Distributed Cognition and Reality

How Pilots and Crews Make Decisions

Katherine L. Plant, Neville A. Stanton

• Publisher: CRC Press
• Year: 2016
• Print ISBN: 9781472482983
• eBook ISBN: 9781317149408
• Edition: 1
• Topic: Technology & Engineering
• Subtopic: Industrial Engineering
• Index: Technology & Engineering

Distributed Cognition and Reality puts theory into practice, as the first book to show how to apply the Perceptual Cycle Model in aviation decision making. Based on case studies, critical incident interviews and live observations in cockpits, the authors develop a new way to understand how pilots and crews make decisions. This book will be useful for practitioners involved in accident and incident investigations and decision-making training, researchers and students within the disciplines of Aviation, Human Factors, Ergonomics, Engineering, Computer Science, and Psychology.

Dr Katherine L Plant is a New Frontiers Fellow in Human Factors Engineering at the University of Southampton in the UK. In 2014 she was awarded the Honourable Company of Air Pilots Prize for Aviation Safety Research.

Professor Neville A Stanton holds the Chair in Human Factors Engineering at the University of Southampton in the UK. In 2007 The Royal Aeronautical Society awarded him the Hodgson Medal for his work on flight-deck safety.

1	Introduction

1.1 BACKGROUND

It has been argued that advances in automation and technology have increased, rather than reduced, cognitive demands on humans (Militello and Hutton, 1998). In the highly automated aviation environment, more procedural and predictable tasks are handled by machines, whilst humans are left responsible for tasks that require diagnoses, judgement and decision-making (Militello and Hutton, 1998). This judgement and decision-making in the handling of emergency situations is usually the deciding factor as to whether an incident turns into an accident (McFadden and Towell, 1999). Decisional errors have consistently been found to account for a high proportion of pilot error (Diehl, 1991; O’Hare et al., 1994; Orasanu and Martin, 1998; Shappell and Wiegmann, 2009). Decisional factors are usually lumped together with other human factors under the umbrella term of ‘human error’, which is consistently implicated as a major contributor to accidents in safety critical systems and is seen by many as the principal threat to flight safety (Harris and Li, 2010).

Accidents often arise from an interaction between technical, systemic and human factors. What stands out, and is intensified by the media, is the human element as there is a desire to ascribe blame (Woods et al., 2010). However, human operators in safety critical systems do not intentionally set out to make mistakes. Aside from the extremely rare incidents of deliberate intent to cause damage and harm, the situation surrounding the human contribution to accidents is much more complicated than it might initially appear. The contemporary perspective of human error does not blame individuals or use the term as a causal attribute. Instead, human error is considered the starting point for any investigation, rejecting the notion of faulty reasoning, and seeks to explore why certain decisions were made over others. Dekker (2006, p. 68) summarised this:

Human error is not an explanation of failure, it demands an explanation.

With that in mind, human error cannot begin to be understood without understanding the situation and precursors to erroneous actions, that is the decision-making processes that underpinned them. Traditional decision-making research has focused on the output product of whether a good or bad decision was made to establish the effectiveness of decision-making (Orasanu and Martin, 1998). However, understanding whether an effective decision-making process was employed, regardless of the observable manifestation of the decision, is arguably more important in order that potential training and mitigation strategies can be proposed. Furthermore, it is only with the benefit of hindsight that a label of ‘bad decision-making’ or ‘human error’ can be prescribed. What should be of interest to researchers and accident investigators is achieving an understanding of why actions and assessments made sense to an operator at the time they were made (i.e. local rationality).

The Federal Aviation Authority (FAA, 1991, p. ii) described aeronautical decision-making (ADM) as

…the mental process used by pilots to consistently determine the best course of action in response to a given set of circumstances. It is what a pilot intends to do based on the latest information he or she has…It is important to understand the factors that cause humans to make decisions and how the decision-making process not only works, but can be improved.

ADM forms the basis of the research presented in this book and is structured around three contexts. First, the primary focus will be from the perspective of aeronautical critical decision-making (ACDM), that is how decisions are made when dealing with critical incidents. Second, the research will focus on incidents as opposed to accidents. Fortunately, aeronautical accidents are rare and their nature means that first-hand accounts are difficult to obtain. It is acknowledged that incidents are the precursors to accidents and occur more often (Weigmann and von Thaden, 2003). Therefore, insights gained into ACDM will have the potential to prevent accidents and will have theoretical or practical relevance for the majority of aviators. Third, the majority of this research will be conducted in the context of rotary wing aviation.

Rotorcrafts, or helicopters, are to aeroplanes what motorcycles are to automobiles; there are fewer of them, but they have disproportionately higher accident rates (Stanton et al., 2015). Estimates suggest that accident rates for helicopters are 10 times higher than for fixed-wing operations (Nascimento et al., 2014). The operational environment for helicopters varies greatly with role, but helicopters generally operate outside of direct air traffic control, at low altitudes and under visual flying conditions (British Helicopter Association, 2014b). These operational advantages mean that helicopters are used in operational contexts that are not suitable for fixed-wing aircraft, including medical rescue over land, search and rescue (SAR) over water or mountains, rapid corporate passenger transfer, oil platform transfer, police search, television broadcasting, facility inspections and firefighting. However, this also means that they are vulnerable to accidents caused by degraded visibility conditions and their low-altitude operational environments (Greiser et al., 2014).

1.2 DISTRIBUTED COGNITION APPROACH

The work presented in this book is grounded in the context of distributed cognition. Stanton (2014) described distributed cognition as the process by which multiple individuals or teams work together in pursuit of a common goal, which comprises multiple interacting subgoals, in which technology often plays a central role to facilitate this. Hutchins’ (1995a,b) view of distributed cognition takes the system as the unit of analysis, rather than the individuals within that system, arguing that the agents and artefacts within a system form a joint cognitive system as cognitive processes are distributed throughout it. As such, Hutchins (2000) proposed that cognitive processes may be distributed in three ways: (1) socially across team members, (2) distributed across internal and external information and (3) distributed through time such that products of earlier events can transform the nature of later events.

Identifying cognition as a systemic endeavour led Stanton and colleagues to develop the idea of distributed situation awareness (DSA); this is founded upon the theoretical concepts of Schema Theory, the perceptual cycle model (PCM) and the distributed cognition approach (Stanton et al., 2006, 2009a; Salmon et al., 2008a). Stanton et al. (2015) described systems as a network of information elements, activated by tasks and belonging to different agents. Within this network, nodes are activated and deactivated as time passes in response to changes in the task, environment and interactions. This approach argues that awareness is distributed across human and technological agents involved in collaborative activity; it does not matter if humans or technology own this information just that the right information is activated and passes to the right agent at the right time. Salmon et al. (2015) argued that it is systems, not individuals, that lose situation awareness, and therefore, these systems should be the focus when attempting to improve performance. The PCM acknowledges the distributed nature of cognition by its emphasis on the internal schemata of the decision maker interacting with the external information in the environment and as such how cognition is distributed between people and the world. Whilst the distributed cognition perspective advocates the whole system as the unit of analysis, the research presented in this book initially focuses on individual pilots, broadening into a team-wide view. These are the units of analysis which are most relevant and accessible for the target audience of this book. However, we consistently encourage the reader to go beyond the individual, by deliberately avoiding terms such as ‘pilot error’, and instead encourage the consideration of the interaction between the person and their work environment.

1.3 AIMS AND OBJECTIVES

This research will explore rotary wing ACDM through Neisser’s (1976) PCM and the associated Schema Theory in order to understand why actions and assessments made sense to the operator at the time they were made. The book aims to address whether the PCM can be used to study the processes of ACDM. This is structured around three key research objectives: first, the book will contribute to the understanding of the role of the perception decision action cycle in ACDM. This objective seeks to explore whether the PCM is a suitable framework to apply to gain a detailed understanding of critical decision-making processes. Within this, the role of Schema Theory, a central tenant of the PCM, will be reviewed in the context of Ergonomics research. Schema Theory has faced criticism, due to its mentalistic nature, that it cannot be considered a true theory. A literature review will establish if Schema Theory can be considered a valid theoretical explanation of behaviour. The construct validity of the PCM as an explanation of behaviour will also be explored. The PCM assumes that information processing occurs in a unidirectional cycle. It will be established whether this is an accurate representation of interaction between schemata, actions and world information. Second, the development of elicitation and analysis methodologies will produce an approach that is capable of eliciting and depicting the manifestation of perceptual cycle processes with qualitative data. The reliability of the approach will also be established. Third, the book will investigate how teams interact in the perceptual-decision-action cycle. The PCM is an individual model of cognition; however, previous research has used it as the theoretical underpinning to explain team processes such as DSA, although a team PCM has not been explored. As such, the final objective will explore team perceptual cycle processes.

1.4 STRUCTURE OF THE BOOK

The book has been constructed so that it can be read from start to finish by readers that are new to the subject area. Each chapter builds on information presented in the one before it. However, the chapters are also designed to be read individually for those that want to dip into specific areas. As such, some of the key theoretical constructs or central arguments are briefly refreshed at the appropriate point in each chapter. The central crux of this work is Neisser’s (1976) PCM, and this is the foundation for the research presented in every chapter. Chapters 2 and 3 provide a theoretical overview of the PCM and the associated Schema Theory; in Chapter 3, this is applied in the context of the Kegworth plane crash. Chapter 4 identifies the critical decision method (CDM) as an interview technique to elicit perceptual cycle information. A coding scheme to thematically analyse the data is also presented via a critical incident case study. Chapter 5 explores the construct validity of the PCM by investigating the flow of information in the model with interviews from 20 helicopter pilots. Chapters 6 and 7 develop methodologies by presenting a more detailed PCM coding scheme for analysing qualitative data and developing an interview schedule specific to the PCM. Chapters 8 and 9 move away from individual pilots and explore the theoretical principles team contexts within the SAR domain. Finally, Chapter 10 summarises the work in light of the book’s objectives and looks at possible lines of further enquiry. An overview of each chapter is provided in slightly more detail below.

Chapter 1: Introduction – This initial chapter introduces the area of ADM, the aims of the research and a summary of each chapter.

Chapter 2: Schema theory: Past, present and future – Schema Theory is intuitively appealing, although it has not always received positive press; critics of the approach argue that the concept is too ambiguous and vague and there are inherent difficulties associated with measuring schemata. Schema Theory and the associated PCM are the theoretical underpinnings of this book. This chapter addresses the criticisms of the theory by demonstrating how Schema Theory has been utilised in Ergonomics research, particularly in the key areas of situation awareness, naturalistic decision-making and error.

Chapter 3: A case study of the Kegworth plane crash: Understanding local rationality with the perceptual cycle model – Introduces the reader to the human error literature. Human error is a significant contributory factor to aviation incidents and accidents, but the most common criticism levelled at human error research is that inadequate causal explanations are provided. This chapter demonstrates how applying the PCM to understand ‘error’ can enable local rationality to be understood (i.e. why actions and assessments made sense at the time they were made). This is exemplified through an analysis of the Kegworth plane crash.

Chapter 4: A pilot study: Using the perceptual cycle model and critical decision method to understand decision-making processes in the cockpit – In this chapter, the CDM is utilised to collect data about an aeronautical critical incident. The data are qualitatively analysed using the principles of thematic analysis based on a coding scheme of the PCM. It is demonstrated that the approach can be used to understand the critical decision-making process and highlights how influential schemata can be at informing decision-making. The reliability of this approach is established, as well as the test–retest reliability of the CDM.

Chapter 5: Examining the validity of Neisser’s perceptual cycle model with accounts from critical decision-making in the cockpit – The PCM assumes information processing occurs in a cyclical manner drawing on top-down and bottom-up influences to produce perceptual exploration and actions. However, the validity of the model has not been addressed. This chapter explores the construct validity of the PCM in the context of ACDM. Data from critical decision-making interviews were used to construct composite PCMs for different phases of dealing with critical incidents. A countercycle was discovered which has been attributed to skill-based behaviour, characteristic of experts. The practical applications and implications for future research are discussed.

Chapter 6: The development of a perceptual cycle classification scheme – In its current form, the PCM only provides a very high level of explanation. In this chapter, data from critical decision-making interviews were used to deconstruct the three high-level categories of the PCM. This resulted in the development of a 28-item taxonomy; the Schema-Action-World (SAW) classification scheme. In doing so, a more detailed description of ACDM was provided by demonstrating the relevance of different concepts in different phases of dealing with a critical incident.

Chapter 7: The schema world action research method (SWARM) for understanding perceptual cycle processes – The SAW taxonomy presented in the previous chapter was used to develop an interview technique that can be utilised to gain a more detailed understanding of the perceptual cycle process. A test–retest study is presented in which the reliability of the interview schedule is established. The method can be used by researchers and practitioners who want to gain a more detailed understanding of perceptual cycle processes and full procedural guidance is provided.

Chapter 8: Team perceptual cycle processes – As described in the previous chapters, the PCM has been successfully applied to explain individual decision-making; however, distributed decision-making in teams is the focus of much research as it is more relevant in understanding complex sociotechnical systems. This chapter explores team perceptual cycle processes in the context of SAR. It was demonstrated that the traditional perceptual cycle representation could not model the interconnectivity of teamwork effectively. As such, a networ...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• List of Figures
• List of Tables
• Preface
• Acknowledgements
• Authors
• List of Abbreviations
• Chapter 1 Introduction
• Chapter 2 Schema Theory: Past, Present and Future
• Chapter 3 Case Study of the Kegworth Plane Crash: Understanding Local Rationality with the Perceptual Cycle Model
• Chapter 4 A Pilot Study: Using the Perceptual Cycle Model and Critical Decision Method to Understand Decision-Making Processes in the Cockpit
• Chapter 5 Examining the Validity of Neisser’s Perceptual Cycle Model with Accounts from Critical Decision-Making in the Cockpit
• Chapter 6 Development of a Perceptual Cycle Classification Scheme
• Chapter 7 Schema World Action Research Method for Understanding Perceptual Cycle Processes
• Chapter 8 Team Perceptual Cycle Processes
• Chapter 9 Exploring Distributed Cognition in Search and Rescue Teams
• Chapter 10 Conclusions
• Appendix: SWARM Prompts and Procedure
• References
• Author Index
• Subject Index