The research, published in the American Journal of Criminal Justice by Jason R. Silva of William Paterson University and Adam Lankford of the University of Alabama, is the most comprehensive assessment of its kind to date. The authors examined every United States mass shooting from 2000 to 2023 in which four or more victims were killed, a total of 483 incidents. Drawing primarily on a National Institute of Justice funded mass shooting database compiled by Jillian Turanovic and colleagues and updated through 2023, the team verified and supplemented the records using seven other open-source datasets, The Violence Project database, official documents such as court records and government reports, and extensive news coverage. Official records were located for 49 percent of cases, and information was corroborated across multiple sources wherever possible, with official documentation given precedence when accounts differed. Three researchers independently reviewed every case for evidence of pre-attack communication, and the final coding decisions were unanimous after discussion and third-party review of any contested cases.

The central distinction the researchers drew is conceptually important. Leakage refers to violent intent communicated to a third party, such as when person A tells person B about an interest in harming person C. A threat refers to violent intent communicated to the potential victim directly. The team also coded whether perpetrators revealed a specific interest in committing a mass killing, or only a general interest in violence such as homicide, along with the mode of communication, the type of recipient, whether the public reported the information, and whether the criminal justice system responded. In total, 197 of the 483 mass shootings, or 41 percent, were preceded by the perpetrator revealing violent intent through leakage or threats.

The type of mass shooting mattered enormously. Public mass shootings, those occurring in populated locations with victims chosen at random or for symbolic value, showed by far the highest rates of pre-attack communication. Leakage or threats preceded 71 percent of public mass shootings, with leakage occurring in 56 percent of cases and threats in 42 percent. By contrast, leakage and threats were less common among family mass shootings and least common among felony mass shootings, those connected to underlying criminal activity such as robbery, gang violence, or drug deals. In fact, family mass shooters were the group most likely to have never leaked, threatened, or discussed their violent intent with anyone, a pattern found in 61 percent of those cases. Notably, the researchers’ hypothesis that family and felony shooters would favor direct threats over leakage was not supported. Public shooters exceeded the other subtypes on every measure of pre-attack communication, largely because they were far more likely to reveal a specific interest in committing a mass killing, which occurred before 35 percent of public shootings but only 8 percent of family shootings and 2 percent of felony shootings.

Why would perpetrators deliberately sabotage the secrecy their attacks require? The study tested three theories using logistic regression models that controlled for the incident year, the volume of news coverage, and the availability of official records. The first is criminal propensity, rooted in Gottfredson and Hirschi’s general theory of crime, which holds that some individuals are so prone to risk-taking that the long-term consequences of talking about violence seem trivial. The second is the cry for help theory, adapted from suicidality research, which suggests that some perpetrators consciously or unconsciously wish to be stopped or rescued before committing acts that will almost certainly end in their death or imprisonment. The third is attention-seeking, grounded in Merton’s anomie theory, which proposes that individuals who lack legitimate means to attain valued goals such as recognition may resort to extreme or provocative statements.

All three theories received strong empirical support. When leakage and threats were analyzed as a composite measure, prior arrest records and histories of domestic violence, markers of criminal propensity, mental health problems and suicidal ideation, markers of a cry for help, and fame-seeking and ideological motives, markers of attention-seeking, were all significantly associated with pre-attack communication. When the behaviors were disaggregated, an instructive pattern emerged. Mental health problems, fame-seeking, and ideological motives predicted leakage specifically, while prior arrests, domestic violence history, mental health problems, and suicidal ideation predicted direct threats. The authors suggest this split makes intuitive sense: attention-seeking individuals may recognize that leaking to a third party garners attention without the added risk of making a threat that is more likely to be reported, whereas individuals with criminal propensity may care less about risks and may be accustomed to using threats coercively from past behavior.

The study also traced precisely where prevention efforts break down, and the results point more toward the public than the system. Of the 197 shootings involving leakage or threats, 54 percent were reported by the public before the attack, meaning nearly half went unreported. Yet when reports were made, 83 percent produced a formal criminal justice response, usually involving law enforcement being dispatched or the eventual perpetrator being interviewed or investigated. The attrition varied sharply by subtype. Public mass shootings were reported most often, at 61 percent of cases with leakage or threats, but those reports were least likely to trigger a response, at 70 percent. Felony mass shootings were reported only 20 percent of the time, though when they were reported, the criminal justice system responded in every single case. Family mass shootings followed a similar pattern, with reports producing a response 98 percent of the time. Even when responses occurred, escalation to arrest happened in only 21 percent of cases with leakage or threats, and incarceration in only 6 percent.

Perhaps the most counterintuitive finding concerns what motivates people to report. The researchers had hypothesized that fear of retribution would suppress reporting when shooters had prior arrests or histories of domestic violence, and that perceived attention-seeking behavior would likewise dampen reports. Both predictions failed. Shooters who directly threatened their victims were reported 71 percent of the time, compared with 48 percent for those who merely leaked, and 80 percent when both occurred. Perpetrators with prior arrest records, domestic violence histories, mental health problems, or pre-attack suicidal ideation were all reported at or above average rates. Fame-seeking shooters were reported 74 percent of the time, well above the 54 percent baseline. The authors conclude that public uncertainty about whether someone is genuinely dangerous, rather than fear of the dangerous person, appears to be the greatest barrier to reporting. Clearer signals, specific interest in mass violence, communication through multiple modes or to multiple recipients, and the expression of both leakage and threats, all raised reporting rates significantly, sometimes to 73 or 78 percent.

Only one hypothesis about reporting was confirmed in the unexpected direction: ideologically motivated shooters were reported less often than average, at 46 percent. The researchers suggest that members of the public may struggle to differentiate between people who intend to commit ideologically driven attacks and others who hold extremist views but are not violent, a challenge that even professional threat assessors face, since radicalized beliefs are themselves a weak predictor of attacks. Overall, the pattern aligns closely with the classic bystander effect described by Latané and Darley, in which diffusion of responsibility, pluralistic ignorance, and evaluation apprehension lead people to assume someone else will act or that the situation is not truly an emergency. The authors recommend that law enforcement agencies educate the public about evidence-based warning signs, encourage reporting even when people are uncertain, provide easy and anonymous reporting channels, and publicize success stories showing how community members’ reports helped avert violence. The findings suggest that while it may be impossible to prevent all mass shootings, a substantial share of them broadcast their intentions in advance, and the greatest room for improvement lies in convincing ordinary bystanders that what they are hearing is worth taking seriously.

Subject of Research: Pre-attack leakage, threats, public reporting, and criminal justice responses in U.S. mass shootings

Article Title: A Comprehensive Assessment of Leakage, Threats, Public Reporting, and Responses to Reports among Public, Family, and Felony Mass Shootings

Article References: A Comprehensive Assessment of Leakage, Threats, Public Reporting, and Responses to Reports among Public, Family, and Felony Mass Shootings. (n.d.). https://doi.org/10.1007/s12103-026-09946-8

Image Credits: AI Generated

DOI: 10.1007/s12103-026-09946-8

Keywords: mass shootings, leakage, threats, violence prevention, threat assessment, criminal justice, bystander effect, public reporting, criminology, fame-seeking, suicidal ideation, open-source research

The first component of the model concerns the emergence of potential perpetrators. Using demographic data drawn from the U.S. Census, the researchers estimated a per-capita rate at which motivated individuals arise in a given population, and this rate was estimated jointly with the other model parameters rather than fixed in advance. This joint estimation is a key strength of the Bayesian approach: uncertainty in one parameter, such as perpetrator emergence, can be constrained by observations that also bear on other parameters, such as firearm availability or population structure. By allowing all parameters to be inferred simultaneously from the historical record of events, the framework avoids the pitfalls of chaining together independently derived estimates and produces a coherent probabilistic picture of shooting risk across the United States.

The second component, access to firearms, proved to be the dominant predictor of risk in the model. Because comprehensive data on private gun ownership are difficult to obtain, the researchers used the local density of Federal Firearms License (FFL) holders as a proxy for firearm availability. This proxy emerged from the calibration as the single most influential factor shaping the geographic distribution of expected shootings, outweighing the contributions of population density alone or the estimated rate of perpetrator emergence. In practical terms, the model implies that the availability of a firearm may function as an important limiting factor in mass shooting risk: where guns are more accessible, the probability that a motivated individual can translate intent into a high-fatality event rises substantially, holding other conditions constant.

The third component captures the presence of suitable target populations, reflecting the observation that mass shootings typically occur in places where potential victims gather, such as schools, workplaces, retail spaces, and public venues. By combining these three ingredients within an agent-based architecture, the simulation generates stochastic realizations of shooting risk across counties, producing not just a single point estimate but a distribution of plausible outcomes. This probabilistic character is central to the study’s forecasting ambitions. Just as epidemic models produce ranges of possible outbreak trajectories rather than deterministic predictions, the shooting model yields expected event counts and their uncertainty, allowing researchers and policymakers to reason about risk in a statistically disciplined way.

To test the model, the authors compared its predictions against mass shooting events recorded in the Mother Jones database covering 1982 through 2024. This widely used database includes approximately 160 events with three or more fatalities, excluding the perpetrator, and deliberately excludes incidents categorized as terrorism, gang-related violence, or domestic disputes. The exclusion criteria matter: they define a specific and arguably more preventable class of public mass shootings, distinct from forms of gun violence with different underlying dynamics. The researchers found that the models assigned higher risks of shootings precisely to areas where shootings had historically occurred, demonstrating that the framework can discriminate between geographic areas with genuinely different levels of risk rather than merely reproducing population counts.

Beyond retrospective validation, the calibrated model identifies the areas facing the greatest expected burden in the coming years. The ten highest-risk areas by expected events per decade are Maricopa County, Arizona; Harris County, Texas; Dallas County, Texas; Los Angeles County, California; Kings County, New York; Clark County, Nevada; Orange County, California; Queens County, New York; Cook County, Illinois; and Philadelphia County, Pennsylvania. The list spans large metropolitan counties across the Sun Belt, the coasts, and the industrial Midwest, and reflects the combined influence of substantial populations, dense networks of firearm dealers, and the urban characteristics that concentrate suitable targets. Notably, several of these counties also rank among the most populous in the nation, but the model’s ranking is not a simple population map; the dominance of FFL density as a predictor reshapes the geography of expected risk in ways that pure demographic extrapolation would not.

According to the authors, agent-based models of this kind can outperform standard statistical baselines in identifying areas at elevated risk. The improvement arises from the models’ structural realism: instead of fitting a regression to correlate inputs with outcomes, the simulation encodes a causal narrative, a motivated perpetrator meeting a firearm and a target population, and lets the data determine how strongly each element drives observed events. When the fitted mechanism reproduces historical patterns better than atheoretical alternatives, confidence grows that the model captures something real about the process, not merely a statistical coincidence. That realism also makes the framework adaptable; as new data arrive or as policy environments change, parameters can be re-inferred and projections updated within the same Bayesian machinery.

The potential applications extend to the allocation of prevention resources. If the geographic distribution of risk can be estimated with quantified uncertainty, interventions such as threat assessment programs, community-based violence prevention, extreme risk protection orders, and law enforcement preparedness could, in principle, be targeted toward the communities where expected benefits are largest. The model’s emphasis on firearm access as the dominant predictor also speaks directly to ongoing policy debates. If access to a firearm is an important limiting factor in whether mass shootings occur, then measures that reduce availability in high-risk areas, or that introduce friction into the acquisition process, may carry disproportionate preventive value relative to interventions focused solely on identifying perpetrators in advance, a task that remains notoriously difficult given the rarity and unpredictability of individual events.

The study also carries important caveats and a broader scientific significance. An agent-based model calibrated on roughly 160 events over four decades cannot predict when or where any specific attack will happen, and the use of FFL density as a proxy for gun availability introduces measurement uncertainty that the Bayesian framework represents but cannot eliminate. Mass shootings are rare, socially complex events shaped by factors, including mental health, social contagion, media dynamics, and economic conditions, that no single model fully captures. Yet the value of the work lies in demonstrating that even deeply troubling social phenomena can be brought within the reach of formal, mechanistic, data-constrained science. By combining agent-based simulation with Bayesian calibration, the researchers offer a template for rigorously reasoning about rare violent events, and a tool that, its authors argue, may help direct scarce prevention resources toward the places where they can do the most good.

Subject of Research: Bayesian agent-based modeling of the geographic risk and potential severity of mass shootings in the United States

Article Title: Forecasting mass shootings

Article References: Forecasting mass shootings. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: mass shootings, agent-based modeling, Bayesian inference, firearm availability, geographic risk, PNAS Nexus, gun violence prevention, Federal Firearms License, Mother Jones database, risk forecasting, public health, United States