Drawing practical lessons from punctuated equilibrium theory
從間斷平衡理論中汲取實際教訓

Promoting informed decision-making is a recurring theme in improving the efficacy and responsiveness of government. Consistent with this theme is the assumption that decision-makers face an information deficit - that if decision-makers simply had access to the right knowledge that they would make better decisions. There exists an overarching normative consensus that more information in the policy process is generally good, but there are limits to how much decision-makers can process. Scholars of public policy, and particularly agenda setting, have tended to find that the central problem of policymaking is not a deficit of information, but instead, an oversupply of information. Very often, then, the central problem confronting decision-making is prioritising among this information. Information-processing, specifically attention, is seen as the driver of policy change in theories of agenda-setting such as multiple streams (see Cairney, this issue) and particularly so in punctuated equilibrium theory (PET). PET scholars have long understood that policy and institutional design are critical to information-processing (Baumgartner and Jones, 1993). This chapter represents a guide for scholars wishing to bridge the divide between PET theory and the practice of institutional design.
促進知情決策是提升政府效能和反應能力的反覆出現的主題。與這一主題一致的是決策者面臨信息不足的假設——即如果決策者能夠獲得正確的知識，他們就會做出更好的決策。存在一個普遍的規範共識，即政策過程中的更多信息通常是好的，但決策者能夠處理的信息量是有限的。公共政策學者，特別是議程設置學者，往往發現政策制定的核心問題不是信息不足，而是信息過剩。因此，決策面臨的核心問題往往是對這些信息進行優先排序。信息處理，特別是注意力，在諸如多流理論（見本期 Cairney）的議程設置理論中被視為政策變革的驅動力，特別是在斷點平衡理論（PET）中。PET 學者長期以來理解到政策和制度設計對信息處理至關重要（Baumgartner 和 Jones，1993）。 本章為希望彌合理論與制度設計實踐之間鴻溝的學者提供了一份指南。

The engine of information processing in policy systems are subgovernments. Subgovernments are defined collections of policy actors in government, and around government, who develop and make policy within substantively specific issues. Usually, these subgovernments will contain an authoritative body (for example, a city council or congressional committee), an administrative unit for implementing policy (for example, the ministry of transportation or the police department), and supportive constellations of those interested in the policy issue (for example, Greenpeace or the Chamber of Commerce). Finally, subgovernments may also operate within specific policy areas (for example, energy policy), but be delineated along functionally differentiated lines (for example, regulate energy or subsidise the industry).
政策系統中的信息處理引擎是次級政府。次級政府被定義為政府內外的一群政策行動者，他們在特定實質問題上制定和實施政策。通常，這些次級政府會包含一個權威機構（例如，市議會或國會委員會）、一個負責實施政策的行政單位（例如，交通部或警察局），以及對該政策問題感興趣的支持團體（例如，綠色和平組織或商會）。最後，次級政府也可能在特定政策領域（例如，能源政策）內運作，但根據功能區分來劃分（例如，監管能源或補貼產業）。

It is commonly understood that more information in the policy process is generally a good thing, but building the capacity to process that information creates classic challenges of delegation, coordination, accountability and organisation. Subgovernments have their own information processing limitations, irrespective of how narrow their focus. Information processing limitations and oversight challenges lead to inadequate problem redress even on the parts of subgovernments. For governments, centralisation of decision-making is a natural impulse to contend with the daunting twin challenges of corralling myriad subgovernments coupled with still suboptimal outcomes. Where it concerns the process of problem definition, or how we come to define and understand particular policy problems, the centralisation impulse is wrong and often counterproductive. In particular, it leaves policy systems vulnerable to leaving out important dimensions of problems, dimensions that left unattended, will lead to policy failures and volatility in governance outcomes (Jones and Baumgartner, 2005).
普遍認為，在政策過程中擁有更多資訊通常是件好事，但建立處理這些資訊的能力卻帶來了委託、協調、問責和組織等經典挑戰。次級政府無論其關注點多麼狹窄，都有其自身的資訊處理限制。資訊處理的限制和監督挑戰導致即使是次級政府也無法充分解決問題。對於政府而言，集中決策是應對眾多次級政府管理與仍然不理想的結果這雙重挑戰的自然衝動。然而，在問題定義的過程中，或者說我們如何定義和理解特定政策問題時，這種集中化的衝動是錯誤的，並且常常適得其反。特別是，它使政策系統容易忽略問題的重要層面，這些層面若被忽視，將導致政策失敗和治理結果的不穩定（Jones 和 Baumgartner，2005）。

The trade-offs between organisational impulses to delegate or to centralise are widely discussed in public administration and public policy circles (May et al, 2008; Ostrom, 2008; Selznick, 1949). Governments understand that there is not one strategy of delegation that solves decisionmaking dilemmas; further, public officials know that such dilemmas are rarely, if ever, solved. However, we argue that thinking of these organisational choices in the context of broader theories of information processing can help to not only structure a better decision-making apparatus in government - but also to work with the structures that currently exist. The centralisation-decentralisation argument is fantastical in terms of many of the solutions it offers for policymaking: restructuring government is hard and, like any policy choice, may produce more problems than it solves.
在公共行政和公共政策領域中，組織傾向於授權或集中化的權衡被廣泛討論（May 等，2008；Ostrom，2008；Selznick，1949）。政府明白並沒有一種授權策略能解決決策困境；此外，公共官員知道這些困境很少（如果有的話）被解決。然而，我們認為，在更廣泛的資訊處理理論背景下思考這些組織選擇，不僅有助於在政府中構建更好的決策機制，還能與現有結構協同工作。集中化與分散化的論點在政策制定提供的許多解決方案中是幻想的：重組政府是困難的，並且像任何政策選擇一樣，可能產生的問題比解決的更多。

If PET shows that government efficacy and efficiency is hindered by information processing challenges, then the unstated practical contribution of PET and other information-processing scholarship is to make governments better at generating appropriate information for problem solving and managing the flow of information. Our work offers information processing scholars, particularly PET scholars, five ways of thinking about matching future policy research with solutions to underlying problems that this literature has identified as impediments to policymaking. We caution that (1) centralisation does not solve problems of information search, instead, centralisation creates bottlenecks, and (2) multiple venues offer more representation and opportunities for citizen influence, but suffer from attention limitations. Given these cautions, we suggest that (3) institutions be explicitly designed to be information- seeking, (4) issue
如果 PET 顯示政府效能和效率因信息處理挑戰而受阻，那麼 PET 及其他信息處理學術研究的未言明實際貢獻，便是使政府更擅長生成解決問題所需的適當信息並管理信息流。我們的工作為信息處理學者，特別是 PET 學者，提供了五種思考方式，以將未來的政策研究與解決根本問題的方案相匹配，這些問題已被該文獻識別為政策制定的障礙。我們警告說，（1）集中化並不能解決信息搜索的問題，反而會造成瓶頸，（2）多個場所提供了更多的代表性和公民影響的機會，但卻受到注意力限制的困擾。鑑於這些警告，我們建議（3）機構應明確設計為尋求信息的，（4）問題

The classic economics of information views it as a good held private, necessitating rents (Stigler, 1961). Previous understandings of government decision-making have assumed an information-poor environment in which decision-makers’ tasks were focused on information-gathering, or building incentives to reveal information. In the competitive political environments, information is provided freely in an effort to steer the course of policy (Workman, 2015). Conceptualisations of governmental, non-governmental, firm and individual decision-making rested on a now softened conceptualisation of choice that assumed information flows to governance units were controllable through some kind of intentional search.
經典的資訊經濟學將資訊視為一種私有商品，需要通過租金來獲取（Stigler, 1961）。以往對政府決策的理解假設了一種資訊貧乏的環境，決策者的任務集中在資訊收集或建立激勵機制以揭示資訊。在競爭的政治環境中，資訊被免費提供以引導政策走向（Workman, 2015）。政府、非政府組織、企業和個人決策的概念化建立在一個現已軟化的選擇概念上，該概念假設通過某種有意的搜索可以控制流向治理單位的資訊。

Here we leverage Baumgartner and Jones’ (1993) widely applied PET to characterise the information processing environment of governments, non-governmental organisations, and other groups that are charged with solving problems via policy. PET profoundly challenges the convention that governments cannot make good decisions because they don’t have enough information. It contends that governments have too much information, that they are cognitive misers, and that subgovernments
在這裡，我們利用 Baumgartner 和 Jones（1993 年）廣泛應用的 PET 來描述政府、非政府組織及其他被賦予通過政策解決問題任務的群體的信息處理環境。PET 深刻挑戰了政府因信息不足而無法做出良好決策的傳統觀念。它主張政府擁有過多信息，他們是認知吝嗇者，而次政府

In the sections that follow, we outline the major features of PET. As scholars who are part of the PET research community, we are struck by two features: 1) the fact that most of the policy process community beyond PET scholars is aware of the theory and 2) the simple parts of the metaphor give way to vague understandings of the inner workings of the informational processing theory that underlie the actual theoretical mechanisms that give rise to a predicted pattern of change. Our experience is that many scholars think of PET as a theory that attempts to explain major change that results from exogenous shocks. While PET can help explain large changes, some of which emerge from exogenous shocks, such a limited understanding is, well, wrong.
在接下來的章節中，我們概述了 PET 的主要特點。作為 PET 研究社群的一部分，我們對兩個特點印象深刻：1）大多數政策過程社群對 PET 理論有所了解，2）比喻的簡單部分讓位於對信息處理理論內部運作模糊的理解，這些理論機制產生了預測的變化模式。我們的經驗是，許多學者認為 PET 是一種試圖解釋由外生衝擊引起的重大變化的理論。雖然 PET 可以幫助解釋一些由外生衝擊引起的大變化，但這種有限的理解是錯誤的。

From a methodological perspective, PET is a unified theory of information processing and policy change. This is to say that PET incorporates a model of decision-making that is intended to be broadly applied to a variety of policy venues and integrates individual and organisational decisionmaking. For PET scholars, shifting attention is the causal driver of choice. Limited attention is key to understanding human cognition (Jones, 2017), and the strategies that human beings use to deal with an information- rich environment (for example, Thomas (2017) on cue-taking). Under the strain of policy failures, focusing events, or salient issues and the increased cognitive load they bring, focusing attention is even more difficult for individuals (Shaffer, 2017). Organisations can be seen as one way to address the attention limits of individuals (Jones, 2001), and the delegation of attention defines institutional organisation (Workman, 2015). Institutional capacity renders more stable policymaking, even in the face of a complex stream of information that might otherwise produce volatility in attention (Epp and Baumgartner, 2017).
從方法論的角度來看，PET 是一種關於信息處理和政策變化的統一理論。這意味著 PET 包含了一個決策模型，該模型旨在廣泛應用於各種政策場合，並整合了個人和組織的決策過程。對於 PET 學者來說，注意力轉移是選擇的因果驅動因素。有限的注意力是理解人類認知的關鍵（Jones, 2017），以及人類在信息豐富的環境中使用的策略（例如，Thomas (2017) 關於線索採納的研究）。在政策失敗、聚焦事件或突出問題及其帶來的認知負荷增加的壓力下，個人更難集中注意力（Shaffer, 2017）。組織可以被視為解決個人注意力限制的一種方式（Jones, 2001），而注意力的委託則定義了機構組織（Workman, 2015）。機構能力使得決策更加穩定，即使面對可能導致注意力波動的複雜信息流（Epp and Baumgartner, 2017）。

To this end, PET is a distributional theory of policy change in that it intends to understand a wide range of outcomes from policy venues, and the processes that underlie these. Among policy scholars, the importance of incrementalism in explaining policy punctuations is a never-ending source of irony and confusion. Charles Lindblom (1959) is ‘alive and well’ in PET (Howlett and Migone, 2011). Plainly, the theory seeks to explain long periods of stability, short periods of rapid change, and the lack of much in between - moderate change. It is here where the concept of policy punctuations is perhaps the most intuitively interesting part of the theory to most casual consumers of policy process research, but, is simply one component of understanding policy change for scholars that apply PET.
為此，PET 是一種政策變遷的分配理論，旨在理解來自政策場域的廣泛結果及其背後的過程。在政策學者中，漸進主義在解釋政策斷點方面的重要性是一個永無止境的諷刺和困惑來源。查爾斯·林德布洛姆（1959）在 PET 中「活得好好的」（豪利特和米戈內，2011）。簡單來說，該理論試圖解釋長期的穩定、短期的快速變遷，以及中間缺乏的——適度變遷。正是在這裡，政策斷點的概念對於大多數政策過程研究的普通消費者來說，可能是理論中最直觀有趣的部分，但對於應用 PET 的學者來說，它僅僅是理解政策變遷的一個組成部分。

Policymaking venues must prioritise issues and information about them. Policy venues have much information that they could use for decision-making, but ultimately use a systematically biased subset, potentially in the face of dramatically changing problem and political conditions. The path by which policy venues, policymaking systems, or governing institutions seek or sort information is a function of the relative strength of subgovernments to allow for alternative problem definitions. Subgovernments attend to issues as a function of their preferences and capacity, but occasionally are overwhelmed by macro-political forces that focus the attention of the broader institutions in which they are situated.
政策制定場所必須優先考慮問題及其相關信息。政策場所擁有大量可用於決策的信息，但最終使用的是系統性偏見的子集，這可能面臨急劇變化的問題和政治條件。政策場所、政策制定系統或治理機構尋求或分類信息的途徑是子政府相對強度的函數，以允許替代性問題定義。子政府根據其偏好和能力處理問題，但有時會被宏觀政治力量所壓倒，這些力量集中於它們所處的更廣泛機構的注意力。

As a theory that explains information processing, PET relies on a mixture of empirical assessments of policy venue outputs and assumptions regarding causal mechanisms. The vast majority of PET research asserts the
作為一種解釋信息處理的理論，PET 依賴於對政策場所輸出的實證評估和對因果機制的假設的混合。絕大多數 PET 研究斷言
mechanisms (described in the sections that follow) that create conditions for institutionally-induced stability and change. We characterise the major assumptions of PET in the next section.
機制（在接下來的章節中描述）為制度誘導的穩定和變革創造了條件。我們將在下一節中描述 PET 的主要假設。

We begin with a discussion of two key underlying assumptions of PET: an understanding of how individuals and institutions make decisions and the role of attention in decision-making. The basic theoretical understanding of decision-making is found in Jones’ early works on decision-making (1994; 2001) in which the rational model of the individual is challenged. Based in works of cognitive psychology and behavioural economics, PET envisions individuals who are limited in their capacity to seek and make sense of information. This makes individuals ‘cognitive misers’ in the face of multiple decisions. While political behaviourists (Rugeley and Gerlach, 2012) and other theorists of the policy process are concerned with the formation and change of beliefs (see also Weible and Ingold this issue), PET assumes preferences are relatively fixed - or at least fixed enough that shifts in choice are not the result of changes in beliefs or preferences.
我們從討論 PET 的兩個關鍵基本假設開始：對個人和機構如何做出決策的理解，以及注意力在決策中的作用。對決策的基本理論理解可以在 Jones 早期關於決策的著作（1994；2001）中找到，其中挑戰了個人的理性模型。基於認知心理學和行為經濟學的著作，PET 設想個人在尋求和理解信息的能力上是有限的。這使得個人在面對多重決策時成為「認知吝嗇者」。雖然政治行為學家（Rugeley 和 Gerlach，2012）和其他政策過程理論家關注信念的形成和變化（另見 Weible 和 Ingold 本期），PET 假設偏好是相對固定的——或者至少固定到足以使選擇的轉變不是信念或偏好變化的結果。
Practically, this means that there is a fixed range of issues to which individuals can attend coupled with a cognitive limitation regarding how much time individuals can attend to those issues. Individuals largely rely on heuristics to make choices rather than engage in a full information search in order to address choices with which they are faced. Individuals are consistently attuned to certain issues (May, 1991), but occasionally are forced to make choices on issues they do not regularly attend. Choices individuals make on these issues are a function of the preferences they currently hold rather than preferences that are developed in problem solving the issue over time. In other words, attention activates preferences, and is the precursor of choice - a necessary, but not sufficient condition for action.
實際上，這意味著個人能夠關注的問題範圍是固定的，並且在關注這些問題的時間上也存在認知限制。個人很大程度上依賴啟發法來做出選擇，而不是進行全面的信息搜索來應對所面臨的選擇。個人始終對某些問題保持敏感（May, 1991），但偶爾被迫對他們不常關注的問題做出選擇。個人對這些問題做出的選擇是他們當前持有的偏好而非隨著時間推移在問題解決過程中形成的偏好的函數。換句話說，注意力激活了偏好，並且是選擇的前兆——行動的必要但不充分條件。
Organisations are designed to alleviate the limits of attention in individuals (Jones, 2001). The concept of ‘institutions’ can take on a variety of meanings in the study of politics and policy, even within the narrow field of public policy process research. Here, we understand institutions to mean the rules, norms and units along which lines of delegation, accountability and authority operate. Institutions help alleviate individual decision-making limitations in two ways. First, institutions contain numerous individuals, extending the information processing capacity of individuals. Second, institutions are designed to attend to a finite set of issues; that is, they are designed to allocate attention (Hammond, 1986).
組織的設計旨在減輕個人注意力的限制（Jones, 2001）。在政治與政策研究中，「制度」的概念可以有多種含義，即使在公共政策過程研究的狹窄領域內也是如此。在這裡，我們將制度理解為規則、規範和單位，這些要素沿著授權、問責和權力的運作線路展開。制度通過兩種方式幫助減輕個人決策的限制。首先，制度包含眾多個體，擴展了個人的信息處理能力。其次，制度被設計來關注一組有限的問題；也就是說，它們被設計來分配注意力（Hammond, 1986）。

As the rational model of the individual is challenged by the foundational assumptions of PET, so is the foundation of the rational model of institutions. At some level of aggregation, institutions ultimately face the same limits of attention and information processing as individuals. Institutions face severe problems when forced to allocate attention to issues for which they are not designed, or did not anticipate. Furthermore, institutions are often populated by either the same individuals or the same interests over long time periods. Thus, institutions privilege certain types of information over time, engage in constrained patterns of information search, and have similar agendas. Institutions often are nested within other institutions, which is the result of a macro- institution attempting to delegate attention to a smaller subunit (Workman, 2015; Workman et al, 2009). The resulting delegation of attention can be thought of as an analogue to organisational mission, jurisdiction, or turf.
隨著個人理性模型受到 PET 基本假設的挑戰，機構的理性模型基礎也同樣受到質疑。在某種聚合層面上，機構最終面臨與個人相同的注意力和信息處理限制。當被迫將注意力分配給它們未設計或未預見的問題時，機構會面臨嚴重問題。此外，機構通常由相同的個人或相同的利益群體長期主導。因此，機構隨著時間推移會優先處理某些類型的信息，進行受限的信息搜索模式，並擁有相似的議程。機構往往嵌套在其他機構之中，這是宏觀機構試圖將注意力委派給較小單位的結果（Workman, 2015; Workman et al, 2009）。這種注意力的委派可以被視為組織使命、管轄範圍或領域的類比。

Subgovernments represent a more restricted and institutionalised form of subsystems in policymaking. While subsystems may include multiple policymaking bodies, jurisdictions and levels of government; subgovernments are subordinate units of government within policymaking institutions. Subgovernments emerge because of the delegation of attention (Baumgartner et al, 2014). Institutions that intentionally delegate authority to smaller groups in order to both create agenda space and focus attention on specific problems do not have the capacity to monitor the activities they have delegated with great frequency. Thus, the institutions created to attend to specific problems often do so with great independence (Workman et al, 2009). Delegation, not control, is the norm in policymaking.
次級政府代表了政策制定中更為受限且制度化的子系統形式。雖然子系統可能包含多個政策制定機構、管轄權和政府層級；次級政府則是政策制定機構內的下屬單位。次級政府的出現源於注意力的委派（Baumgartner 等，2014 年）。那些為了創造議程空間並將注意力集中於特定問題而有意將權力委派給較小團體的機構，並無能力頻繁監控其所委派的活動。因此，那些為處理特定問題而創建的機構通常具有很大的獨立性（Workman 等，2009 年）。在政策制定中，委派而非控制是常態。

Generally speaking, as governments cede authority to subgovernments, a number of actors in positions of power seek to control the outputs that emerge from the policy venue. These outputs are ultimately dependent upon the macro-institution for passage and legitimacy, but the government has, by design, created a subgovernment for which little attention is required. Insofar as the subgovernment does not attract outside interest, conditions exist for actors within the institution to leverage their position to control what Baumgartner and Jones (1993) refer to as the ‘policy image’ (Pralle, 2003; Vaughan and Arsneault, 2008).
一般而言，隨著政府將權力下放給次級政府，許多掌權的行動者便試圖控制從政策場域中產出的結果。這些結果最終依賴於宏觀機構的通過與合法性，但政府通過設計創造了一個幾乎不需要關注的次級政府。只要次級政府不吸引外部利益，機構內的行動者就有條件利用其職位來控制 Baumgartner 和 Jones（1993）所稱的「政策形象」（Pralle, 2003; Vaughan 和 Arsneault, 2008）。
Empirical examples of subgovernment politics are found in the foundation of the PET literature. An enduring story from Baumgartner and Jones’ (1993) original work is the story of problem definition and pesticides. Following the Second World War, pesticide development in the United States and across the world was viewed as a positive step toward
次政府政治的實證例子可以在 PET 文獻的基礎中找到。Baumgartner 和 Jones（1993）原創作品中的一個持久故事是關於問題定義和殺蟲劑的故事。第二次世界大戰後，美國及全球的殺蟲劑發展被視為邁向積極的一步。
food production. In the US, the federal government largely delegated attention to issues related to pesticides to agricultural committees within Congress. Agricultural committees sought and prioritised problems associated with pesticides through the lens of production rather than human or environmental health. As such, for decades following the Second World War, pesticide policy was largely quiet and limited to a few players who maintained a status quo of pesticide promotion. The policymaking environment associated with pesticide policy in the United States resonates with the information processing pathologies of any organisation. All organisations face limited capacity for attention and information processing; at some point they become overwhelmed in processing information for a particular problem. Organisations can simply produce suboptimal outcomes, or choose to create alternative information processing structures and processes. In addition to creating expertise in information processing, this new sub-organisation is delegated attention.
食品生產。在美國，聯邦政府很大程度上將與農藥相關的注意力委託給國會內的農業委員會。農業委員會從生產的角度而非人類或環境健康的角度來尋求和優先處理與農藥相關的問題。因此，在第二次世界大戰後的幾十年裡，農藥政策基本上保持沉默，僅限於少數維護農藥推廣現狀的參與者。美國與農藥政策相關的決策環境與任何組織的信息處理病理學產生共鳴。所有組織都面臨著注意力和信息處理能力的限制；在某個時刻，它們在處理特定問題的信息時會不堪重負。組織可能僅僅產生次優結果，或選擇創建替代的信息處理結構和流程。除了在信息處理方面創造專業知識外，這個新的子組織還被委託了注意力。

PET envisions information processing similar to a circuit. Given the attention dynamics of PET that we previously describe, the delegation of authority from a government to a subgovernment creates policy monopolies processing information largely in parallel with one another. However, each policy venue - subgovernments and governments processes information serially (Jones, 1994; 2001). This is particularly important for macro-organisations that ultimately have the final decisionmaking authority over programmatic changes. In general, governments attend little to the inner-workings, information-processing and outputs of subgovernments - a process that buttresses favoured policy images over time.
PET 設想資訊處理類似於電路。根據我們之前描述的 PET 注意力動態，從政府到次級政府的授權創造了政策壟斷，這些壟斷在很大程度上並行處理資訊。然而，每個政策場域——次級政府和政府——則以串行方式處理資訊（Jones, 1994; 2001）。這對於最終擁有程式變更最終決策權的宏觀組織尤其重要。總的來說，政府很少關注次級政府的內部運作、資訊處理和產出——這一過程隨著時間的推移支持了偏好的政策形象。

In Baumgartner and Jones’ (1993) case of pesticides, concerned groups began to attempt to frame pesticide problems as human or environmental harms. In addition to agricultural committees, health and welfare committees began seeking and sorting information related to pesticides. Agricultural production is not the priority of health advocates - rather these committees focus on issues related to toxicity and exposure rates. Thus, alternative problem definitions emerged for the same problem across two delegated subgovernments. As these definitions clashed, the policy monopoly of agriculture began to weaken and policy caught up with shifting opinion.
在 Baumgartner 和 Jones（1993 年）關於農藥的案例中，相關團體開始嘗試將農藥問題塑造成對人類或環境的傷害。除了農業委員會，健康和福利委員會也開始尋找和整理與農藥相關的資訊。對於健康倡導者而言，農業生產並非首要任務——這些委員會更關注的是與毒性和暴露率相關的問題。因此，對於同一問題，兩個委派的次政府出現了不同的問題定義。隨著這些定義的衝突，農業的政策壟斷開始削弱，政策也隨之跟上了轉變的民意。
Rapid policy change need not be the fate for all policy change. Complex problems of modern government often demand that authoritative and deliberative organisations synthesise information from multiple substantive
快速政策變革未必是所有政策變革的命運。現代政府的複雜問題往往要求權威性和審議性組織綜合來自多個實質性領域的信息。
issues in order to make policy. The challenge of modern public institutions is to arrive at integrative problem definitions derived from a synthesis of potentially relevant streams of information across substantive and functional dimensions (Jones and Baumgartner, 2015). If policy responses end up falling along one of these substantive dimensions, it should be because of the choice or collective will of the decision-making organisation, and the process is designed to amplify only one among many streams of information.
為了制定政策而處理問題。現代公共機構的挑戰在於，從跨實質與功能維度的潛在相關信息流中，綜合出整合性的問題定義（Jones 和 Baumgartner，2015）。如果政策回應最終沿著這些實質維度之一展開，那應該是因為決策組織的選擇或集體意志，並且該過程被設計為僅放大眾多信息流中的一種。

What follows from changes in macro-attention to issues within the purview of the policy subsystem is a function largely of features specific to the policy monopoly itself. In other words, whether ‘external perturbations’ (Cashore and Howlett, 2007; Howlett and Cashore, 2009) influence otherwise stable policymaking arrangements is counterintuitively a result of endogenous features. Most recent scholarship in punctuated equilibrium is less concerned with causes of shifts in attention and more concerned with how policymakers and policymaking institutions respond. A robust literature on policy entrepreneurs beginning with Kingdon (1995) and expanding across multiple studies describe mechanisms by which politically motivated individuals or groups can act as attention focusers and policy fixers (Mintrom and Norman, 2009). Issue expansion can occur whether generated by specific policy entrepreneurs, political circumstances, events that are impossible to ignore, or all the above.
宏觀關注的變化對政策子系統範圍內議題的影響，很大程度上取決於政策壟斷本身的特徵。換句話說，「外部干擾」（Cashore and Howlett, 2007; Howlett and Cashore, 2009）是否會影響原本穩定的政策制定安排，直觀上看是內生特徵的結果。最近關於間斷平衡的研究較少關注注意力轉移的原因，而更多關注政策制定者及政策制定機構如何應對。從 Kingdon（1995）開始並擴展到多項研究的政策企業家文獻，描述了政治動機的個人或團體如何作為注意力聚焦者和政策修復者的機制（Mintrom and Norman, 2009）。議題擴展可能由特定的政策企業家、政治環境、無法忽視的事件或以上所有因素引發。
The extent to which a subgovernment’s maintenance of a policy image is robust to macro-political attention is a function of two interrelated features. First, the relative unity of preferences within a subgovernment can create a powerful counterforce against any new information brought into the policy discourse. We will refer to this as the ‘strong subgovernment’ scenario. Second, the extent to which the subgovernment has considered an expanded agenda of information over previous iterations of policymaking lays the groundwork for an adaptive policy system. We refer to this as the ‘weak subgovernment’ scenario. Each set of circumstances presents different outputs of policy change. In the first scenario, it is possible that the endogenous strength of the monopoly can fend off challenges, which will maintain stability even in time of new information. In the second scenario, the subgovernment is more open to changes to the policy image by macro-political forces; but it may have already changed sufficiently over time, such that change or reform, even with new attention, is stunted.
次政府維持政策形象的穩健程度對宏觀政治關注的抵抗力，取決於兩個相互關聯的特徵。首先，次政府內部偏好的相對統一能夠對抗任何引入政策討論的新信息，形成強大的反作用力。我們將此稱為「強次政府」情境。其次，次政府在過去的政策制定過程中考慮過擴展信息議程的程度，為適應性政策系統奠定了基礎。我們將此稱為「弱次政府」情境。每種情境都呈現出不同的政策變革結果。在第一種情境中，壟斷的內生力量可能抵禦挑戰，即使在新信息出現時也能保持穩定。在第二種情境中，次政府對宏觀政治力量改變政策形象更加開放；但隨著時間的推移，它可能已經發生了足夠的變化，以至於即使有新關注，變革或改革也會受到阻礙。

The two features of these policy monopolies describe the precursors to different distributions of policy change over time. PET expects policy
這兩項政策壟斷的特徵描述了政策變化隨時間不同分布的前兆。PET 預期政策
change to exhibit ‘punctuations’ under the strong policy monopoly scenario, primarily through the mechanisms of negative and positive feedback. As issues expand, macro-political attention brings in new actors and institutions which have competing definitions of the policy image. Issue expansion is more likely to occur as a function of the distance between the policy monopoly’s conceptualisation of the policy image versus actual problem characteristics at the time of macro-attentional focus. As subgovernments are unresponsive to real world changes in the nature of the problem, this structurally-induced stability will itself lead to dramatic policy change as the policy system is increasingly ill-fitted and maladapted to the real problem and the subgovernment succumbs to the disconnect between policy and reality. For practitioners, one of the central lessons of PET is that ‘locking-in’ policies or governing arrangements, particularly such that they are maladaptive as the problem changes, guarantees policy volatility and disjointedness at some point in the future.
在強烈的政策壟斷情境下，轉變為展示「標點符號」，主要透過負面與正面反饋的機制。隨著議題擴大，宏觀政治關注引入了新的參與者和機構，這些參與者和機構對政策形象有著競爭性的定義。議題擴張更可能發生，作為政策壟斷對政策形象的概念化與宏觀關注時實際問題特徵之間距離的函數。由於次級政府對問題性質的現實變化反應遲鈍，這種結構性誘導的穩定性本身將導致戲劇性的政策變化，因為政策系統越來越不適應實際問題，次級政府屈服於政策與現實之間的脫節。對於實踐者來說，PET 的一個核心教訓是，「鎖定」政策或治理安排，特別是在問題變化時這些安排變得適應不良，保證了未來某個時點的政策波動和不連貫性。

Thus, in general, two types of systems characterise these dynamics of policy change. The first is a system of error correction, in which policymaking is approximal. As the system responds to information, policy is adjusted in response to the information. This adjustment, however, never quite meets the demands of the information at hand. Thus, most policy systems are characterised by disproportionate information processing - that is, the policy system never responds quite in proportion to the new information about a problem. There is both over- and underresponse. The second type of system is one of error accumulation. In this system, policy is increasingly out of step with the current stream of information about a problem - representing the real-world status of the issue. Usually this will be owing to unresponsive institutions, which are loathe to move from the current status quo policy - either because of preferences for the status quo, or because of high decision-making costs within the institution or decision-making body. In the latter instance, pressure builds to respond to an ever-divergent stream of information until a drastic policy change is needed. For pesticides, the creation of a new regulatory agency in the Environmental Protection Agency (EPA) in 1970 and resulting committees within Congress provided competing subgovernments which eventually supplanted the agricultural production framing of pesticide problems. This laid the foundation for major agenda and policy change. In some cases, problems span the boundaries of previously defined subgovernments (Workman and Shafran, 2015), which leads to dynamic interactions between groups that previously worked in parallel. Critically, as the expanded issue is addressed, macro-institutional attention eventually recedes leaving in place the new entrants - often in the form of institutions or constellations of actors - to maintain attention
因此，一般而言，兩種系統類型刻畫了政策變化的這些動態。第一種是錯誤修正系統，其中政策制定是近似性的。隨著系統對信息作出反應，政策會根據信息進行調整。然而，這種調整從未完全滿足手頭信息的需求。因此，大多數政策系統的特徵是信息處理的不成比例——即政策系統從未完全按比例回應關於問題的新信息。存在過度反應和反應不足的情況。第二種類型的系統是錯誤積累系統。在這個系統中，政策越來越與當前關於問題的信息流脫節——代表著該問題的現實世界狀況。通常，這將歸因於反應遲鈍的機構，這些機構不願從現狀政策中移動——無論是因為對現狀的偏好，還是因為機構或決策機構內的高決策成本。在後一種情況下，壓力逐漸增大，以回應日益分歧的信息流，直到需要進行劇烈的政策變更。 對於農藥問題，1970 年在環境保護局（EPA）內成立的新監管機構以及隨之在國會內設立的委員會，提供了相互競爭的次級政府體系，最終取代了將農藥問題框定在農業生產範疇的做法。這為主要的議程和政策變革奠定了基礎。在某些情況下，問題跨越了先前定義的次級政府體系的邊界（Workman 和 Shafran，2015 年），這導致了原本平行運作的群體之間產生了動態互動。關鍵的是，隨著擴展的問題得到處理，宏觀機構的關注最終會消退，留下新進入者——通常以機構或行動者群體的形式——來維持關注。
on the policy area. Depending on institutional structure, these actors will begin to mould the policy image in the scope of their own preferences, through a process of negative feedback and the approximal policymaking described above. This is the classic punctuated equilibrium scenario.
在政策領域上。根據制度結構，這些行為者將開始在其偏好範圍內塑造政策形象，通過負面反饋和上述近似政策制定的過程。這是典型的間斷平衡情境。

A less punctuated distribution can occur, however, for policy monopolies whose attention is broader or that serve a variety of competing rather than cooperating interests, or, alternatively in policies for which there are no discernable set of actors or publics (May, 1991). A weak monopoly may be more volatile in its policy outputs, but will have made marginal adjustments over time that maintain a small distance between the ‘actual’ problem definition and that favoured by the policy image of the monopoly. In this scenario, it is possible for large swings in attention associated with an issue, but with a policy monopoly to maintain its policy image. Empirical studies find that all policy systems are subject to these dynamics of attention and resultant policy change, but some less than others. However, institutional design can mitigate policy volatility by structuring information processing to be adaptive (we come to this point later).
然而，對於那些關注範圍更廣或服務於多種競爭而非合作利益的政策壟斷，或者對於那些沒有明顯的行動者或公眾的政策，可能會出現一種較少標點符號的分佈（May, 1991）。一個弱壟斷可能在政策輸出上更加波動，但會隨著時間的推移進行邊際調整，以保持「實際」問題定義與壟斷所偏好的政策形象之間的小距離。在這種情況下，可能出現與某個問題相關的注意力大幅波動，但政策壟斷仍能維持其政策形象。實證研究發現，所有政策系統都受到這些注意力動態和由此產生的政策變化的影響，但有些系統受到的影響較小。然而，制度設計可以通過結構化信息處理來適應性減輕政策波動（我們稍後會討論這一點）。

In this chapter we have presented an account of PET to motivate future scholars to structure research around practical applications of policy process theories from an information processing approach. We show that theoretical and empirical applications of PET suggest that political systems of policymaking typically are punctuated. However, the degree to which political systems are punctuated is a function of institutional design. Thus, we offer five axioms for scholars engaging in research related to the two audiences that must contend with policy problems that are troubled to process the inherently complex array of problems and solutions that face policymaking systems. The first audience is comprised of entrepreneurs or other political actors who seek strategies to gain precious agenda space. The second audience includes policymakers who hope to devise institutions that have increased information processing capacity - both in terms of information search and attention - to avoid imperfect policymaking that accompanies the large episodic shifts in attention over time.
在本章中，我們介紹了 PET 的概述，以激勵未來學者從信息處理的角度圍繞政策過程理論的實際應用來構建研究。我們展示了 PET 的理論和實證應用，表明政治系統的決策過程通常是間斷的。然而，政治系統的間斷程度是制度設計的函數。因此，我們為從事與兩類受眾相關研究的學者提供了五個公理，這些受眾必須應對難以處理的複雜問題和解決方案，這些問題和解決方案是決策系統所面臨的。第一類受眾包括企業家或其他政治行動者，他們尋求策略以獲得寶貴的議程空間。第二類受眾包括政策制定者，他們希望設計出具有更高信息處理能力的制度——無論是在信息搜索還是注意力方面——以避免伴隨著時間推移注意力大幅間歇性轉移所帶來的不完美決策。

The practical lessons to be gleaned from information processing perspectives on governance and policymaking are increasingly relevant due to the nature of changing institutional arrangements, the onslaught of globalisation, and the emergence of complex, boundary-spanning problems (Nicholson and Orr, 2016). Given the difficulties of reaching consensus at the federal level, or apex, of modern democracies, actors at other levels
從資訊處理的角度看待治理和決策所獲得的實際教訓，由於制度安排的變革、全球化的衝擊以及跨領域複雜問題的出現（Nicholson 和 Orr，2016），變得越來越相關。考慮到在現代民主國家的聯邦層級或頂點達成共識的困難，其他層級的參與者
of government face increasing pressure to meet modern problems that stretch their capacity. One needs look no further for relevance than the polarisation and gridlock characterising the dysfunctional US Congress (Lewallen et al, 2016). Additionally, the politics of democratic electorates suggests increasing divergence between rural and urban governance reflected in elections in the US and between old and young in the UK (Brexit). This divergence, division and recalcitrance is occurring in an era of policymaking where globalisation is increasingly relevant, and where problems span substantive and jurisdictional boundaries - taxing the limits of divergent institutions to address them. In these contexts, the processes leading to disjointed policy change are potent forces for instability in governing arrangements and for volatility in policy responses.
政府面臨越來越大的壓力，以應對那些超出其能力的現代問題。我們無需遠尋，就能看到美國國會的功能失調所表現出的極化和僵局（Lewallen 等，2016）。此外，民主選民的政治顯示出美國選舉中城鄉治理之間的分歧，以及英國脫歐中老一代與年輕一代之間的分歧。這種分歧、分裂和頑固性發生在一個政策制定的時代，其中全球化越來越重要，問題跨越實質性和管轄權的界限，挑戰著不同機構應對這些問題的極限。在這些背景下，導致政策變革不連貫的過程是治理安排不穩定和政策反應波動的強大力量。
Our suggestions take the form of two cautions and three suggestions. Many reading this chapter might infer from the structural analysis we present that constantly churning government structure is the only way to ensure governments that maintain a broad focus. We state early on in this chapter that restructuring governments is a natural impulse when things are not going well, but that restructuring is hard and may create more problems than it solves. Our cautions address classic problems of centralisation and delegation using the lens of information processing. We suggest the design of explicitly information-seeking institutions to contend with new problems, for governments to yoke the disparate flows of information from subgovernments through issue bundling and consider the value of convergence of policy signals from subgovernments as policy evidence itself.
我們的建議以兩項警示和三項建議的形式呈現。許多閱讀本章的讀者可能會從我們提出的結構分析中推斷，不斷變革政府結構是確保政府保持廣泛關注的唯一方法。我們在本章開頭即指出，當事情不順利時，重組政府是一種自然的衝動，但重組是困難的，並且可能創造出比解決更多的問題。我們的警示從信息處理的角度來探討集中化和授權的經典問題。我們建議設計明確尋求信息的機構來應對新問題，讓政府通過議題捆綁來協調來自次級政府的不同信息流，並考慮次級政府政策信號的匯聚作為政策證據本身的價值。

Given that venue shopping is identified as a mechanism by which issue expansion and, therefore, policy punctuations occur, designers of political institutions would do well to assume that entrepreneurs will seek venues receptive to their problem definitions. One implication from findings on venue shopping might be to simply create fewer venues; therefore, centralising decision-making into fewer venues offers fewer opportunities for policy entrepreneurs to seek policy change. However, findings from the venue-shopping literature suggest that centralisation is no universal solution to issues related to punctuated policymaking and, indeed, can exacerbate information and attention bottlenecks that lead to intermittent, and punctuated, policymaking. Centralisation of decision-making is shown in the literature to lead to distributions of policy change that are more volatile rather than less.
鑑於場域購物被視為問題擴張從而導致政策中斷的機制，政治制度的設計者最好假設企業家會尋找對其問題定義持開放態度的場域。場域購物研究結果的一個啟示可能是簡單地減少場域數量；因此，將決策集中到較少的場域中，為政策企業家尋求政策變革提供了更少的機會。然而，場域購物文獻的發現表明，集中化並非解決與中斷式政策制定相關問題的萬能方案，實際上可能加劇信息和注意力的瓶頸，導致間歇性和中斷式的政策制定。文獻顯示，決策的集中化會導致政策變化的分佈更加波動，而非減少。

Centralisation of authority is typically not accompanied by a concomitant expansion of information processing or seeking capacity. In other words, the same challenges that befall processing in serial affect centralised authority wherein attention is limited to a few items at a time while the remainder of issues under the umbrella of an organisation go largely unchecked. The Federal Emergency Management Agency’s (FEMA) transition to an ‘all-hazards’ approach to disaster management provides a useful example. May et al (2008) show that the centralisation of disaster management under the top levels of the Department of Homeland Security destabilised attention to natural hazards. Instead, the all-hazards approach was synonymous with man-made disaster - terrorism, with dire consequences for the people of New Orleans at the arrival of Hurricane Katrina. This study highlights the consequences of attention shifting and centralisation within organisations. Attention is limited and attending to terrorism necessarily meant the traditional disaster management agenda was destabilised (May et al, 2009).
權力的集中通常並不會伴隨著信息處理或尋求能力的相應擴展。換句話說，影響串行處理的相同挑戰也影響著集中化的權威，其中注意力一次只能集中在少數項目上，而組織傘下的其餘問題則大多未受檢查。聯邦緊急事務管理署（FEMA）向「全災害」災害管理方法的轉變提供了一個有用的例子。May 等人（2008 年）表明，在國土安全部高層集中災害管理破壞了對自然災害的關注。相反，全災害方法與人為災害——恐怖主義——同義，這在颶風卡特里娜到來時對新奧爾良人民造成了嚴重後果。這項研究強調了組織內注意力轉移和集中化的後果。注意力是有限的，關注恐怖主義必然意味著傳統的災害管理議程被破壞（May 等人，2009 年）。

Observers might argue that centralisation allows for responses that demand urgency and speed of action, and that it forces the policy system to focus. This is indeed true, but the urgency and speed of action that centralisation enables is also at the root of maladaptive responses and policy failures when the system fails to incorporate relevant streams of information (for example, natural disasters). Long ago, Heclo (1974) noted that politics is about ‘powering and puzzling’. The vast majority of lessons to be drawn from the study of public policy, both in political science and economics, pertain to Heclo’s ‘powering’ - arriving at a decision. With this chapter, we aim to refocus attention on his ‘puzzling’ which is far less understood, at least in the context of lessons one might draw from policy scholarship.
觀察者可能會認為，集中化允許對需要緊急和快速行動的情況做出反應，並且它迫使政策系統集中注意力。這確實是事實，但集中化所帶來的緊急性和快速行動能力，也是當系統未能納入相關信息流（例如自然災害）時，適應不良反應和政策失敗的根源。很久以前，Heclo（1974）指出，政治是關於「權力與困惑」。從公共政策研究中得出的絕大多數教訓，無論是在政治學還是經濟學中，都與 Heclo 的「權力」相關——即做出決定。通過本章，我們旨在重新聚焦於他的「困惑」，這在從政策學術中可能得出的教訓背景下，遠未得到充分理解。

This research offers a critical lesson for organisations as they attempt to manage their agendas. In general, organisations economise on the scarcity of attention plaguing individual decision-makers. By choosing centralised designs, the attention capacity of organisations is reduced to that of one or a few individuals or offices. While this focus often allows for coordinated and timely action, it should be understood that it will necessarily lead to the loss of attention for other items. Centralised organisations are also stripped of delegation. In the context of a complex, diversified agenda, this means that the organisation is also stripped of much of its analytical capacity. To assume otherwise, is to assume that those who are the beneficiaries of centralisation are experts on the problem, or set of problems, at hand. Under circumstances where this amounts to a ‘heroic assumption’, organisations not only will have difficulty maintaining attention to other important items, but the solutions developed to address problems will likely reflect less expertise than they would under a delegated arrangement.
這項研究為組織在嘗試管理其議程時提供了一個重要的教訓。一般而言，組織會因應個別決策者面臨的注意力稀缺問題而採取節約措施。通過選擇集中化的設計，組織的注意力容量被縮減為一個或少數幾個個人或辦公室的容量。雖然這種集中化通常能夠促成協調一致且及時的行動，但應理解的是，這必然會導致對其他事項的注意力喪失。集中化的組織也失去了委派的機會。在一個複雜且多元化的議程背景下，這意味著組織也失去了其大部分的分析能力。如果假設情況並非如此，那就是假設那些集中化的受益者是當前問題或一系列問題的專家。在這種情況下，如果這被視為一種「英雄式的假設」，組織不僅難以維持對其他重要事項的注意力，而且針對問題開發的解決方案可能反映出的專業知識將比在委派安排下要少。

Social organisations, of which political and policymaking institutions are a variant, face a fundamental tension that no institutional design, in fact no system of government, can alleviate. Centralised organisations trade off expertise and capacity to address more problems, in return for urgency and coordination. Delegated organisations tend towards expertise and increased agenda capacity at the expense of urgency and coordinated responses. This trade-off is key to understanding how organisations process information and how these designs map onto agenda and policy change. As a design feature, it is perhaps the most important element of understanding the likely consequences of organisational information processing.
社會組織，其中政治和決策機構是其變體，面臨著一種根本性的緊張關係，沒有任何制度設計，實際上沒有任何政府體系能夠緩解這種緊張。集中化的組織在解決更多問題的專業知識和能力與緊迫性和協調性之間進行權衡。委託化的組織則傾向於專業知識和增加的議程能力，但以犧牲緊迫性和協調回應為代價。這種權衡是理解組織如何處理信息以及這些設計如何映射到議程和政策變化的關鍵。作為一種設計特徵，它可能是理解組織信息處理可能後果的最重要元素。

There is some evidence from the literature on state budgeting (Breunig and Koski, 2009) that suggests that certain kinds of centralisation of authority can produce more adaptive policymaking structures, however, this is inherently a relativistic impression. In comparison to other institutions - such as citizen legislatures - stronger centralised decisionmaking (for example, strong governors) may contain a greater capacity to address policy problems. Baumgartner et al’s (2017) study of authoritarian regimes and budgeting is suggestive of the more general trend that we argue. Thus, when considering centralisation of decision-making in institutional design, which is often a temptation for many political actors frustrated with intransigence associated with diffuse policymaking, there are caveats.
來自州預算文獻（Breunig 和 Koski，2009 年）的一些證據表明，某種形式的權力集中化可以產生更具適應性的政策制定結構，然而，這本質上是一種相對主義的印象。與其他機構（如公民立法機構）相比，更強的集中決策（例如，強勢的州長）可能具備更大的能力來應對政策問題。Baumgartner 等人（2017 年）對威權體制和預算的研究暗示了我們所主張的更普遍的趨勢。因此，在考慮制度設計中的決策集中化時，這通常是許多政治行為者因分散政策制定所帶來的頑固性而感到沮喪的誘惑，但這其中也有需要注意的地方。

Policy practitioners rarely find themselves able to create new institutions to solve problems or that policy problems fit neatly within an existing boundary of policymaking. For many policies, policymakers rely on several existing institutions. In creating partnerships to achieve policy outcomes (Koski, 2011), policy designers would do well to understand the resources and goals of each participant. Organisations with little goal overlap and few slack resources may only intermittently attend to the problems of the partnership (Koski, 2015), making such groups ripe for dramatic shifts in policy image definitions that reverberate throughout the partnership.
政策實踐者很少能夠創建新機構來解決問題，或者政策問題能夠完美地契合於現有政策制定的邊界內。對於許多政策，政策制定者依賴於幾個現有機構。在建立合作夥伴關係以實現政策成果時（Koski, 2011），政策設計者最好了解每個參與者的資源和目標。目標重疊少且鬆弛資源少的組織可能只會間歇性地關注合作夥伴關係的問題（Koski, 2015），這使得這些團體容易發生政策形象定義的劇烈變化，這些變化會在整個合作夥伴關係中產生迴響。

Multiple-venues are often viewed as advantages from the perspective of policy entrepreneurs, however, policy designs may explicitly create multiple venues to achieve a variety of policy goals. A multi-venue strategy can be part of a general effort to expand the agenda capacity of an existing institution. Additionally, multi- venue strategies serve to offer greater democratic representation; while perhaps not maximising a policy goal, there are other significant benefits associated with
多場地常被政策企業家視為優勢，然而，政策設計可能會明確創造多場地以實現多樣化的政策目標。多場地策略可以是擴大現有機構議程容量的一部分努力。此外，多場地策略提供更大的民主代表性；雖然可能不會最大化政策目標，但還有其他重要的好處。
citizens feeling as though they have multiple opportunities to access the policy process. In particular, multiple venues act as a democratic ‘pressure valve’, assuring that the system can accommodate public demand, while simultaneously slowing reform that may destabilise the system. Additionally, multiple venues may represent specific attempts to reduce the speed with which policy occurs by structuring attention to different details at different levels of salience for similar problems. The policymakers must consider which venues will have authority over their policies in order to predict the types of information that might be brought to bear on policy decisions.
公民感到他們有多種機會參與政策過程。特別是，多個場所充當了民主的「壓力閥」，確保系統能夠滿足公眾需求，同時減緩可能破壞系統穩定的改革。此外，多個場所可能代表著特定的嘗試，通過在不同顯著層面上對類似問題的不同細節進行結構化關注，來降低政策發生的速度。政策制定者必須考慮哪些場所將對其政策擁有權威，以便預測可能影響政策決策的信息類型。

Crucial to understanding how institutions process information is the distinction between the problem and solution space of issues (Workman, 2015). Problem solving is comprised of two processes. First, an individual or organisation must develop a problem representation. We label this ‘problem definition’. Problem definition involves constructing an understanding of a policy problem that comprises some dimensions of the issue for attention and excludes other dimensions. Problem is important because defining the problem for action is necessary for generating solutions, or policies.
理解機構如何處理資訊的關鍵在於區分問題與解決方案空間（Workman, 2015）。問題解決由兩個過程組成。首先，個人或組織必須建立問題的表徵。我們將此稱為「問題定義」。問題定義涉及構建對政策問題的理解，包括問題的某些維度以供關注，並排除其他維度。問題定義之所以重要，因為為行動定義問題對於生成解決方案或政策是必要的。
Problem definition also involves a different set of organisational costs than choice or implementation. Traditionally, practitioners and scholars alike have troubled over, and emphasised, the importance of transaction and decision costs in policymaking (Jones et al, 2003). When the job of the institution is to generate solutions and choose among them, these costs do indeed loom large. Both groups, however, overlook the importance of informational and cognitive costs. Informational costs pertain to the generation of information about a problem. Cognitive costs pertain to understanding the problem or processing the supply of information. At the front end of the policy process, where issues are prioritised and problems defined, it is informational and cognitive costs that are the most potent stressors of institutional design.
問題定義也涉及一組與選擇或實施不同的組織成本。傳統上，從業者和學者都對政策制定中的交易和決策成本的重要性感到困擾並加以強調（Jones 等，2003 年）。當機構的工作是生成解決方案並在其中進行選擇時，這些成本確實顯得非常重要。然而，這兩群體都忽略了信息和認知成本的重要性。信息成本涉及生成有關問題的信息。認知成本涉及理解問題或處理信息供應。在政策過程的前端，即問題被優先排序和定義的地方，信息和認知成本是制度設計中最強大的壓力源。

In combining our understanding of policy entrepreneurs, constituencies and stakeholders with our understanding of the different costs associated with agenda setting and problem definition as opposed to solution generation and choice, a powerful insight for policy system design emerges. As institutions, or their aggregations in policymaking systems, develop issue agendas and craft problem definitions, information search should be
在將我們對政策企業家、選民和利益相關者的理解與我們對議程設定和問題定義相關的不同成本（相對於解決方案生成和選擇）的理解相結合時，一個強大的政策系統設計洞察應運而生。作為機構或其決策系統中的聚合體，在制定議程和精心設計問題定義時，信息搜索應該被
inclusive and expansive. The real danger at the agenda setting and problem definition stage consists in missing an important issue deemed irrelevant by limited search, or excluding a dimension of an issue that is important for solution generation. Moreover, the inclusiveness of information sources in the problem definition stage mitigates against the bias of any single actor providing information, because the decision-making body or institution can triangulate this bias (Workman, 2015). Redundancy of information provision enhances its reliability (Bendor, 1985; Landau, 1969).
包容且廣泛。在議程設定和問題定義階段，真正的危險在於因有限的搜索而錯過被視為無關緊要的重要問題，或排除對解決方案生成至關重要的問題維度。此外，問題定義階段信息來源的包容性有助於減輕任何單一信息提供者的偏見，因為決策機構或組織可以對這種偏見進行三角測量（Workman, 2015）。信息提供的冗餘性增強了其可靠性（Bendor, 1985; Landau, 1969）。

Inclusive and expansive problem search at the agenda setting and problem definition stage also directly addresses the costs most prominent at this stage. Diverse sources of information, especially if independent, greatly increase the quality and quantity of information available to policymakers. Structuring searches to consider alternative definitions and related issues also reduces the risk that policymakers will be inattentive to potentially important elements of a problem. Examples of this type of search process already exists in federal rulemaking, where diverse institutions and groups provide a flow of information about regulatory problems and policies - no doubt search processes at state and local levels would benefit from similar types of approaches.
在議程設定和問題定義階段，包容且廣泛的問題搜尋直接應對了此階段最顯著的成本。多元的信息來源，尤其是獨立的來源，極大地提高了政策制定者可獲取信息的質量和數量。構建搜尋以考慮替代定義和相關問題，也降低了政策制定者對潛在重要問題元素疏忽的風險。這種搜尋過程的範例已經存在於聯邦規則制定中，其中多元的機構和團體提供了關於監管問題和政策的信息流——無疑，州和地方層面的搜尋過程將受益於類似的方法。

Organisationally, the stage of agenda setting and problem definition demands delegation. In piecing together the issues to be addressed on the agenda, or the components of a particular problem, organisations that delegate attention can count on comparative advantage. Organisational units need not address the broader aggregated problem, and instead, focus on producing expertise-laden problem definitions that are aggregated, or blended (see the discussion of issue bundling below), at the apex of the organisation where coordination is paramount.
在組織層面上，議程設定和問題定義的階段需要進行授權。在拼湊議程上需要解決的問題或特定問題的組成部分時，能夠授權注意力的組織可以依賴比較優勢。組織單位不必處理更廣泛的綜合問題，而是專注於產生富含專業知識的問題定義，這些定義在組織頂端進行匯總或融合（見下文關於問題捆綁的討論），在那裡協調至關重要。

As policy systems set an agenda and develop definitions for problems, decision and transaction costs become ever more relevant. In addition, there is a need for coordinated, authoritative decision-making that reaches across organisational units. At the stage where problem definition is settled, organisational strategies that constrain participation will move proposed solutions to implementation in an expeditious manner. Constraining participation and choice at this stage presumes, of course, that relevant information has not been excluded during problem definition and agenda setting. It is here that centralisation shines. Given a diverse set of solutions generated from an expansive understanding of the problem, centralised organisations are better situated to coordinate action and act with urgency. Two related processes ensure that coordination and authoritative action at this broader level is both responsive and effective (see, for instance, Gerring et al, 2009).
隨著政策系統設定議程並為問題制定定義，決策和交易成本變得越來越重要。此外，還需要跨組織單位進行協調、權威的決策。在問題定義確定的階段，限制參與的組織策略將以迅速的方式推動擬議的解決方案進入實施階段。當然，在此階段限制參與和選擇的前提是，在問題定義和議程設定過程中沒有排除相關信息。正是在這一點上，集中化顯示出其優勢。鑑於從對問題的廣泛理解中產生的多樣化解決方案，集中化的組織更適合協調行動並迅速採取行動。兩個相關的過程確保了在這個更廣泛層面上的協調和權威行動既具有響應性又有效（例如，參見 Gerring 等人，2009 年）。

In his study of federal regulations, Workman (2015, 116-23) finds that Congress is able to shape the contours of information flows generated by bureaucracies through agenda setting, in particular, by considering issues in bundles. At the stage of agenda setting and problem definition, policy entrepreneurs, stakeholders, citizens and bureaucracies all want to steer problem definitions in a way that favours their preferred policies. As policymakers consider issues in bundles instead of separately, these actors are forced to adjust the information they supply in an effort remain relevant in the defining of the problem. This is a key function of upper level organisational units, with the ability to shape information supply such that policymaking is amenable to problems such as boundary- and jurisdiction-spanning issues.
在對聯邦法規的研究中，Workman（2015, 116-23）發現，國會能夠通過議程設置，特別是將問題捆綁考慮，來塑造由官僚機構產生的信息流的輪廓。在議程設置和問題定義的階段，政策企業家、利益相關者、公民和官僚機構都希望以有利於他們首選政策的方式引導問題的定義。當政策制定者將問題捆綁考慮而非單獨處理時，這些行為者被迫調整他們提供的信息，以保持在問題定義中的相關性。這是高層組織單位的關鍵功能，具備塑造信息供應的能力，使得政策制定能夠適應跨邊界和跨管轄區的問題。

Consider the problem of climate change. Understood only as an environmental problem, the definition of climate change limits the range of solutions that may be developed. Now, imagine that policymakers understand climate change to be composed of environmental, agricultural, energy and transportation elements. By considering these in a bundle labelled climate change, three things happen. First, policymakers are privy to information generated in each of these streams, meaning crucial elements are less likely to be overlooked. Second, authoritative decision- makers in the policy system are free to combine and customise these elements into a holistic approach to the problem, if desirable. Third, issue bundles foster competition among bureaucracies, policy entrepreneurs, interest groups and citizens to be the most prominent provider of information about the problem. This competition encourages an information supply that is responsive to important, but otherwise ignored, elements of the problem. A prime example is the EPA and the Department of Agriculture (USDA) aligning on many of the proposed solutions to climate change from 2006 through to 2013.
考慮氣候變遷的問題。如果僅將其理解為一個環境問題，氣候變遷的定義就會限制可能開發的解決方案範圍。現在，想像政策制定者將氣候變遷理解為由環境、農業、能源和交通元素組成。通過將這些元素視為氣候變遷的一部分，會發生三件事。首先，政策制定者能夠獲取每個領域產生的信息，這意味著關鍵元素不太可能被忽視。其次，政策體系中的權威決策者可以根據需要自由地將這些元素結合並定制成一個整體的解決方案。第三，問題的組合促進了官僚機構、政策企業家、利益團體和公民之間競爭，成為問題信息的最主要提供者。這種競爭鼓勵了對問題中重要但被忽視的元素作出反應的信息供應。一個典型的例子是美國環境保護署（EPA）和農業部（USDA）在 2006 年至 2013 年間對許多氣候變遷解決方案的一致性。
In practice, issue bundling occurs through agenda setting and delegation and has two virtues. First, any deliberative body, be it a city council or congressional committee, is able to alter the flow of information from actors in an issue area by simply changing the substantive issues considered in a discussion of policy change. Why? Arriving at this conclusion requires only the assumptions that the deliberative body has some policymaking authority, and that the actors in the process beneath those with authority want to influence policy. The second virtue of issue bundling is that it does not require the passage of a policy directive, legislation, ordinance or bill. It occurs through the process of agenda setting, such that the direction of
在實踐中，議題捆綁通過議程設置和授權發生，並具有兩個優點。首先，任何審議機構，無論是市議會還是國會委員會，都能夠通過簡單地改變政策變更討論中考慮的實質性問題來改變來自某一議題領域行動者的信息流。為什麼？得出這一結論僅需假設審議機構具有一定的政策制定權威，且該過程中的行動者希望影響政策。議題捆綁的第二個優點是它不需要通過政策指令、立法、法令或法案的通過。它通過議程設置的過程發生，從而引導
the deliberative body is innate to the process. Put simply, to change the stream of information, the deliberative body only needs to change the way it discusses an issue. Actors in the process will respond with information addressing the current understanding of the problem, most simply, because they want to steer the problem definition, and hence, solutions generated.
審議機構是這一過程的固有部分。簡而言之，要改變信息的流向，審議機構只需要改變其討論問題的方式。過程中的參與者會根據當前對問題的理解提供信息，最簡單的原因是因為他們希望引導問題的定義，從而影響所生成的解決方案。
By bundling issues on the agenda, policymakers are able to shape the overall supply of information about policy problems without choking off the expertise at lower levels of the organisation. In such a process, the expertise and delegation at the problem definition and agenda setting stage works in lockstep with the coordination and urgency elements of centralised units in the system. By bringing actors in the policy system into competition with one another, policymakers also increase the value of signalling and what might be learned from information about a policy problem.
通過將議程上的議題捆綁在一起，政策制定者能夠塑造有關政策問題的整體信息供應，而不會扼殺組織低層級的專業知識。在這樣的過程中，問題定義和議程設定階段的專業知識與授權，與系統中集中單位的協調和緊迫性要素同步運作。通過使政策系統中的參與者相互競爭，政策制定者還提高了信號傳遞的價值，以及從政策問題信息中可能學到的內容。

By fostering problem search that is expansive and inclusive, policymakers retain two sources of information. The first of these is the information generated and provided by actors in the system. The second is the more subtle information gleaned from the messengers themselves. In constructing problem search such that diverse flows of information are generated about problems and how they are to be understood, policymakers are able to triangulate among various definitions of the problem. That is to say, policymakers at the upper reaches of the system can blend various problem definitions to address boundary-spanning, complex problems. They are able to triangulate among definitions imbued with expertise to arrive at an understanding that incorporates important elements.
通過促進廣泛且包容性的問題搜尋，決策者保留了兩種資訊來源。第一種是由系統中的參與者生成和提供的資訊。第二種則是從傳遞者本身獲得的更為微妙的資訊。在構建問題搜尋時，決策者能夠生成關於問題及其理解方式的多樣資訊流，從而能夠在各種問題定義之間進行三角定位。也就是說，位於系統上層的決策者可以融合各種問題定義，以應對跨界的複雜問題。他們能夠在蘊含專業知識的定義之間進行三角定位，從而得出一個包含重要元素的理解。
The more subtle form of information conveyed by search processes constructed in this way pertains to the messengers themselves. When suppliers of information about problems have diverse and divergent preferences and problem definitions, their convergence on a problem definition or a set of solutions is a powerful signal to policymakers at the top of organisations or policy systems. For example, the EPA and USDA’s convergence on tax incentives for reforestation offer a powerful signal about a policy that is likely not only deemed effective, but also politically feasible. This is, however, not how we normally contemplate organisations or policymaking systems. Especially at state and local levels, where resources and power are more unequal, great effort is taken to constrain participation and dampen competition. This inclination robs the system, and especially those at the top, of powerful signals about the likely benefits and feasibility
透過這種方式構建的搜索過程所傳達的更為微妙的信息，與信息傳遞者本身有關。當提供問題信息的供應商具有多樣化且分歧的偏好和問題定義時，他們在問題定義或一組解決方案上的匯聚，對於組織或政策系統頂層的決策者來說是一個強有力的信號。例如，EPA 和 USDA 在重新造林稅收優惠上的共識，提供了一個強有力的信號，表明該政策不僅可能被視為有效，而且在政治上也具有可行性。然而，這並不是我們通常思考組織或政策制定系統的方式。特別是在州和地方層面，資源和權力更加不平等的情況下，人們會付出巨大努力來限制參與和抑制競爭。這種傾向剝奪了系統，尤其是頂層決策者，關於可能收益和可行性的強有力信號。
of various ways of thinking about problems, or solutions to them. This inclination is rampant at all levels of government, but is particularly virulent in local governance systems and within organisations. It severely undermines institutional information processing in these settings.
對於思考問題或解決方案的各種方式的偏好。這種傾向在各級政府中普遍存在，但在地方治理體系和組織內部尤為嚴重。它嚴重削弱了這些環境中的機構信息處理能力。

Finally, we would note that the dimensions of problem solving and decision- making in government cannot be maximised simultaneously. The organisational ramifications of limited attention and institutional information processing represent a tension in governance structures, indeed in all social organisations, that can never be optimised. The balance in any decision context should be reflective of the cautions and suggestions we put forth here. Far from the definitive words on this subject, we fully intend for future scholars to investigate, corroborate and challenge the practical lessons we draw here.
最後，我們要指出的是，政府中的問題解決和決策制定的維度無法同時最大化。有限的注意力和機構信息處理的組織影響代表著治理結構中的一種緊張關係，實際上在所有社會組織中都是如此，這種緊張關係永遠無法被優化。在任何決策情境中的平衡應該反映出我們在此提出的警告和建議。遠非這一主題的最終定論，我們完全期待未來的學者能夠調查、證實並挑戰我們在此得出的實際教訓。

Baumgartner, FR, Jones, BD, 1993, Agendas and Instability in American Politics, Chicago, IL: University of Chicago
Baumgartner, FR, Jones, BD, 1993, 《美國政治中的議程與不穩定性》, 芝加哥, 伊利諾伊州: 芝加哥大學出版社
Baumgartner, FR, Jones, BD, Mortensen, PB, 2014, Punctuated equilibrium theory: Explaining stability and change in public policymaking, Theories of the Policy Process, 59-103
Baumgartner, FR, Jones, BD, Mortensen, PB, 2014, 斷點平衡理論：解釋公共政策制定中的穩定與變化, 政策過程理論, 59-103
Baumgartner, FR, Carammia, M, Epp, DA, Noble, B, Rey, B, Yildirim, TM, 2017, Budgetary change in authoritarian and democratic regimes, Journal of European Public Policy 24, 6, 1-17
Baumgartner, FR, Carammia, M, Epp, DA, Noble, B, Rey, B, Yildirim, TM, 2017, 專制與民主政體下的預算變遷, Journal of European Public Policy 24, 6, 1-17
Bendor, J, 1985, Parallel systems in government, Berkeley, CA: University of California Press
Bendor, J, 1985, 政府中的平行系統, 柏克萊, 加州: 加州大學出版社
Breunig, C, Koski, C, 2009, Punctuated budgets and governors’ institutional powers, American Politics Research 37, 6, 1116-38
Cashore, B, Howlett, M, 2007, Punctuating which equilibrium? Understanding thermostatic policy dynamics in Pacific Northwest forestry, American Journal of Political Science 51, 3, 532-51
Cashore, B, Howlett, M, 2007, 標點哪種均衡？理解太平洋西北林業中的恒溫政策動態, American Journal of Political Science 51, 3, 532-51
Epp, DA, Baumgartner, FR, 2017, Complexity, capacity, and budget punctuations, Policy Studies Journal 45, 2, 247-64
Epp, DA, Baumgartner, FR, 2017, 複雜性、能力與預算間斷, Policy Studies Journal 45, 2, 247-64
Gerring, J, Thacker, SC, Moreno, C, 2009, Are parliamentary systems better? Comparative Political Studies 42, 3, 327-59
Gerring, J, Thacker, SC, Moreno, C, 2009, 議會制是否更優？《比較政治研究》42, 3, 327-59
Hammond, TH, 1986, Agenda control, organizational structure, and bureaucratic politics, American Journal of Political Science 30, 2, 379-420
Hammond, TH, 1986, 議程控制、組織結構與官僚政治, 美國政治科學期刊 30, 2, 379-420
Heclo, H, 1974, Modern social politics in Britain and Sweden: From relief to income maintenance, New Haven, CT: Yale University Press
Heclo, H, 1974, 現代社會政治在英國和瑞典：從救濟到收入維持, 紐黑文, CT: 耶魯大學出版社

Thomas, HF, 2017, Modeling contagion in policy subsystems, Cognitive Systems Research, 44, 74-88, August
Thomas, HF, 2017, 政策子系統中的傳染模型, Cognitive Systems Research, 44, 74-88, 八月
Howlett, M, Cashore, B, 2009, The dependent variable problem in the study of policy change: Understanding policy change as a methodological problem, Journal of Comparative Policy Analysis: Research and Practice 11, 1, 33-46
Howlett, M, Cashore, B, 2009, 政策變遷研究中的依賴變量問題：理解政策變遷作為方法論問題, Journal of Comparative Policy Analysis: Research and Practice 11, 1, 33-46
Howlett, M, Migone, A, 2011, Charles Lindblom is alive and well and living in punctuated equilibrium land, Policy and Society 30, 1, 53-62
Howlett, M, Migone, A, 2011, 查爾斯·林德布洛姆依然健在並生活在間斷平衡之地, Policy and Society 30, 1, 53-62
Jones, BD, 1994, Reconceiving decision-making in democratic politics: Attention, choice, and public policy, Chicago, IL: University of Chicago Press
Jones, BD, 2001, Politics and the architecture of choice: Bounded rationality and governance, Chicago, IL: University of Chicago Press
瓊斯，BD，2001，《政治與選擇的架構：有限理性與治理》，芝加哥，IL：芝加哥大學出版社
Jones, BD, Baumgartner, FR, 2005, The politics of attention: How government prioritizes problems, Chicago, IL: University of Chicago Press
Jones, BD, Baumgartner, FR, 2005, 注意力政治：政府如何優先處理問題, 芝加哥, IL: 芝加哥大學出版社
Jones, BD, Baumgartner, FR, 2015, The politics of information: Problem definition and the course of public policy in America, Chicago, IL: University of Chicago Press
Jones, BD, Baumgartner, FR, 2015, 資訊的政治：問題定義與美國公共政策的進程, Chicago, IL: University of Chicago Press
Jones, BD, 2017, Behavioral rationality as a foundation for public policy studies Cognitive Systems Research, 43, 63-75, June
Jones, BD, 2017, 行為理性作為公共政策研究的基礎，《認知系統研究》，43 卷，63-75 頁，6 月
Jones, BD, Sulkin, T, Larsen, HA, 2003, Punctuations in political institutions, American Political Science Review 97, 1, 1-23
Jones, BD, Sulkin, T, Larsen, HA, 2003, 政治制度中的標點符號, American Political Science Review 97, 1, 1-23
Kingdon, JW, 1995, Agendas, alternatives, and public policies, New York: Harper Collins
Kingdon, JW, 1995, 《議程、替代方案與公共政策》，紐約：哈珀柯林斯
Koski, C, 2011, Committed to protection? Partnerships in critical infrastructure protection, Journal of Homeland Security and Emergency Management 8, 1
Koski, C, 2011, 致力於保護？關鍵基礎設施保護中的夥伴關係, 國土安全與應急管理期刊 8, 1
Koski, C, 2015, Does a partnership need partners? Assessing partnerships for critical infrastructure protection, The American Review of Public Administration 45, 3, 327-42
Koski, C, 2015, 合作關係需要合作夥伴嗎？評估關鍵基礎設施保護的合作關係, The American Review of Public Administration 45, 3, 327-42
Landau, M, 1969, Redundancy, rationality, and the problem of duplication and overlap, Public Administration Review 29, 4, 346-58
Landau, M, 1969, 冗餘、理性以及重複與重疊的問題, 公共行政評論 29, 4, 346-58
Lewallen, Jo, Theriault, SM, Jones, BD, 2016, Congressional dysfunction: An information processing perspective, Regulation $\mathcal{E}$$\mathcal{E}$E\mathcal{E} Governance 10, 2, 179-90
Lewallen, Jo, Theriault, SM, Jones, BD, 2016, 國會功能失調：資訊處理的視角, Regulation $\mathcal{E}$$\mathcal{E}$E\mathcal{E} Governance 10, 2, 179-90
Lindblom, C, 1959, The science of muddling through, Public Administration Review 19, 2, 79-88
May, PJ, 1991, Reconsidering policy design: Policies and publics, Journal of Public Policy 11, 2, 187-206
May, PJ, 1991, 重新考慮政策設計：政策與公眾, Journal of Public Policy 11, 2, 187-206
May, PJ, Workman, S, Jones, BD, 2008, Organizing attention: Responses of the bureaucracy to agenda disruption, Journal of Public Administration Research and Theory 18, 4, 517-41
May, PJ, Workman, S, Jones, BD, 2008, 組織注意力：官僚機構對議程中斷的回應, Journal of Public Administration Research and Theory 18, 4, 517-41

May, P, Sapotichne, J, Workman, S, 2009, Widespread policy disruption: Terrorism, public risks, and homeland security, Policy Studies Journal 37, 2, 171-94
五月, P, Sapotichne, J, Workman, S, 2009, 廣泛的政策中斷：恐怖主義、公共風險與國土安全, 《政策研究期刊》37, 2, 171-94
Mintrom, M, Norman, P, 2009, Policy entrepreneurship and policy change, Policy Studies Journal 37, 4, 649-67
Mintrom, M, Norman, P, 2009, 政策創業與政策變遷, Policy Studies Journal 37, 4, 649-67
Nicholson, J, Orr, K, 2016, Local government partnership working: A space odyssey. Or, journeys through the dilemmas of public and private sector boundary-spanning actors, Policy & Politics 44, 2, 269-87
Nicholson, J, Orr, K, 2016, 地方政府合作工作：一場太空奧德賽。或者，穿越公共和私營部門跨界行動者困境的旅程，Policy & Politics 44, 2, 269-87
Ostrom, V, 2008, The intellectual crisis in American public administration, 3rd edn, Tuscaloosa, AL: University of Alabama Press
Ostrom, V, 2008, 美國公共行政的知識危機, 第三版, Tuscaloosa, AL: 阿拉巴馬大學出版社
Pralle, SB, 2003, Venue shopping, political strategy, and policy change: The internationalization of Canadian forest advocacy, Journal of Public Policy 23, 3, 233-60
Pralle, SB, 2003, 場地選擇、政治策略與政策變更：加拿大森林倡議的國際化, Journal of Public Policy 23, 3, 233-60
Rugeley, CR, Gerlach, JD, 2012, Understanding environmental public opinion by dimension: How heuristic processing mitigates high information costs on complex issues, Politics & Policy 40, 3, 444-70
Rugeley, CR, Gerlach, JD, 2012, 通過維度理解環境公眾意見：啟發式處理如何減輕複雜議題的高信息成本, Politics & Policy 40, 3, 444-70
Shaffer, R, 2017, Cognitive load and issue engagement in congressional discourse, Cognitive Systems Research 44, 89-99, August
Shaffer, R, 2017, 認知負荷與國會話語中的議題參與, Cognitive Systems Research 44, 89-99, 八月
Selznick, P, 1949, TVA and the grass roots: A study of politics and organization, Berkeley, CA: University of California Press
塞爾茲尼克, P, 1949, TVA 與草根: 政治與組織研究, 伯克利, 加州: 加州大學出版社
Stigler, GJ, 1961, The economics of information, Journal of Political Economy 69, 3, 213-25
Stigler, GJ, 1961, 資訊的經濟學, Journal of Political Economy 69, 3, 213-25
Vaughan, SK, Arsneault, S, 2008, Not-for-profit advocacy: Challenging policy images and pursuing policy change, Review of Policy Research 25, 5, 411-28
Vaughan, SK, Arsneault, S, 2008, 非營利倡導：挑戰政策形象與追求政策變革, Review of Policy Research 25, 5, 411-28
Workman, S, 2015, The dynamics of bureaucracy in the US government: How Congress and federal agencies process information and solve problems, New York: Cambridge University Press
Workman, S, 2015, 美國政府中的官僚動態：國會與聯邦機構如何處理信息與解決問題, 紐約: 劍橋大學出版社
Workman, S, Shafran, JS, 2015, Communications frameworks and the supply of information in policy subsystems, in M Howlett, J Hogan (eds) Policy Paradigms in Theory and Practice: Discourses, Ideas, and Anomalies in Public Policy Dynamics, New York: Palgrave MacMillan
Workman, S, Shafran, JS, 2015, 通訊框架與政策子系統中的信息供應, 載於 M Howlett, J Hogan (編) 《政策範式在理論與實踐中的應用：公共政策動態中的話語、思想與異常》, 紐約: 帕爾格雷夫·麥克米倫
Workman, S, Jones, BD, Jochim, AE, 2009, Information processing and policy dynamics, Policy Studies Journal 37, 1, 75-92
Workman, S, Jones, BD, Jochim, AE, 2009, 信息處理與政策動態, Policy Studies Journal 37, 1, 75-92