Decision Making and Crisis Management

Research Question What is the relation between planning and decision-making processes in crisis management? Introduction “Once in a lifetime” events are far more common and dangerous than we ever realized (Taleb, 2010). Translating into numbers, the monetary costs of these events have consistently increased over the last decades. “Disasters in 2011 set a new record of $366 billion for economic losses” (UNISDR, 2012). In terms of insurance payouts, two-thirds of the twenty-five most expensive disasters in the last forty years have taken place since 2001 (Kunreuther, 2010). As the global climate changes and more people move to overbuilt and hence more vulnerable cities, the pace of cataclysmic “extreme events” is liable to increase (Kunreuther, 2010). As a result of these economic -- and especially human -- losses, public interest in issues of safety and risk is growing in intensity, legislation in the health and safety area is becoming more extensive and expectations about how decision-makers should perform when confronting a variety of different risks are rising (Turner, 1994). All of these developments serve to give more contemporary prominence to crisis management; a prominence which brings several difficulties in its train (Turner, 1994). In this context and setting a foundation for an answer to the research question above, this paper discusses the definitions of crises, the role of planning in dealing with increased risks, followed by a brief overview of the advantages and disadvantages of decision-making models, and then highlights challenges and opportunities related to decision-making processes in crisis management. The relevance of this paper is not in its conclusion but in advancing a discussion that “unsurprisingly, has developed political and cultural overtones, and the more widely studied it becomes, the more highly is it contested” (Turner, 1994). Crisis Management -- Definitions, Variables, Causes, and Consequences The systematic study of crises started early 1960. Charles McClelland noted five approaches in this study i) definitions, ii) classifications, iii) objectives, iv) decision-making, and v) planning and management (Shlaim, 1983). This paper focuses on the fourth approach -- decision-making under crisis -- by highlighting how crisis is defined in terms of decision-making processes within a state or in terms of the interaction process between states. Oran Young views crisis as a situation which involves change in the normal interaction patterns between states or in the international system as a whole (Shlaim, 1983). Based on these internal and external interactions, Charles Hermann elaborated the most widely accepted definition of crisis: “it is a situation that threatens the high priority goals of the decision-making unit, restricts the amount of time available for response before the situation is transformed, and surprises the members of the decision-makers who are experiencing crisis” (Shlaim, 1983). Along Hermann’s lines, Uriel Rosenthal defines crises as “periods of upheaval and collective storms, disturbing everyday patterns and threatening core values and structures of a social system in unexpected, often unconceivable ways” (Rosenthal, 2001). Arjen Boin agrees with Rosenthal and affirms that “crises are a threat to important values, norms, and structures of society since they create uncertainty and a sense of urgency” (Boin, 2006). In addition to uncertainty, the study of crisis also takes into account the broad definition of “problem” and its related variables. Problem is “any difference between an actual situation and a desired situation” (Shlaim, 1983). Problem is perceived as a deficiency when an outcome is realized as “less than desired” and becomes an opportunity if its desired level is “better than anticipated” (Shlaim, 1983). In the study of “how to solve problems”, the academic challenge has been “identifying the discrepancy between an actual and desired state of affairs” since this identification process requires “the right information being available to the right people at the right times” (Shlaim, 1983). Management is another related variable of “problem solving”. Managerial problems are classified as structured -- straightforward-routine decision --, unstructured -- unexpected non-programmed decision --, and crisis -- unexpected problem that leads to disaster if not resolved quickly (Shlaim, 1983). Finally, risk and uncertainty are problem-solving conditions and environments that managers have to deal with in complex and unstructured problems. Risk is when there is lack of complete information on action alternatives and their consequences but there is some sense of the probabilities associated with their occurrence (Shlaim, 1983). Uncertainty is when information is so poor that managers are unable even to assign probabilities to the likely outcomes of alternatives that are known (Shlaim, 1983). Crisis management is a quite complex issue since its nature varies largely. “Modern crises are increasingly characterized by complexity, interdependence, and politicization” (Rosenthal, 2001). Crises are ever changing phenomena: “tomorrow’s crises will look different from today and yesterday’s crises” (Rosenthal, 2001). Also, people do not perceive crises in the same way: “what is crisis to some, may be opportunity to others” (Rosenthal, 2001). Other conceptual difficulties in the study of crisis management are that “crises cannot be studied in absolute terms”, “the ability of politicians, administrators, and managers to understand what goes on inside or outside the organization at hand is rather limited”, and “the most commanding dimension of uncertainty may be surprise” (Rosenthal, 2001). Idiosyncrasy and vicissitude exacerbate the complexity of crises since “crises are culturally and politically defined events that contain levels of conflict and arouse string emotional responses” and “are, at least partially, the outcome of organizational or societal “sensemaking”” (Rosenthal, 2001). Moreover, crises cannot be studied in isolation as they “are part of a dynamic process that leads a system from one temporary state to another” (Rosenthal, 2001). The causes and consequences of crises add a great deal of complexity to crisis research. The causes of crises are rooted at the microlevel such as i) human errors -- failed judgment, negligence, blunders, or sabotage -- and at the macrolevel -- technical complexity and environmental pressures (Boin, 2006). The general consequences of crises are i) the restoration of the status quo in the short-term and ii) an opportunity for learning in the long-term (Boin, 2006). From an institutional standpoint, crises put structures and existing policies to the test (Boin, 2006). Institutions face a great difficulty to return to normalcy as crises emerge because crises are usually i) endemic, ii) compounded in nature, and iii) self-perpetuating. Because of this vicious circle fed by uncertainty, crises tend to recreate themselves in mutating forms (Boin, 2006). This recreation form happens because crisis management has experienced the following trends i) transnationalization: global scale developments, ii) mediazation: the subjective notion of disaster -- “If CNN defines a situation as a crisis, it will indeed be a crisis in all its consequences” --, iii) technological developments: complexity to fix, and iv) dissipation of state authority: declining role of public authorities (Boin, 2006). The prevention of crises is a challenge. Preventive and preparatory policies are often difficult to pursue because there is i) a widespread tendency to undermine hazard risks and the probability of crisis and ii) uncertainty whether we should put all our money on prevention (Rosenthal, 2001). In the context of prevention, the discussion over resilience -- “the capacity to cope with unanticipated dangers after they have become manifest, learning to bounce back” -- has become a defining feature of contemporary governance (Boin, 2006). However, policy makers have difficulties to enact preventive policies because in times of crisis, communities and members of organizations tend to minimize the impact of the crisis at hand while critics and bureaucratic competitors try to seize the moment to blame incumbent rulers and their policies for a phenomenon which is the result of multiple interacting causes (Boin, 2006). In spite of the challenges to enact preventive policies, decision makers must somehow establish a sense of normalcy and foster collective learning from crisis experiences (Boin, 2006). This sense of normalcy is re-established by having governments dealing with acute problems despite i) the obvious disparities between the demand for and supply of public resources, ii) an unclear and volatile situation, and iii) the lack of time to think, consult, and make decisions (Boin, 2006). As for collective learning, it happens when crisis decision making involves both key value tradeoffs and political risks. Successful crisis decision making depends on an extraordinary level of interagency, intergovernmental, and intergroup coordination to prevent miscommunication and overlapping activities and to avoid value or organizational conflicts between state, private, and nongovernmental actors (Boin, 2006). A three-step process to re-establish normalcy and collective learning after crises relies on i) "meaning making": when political leaders make sense of the situation and citizens know what is going on by receiving timely and detailed information, ii) termination: when leaders orchestrate a shift from emergency status back to routine life and render an account of what has happened, and iii) learning process: when those involved in crisis management draw lessons from past events and translate them into organizational practices, policies, and laws (Boin, 2006). Contingency plans deal with crises and accelerate the transition of a critical situation to normalcy. Contingency plans are not meant to be complicated, though (Boin, 2005). It should formulate clear principles about aims to be achieved and the core preconditions that apply; in addition to “being flexible and low on details about how the various professionals are supposed to do their jobs” (Boin, 2005). Planning plays an important role in the development of contingency plans. The secret of “the planning process” while working on response issues are i) sensitizing participants to problems that emerge during crisis ii) the development of an understanding for the needs and capacities of other potential crisis management actors, and iii) the support to build social capital that facilitates smooth interactions in the heat of crisis such as simulation exercises which help foster realistic expectations and build mutual trust (Boin, 2005). Planning & Crisis Management -- Contradiction & Contingent Approach “Crises are often best engaged through a combination of strategic choices and concerted action” but leaders have to deal with a paradox since everybody expects leaders to be in charge but circumstances do not allow them to make free choices (Boin, 2005). Government experiences many problems in the implementation of crisis contingency plans because there is a fundamental tension between the idea of planning and the nature of crisis decision making (Boin, 2005). Planning presupposes knowledge of what will happen (Boin, 2005). Crisis disturbs stable environments and creates uncertainty (Boin, 2005). This paradox between planning and crisis presents authorities with unfamiliar challenges that can never fully be dealt with in preconceived plans (Boin, 2005). The immediate consequence of this paradox is that “any crisis response operation necessarily contains elements of improvisation”, which requires i) flexilibility and ii) resilience rather than plans (Boin, 2005). Crisis decision making and coordination are much more effective when they are not dictated by detailed plans and allow for a healthy degree of improvisation (Boin, 2005). In most crises, lives are saved and interests are protected because of alert and decisive individuals and because organizations worked together in innovative ways (Boin, 2005). Crises threaten collective goals, surprise policy-makers, and leave them with short decision time for planning (Shlaim, 1983). This occurs because “shared risk is nonlinear and dynamic in that small differences in initial conditions, are repeated in actions over time, and lead to unpredictable outcomes” (Comfort, 1999). Shared risk appears to be more problems of collective learning, involving multiple groups at different levels of understanding, commitment, and skill, as well as requiring different types of knowledge, authority, and action for effective resolution (Comfort, 1999). The linkage between information technology and organizational management creates a “sociotechnical system” in which the ability to exchange timely and accurate information among multiple participants facilitates a more open, responsive, and creative approach to solving shared problems (Comfort, 1999). “Sociotechnical systems” also facilitate transition and self-organizing processes (Comfort, 1999). “The whole community needs a decision-making system that can hold and exchange information among multiple actors with different levels of responsibility and vulnerability to risk” (Comfort, 1999). Causality, time, distance, and complexity are fundamental to framing courses for collective action in reference to problems of shared risk (Comfort, 1999). The eight conditions that lead to self-organization -- characteristics of nonlinearity, dynamic systems, and interdependent relationships -- are i) community capacity to mobilize collective action, ii) community collective memory influencing choices and behavior, iii) irreversible effects within the system, iv) “feedback loops”, v) irregular clusters of attraction and avoidance, vi) transition to a different state, vii) unpredictable results in social action, and viii) nonlinear systems reproducing self-similar patterns of behavior (Comfort, 1999). The planning challenge is institutionalizing a contingent way of thinking (Rosenthal, 2001). The fear of surprise makes risk aversion loom large in the planning process, which makes crisis planning increasingly relying on various modes of contingency analysis (Rosenthal, 2001). Thinking contingently includes both worst-case and best-case scenarios with emphasis on reconstructive logics, multiple realities, and alternative futures (Rosenthal, 2001). The challenges for contingency plans are that “planning and preparatory measures may fall prey to routinization -- over institutionalization -- or become part of the high politics game -- under institutionalization” (Rosenthal, 2001). A contingency approach to crisis management examines crisis as i) a process: characters and conditions of a studied phenomena, ii) as a facilitator: facilitating desirable change or reform, iii) as a public-private affair: tendency to shift the burden of organizational activities and implementation from the public to the private sector, iv) as an open-minded approach: eliminate preconceived notions, v) an emphatic approach: endogenous drama, turbulences, memory, vi) a multi-disciplinary approach: eliminate the distinction between natural and man-made disasters, policy dynamics and institutional strain, and vii) addressing theoretical challenges: eliminate near-exclusive military planning such as long-range planning capacity, strategic capabilities, high-quality learning, and new ways of policy reasoning (Rosenthal, 2001). Another planning challenge is the study on how individuals and social groups respond to extreme events in nature. Even though the global toll from extreme events of nature is increasing loss in property and life, technology attracts people to hazardous areas (Burton, 1993). “The ways in which humankind deploys its resources and technology in attempts to cope with extreme events of nature are inducing more rather than less damage” (Burton, 1993). People not only locate in areas of high, recurrent natural hazard; they survive and prosper there (Burton, 1993). However, lay people cannot appraise the magnitude and frequency of extreme events with accuracy: “People are rarely aware of all the alternatives open to them” since “they differ greatly in the way they judge the consequences of particular actions even on the rare occasions when the physical outcomes are known accurately” (Burton, 1993). The comparison of many different consequences is a highly complex operation for a decision analyst armed with precise data and a computer, let alone for a farmer choosing a crop as the rainy season approaches (Burton, 1993). Self-organizing processes help planners deal with the paradox of technology and risk and incorporate contingent thinking in planning. The conditions that encourage the emergence of self-organizing processes and enable communities to identify and pursue common interests occur i) when external conditions or factors force consideration of risk and ii) when communities self-organize to reduce risk (Comfort, 1999). In so doing, communities develop the capacity to respond effectively to future disasters (Comfort, 1999). Key for this development is the sharing of information for the emergence of “sociotechnical systems” to support collective actions. However, throughout this process, it should be considered the differences in time perspectives of the actors and the need of information structure that is accessible to all -- national and local groups (Comfort, 1999). In the context of flood plain, for instance, “managers of unprotected bottom land property make use of alternative measures as far as they feel economically justified” and their “attention to flood problems has centered more upon planning engineering work and judging its economic feasibility than upon the effects of that work on public policy” (White, 1964). To counteract the emphasis on engineering planning, local community planners are advised to consider i) arrangements for physical studies, ii) results available in easy language for a large audience, iii) the organization of modern forecasting and warning systems, iv) regulations to preserve physical capacity, v) attention of private and public responsible officials, vi) the benefits of emergency, structural, and insurance adjustments, vii) building code, subdivision regulations, zoning ordinances, and viii) public acquisition of hazardous locations: coordinated within state and federal agencies (White, 1964). Other adjustments that community planners have to make in order to incorporate a contingent way of thinking and combine technology and risk accordingly are i) the perception of the hazard, ii) the perception of the possible adjustments, iii) the technical feasibility of particular adjustments, iv) the economic efficiency of these adjustments, and v) the timing and incidence of decisions by the private and public managers. These incorporation and combination processes require a technical feasibility study since adjustments differ greatly because of frequency, stage, duration, and velocity of a hazard. Once these comprehensive self-organizing processes are concluded, local planners are then able to propose adjustments on i) emergency evacuation, ii) structural changes, and iii) land use changes (White, 1964). The solution to the paradox between planning and crisis management depends on education and communication. “We need greater and more careful efforts at education and communication between government agencies and residents” (White, 1974). These efforts are necessary because of i) the extremely close attachment of the residents’ to locations, ii) residents’ lack of knowledge about what they personally might be able to do or should do to prevent damages other than trivial palliative adjustments, iii) the lack of awareness by residents of the extent to which government agencies, in fact, do assist them in one way or another with hazards, iv) the perceived poor communications of government agencies with the residents, and v) the low regard which many people have for various units of government. The success of these efforts in the long-term depends on i) the amount of resident involvement in the planning process and ii) the amount of communication with the people living in hazardous areas (White, 1974). Decision-Making Models -- Top-down v. Bottom-up Institutional decision-making involves top-down and bottom-up processes. The four models of collective action in response to risk are: i) Command and Control: these are mechanistic models that have been developed in the physical sciences and engineering. The basic assumption underlying these models is that if the problem is well defined and systems can be closed to outside interference and disturbance, they can function without error. Once the system is carefully designed and functioning, considerable effort is placed on the control of error, or any aberrant disruption or behavior. When this system is in command, all other disturbances need to be controlled or eliminated. This is a typical of military model since it includes hierarchical authority structures and tightly-coupled systems. It is focused on planning and training, robust in routine conditions but weakest in uncertain and dynamic situations. This model is also vulnerable to "lock out": omission of relevant information from decision process and its main difficult is to apply a command and control model to social systems, especially democratic ones. ii) Anarchy: it assumes intelligent individuals can recognize opportunities for action and take advantage of them, and so relax behavioral rules and encourage personal choice. Operating under uncertainty, this model is not rational and planned but opportunistic recognition of a match between problem and solution. This process emphasizes flexibility and loosely-coupled systems -- "relax structural controls on attention that allow the spontaneous matching of problems, solutions, and participants in creative response to perceived opportunities or needs". iii) Redundancy: A method of "hardening" the system against disruptions by providing backups to maintain operation even during sudden failure or unexpected demands. An example of its use in fire-fighting is piling on equipment and manpower until the fire is overwhelmed. Redundancy is generally effective, but often prohibitively expensive. iv) Inquiry: instead of permanent laws governing social relationships, the presence of a system organizing action towards a defined goal. An inquiring system is fundamentally a means of organizing information and communication processes in order to solve problems for a specific group. Difficulty of this approach is balancing openness -- continuing broad information search and process -- with focus required to take action. This system needs to operate on micro and macro scales simultaneously and it quickly encounters the limits of human information-processing capacity when addressing large, complex, and social problems (Comfort, 1999). All these models assume cooperation on a shared goal and that divergence of interests can cause the systems to break down. Command and Control deals with divergence through suppression and discipline. Anarchy uses bargaining. Redundancy partitions roles, assigns buffers, and reduces conflict by allocating "slack" resources. Inquiry uses education and discovery based on error over time to reduce conflict (Comfort, 1999). The classical decision model is a typical top-down and problem solving process that i) finds and defines the problem, ii) generates and evaluates alternative solutions, iii) selects preferred solution, iv) implements the solution, and v) evaluates results: recycling process as necessary. Along this process, decision makers i) do information gathering, processing and deliberation: narrowing down problems studied and focusing on causes rather than symptoms, ii) data analysis considering possible alternatives: being careful to not select a particular solution too quickly and end up choosing an alternative that has damaging side effects, iii) criteria establishment: for evaluating alternatives through benefits, costs, timeliness, acceptability, ethical soundness, iv) anticipate difficulties: such as lack of participation and failure to involve those whose support is necessary, and v) evaluate general pros and cons of the decision to be made. This classical model assumes that decision makers have complete information and clear define problems throughout this process (Shlaim, 1983). In a top-down process, decision makers are the government or private owners. The decision maker selects an interdisciplinary team to carry out an unbiased analysis, assigns a team leader, defines the communities to be assessed, and establishes the organization process. This process usually features the following stages i) project identification: decision maker provides framework for analysis and determines purposed questions such as who, what, why, where, when, ii) scoping: interdisciplinary team informs community and asks for comments and suggestions to identify issues and opportunities, iii) data collection and interpretation, iv) alternatives design: “do nothing” is always an alternative, v) mitigation plan to deal with adverse effects, vi) effects evaluation: of each alternative, vii) alternatives comparison in terms of how well they fulfill project’s purposes and goals while addressing issues and opportunities, and viii) project implementation and monitoring: of alternative chosen (Black, 2007). Idiosyncrasy imposes difficulties for top-down decision-making processes since managers make decisions differently. This occurs because managers perceive problems through their own biases and filters, which results in four major decision styles: directive, analytical, conceptual, and behavioral (Shlaim, 1983). The directive style is about clear-cut solutions, quick decisions, not too much information, one or two alternatives, and relies on existing rules and procedures. The analytical one requires complex solutions, much data, consider alternatives carefully as well as objective and rational data, and the best possible decision is based on information available. The conceptual style is about broad amount of information, it is more socially oriented, and usually shares socially, creative solving techniques. The behavioral decision-making style involves deep concern for others, one-on-one discussions, feelings understanding, and decisions that help others achieve their goals (Shlaim, 1983). Another challenge that top-down decision-making processes encounter is cognitive limitations. Administratively, managers have cognitive limitations and often act with incomplete information in uncertain environments (Shlaim, 1983). This limitation is a result of “bounded rationality”, which states that decisions are rational only within the boundaries defined by the available information (Shlaim, 1983). The main flaw caused by “bounded rationality” is that the first alternative that appears to give a satisfactory resolution is likely to be chosen. This tendency is called sacrificing (Shlaim, 1983). The cognitive limitation has also given room to the study of other decision-making models. Behavioral decision model, for instance, includes judgment heuristics -- speculative -- approaches to decision making. Judgmental heuristics approach is often adopted to simplify managerial decision making by relying on decisions highly influenced by information readily available in memory and comparisons with similar circumstances and current situation -- these cognitive limitations are known as anchoring and adjustment heuristic. Availability heuristic is about memory and representative heuristic uses comparisons with similar circumstances as well as anchoring and adjusting heuristics such as assessment of current situation in the process of decision-making under “bounded rationality” (Shlaim, 1983). The limitations of top-down decision-making processes are compounded by the fact that decisions are individual-based. Individuals not only perceive problems differently while interpreting information and data but they also have difficulties i) to appraise the probability and magnitude of extreme events, ii) canvass the range of possible alternative actions, iii) evaluate the consequences of selected actions, and iv) choose one or a combination of actions (Slovic, 1974). These limitations are offset by the implementation of participative decision-making models. Effective participative decision-making processes depend on managers involving the right people in the right ways in helping them solve problems. The main advantage of this approach is that it provides i) a broader perspective for problem definition and analysis, ii) more knowledge, facts, alternatives evaluated, iii) discussion to clarify ambiguous problems, iv) less uncertainty about alternatives, and v) fosters member satisfaction and support for decision. The disadvantages of this approach is that it is i) time consuming, ii) waste resources if used for programmed decisions, iii) compromise decisions may not satisfy no one, iv) groupthink: group norms may reduce dissent and opinion diversity, and v) no clear focus for decision responsibility. Considering these disadvantages, participative decision-making models are improved by stimulating creative thinking and growing use of information technology to share information (Shlaim, 1983). Bottom-up decision-making processes complements and improves top-down models. Self-organization decision-making processes are spontaneous actions based upon informed choice taken to achieve a collective goal. They exemplify citizens acting together, voluntarily, to meet a common need. They represent the fullest type of adaptation in a complex system that engages participants in collective action to reduce risk. They are likely the most efficient means of risk reduction (Comfort, 1999). The examination of self-organizing processes by which communities act in their own interest to mitigate and reduce risk is a type of disaster response system defined as a set of interactions or communications. The relevance of self-organizing systems lies in the fact that “there is not some well-defined organization that does all the responding”. The response to disasters “is a transient event, and whatever configuration of entities participates in the event is what we should regard as the "system"” (Comfort, 1999). Self-organizing systems operate based on the idea that “there is a continuum of behavior from order to chaos and that adaptive systems live "on the edge of chaos"” (Comfort, 1999). Even though these systems are stable and in a steady state, they are not static or at equilibrium (Comfort, 1999). They balance structure and flexibility. This balance makes adaptive systems the best systems to handle the disruption caused by large events (Comfort, 1999). The three community axes that proved to be relevant to disaster resistance and resilience are i) technical structure, ii) organizational flexibility, and iii) cultural openness to new information. These three dimensions -- evaluated into three categories, high, medium, and low --, are grouped into four subsets based on the path they take towards self-organization in a disaster: i) nonadaptive: low on structure and flexibility, and generally on cultural openness as well, ii) emergent adaptive: low structure, high flexibility, and increasing cultural openness, iii) operative adaptive: have medium structure, medium flexibility, and medium openness, and iv) autoadaptive systems: high on structure, flexibility, and openness (Comfort, 1999). Self-organizing systems require few conditions. The transition to a coherent and cooperative state needs i) articulation of commonly understood meanings among participating organizations and key audiences in the wider environment, ii) sufficient trust among leaders, organizations, and citizens to enable them to accept a shared goal, iii) sufficient resonance between the organizations seeking change and their potential sources of support for action, and iv) sufficient capacity and resources to sustain collective action among participating organizations to achieve a shared goal (Comfort, 1999). In this transition, the key stages of the adaptation response are i) information acquisition: assessment, ii) information sharing: communication, and iii) information retention: organizational learning. The establishment of a self-organizing system also requires adaptive behavior, which means changing not only actions, but priorities, so as to match response to new and immediate environmental demands. Adaptive behavior in a community requires coordination because if one organization changes its behavior, the interdependent organizations that support it must change to continue providing support. Change is, therefore, most effective when the community can "self-organize" and reorder its own priorities spontaneously, without externally-imposed controls. Self-organization is most probable when the various components become aware of how their individual needs and responsibilities interlock and participate voluntarily in the response. Sustaining self-organization requires persistent communication between participants to exchange information, store memory, and perform self-analysis and feedback. It is available as a voluntary response but requires technical and organizational infrastructure that engender organizational learning (Comfort, 1999). Challenges for Decision-Making Processes -- Conceptual, Structural, and Cognitive The very nature of unexpectedness in crisis situations imposes a challenge for decision-making processes. Not knowing when a major emergency might hit and the extent of its impact, gives logical basis for emergency planning but increases its complexity (Taleb, 2010). The key question is how much planning and investment should be made in preparing for something that may never happen or not happen according to a predicted time frame and the consequences of its occurrence may indeed be "unexpected" (Taleb, 2010). In this context, it may not be suitable to do conventional "expected return" or similar analyzes that would yield relatively sound quantitative insights (Taleb, 2010). The responsibility to address this challenge created by emergency planning questions and situations lies in the Federal agencies as well as at the State, County, Municipal and individual household levels. Vetting, "strategizing", and rationalizing what could and should be done is warranted and healthy based on the unexpected nature of things (Taleb, 2010). The unexpectedness nature of crisis creates other problems. Probabilistic decision-making models might fail in the real world. Data collection to produce better models limits inference spaces since then catastrophes appear from places "we didn't expect" (Taleb, 2010). Also, the inadequate appreciation of the uncertainty of the environment inevitably leads economic agents to take risks they should avoid (Taleb, 2010). The main features of highly improbable events also known as “black swans” have three principal characteristics i) unpredictability: rarity, ii) massive impact: extreme event, and iii) retrospective predictability: after explanation that make “black swans” seem less random or more predictable (Taleb, 2010). An example of a “black swan” is the terrorist event on September 11th, 2001 in the United States. The phenomenon “black swan” is only acknowledged until after it occurs (Taleb, 2010). This happens because humans are hardwired to learn specifics when we should be focusing on generalities. The methodology, therefore, to reduce the impact of uncertainty is i) to concentrate on things we do not know and ii) to reward those who can imagine the “impossible”. This approach in dealing with “black swans” is backed by the findings that “we do not learn rules, just facts, only facts” and “we do not seem to be good at getting metarules” and is compounded by the fact that “we scorn the abstract; we scorn it with passion” (Taleb, 2007). In seeking the causes of many modern large-scale accidents it is important to consider organizational failings. So-called “organizational accidents” stem from an incubation of latent errors and events which are at odds with the culturally taken for granted, accompanied by a collective failure of organizational intelligence (Turner, 1978). Theoretical models have also moved from purely post hoc descriptions of accidents and their causes in the attempt to specify “safe” cultures and “high-reliability” organizations. Recent research, however, has shown that while effective learning about crises is a common assumption of such attempts, organizations can be very resistant to learning the full lessons from past incidents and mistakes (Turner, 1978). Two common barriers to learning from past crises are i) information difficulties and ii) organizational politics (Turner, 1978). The 1962 Cuban Missile crisis, for instance, depicted the serious problems on the relationship between the quality of decision-making processes in an international crisis and the desirability of its outcome. The main problem to analyze this relationship is to identify a set of criteria to assess the quality of decision-making and wrong decisions based on incomplete information (Welch, 1989). Another shortcoming of crisis decision making is to assume that situations remain static. This means, for instance, that frequency, magnitude, and intensity of natural hazards are the same over time (White, 1975). These assumptions result in little incentive to find out how the adjustments or conditions of choice change and what accounts for those deviations (White, 1975). Three factors that contribute to situational assessments and influence individual choices are i) experience with hazard: more experience, more susceptible to accept risk assessment, ii) material wealth at risk: more wealth, more to lose, more response, and iii) personality: the way to respond to new information. In addition, the sense of efficacy of a public agency in dealing with a hazard strongly affects the choice of corrective and preventive actions (White, 1975). Communication under uncertainty is fundamental to localities exposed to recurring risk. While performance assessment following devastating events call for stronger coordination among agencies, such coordination cannot occur without an effective process of communication (Comfort, 2007). Communication has both technical and social components, each affecting the other and producing potential failure as well as probable strength (Comfort, 2007). Failure in technical systems of communication almost certainly triggers failure in organizational performance under the rapidly changing conditions of disaster. Such failures lead to asymmetrics in access to communication and timely information exchange that are critical to a community’s capacity to adapt quickly to serious threat (Comfort, 2007). The development of a strong communication system with other organizations before a disaster occurs is pivotal to establish appropriate communication in which effective interagency coordination takes place at the time of a disaster (Kapucu, 2006). Coordinating response operations to extreme events is an extraordinarily complex task for public managers. The performance of an intergovernmental system depends on a model of auto-adaptation to extreme events (Comfort, 2002). This model is based on individual, organizational, and collective learning in environments exposed to recurring risk (Comfort, 2002). Such a model requires public investment in the development of an information infrastructure that can support the intense demand for communication, information search, exchange, and feedback that characterizes an auto-adaptive system (Comfort, 2002). However, when a collectivity encounters a major crisis, political rivalries about the interpretation of fast-moving events intensify and become part of the drama that crisis management entails (Comfort, 2002). Crises make and break political careers, shake bureaucratic orders, and define organizational destinies. Five core tasks of crisis leadership, in a political context, are i) sensemaking: recognize signals that something out of ordinary is developing, ii) decision making: interagency and intergovernmental coordination, iii) meaning making: accept a definition of the situation, iv) terminating: return to a sense of normalcy, and v) learning: lessons feed, organizational practices, policies, and laws (Boin, 2005). These characteristics of crisis leadership are important because “when emerging vulnerabilities and threats are inadequately assessed and addressed, some potentially devastating contingencies simply do not happen”. Also, “misperceptions and negligence allow crises to occur” and “the mass media continuously scrutinize and assess leaders” in crisis moments. To improve crisis decision making, leaders should be aware that “the rhetoric of “the leader in charge” has little to do with the reality of effective crisis decision making and coordination and more about how facts are communicated and normalcy restored (Boin, 2005). Leader who lacks the ability to communicate cannot lead in crises since “a crisis never speak for itself”. These leadership skills are especially important because through effective communication leaders share perceptions that channel behavior and manage both the operational and the political dimensions of crisis (Boin, 2005). The challenge of crisis communication arises from the particularities of disaster management and individual differences. Discrepancies between official conceptualization and political reality, tensions between official and sociological designs, organizational infrastructure, and emergency management paradox raise difficulties for decision makers managing disasters that come as a surprise (Rosenthal, 2001). To address these particularities, crisis leaders must avoid the syndrome of “winning the last war” -- when disaster policies become oriented to the past instead of the future based on supposedly known facts instead of creative imagination (Rosenthal, 2001). Disaster scenarios usually do not reach beyond extrapolations of the most recent calamity, and so require incremental solutions upon a typically nonincremental context (Rosenthal, 2001). Governments that fall short of being smooth cannot find rational solutions to sudden problems; therefore, policy-makers and planners “should plan for reality, not for a pseudoreality made of rules and rigid structures since disasters are real events, not occurrences on paper” (Rosenthal, 2001). As for the role of individual differences in crisis communication, decision makers should take into account that crisis managers are not immune to some of the organizational, political, and psychological pressures that are heightened under crisis conditions (Hart, 1991). Political processes corrupt possibilities for organizational learning. Nevertheless, political problems, and in particular that of blame in organizations, ultimately require political solutions (Pidgeon, 1997). For instance, long-term contracts for reducing losses from future catastrophes provide crisis leaders a comprehensive strategic framework for assessing, responding to, and managing extreme risk (Kunreuther, 2010). Even though incumbent leaders resist to contemplate lessons learned from past crises since they risk revealing the deficiencies in prevention, preparedness, and response crisis plans, they also agree that these institutional barriers to learning and change should be overcome (Boin, 2006). 1989 Exxon Valdez oil spill would have been dealt more effectively if previously management shortcomings were learned and then properly addressed at the beginning of this crisis (Williams, 1994). Also, the military decision model applied during the Cuban Missiles Crisis replicated a common decision-making shortcoming by overestimating information-processing capacity of individuals and by misrepresenting the social nature of decisions in organizational contexts (Anderson, 1983). The understanding of crisis decision-making processes is at the heart of systematic improvements in public policy. Two theories of decision making under risk that streamline crisis political process are i) maximization of expected utility which features two principles “transitivity” -- if A is better than C, and B is also better than C, a rational decision maker would never choose C -- and “the extended sure-thing principle” -- outcomes that are not affected by your choice should not influence your decision -- and ii) bounded rationality which deals with “cognitive limitations of decision makers forcing them to construct a simplified model of the world to deal with it” (Slovic, 1974). This simplified model is based on the notion of “satisficing” whereby an organism strives to attain some satisfactory, though not necessarily maximal, level of achievement which drastically simplifies decisions by comparing only those situations that differ in relatively small degree from realities already in effect. While coping with crises, leaders must enlarge the social costs of disasters to make themselves more vulnerable to the consequences of the great extremes before a decision is made and to avoid complacency (Slovic, 1974). Crisis decision-making is inter- and multi-disciplinary (Pearson, 1998; Alexander, 1997). Grasping this concept is important to acknowledge that “attempts to control nature and determine government policy will not succeed without a better understanding of the interplay among psychological, economic, and environmental factors as they determine the adjustment process” (Slovic, 1974). Adjustments refer to the many courses of action available to crisis managers such as insurance systems, land policies, structural works, and public relief. Also, the psychological study of decision making should consider “bounded rationality”, which takes into account the limitations of decision makers’ perception and cognitive capabilities (Slovic, 1974). “In the real world people don’t always behave rationally” affirmed Herbert Simon (McCarty, 2001). The reason for this irrationality is “scarcity of attention”. Simon, thus, proposes that instead of aiming the optimal information, we focus on accessing the best information we can for making the most practical choice (McCarty, 2001). Other difficulties to gather the best information -- or “cognitive obstacles to rational decisions” -- are i) “gambler’s fallacy”: when someone assumes that if an event has occurred one year, it is less likely to occur the following year and ii) “anchoring”: faced with masses of information, people began with one rough estimate and then adjust it as more information comes in (Slovic, 1974). The immediate consequence of “gambler’s fallacy” and “anchoring” is a psychological heuristic that explains how people believe they have a much better picture of the truth than they really do and as a result end up underestimating phenomena (Slovic, 1974; Kanehmann, 1974). Framing also influences decision making. “Different windows, lead to different decisions -- even for what was actually the same decision problem” (Mook, 2004). We are “risk seeing” -- accept responsibility -- in cases where we are contemplating losses and we are “risk averse” -- take fixed number -- in cases where we are contemplating gains. The difficulties in studying these irrationalities are that “there is no theory that tells us how we ought to look at the problem” that constantly violate the most fundamental criterion of rationality: consistency (Mook, 2004). For learning purposes, cognitive processes are divided into two systems. System one is unconscious, intuitive, and automatic and system two is a conscious, analytic, and controlled decision-making model (Pierre, 2011). In the case of optimistic bias, people tend to use fast and intuitive thinking -- system one -- whenever possible, and switch over to a more deliberate and effortful system two when they recognize that the problem at hand is not an easy one (Kanehmann, 2011). Natural decision-making explains these systems through the role of experience. “In routine situations and common emergencies, decision makers size up the situation and quickly recognize which actions make sense -- system one (Klein, 1998). Thus, by relying on their holistic assessment of both the situation and the available options, experts actually do not think when they decide -- they “just do what is right” --; however, every time unfamiliar or more complex elements arise, system two exerts oversight on automatic decisions among choices that can be pre-played using mental simulations of future situations (Klein, 1998). Even experts in a given domain cannot rely solely on unconscious decision-making since situations occasionally demand conscious thinking to achieve a “good decision”“ (Klein, 1998). Five steps helping crisis managers navigate between systems one and two are i) preparedness, ii) situation assessment: defining the problem, gathering information, and building of mental models, iii) planning of actions: formulating goals, assessing risk, planning, making decisions, iv) execution of action, and v) review of effects: reviewing actions, revising strategy, and self-reflection (Klein, 1998). Rational individuals do not pursue common interests unless the community is small or otherwise coerced into doing so. In addition, parochial interests discourage attention to broader community needs (Olson, 1965). In crisis situations, key decisions are often made by small, tightly knit group of individuals (Hart, 1990). Dominant leaders believe that by strengthening the “cognitive process” of small groups they are improving institutional decision-making processes (Kellerman, 1983). Groupthink, however, is a cause of policy failure because of its tendency for premature and concurrence seeking among group members (Hart, 1990). These particularities of groupthinking have been identified as the reasons for major fiascoes in foreign policy, large planning projects, and strategic management (Hart, 1990). Crisis-produced stress also leads to the simplification of information processing -- diminishing bureaucratic politics -- and reinforces small group decision-making (Levi, 1980; Monten, 2010). In this context, organizations -- despite being organized in small groups and simplifying decision-making processes -- rather than individuals are still in a better position to tame optimism, anticipate a plan’s weaknesses, and make a decision since organizations are a more conducive environment for sharing experts’ opinions than vis-à-vis arguments (Klein, 1998). Crisis leaders have to deal with stress. They do so by escalating or de-escalating the decision-making processes of i) information: cognitive performance, the perceived need and consequent quest for information, the receptivity and size of the information-processing group, ii) consultation: the type and size of the consultative units, group participation in the consultative process, iii) decisional forums: the size and structure of the decisional forums, authority patterns within decisional units, and iv) alternatives: search for and evaluation of alternatives, the perceived range of available alternatives (Shlaim, 1983). The impact of stress on crisis leaders are on i) contraction, ii) concentration, and iii) activation (Sigelman, 1990). The high level of emotional and physical stress restrict the amount of time for response, threat high-priority goals, and make implementation processes vulnerable to malfunctions (Smart, 1977). Opportunities for Decision-Making Processes -- Political, Information Systems, and Traditional Knowledge Crises are opportunities for political reform. These opportunities, however, are limited (Boin, 2003). Even though crises expose the status quo as problematic, making it easier to gain momentum for alternative policies and institutions, the opportunities for reform in the wake of crisis are smaller than often thought because the requisites of crisis leadership are at odds with the requirements of effective reform (Boin, 2003). Moreover, “it requires political skills to turn disaster and crisis from threat and adversity into an opportunity and a political asset” (Boin, 2006). Communication strategies are political skills that can greatly enhance efficiency and professionalism of delivery of client services (Ferrara, 1996). Even in the absence of communication skills to turn crises into opportunities, crisis agencies usually welcome crises as the moment they can prove their right to exist and as a precondition for changes in conservative systems (Rosenthal, 2001). Crises represent “occasion for decisions” (Rosenthal, 2001). Because crises induce a sense of “urgency”, administrative decision making becomes centralized (Hart, 1993). This centralization is particularly important for emergency managers as they often need to take decisions quickly on very inadequate information (Cosgrave, 1992). An effective emergency management, because of limited information, is often connected with the goals to be achieved in response activities. The on-scene commanding structure and the on-scene commander play an important role in fighting emerging crises. This dynamic decision making at the "sharp end", however, presents severe demands on the commanders due to sudden scenario occurrences and the vast number of variables involved and the apparent uniqueness of each scenario (Rake, 2004). These severe demands are offset by emphasizing the importance of recognition-primed decision making and the vital role of on-scene commanders on crisis developments (Flin, 1996). Operating with very little information, on-scene crisis leaders have to rely on creativity in solving problems (Shlaim, 1983). This creative process is only valid if during crisis management plans reduce the steps required to make critical response decisions (Hale, 2006). The common notion that during crises decision making should be done in line with a command and control structure is invalid (Santen, 2009). The best way for crisis decision making teams in a bureaucratic political context is to follow an integrative negotiation approach as the shared mental model of decision making (Santen, 2009). In Finland, for instance, decision-making is often left in the hands of a small elite group, except in the case of civilian emergencies when key decisions are made at the operational level (Forsberg, 2006). Finns generally avoid defining difficult situations as a “crisis” to prevent the government using this event as a vehicle for acquiring more power (Forsberg, 2006). This situation only occurs because the political elite, the business community, and the civil society have been fairly united and have rejected critical opinions during a crisis and they would rather react to “crisis” cautiously rather than in panic (Forsberg, 2006). This social cohesion observed in the Finn society is made possible because the higher the socioeconomic status of the population, the greater the level of citizen participation in day-to-day community decision-making (Crain, 1967). The main effect of this relationship is the increase of the power of the citizens vis-à-vis the local government and the elite (Crain, 1967). Successful crisis avoidance can simply be a matter of watching the numbers, assigning a meaning, and considering the alternatives (Young, 2008). In crisis management planning, organizations must be positioned to capture and examine nearly every out-of-the-ordinary indicator relevant to their enterprise (Young, 2008). By acknowledging the cognitive and technical limitations imposed by “bounded rationality” -- information, time, and budget constraints -- and the recognition-primed decision model -- when crisis managers assess the situation to seek recognizable scenarios --, it can be argued that a formalized decision making template should be developed to avoid information overload during the most critical of incidents (Young, 2008). In addition, the contingency model of decision-making may offer value to crisis managers since it considers ambiguity, complexity, instability, and accountability along the decision process. By blending these options -- rule-based: following procedures -- and knowledge-based: creative in nature --, additional research would help find a balance between procedural and creative decision-making models (Young, 2008). Insurance is an example of rule-based model since it i) distributes losses, ii) reconstructs loans, iii) relieves welfare payments, and iv) exercises guidance over the extent to which people expose themselves to risks from hazards (White, 1975). Crisis simulation illustrates creative models as this is an exercise in cognitive structuring that entails a search for heuristics and organizational decisions which might render complex, highly interactive social and technological situations comprehensible and controlled (Jarman, 1990). New decision-making models must be developed to accommodate greater flexibility and creativity (Mendonça, 2008). The improvisation in emergency management is based on the paradigm of operational risk management and is motivated by the observation that emergency response organizations must be prepared to improvise during response activities (Mendonça, 2008). The process of emergency response in light of this new concept must discuss the opportunities for supporting the process identified since it deals with complex and unpredictable events (Mendonça, 2004). The fuzzy decision support system (FDSS) is a decision making structure which can be applied to manage the crisis conditions in any large scale systems with many parameters (Menhaj, 2010). After receiving both functional variables and fault signals, the FDSS makes proper decisions to make up and repair the distorted situation and the affected elements of the network according to its data base established through experience gathered from expert managers and decision models properly developed. These decisions are expressed in the form of some scenarios with different desirability degrees, which are determined by some properly developed fuzzy multi-criteria decision making methods, helping the manager choose the best one according to his discretion (Menhaj, 2010). Decisions to invest in information systems (IS) are made by many organizations on a very regular basis. Such decisions can vary from quickly identifying the problem, screening options and choosing a solution in a very straightforward way, to very extensive and repeated search, screen, design and negotiation activities that can take many years (Boonstra, 2003). The common issues involving IS decision-making processes are i) whether there is scope to design a solution, ii) whether distinct alternatives have to be searched for, iii) the degree of urgency and necessity from the perspective of decision-makers, iv) whether the decision can be subdivided in order to follow a gradual process path -- planned v. incremental --, and v) the number and power of stakeholders involved in the process and the extent that their interests vary and contrast (Boonstra, 2003). IS applications should be aware of these factors in order to design a process that fits best with the specific circumstances: no single process should be considered universally applicable (Boonstra, 2003). This conclusion is in contrast with many decision-making models rooted in the IS-field, which suggest to use prescriptive and rational approaches to organize IS decision-making processes (Boonstra, 2003). The understanding that IS applications should be flexible to meet the requirements of strategic crisis response models to check whether it is necessary to expand response structure and lower the cost of configuration that is adequate to the new situation, fulfilling the legislation requirements (Costa, 2005). Three insights over iterative models are i) action-oriented problem solving such as acting, interpreting, and cultivating diagnoses, ii) feedback among these processes opens and closes windows of adaptive problem solving, and iii) reinforcing feedback and confirmation bias, usually considered dysfunctional, for adaptive problem solving (Rudolph, 2009). Local knowledge -- or traditional ecological knowledge -- provides insights on how communities should interact with their environment. Local knowledge is “a cumulative body of knowledge and beliefs, evolving by adaptive processes and handed down through generations by cultural transmission, about the relationship of living beings with one another and with their environment”(Gunderson, 2002). It is a broad concept that covers all forms of knowledge -- technologies, knowhow, skills, practices, and beliefs -- that enable the community to achieve stable livelihoods in their environment (UNEP, 2008). Various communities rely on indigenous knowledge to sustainably utilize their natural resources. The indigenous knowledge systems provide a variety of options and innovations to deal with the challenges of nature conservation and disaster management in the course of making a living (UNEP, 2008). For example, a community living on the shores of Lake Victoria had a well-organized system for mitigating impeding disasters. There were elders who dealt with rainfall prediction and early warning. Each homestead had a dugout canoe ready for transport in case of heavy flooding. Each community was also required to dig trenches to control the water around the homestead and around farmlands. In addition, they were required to avoid ploughing along the lake shores when heavy flooding was predicted and were advised to catch fish during rainy period when they were plentiful and preserve them by drying and smoking for use in times of scarcity (UNEP, 2008). Other indigenous methods are the ones used by the Swazis to predict natural hazards using wind direction, the shape of the crescent moon, and the behavior of certain animals. Popularizing these and other indigenous pieces of knowledge not only alleviates poverty but also functions as disaster risk reduction efforts. Furthermore, indigenous knowledge should be incorporated into national development plans, taught in schools, and be protected by intellectual property laws (UNEP, 2008). Conclusion The relation between planning and decision making during crises should be viewed “as a form of policy making under adversity and of societal problem-handling processes” (Dror, 1988). Due to the inconsistency of this relation we must acknowledge that “the logic of decision making in emergency environments is “fuzzy” or imprecise, and wisely so, recognize the necessity for innovative problem-solving in unforeseen situations” (Comfort, 1988). Despite planning and decision making relations being irregular, the concepts, problems, and solutions presented in this paper show that planning and decision-making processes complement each other during crisis management. Physical, community, and policy planning offset the shortcomings and maximize the merits of decision-making processes. Planners prepare decision-makers for the uncertainties, risks, and urgencies associated to crisis management and decision-makers utilize critical events to reinforce political and economic interests based on planning principles and practices. This joint operation mitigates social and cognitive challenges and explores financial and cultural opportunities related to crisis management. This positive correlation, however, relies on political will, monetary resources, fluid communication, and technical expertise to thrive and as a result to reduce the impact of and accelerate response to crises. Communication, education, and technology are the forces driving this complementary work between planners and decision-makers throughout the disaster cycle. References: Alexander, D. “The Study of Natural Disasters, 1977-97: Some Reflections on a Changing Field of Knowledge”, Disasters, v. 21, i. 4, pp. 284-304, 1997. Anderson, P. “Decision Making by Objection and the Cuban Missile Crisis”, Administrative Science Quaterly, v. 28, n. 2, pp. 201-222, 1983. Black, D. “Environmental analysis: a decision-making process”, US Forest Service, 2007. Boin, A. et al. “The Politics of Crisis Management: Public Leadership Under Pressure”, Cambridge, 2006. Boin, A.; Hart, P. “Public Leadership in Times of Crisis: Mission Impossible?”, Public Administration Review, v. 63, i. 5, pp. 544-553, 2003. Boonstra, A. “Structure and analysis of IS decision-making processes”, European Journal of Information Systems, v. 12, n. 3, pp. 195-209, 2003. Burton, I. et al. “The Environment as Hazard”, The Guilford Press, 2nd edition, 1993. Comfort, L. “Asymmetric Information Processes in Extreme Events”, Chapter 5 in Communicable crises: prevention, response, and recovery in the global arena, Research in public management, 2007. Comfort, L. “Designing Policy for Action: The Emergency Management System”, University of Pittsburgh, 1988. Comfort, L. “Managing Intergovernmental Responses to Terrorism and Other Extreme Events”, Publius, v. 32, i.4, pp. 2002 Comfort, L. “Shared Risk: Complex Systems in Seismic Response”, Pergamon Press, 1999. Cosgrave, J. “Decision making in emergencies”, Disaster Prevention and Management, v. 5, I, 4, pp. 28-35, 1992. Costa, L. et al. “Strategic Optimization and Contingency Planning Model for Oil Spill Response”, Society of Petroleum Engineers, 2005. Crain, R.; Rosenthal, D. “Community Status as a Dimension of Local Decision-Making”, American Sociological Review, v. 32, n. 6, 1967. DeRouen Jr., K.; Sprecher, C. “Initial Crisis Reaction and Poliheuristic Theory”, v. 48, n. 1, pp. 56-68, 2004. Dror, Y. “Decisionmaking under disaster conditions” in “Managing Disaster: Strategy and Policy Perspectives”, Comfort, L. Indiana University, 1988. Ferrara, F. “Interagency Communication Measure (ICM): A Survey Instrument to Enhance Interagency Collaboration”, The Clinical Supervisor, v. 14, n. 1, pp. 175-186, 1996. Flin, R. et al. “Emergency Decision Making in the Offshore Oil and Gas Industry”, Human Factors: the Journal of the Human Factors and Ergonomics Society, v. 38, n. 2, pp. 262-277, 1996. Forsberg, T.; Pursiainen, C. “Crisis Decision-Making in Finlan: Cognition, Institutions and Rationality”, Cooperation and Conflict, v. 41, n. 3, pp. 235-260, 2006. Gunderson, L.; Holling, C. “Panarchy: Understanding Transformations in Human and Natural Systems”, Island Press, 2002. Hale, J. et al. “Decision processes During Crisis Response: An Exploratory Investigation”, Journal of Managerial Issues, v. XVIII, n. 3, pp. 301-320, 2006. Hart, P. “Groupthink in government: A study of small groups and policy failure”, Swets & Zeitlinger Publishers, Netherlands, 1990. Hart, P. et al. “Crisis Decision Making: The Centralization Thesis Revisited”, Administration & Society, v. 25, n. 1, pp. 12-45, 1993. Hart, P.; Rosenthal, U. “Experts and Decision Makers in Crisis Situations”, Science Communication, v. 12, n. 4, pp. 350-372, 1991. Jarman, A.; Kouzmin, A. “Decision pathways from crisis”, Crime, Law and Social Change, v. 14, n. 4, pp. 399-433, 1990. Kahneman, D. “Bias, Blindness and How We Truly Think”, Bloomberg, Oct 24, 2011. Accessed on Feb 28, 2012. http://www.bloomberg.com/news/2011-10-24/bias-blindness-and-how-we-truly-think-part-1-daniel-kahneman.html Kahnemann, D.; Tversky, A. “Judgement Under Uncertainty: Heuristics and Biases”, Science, v. 185, 1974. Kapucu, N. “Interagency Communication Networks During Emergencies: Boundary Spanners in Multiagency Coordination”, American Review of Public Administration, v. 36, n. 2, pp. 207-225, 2006. Kellerman, B. “Allison Redux: Three More Decision-Making Models”, Polity, v. 15, n. 3, 1983. Klein, G. “Sources of Power, How People Make Decisions”, MIT, 1998. Kunreuther, H.; Useem, M. “Learning from catastrophes: strategies for reaction and response”, Wharton School Publishing, New Jersey, 2010. Levi, A.; Tetlock, P. “A Cognitive Analysis of Japan’s 1941 Decision for War”, Journal of Conflict Resolution, v. 24, n. 2, pp. 195-211, 1980. McCarty, M. “The Nobel laureates: how the world’s greatest economic minds shaped modern thought”, McGrw-Hill, New York, 2001. Mendonça, D. et al. “Decision support for improvisation during emergency response operations”, International Journal of Emergency Management, v. 1, n. 1/2001, pp. 30-38, 2004. Menhaj, M. et al. “Fuzzy decision support system for crisis management with a new structure for decision making”, Expert Systems with Applications, v. 37, i. 5, pp. 3545-3552, 2010. Monten, J.; Bennett, A. “Models of Crisis Decision Making and the 1990-91 Gulf War”, Security Studies, vl. 19, i. 3, pp. 486-520, 2010. Mook, D. “Classic Experiments in Psychology”, Greenwood Press, 2004. Olson, Mancur. “The Logic of Collective Action: Public Goods and the Theory of Groups”, Harvard University Press, 1965. Patel, A. “Crisis communication under transformative change: the emergent context and roles of primary and secondary organizations”, PhD thesis, Queensland University of Technology, 2011. Pearson, C.; Clair, J. “Refraiming Crisis Management”, The Academy of Management Review, v. 23, n. 1, pp. 59-76, 1998. Pidgeon, N. “The Limits to Safety? Culture, Politics, Learning and Man-Made Disasters”, Journal of Contingencies and Crisis Management, v. 5, i. 1, pp. 1-14, 1997. Pierre, M. et al. “Crisis Management in Acute Care Settings”, Second Edition, Springer, 2011. Rake, E. “Emergency management and decision making on accident scenes: taxonomy, models and future research”, v. 1, n. 4., pp. 397-409, 2004. Remus, W.; Kottemann, J. “A test of the Kahnemann and Tversky model for decision making”, University of Hawaii at Manoa, 1988. Rosenthal, U. “Managing Crises: Threats, Dilemmas, Opportunities”, Charles C Thomas, 2001. Rudolph, J. et al. “The Dynamics of Action-Oriented Problem Solving: Linking Interpretation and Choice”, v. 34, n. 4, pp. 733-756, 2009. Santen et al. “Crisis Decision Making Through a Shared Integrative Negotiation Mental Model”, 6th International ISCRAM Conference, Sweden, 2009. Shlaim, A. “The United States and the Berlin Blockade, 1948-1949”, University of California Press, Berkeley, 1983. Sigelman, L.; McNeil, D. “White House Decision-Making under Stress”, American Journal of Political Science, v. 24, n. 4, 1980. Slovic, P. “Individual decision-making model for disasters”, in “The Environment as Hazard”, Burton, I. et al, The Guilford Press, 2nd edition, 1993. Slovic, P. et al. “Decision Processes, Rationality, and Adjustment to Natural Hazards”, in “Natural Hazards: Local, National, Global”, White, G. Oxford University Press, pp. 187-205, 1974. Smart, C.; Vertinsky, I. “Designs for Crisis Decision Units”, Administrative Science Quaterly, v. 22, n. 4, 1977. Taleb, N. “The Black Swan: The Impact of the Highly Improbable”, Random House Trade Paperbacks, 2010. Turner, B. “The Future for Risk Research”, Journal of Contingencies and Crisis Management, v. 2, i. 3, pp. 146-156, 1994. Turner, B.; Pidgeon, N. “Man-Made Disasters”, Butterworth-Heinemann, 1978. UNEP. “Indigenous Knowledge in Disaster Management in Africa”, United Nations Environment Programme, 2008. UNISDR. “Evidence: Monthly UNISDR Newsletter”, i. 4, January, 2012. Welch, D. “Crisis Decision Making Reconsidered”, The Journal of Conflict Resolution, vl. 33, n. 3, pp. 430-445, 1989. White, G. “Choice of Adjustment to Floods”, The University of Chicago, 1964. White, G; Haas, E. “Assessment of Research on Natural Hazards”, The MIT Press, 1975. Williams, D.; Olaniran B. “Exxon’s decision-making flaws: the hypervigilant response to the Valdez grounding”, Public Relations Review, v. 20, i. 1, pp. 5-18, 1994. Young, R. “Crisis Prediction and Effective Decision-Making Models”, Continuity Central, accessed on Feb 27, 2012. http://www.continuitycentral.com/feature0600.html.