a consistent direction: greater cyberbullying involvement predicted more severe psychotic experiences, and more severe psychotic experiences predicted higher suicidal ideation. Direct associations between cyberbullying and suicidal ideation were also confirmed, independent of the mediation.
Why Perpetrators Matter Too
One of the less discussed aspects of the study is its finding that the mediation pathway holds for perpetrators as well as victims. Prevention conversations around cyberbullying tend to be focused on those who are targeted. The research consistently shows that victims face elevated suicide risk, and intervention programs are typically designed with victims in mind.
But the data here suggest that young people who engage in cyberbullying toward others also experience elevated psychotic symptoms and elevated suicidal ideation. This complicates the victim-perpetrator framing that dominates most prevention literature and points toward the need for interventions that address all young people involved in cyberbullying dynamics, not only those on the receiving end.
The Prevention Gap
The study's most direct practical contribution is its argument that suicide risk assessment for young people involved in cyberbullying should routinely include screening for subclinical psychotic symptoms. This is not standard practice. Most prevention frameworks center on depression, anxiety, and perceived stress as the psychological variables to assess and address in this context — and while those factors remain relevant, this research suggests that attenuated psychotic symptoms represent an additional and underappreciated risk marker that clinicians and school counselors are currently likely to miss.
Alongside individual screening, the researchers call for multilevel prevention approaches combining school-based, community, and clinical interventions. Programs built around digital citizenship, communication skills, empathy training, and coping strategies are identified as evidence-based tools with demonstrated potential to reduce suicide risk in young people affected by cyberbullying.
The researchers are also realistic about the limits of cyberbullying reduction as a primary prevention strategy. In a world where digital platforms are woven into every aspect of young people's social lives, the goal of meaningfully reducing cyberbullying prevalence faces obvious constraints. Addressing the internal factors that determine why cyberbullying raises suicide risk in some individuals — and not in others — is both a more tractable target and a more clinically useful one.
Limitations and Next Steps
The study's authors are explicit about what the research cannot conclude. The cross-sectional design means that the direction of causality cannot be established from these findings alone. It is possible, for example, that individuals who already experience psychotic symptoms are more likely to perceive neutral online interactions as hostile, or more likely to engage in cyberbullying behavior as a response to those symptoms — a direction of effect that cannot be ruled out from this data.
The sample's demographic profile also limits generalizability. Participants were predominantly female, unmarried, tertiary-educated, and living with family — characteristics that do not represent the full breadth of the young adult population. Future longitudinal studies in larger and more diverse samples are the identified priority for confirming and extending these findings.