Lies, Damned Lies, and Social Media Following Extreme Events.

With the increased use of social media in crisis communication following extreme events, it is important to understand how the public distinguishes between true and false information. A U.S. adult sample (N = 588) was presented 20 actual social media posts following a natural disaster or soft-target terrorist attack in the United States. In this study, social media posts are conceptualized as truth signals with varying strengths, either above or below each individual's threshold for believing the post is true. Optimally, thresholds should be contingent on the (incentivized) error penalties and base-rate of true posts, both of which were manipulated. Separate receiver operating characteristic (ROC) analyses indicate that participants performed slightly better than chance for natural disasters and moderately better than chance for terror attacks. While the pooled thresholds are ordinally consistent with the base-rate and error penalty manipulations, they are underadjusted compared to the optimal thresholds. After accounting for demographic and cognitive variables, the base-rate manipulation significantly predicted sensitivity, specificity, and true response rates in the expected direction for both content domains, while the error penalty manipulation had no significant effect in either domain. Self-identified political conservatives performed worse at classifying false content as false for natural disasters, but better for terror attacks.

Behavioral Determinants of Target Shifting and Deterrence in an Analog Cyber-Attack Game.

This study examines how exploiting biases in probability judgment can enhance deterrence using a fixed allocation of defensive resources. We investigate attacker anchoring heuristics for conjunctive events with missing information to distort attacker estimates of success for targets with equal defensive resources. We designed and conducted a behavioral experiment functioning as an analog cyber attack with multiple targets requiring three stages of attack to successfully acquire a target. Each stage is associated with a probability of successfully attacking a layer of defense, reflecting the allocation of resources for each layer. There are four types of targets that have nearly equal likelihood of being successfully attacked, including one type with equally distributed success probabilities over every layer and three types with success probabilities that are concentrated to be lowest in the first, second, or third layer. Players are incentivized by a payoff system that offers a reward for successfully attacked targets and a penalty for failed attacks. We collected data from a total of 1,600 separate target selections from 80 players and discovered that the target type with the lowest probability of success on the first layer was least preferred among attackers, providing the greatest deterrent. Targets with equally distributed success probabilities across layers were the next least preferred among attackers, indicating greater deterrence for uniform-layered defenses compared to defenses that are concentrated at the inner (second or third) levels. This finding is consistent with both attacker anchoring and ambiguity biases and an interpretation of failed attacks as near misses.

Assessing the Benefits and Costs of Homeland Security Research: A Risk-Informed Methodology with Applications for the U.S. Coast Guard.

This article describes a methodology for risk-informed benefit-cost analyses of homeland security research products. The methodology is field-tested with 10 research products developed for the U.S. Coast Guard. Risk-informed benefit-cost analysis is a tool for risk management that integrates elements of risk analysis, decision analysis, and benefit-cost analysis. The cost analysis methodology includes a full-cost accounting of research projects, starting with initial fundamental research costs and extending to the costs of implementation of the research products and, where applicable, training, maintenance, and upgrade costs. The benefits analysis methodology is driven by changes in costs and risks leading to five alternative models: cost savings at the same level of security, increased security at the same cost, signal detection improvements, risk reduction by deterrence, and value of information. The U.S. Coast Guard staff selected 10 research projects to test and generalize the methodology. Examples include tools to improve the detection of explosives, reduce the costs of harbor patrols, and provide better predictions of hurricane wind speeds and floods. Benefits models and estimates varied by research project and many input parameters of the benefit estimates were highly uncertain, so risk analysis for sensitivity testing and simulation was important. Aggregating across the 10 research products, we found an overall median net present value of about $385 million, with a range from $54 million (5th percentile) to $877 million (95th percentile). Lessons learned are provided for future applications.

The dark figure of sexual recidivism.

Empirical studies of sexual offender recidivism have proliferated in recent decades. Virtually all of the studies define recidivism as a new legal charge or conviction for a sexual crime, and these studies tend to find recidivism rates of the order of 5-15% after 5 years and 10-25% after 10+ years. It is uncontroversial that such a definition of recidivism underestimates the true rate of sexual recidivism because most sexual crime is not reported to legal authorities, a principle known as the "dark figure of crime." To estimate the magnitude of the dark figure of sexual recidivism, this paper uses a probabilistic simulation approach in conjunction with the following: (i) victim self-report survey data about the rate of reporting sexual crime to legal authorities; (ii) offender self-report data about the number of victims per offender; and (iii) different assumptions about the chances of being convicted of a new sexual offense given that it is reported. Under any configuration of assumptions, the dark figure is substantial, and as a consequence the disparity between recidivism defined as a new legal charge or conviction for a sex crime and recidivism defined as actually committing a new sexual crime is large. These findings call into question the utility of recidivism studies that rely exclusively on official crime statistics to define sexual recidivism, and highlight the need for additional, long-term studies that use a variety of different measures to assess whether or not sexual recidivism has occurred.

Media Disaster Reporting Effects on Public Risk Perception and Response to Escalating Tornado Warnings: A Natural Experiment.

Previous research has evaluated public risk perception and response to a natural hazards in various settings; however, most of these studies were conducted either with a single scenario or after a natural disaster struck. To better understand the dynamic relationships among affect, risk perception, and behavioral intentions related to natural disasters, the current study implements a simulation scenario with escalating weather intensity, and includes a natural experiment allowing comparison of public response before and after a severe tornado event with extensive coverage by the national media. The current study also manipulated the display of warning information, and investigated whether the warning system display format influences public response. Results indicate that (1) affect, risk perception, and behavioral intention escalated as weather conditions deteriorated, (2) responses at previous stages predicted responses at subsequent stages of storm progression, and (3) negative affect predicted risk perception. Moreover, risk perception and behavioral intention were heightened after exposure to the media coverage of an actual tornado disaster. However, the display format manipulation did not influence behavioral responses. The current study provides insight regarding public perception of predisaster warnings and the influence of exposure to media coverage of an actual disaster event.

Public Response to a Near-Miss Nuclear Accident Scenario Varying in Causal Attributions and Outcome Uncertainty.

Many studies have investigated public reactions to nuclear accidents. However, few studies focused on more common events when a serious accident could have happened but did not. This study evaluated public response (emotional, cognitive, and behavioral) over three phases of a near-miss nuclear accident. Simulating a loss-of-coolant accident (LOCA) scenario, we manipulated (1) attribution for the initial cause of the incident (software failure vs. cyber terrorist attack vs. earthquake), (2) attribution for halting the incident (fail-safe system design vs. an intervention by an individual expert vs. a chance coincidence), and (3) level of uncertainty (certain vs. uncertain) about risk of a future radiation leak after the LOCA is halted. A total of 773 respondents were sampled using a 3 × 3 × 2 between-subjects design. Results from both MANCOVA and structural equation modeling (SEM) indicate that respondents experienced more negative affect, perceived more risk, and expressed more avoidance behavioral intention when the near-miss event was initiated by an external attributed source (e.g., earthquake) compared to an internally attributed source (e.g., software failure). Similarly, respondents also indicated greater negative affect, perceived risk, and avoidance behavioral intentions when the future impact of the near-miss incident on people and the environment remained uncertain. Results from SEM analyses also suggested that negative affect predicted risk perception, and both predicted avoidance behavior. Affect, risk perception, and avoidance behavior demonstrated high stability (i.e., reliability) from one phase to the next.

Valuing Equal Protection in Aviation Security Screening.

The growing number of anti-terrorism policies has elevated public concerns about discrimination. Within the context of airport security screening, the current study examines how American travelers value the principle of equal protection by quantifying the "equity premium" that they are willing to sacrifice to avoid screening procedures that result in differential treatments. In addition, we applied the notion of procedural justice to explore the effect of alternative selective screening procedures on the value of equal protection. Two-hundred and twenty-two respondents were randomly assigned to one of three selective screening procedures: (1) randomly, (2) using behavioral indicators, or (3) based on demographic characteristics. They were asked to choose between airlines using either an equal or a discriminatory screening procedure. While the former requires all passengers to be screened in the same manner, the latter mandates all passengers undergo a quick primary screening and, in addition, some passengers are selected for a secondary screening based on a predetermined selection criterion. Equity premiums were quantified in terms of monetary cost, wait time, convenience, and safety compromise. Results show that equity premiums varied greatly across respondents, with many indicating little willingness to sacrifice to avoid inequitable screening, and a smaller minority willing to sacrifice anything to avoid the discriminatory screening. The selective screening manipulation was effective in that equity premiums were greater under selection by demographic characteristics compared to the other two procedures.

Worth the risk? Terrorism-induced fear of flying / ¿Vale la pena el riesgo? Miedo inducido por el terrorismo a volar

We conducted two bi-national experiments regarding emotional and behavioral responses to a terrorist plot against commercial flights, examining both feelings and projected action. The studies employed hypothetical scenarios in which terrorists attacked airplanes with shoulder-fired missiles as they were landing or taking off from an international airport. The scenarios were built around two factorially crossed manipulated variables, each with three levels: (1) government announcements or actions (2) social norm, expressed as variation in airline ticket sales. Each respondent read a questionnaire containing only one of the nine scenarios. Experiment 1 was conducted in Spain and California (n = 360, 50% female), Experiment 2 in Israel and California (n = 504, 50% female). In both studies, fear and flight plans were not differentially affected by governmental response or social norm. Women expressed more fear than men. Experiment 1 examined the purpose of the trip. Most respondents would not change a planned flight to attend a close friend's wedding or important job interview, but a substantial number would postpone a vacation or drive to a different location. Experiment 2 featured escalating attacks. These yielded increased fear and more canceled trips. Within both studies, responses were similar across countries despite national differences in direct experience with terrorism.

Se realizaron dos experimentos bi-nacionales con respecto a las respuestas emocionales y conductuales a un plan terrorista contra vuelos comerciales, examinando ambas cosas: sentimientos y proyectos de acción. Los estudios emplearon escenarios hipotéticos en los que los terroristas atacaron los aviones con misiles disparados desde el hombro, ya que estaban aterrizando o despegando desde un aeropuerto internacional. Los escenarios fueron construidos alrededor de dos variables factoriales cruzadas, cada uno con tres niveles: (1) los anuncios o acciones del gobierno (2) norma social, expresada como la variación en la venta de billetes de avión. Cada encuestado leyó un cuestionario que contenía sólo uno de los nueve escenarios. El experimento 1 se realizó en España y California (n = 360, 50% mujeres), el experimento 2 en Israel y California (n = 504, 50% mujeres). En ambos estudios, el miedo y los planes de vuelo no fueron diferencialmente afectados por la respuesta gubernamental o la norma social. Las mujeres expresaron más miedo que los hombres. El experimento 1 examinó el propósito del viaje. La mayoría de los encuestados no cambiarían un vuelo planeado para asistir a la boda de un amigo cercano o para una importante entrevista de trabajo, pero un número considerable de participantes podrían posponer unas vacaciones o ir en coche a un lugar diferente. El experimento 2 contó con la escalada de los ataques. Estas aumentaron el temor y la cantidad de viajes cancelados. En ambos estudios, las respuestas fueron similares en todos los países a pesar de las diferencias nacionales y la experiencia directa con el terrorismo.

Cumulative Response to Sequences of Terror Attacks Varying in Frequency and Trajectory.

There is a paucity of research examining public response to the cumulative effects of multiple related extreme events over time. We investigated the separate and combined effects of frequency and trajectory of terrorist attacks. A scenario simulation of a series of gas station bombings in Southern California was developed to evaluate respondents' affect, risk perception, and intended avoidance behavior using a 3 (frequency; low vs. medium vs. high) by 3 (trajectory; increasing vs. constant vs. decreasing) factorial design. For each of the nine conditions, three videos were created to simulate news broadcasts documenting the attacks over a three-week period. A total of 275 respondents were included in the analysis. Results from analysis of covariances (ANCOVAs) indicate that trajectory of the sequential attacks (increasing or decreasing in frequency) predicts negative affect, risk perception, and avoidance behavior. In contrast, frequency predicts neither negative affect, positive affect, risk perception, nor intended avoidance behavior. Results from structural equation modeling (SEM) further indicate that the effect of negative affect on behavioral intention is mediated by risk perception and the effect of trajectory on risk perception is partially mediated by negative affect. In addition, both ANCOVAs and SEM model results suggest that (1) females experience less positive affect and perceive more risk than males, (2) respondents with higher income perceive more risk, and (3) younger respondents are more likely to modify their behavior to avoid the risk of future attacks.

Attacker Deterrence and Perceived Risk in a Stackelberg Security Game.

In Stackelberg security games, a defender must allocate scarce resources to defend against a potential attacker. The optimal defense involves the randomization of scarce security resources, yet how attackers perceive the risk given randomized defense is not well understood. We conducted an experiment where attackers chose whether to attack or not attack targets protected by randomized defense schemes, the key treatment variable being whether the defender picks one target at random to guard or imperfectly guards all targets. The two schemes are expected-payoff equivalent, and when provided separately we found no effect of having one scheme or the other. Yet, when both are present, we found that subjects had a preference for the fixed scheme, a preference that cannot be reduced to differences in beliefs. Overall, our results suggest that understanding how individuals perceive risk is vital to understand the behavior of attackers.

Perceptions of randomized security schedules.

Security of infrastructure is a major concern. Traditional security schedules are unable to provide omnipresent coverage; consequently, adversaries can exploit predictable vulnerabilities to their advantage. Randomized security schedules, which randomly deploy security measures, overcome these limitations, but public perceptions of such schedules have not been examined. In this experiment, participants were asked to make a choice between attending a venue that employed a traditional (i.e., search everyone) or a random (i.e., a probability of being searched) security schedule. The absolute probability of detecting contraband was manipulated (i.e., 1/10, 1/4, 1/2) but equivalent between the two schedule types. In general, participants were indifferent to either security schedule, regardless of the probability of detection. The randomized schedule was deemed more convenient, but the traditional schedule was considered fairer and safer. There were no differences between traditional and random schedule in terms of perceived effectiveness or deterrence. Policy implications for the implementation and utilization of randomized schedules are discussed.

Mock jurors' use of error rates in DNA database trawls.

Forensic science is not infallible, as data collected by the Innocence Project have revealed. The rate at which errors occur in forensic DNA testing-the so-called "gold standard" of forensic science-is not currently known. This article presents a Bayesian analysis to demonstrate the profound impact that error rates have on the probative value of a DNA match. Empirical evidence on whether jurors are sensitive to this effect is equivocal: Studies have typically found they are not, while a recent, methodologically rigorous study found that they can be. This article presents the results of an experiment that examined this issue within the context of a database trawl case in which one DNA profile was tested against a multitude of profiles. The description of the database was manipulated (i.e., "medical" or "offender" database, or not specified) as was the rate of error (i.e., one-in-10 or one-in-1,000). Jury-eligible participants were nearly twice as likely to convict in the offender database condition compared to the condition not specified. The error rates did not affect verdicts. Both factors, however, affected the perception of the defendant's guilt, in the expected direction, although the size of the effect was meager compared to Bayesian prescriptions. The results suggest that the disclosure of an offender database to jurors might constitute prejudicial evidence, and calls for proficiency testing in forensic science as well as training of jurors are echoed.

A Bayesian approach to the group versus individual prediction controversy in actuarial risk assessment.

Recent attempts to indict the use of actuarial risk assessment instruments have relied on confidence intervals to demonstrate that risk estimates derived at the group level do not necessarily apply to any specific individual within that group. This article contends that frequentist confidence intervals are inapposite to the current debate. Instead, Bayesian credible intervals are necessary-in principle-to accomplish what commentators are concerned about: describing the precision of an actuarial risk estimate. After illustrating both the calculation and interpretation of credible intervals, this article shows how such intervals can be used to characterize the precision of actuarial risk estimates. It then explores the legal implications of wide and overlapping intervals. Contrary to what detractors claim, the fact that risk estimate intervals overlap is not a germane to legal (logical) relevance, and therefore actuarial risk estimates cannot be per se "inadmissible" on this basis.

Innumeracy and unpacking: bridging the nomothetic/idiographic divide in violence risk assessment.

Structured methods to assess violence risk have proliferated in recent years, but such methods are not uncontroversial. A "core controversy" concerns the extent to which an actuarial risk estimate derived at the group level should-morally, logically, or mathematically-apply to any particular individual within the group. This study examines the related psychological question: When do people apply group-level risk estimates to the individual case? We manipulated whether an actuarial risk estimate is "unpacked;" that is, whether the risk factors on which the estimate is based are articulated. Our findings indicate that the degree of unpacking (e.g., listing six vs. three risk factors) increased the likelihood that jury-eligible citizens will apply an actuarial risk estimate in their decision to civilly commit a particular respondent. Unpacking also increased the perceived relevance of the group-level risk estimate to the individual case. We then split the sample based on self-reported numeracy, defined as "ability with or knowledge of numbers." Numeracy moderates the unpacking effect in that unpacking only made a difference for the innumerate participants. The psychological approach we take to the question of group-to-individual inference is not limited to violence-risk assessment, and may apply to many other areas of law in which group-to-individual inferences are frequently, if controversially, made.

The dynamics of evolving beliefs, concerns emotions, and behavioral avoidance following 9/11: a longitudinal analysis of representative archival samples.

September 11 created a natural experiment that enables us to track the psychological effects of a large-scale terror event over time. The archival data came from 8,070 participants of 10 ABC and CBS News polls collected from September 2001 until September 2006. Six questions investigated emotional, behavioral, and cognitive responses to the events of September 11 over a five-year period. We found that heightened responses after September 11 dissipated and reached a plateau at various points in time over a five-year period. We also found that emotional, cognitive, and behavioral reactions were moderated by age, sex, political affiliation, and proximity to the attack. Both emotional and behavioral responses returned to a normal state after one year, whereas cognitively-based perceptions of risk were still diminishing as late as September 2006. These results provide insight into how individuals will perceive and respond to future similar attacks.

Flu, risks, and videotape: escalating fear and avoidance.

While extensive risk perception research has focused on emotions, cognitions, and behavior at static points in time, less attention has been paid to how these variables might change over time. This study assesses how negative affect, threat beliefs, perceived risk, and intended avoidance behavior change over the course of an escalating biological disaster. A scenario simulation methodology was used that presents respondents with a video simulation of a 15-day series of local news reports to immerse respondents in the developing details of the disaster. Systemic manipulation of the virus's causal origin (terrorist attack, medical lab accident, unknown) and the respondent's proximity to the virus (local vs. opposite coast) allowed us to investigate the dynamics of public response. The unfolding scenario was presented in discrete episodes, allowing responses to be tracked over the episodes. The sample includes 600 respondents equally split by sex and by location, with half in the Washington, DC area, and half in the Los Angeles area. The results showed respondents' reactions to the flu epidemic increased as the disaster escalated. More importantly, there was considerable consistency across respondents' emotional, cognitive, and behavioral responses to the epidemic over the episodes. In addition, the reactions of respondents proximally closer to the epidemic increased more rapidly and with greater intensity than their distant counterparts. Finally, as the flu epidemic escalated, both terrorist and accidental flu releases were perceived as being less risky and were less likely to lead to avoidance behavior compared to the unknown flu release.

Differentiating conflicts in beliefs versus value tradeoffs in the domestic intelligence policy debate.

Since the 9/11 terrorist attacks, there has been an increase in public discussion regarding U.S. domestic intelligence activities. Domestic intelligence activities focus on gathering information about potential threats from individuals within the United States, and completely rational members of the public can have different opinions about the acceptability of various alternatives depending on one's values toward privacy, civil liberty, and security. Past studies have demonstrated that construction of a multiobjective value model can help clarify public values in controversial risk debates. This research explores a range of domestic intelligence alternatives that vary on multiple objectives, and applies value-focused thinking to develop a multiattribute utility model to evaluate and compare the alternatives. The process demonstrates the feasibility of eliciting model parameters from individuals and provides a method for identifying the locus of possible disagreements among individuals. The development of the model is described first, followed by insights found from participants who provided both value tradeoffs and performance scores for six different domestic intelligence alternatives. The participants were two student groups and a group of police officers. The analysis showed differences among weights for an additive model for different stakeholder groups and differences among the performance scores. In particular, there is a "halo" effect for alternatives, such that its supporters ranked the alternative higher on all attributes compared to respondents who find the alternative unacceptable. This modeling approach and results offer organizations such as the Department of Homeland Security insights into the debate surrounding new policy initiatives, particularly those requiring sensitive value tradeoffs.

The effect of framing actuarial risk probabilities on involuntary civil commitment decisions.

Despite a proliferation of actuarial risk assessment instruments, empirical research on the communication of violence risk is scant and there is virtually no research on the consumption of actuarial risk assessment. Using a 2 × 3 Latin Square factorial design, this experiment tested whether decision-makers are sensitive to varying levels of risk expressed probabilistically and whether the framing of actuarial risk probabilities is consequential for commitment decisions. Consistent with research on attribute framing, in which describing an attribute in terms of its complement leads to different conclusions, this experiment found that the way actuarial risk estimates are framed leads to disparate commitment decisions. For example, risk framed as 26% probability of violence generally led decision-makers to authorize commitment, whereas the same risk framed in the complement, a 74% probability of no violence, generally led decision-makers to release. This result was most pronounced for moderate risk levels. Implications for the risk communication format debate, forensic practice and research are discussed.