REFLECTIVE INTELLIGENCE INFORMATION

From the History of Human Thought to Reflective Artificial Intelligence

The history of intelligence has traditionally been understood as the history of human thought. For more than two millennia philosophers, scientists and educators have sought to explain how knowledge is acquired, how judgement develops and how individuals learn from experience. During the twentieth century these questions increasingly intersected with advances in mathematics, engineering and computer science, culminating in the emergence of Artificial Intelligence as a formal academic discipline. Since that time remarkable progress has been achieved in machine learning, natural language processing, robotics and generative modelling, fundamentally transforming the technological landscape of the modern world. Yet despite these extraordinary advances, many of the most significant intellectual challenges associated with Artificial Intelligence remain deeply human. Questions concerning trust, interpretation, ethics, accountability, governance and professional judgement continue to resist purely computational solutions because they concern not only how machines process information but how people understand, evaluate and apply knowledge. It is within this context that Reflective Intelligence has begun to emerge as a broader conceptual framework through which Artificial Intelligence may be understood not simply as an autonomous technology but as an extension of human reflective capability. Rather than measuring intelligence exclusively according to computational performance, Reflective Intelligence emphasises the quality of reasoning, the capacity for continual learning and the integration of technological capability with ethical and organisational responsibility. Its significance therefore lies not in replacing existing theories of Artificial Intelligence but in expanding them to encompass the intellectual processes through which intelligent systems support rather than supplant human cognition.

From Autonomous Performance to Reflective Collaboration

The emergence of Reflective Intelligence reflects a broader transition within contemporary science. During the early decades of Artificial Intelligence, research was dominated by questions concerning whether machines could perform tasks traditionally associated with human intelligence. More recently, however, attention has increasingly shifted towards understanding how intelligent systems should collaborate with people, support organisational learning and contribute to responsible decision-making. This transition mirrors wider developments within systems theory, organisational science and cognitive psychology, all of which increasingly recognise intelligence as an adaptive property emerging through interaction rather than isolation. Reflective Intelligence therefore represents less a discrete technological invention than the convergence of multiple intellectual traditions whose integration has become increasingly necessary as Artificial Intelligence assumes greater influence throughout society.

Philosophy, Reflective Inquiry, Cognitive Science and Cybernetics

Although Reflective Intelligence is a contemporary conceptual framework, its intellectual origins extend deeply into the history of philosophical inquiry. Classical Greek philosophy established many of the ideas that continue to underpin contemporary understandings of reflective reasoning. Aristotle distinguished between theoretical knowledge, technical skill and practical wisdom, arguing that genuine judgement requires more than procedural competence. His concept of phronesis recognised that effective decision-making depends upon contextual understanding, ethical reflection and experience rather than mechanical application of rules. This distinction remains remarkably relevant within modern Artificial Intelligence, where computational optimisation frequently encounters problems requiring moral judgement, uncertainty and contextual interpretation.

Enlightenment Reason and Revisable Knowledge

The Enlightenment subsequently reinforced the importance of rational inquiry through philosophers including René Descartes, Francis Bacon, David Hume and Immanuel Kant, each of whom sought to understand the relationship between observation, evidence and human understanding. Although these thinkers wrote centuries before digital computation, their investigations into reasoning, knowledge and judgement established intellectual foundations upon which later theories of cognition would develop. The scientific revolution further demonstrated that knowledge advances through continual questioning rather than unquestioned certainty, an idea that remains central to Reflective Intelligence because reflective systems continuously reassess evidence, revise understanding and adapt to changing circumstances.

Dewey and Reflective Educational Practice

During the late nineteenth and early twentieth centuries these philosophical traditions increasingly influenced educational theory and psychology. John Dewey's work on reflective thinking represented a particularly significant milestone because it defined reflection as a disciplined process through which individuals actively evaluate evidence before reaching conclusions. Dewey rejected passive learning, arguing instead that genuine intelligence emerges through continual inquiry, experimentation and critical evaluation. His conception of reflective thought subsequently influenced educational practice throughout the twentieth century while providing one of the clearest intellectual antecedents of Reflective Intelligence. Donald Schön later extended these principles through his influential theory of reflective practice, demonstrating that professional expertise develops through continuous reflection both during action and following experience. Professionals become effective not merely because they possess technical knowledge but because they continually reinterpret that knowledge within changing practical situations. Such ideas resonate strongly with contemporary Artificial Intelligence because intelligent systems increasingly operate within environments characterised by uncertainty, adaptation and continual learning.

Cognitive Science and Metacognition

Parallel developments occurred within psychology through the emergence of cognitive science. Researchers investigating perception, memory, problem solving and metacognition increasingly recognised that intelligence extends beyond information processing to include awareness of one's own reasoning processes. Metacognition, frequently described as thinking about thinking, represents one of the most important intellectual influences upon Reflective Intelligence because it introduces the possibility that intelligent systems may not simply generate answers but may also communicate confidence, recognise limitations and recommend further investigation where uncertainty exists. Contemporary developments in explainable Artificial Intelligence, confidence estimation and uncertainty modelling all reflect this broader movement towards systems capable of supporting reflective rather than deterministic decision-making.

Cybernetic Feedback and Self-Regulation

Another decisive influence emerged through cybernetics during the middle of the twentieth century. Norbert Wiener fundamentally transformed scientific understanding by demonstrating that intelligent behaviour depends upon feedback, adaptation and communication. Rather than viewing systems as static mechanisms, cybernetics proposed that effective control requires continual observation of outcomes followed by appropriate adjustment. These principles later influenced control engineering, systems biology, organisational theory and computer science while providing an important conceptual bridge between human learning and computational adaptation. Reflective Intelligence inherits this cybernetic tradition by viewing intelligence as a dynamic process characterised by continual feedback rather than fixed computational procedure.

Converging Foundations of Reflective Intelligence

The convergence of philosophy, psychology and cybernetics therefore established many of the intellectual foundations from which Artificial Intelligence itself would subsequently emerge. Yet these disciplines also preserved an enduring recognition that intelligence cannot be understood exclusively through computation. Reflection, adaptation, ethical reasoning and continual learning remained essential characteristics of intelligent behaviour long before computers existed and it is precisely these characteristics that Reflective Intelligence seeks to integrate with contemporary Artificial Intelligence.

From Symbolic Reasoning to Human-Centred Organisational Intelligence

The formal emergence of Artificial Intelligence as an academic discipline is conventionally dated to the Dartmouth Summer Research Project on Artificial Intelligence, held in Hanover, New Hampshire, during the summer of 1956. Organised principally by John McCarthy, Marvin Minsky, Claude Shannon and Nathaniel Rochester, the conference established the intellectual ambition that machines might be capable of performing functions previously regarded as requiring human intelligence. Although the computational resources available to the participants were extremely limited by contemporary standards, the meeting represented a decisive shift in scientific thought by framing intelligence as a phenomenon that could be investigated through formal models, mathematical reasoning and digital computation. The Dartmouth Conference did not create Artificial Intelligence in isolation, but it provided the conceptual identity around which subsequent decades of research would develop.

Early Optimism in Symbolic Reasoning

The earliest period of Artificial Intelligence research was characterised by considerable optimism. Investigators believed that symbolic reasoning and logical inference would rapidly enable machines to solve many of the intellectual problems traditionally associated with human cognition. Programmes capable of solving mathematical problems, proving logical theorems and playing strategic games appeared to demonstrate that computational intelligence might soon rival human reasoning. Herbert Simon famously predicted during the late 1950s that machines would become capable of performing many forms of professional work within a relatively short period. Although such expectations ultimately proved premature, these early investigations established the theoretical foundations upon which subsequent developments would be constructed.

Symbolic Artificial Intelligence

This first generation of Artificial Intelligence became known as symbolic Artificial Intelligence because knowledge was represented through explicit rules and logical structures. Expert systems, which emerged during the 1970s and early 1980s, represented the practical culmination of this approach. These systems attempted to capture specialist expertise through extensive collections of rules developed in collaboration with human experts. Medical diagnosis, geological exploration and engineering design all became important areas of application. Despite achieving notable successes, expert systems also revealed fundamental limitations. They struggled to adapt to changing environments, required extensive manual maintenance and frequently proved incapable of reasoning beyond their predefined knowledge bases. Intelligence, it became increasingly apparent, involved considerably more than the execution of explicit rules.

Statistical Machine Learning

The limitations of symbolic approaches encouraged increasing interest in statistical methods capable of learning directly from data. Throughout the 1990s machine learning gradually displaced rule-based systems as researchers recognised that many complex problems could be addressed more effectively through pattern recognition than through manually constructed knowledge representation. Statistical learning enabled systems to identify relationships within large datasets without requiring programmers to specify every possible rule explicitly. This transition represented one of the most significant conceptual changes in the history of Artificial Intelligence because it shifted emphasis from programming intelligence towards enabling machines to learn from experience.

Deep Learning and Digital Scale

The beginning of the twenty-first century witnessed an acceleration of this transformation. The exponential growth of digital information, combined with substantial advances in computational processing and data storage, created conditions under which increasingly sophisticated learning algorithms could be developed. Neural networks, which had existed conceptually for several decades, experienced renewed interest as larger datasets and more powerful computing resources enabled substantially improved performance. Geoffrey Hinton, Yoshua Bengio and Yann LeCun demonstrated that deep neural architectures could solve highly complex problems involving speech recognition, computer vision and natural language processing with unprecedented levels of accuracy. Deep learning subsequently became one of the defining technological developments of contemporary Artificial Intelligence.

Transformers and Foundation Models

The emergence of transformer architectures after 2017 fundamentally altered the trajectory of Artificial Intelligence research once again. Large language models demonstrated an extraordinary capacity to generate coherent text, summarise complex documents, translate languages and support increasingly sophisticated forms of human interaction. Generative Artificial Intelligence rapidly expanded beyond language to encompass image generation, software development, scientific research and multimodal reasoning. The public release of accessible conversational systems accelerated global awareness of Artificial Intelligence while simultaneously stimulating unprecedented debate concerning its societal implications.

The Capability–Understanding Paradox

These remarkable achievements nevertheless exposed a paradox that continues to influence contemporary research. Computational capability advanced at extraordinary speed, yet questions concerning interpretation, accountability and human judgement became increasingly prominent. Large language models frequently generated plausible yet inaccurate information, demonstrated inconsistent reasoning and occasionally reflected biases embedded within their training data. Although their computational sophistication continued to improve, organisations increasingly recognised that predictive capability alone could not guarantee trustworthy decision-making. Confidence in Artificial Intelligence therefore became dependent not only upon accuracy but also upon transparency, explainability and responsible governance.

The Emergence of Reflective Intelligence

It is precisely within this historical context that Reflective Intelligence begins to assume greater intellectual significance. Rather than representing a rejection of Artificial Intelligence, Reflective Intelligence may be understood as its philosophical maturation. Earlier generations of research concentrated upon constructing systems capable of performing intelligent tasks. Contemporary scholarship increasingly asks a different question: how should intelligent systems participate within human organisations? This distinction is profound because it shifts attention from computational capability towards the quality of interaction between human cognition and machine intelligence.

Synthesising Symbolic, Statistical and Organisational Learning

Reflective Intelligence therefore represents a continuation of the historical evolution of Artificial Intelligence rather than a competing paradigm. Symbolic reasoning contributed structured knowledge representation; machine learning introduced adaptation through experience; deep learning enabled sophisticated pattern recognition; generative Artificial Intelligence transformed human-computer interaction; Reflective Intelligence extends this progression by emphasising judgement, reflection and responsible collaboration. Each stage builds upon its predecessors while addressing limitations that became increasingly apparent through practical implementation.

Organisations as Intelligent Systems

Another important influence upon the emergence of Reflective Intelligence has been the growing recognition that organisations themselves function as intelligent systems. Organisational theorists increasingly argue that institutions develop collective knowledge through communication, shared experience and continual adaptation. Chris Argyris demonstrated that effective organisations learn not merely by correcting errors but by questioning the assumptions underlying existing practices. Peter Senge similarly proposed that learning organisations continually expand their capacity to understand complex systems through reflection, dialogue and shared vision. These ideas closely parallel developments within Artificial Intelligence because both disciplines increasingly recognise that intelligence emerges through feedback, adaptation and continual learning rather than static optimisation.

Data-Rich Organisations and Fragmented Knowledge

Contemporary digital transformation has reinforced these observations. Organisations now generate unprecedented quantities of operational information through enterprise systems, sensor networks, digital communications and interconnected technologies. The challenge is no longer obtaining information but interpreting it effectively. Reflective Intelligence therefore addresses an increasingly important organisational requirement: enabling intelligent technologies to assist professionals in understanding complexity rather than merely processing data. This objective distinguishes reflective systems from conventional automation by recognising that organisational performance depends fundamentally upon the quality of human judgement supported by computational capability.

Interdisciplinary Research and Organisational Learning

The convergence of Artificial Intelligence with organisational learning has also encouraged renewed interest in interdisciplinary research. Computer scientists increasingly collaborate with psychologists, philosophers, economists, sociologists and legal scholars to address questions extending beyond algorithmic design. Issues concerning fairness, explainability, public trust, governance and ethical accountability require contributions from multiple disciplines because they involve social as well as technical considerations. Reflective Intelligence embodies this interdisciplinary movement by treating intelligence as a socio-technical phenomenon arising through interaction between technology, people and institutions rather than residing exclusively within computational systems.

Four Phases of Artificial Intelligence Development

Consequently, the historical development of Artificial Intelligence may now be viewed as progressing through several broad phases. The first sought to emulate reasoning through symbolic representation; the second enabled learning through statistical methods; the third achieved remarkable advances in perception and language through deep learning and foundation models; the emerging fourth phase increasingly concerns the integration of computational intelligence with reflective human judgement. Reflective Intelligence therefore represents not simply another technological innovation but an intellectual synthesis through which Artificial Intelligence may become more transparent, adaptive, trustworthy and aligned with human values.

Collaborative Rather Than Autonomous Futures

This historical trajectory suggests that the future evolution of Artificial Intelligence will depend less upon constructing increasingly autonomous machines than upon creating increasingly intelligent relationships between computational systems and the human organisations within which they operate. Reflective Intelligence emerges as the natural continuation of this progression because it seeks to ensure that technological capability strengthens rather than diminishes the reflective capacities upon which effective leadership, scientific discovery, public policy and organisational resilience ultimately depend.

Reasoning, Metacognition, Collective Intelligence and Adaptive Governance

The historical development of Artificial Intelligence demonstrates that major advances have rarely resulted solely from improvements in computational performance. Instead, periods of significant progress have emerged when new technological capabilities have converged with broader developments in mathematics, cognitive science, engineering and organisational theory. Reflective Intelligence appears likely to follow a similar pattern. Rather than developing as an isolated scientific discipline, its future trajectory will almost certainly be shaped by continued integration across multiple domains of knowledge. This interdisciplinary character may ultimately become its defining strength, enabling it to provide an intellectual framework capable of reconciling technological innovation with human judgement, ethical responsibility and institutional resilience.

Causal and Contextual Reasoning

One of the most significant future directions concerns the evolution of reasoning within Artificial Intelligence. Contemporary systems have demonstrated extraordinary capability in recognising statistical relationships and generating coherent language; however, genuine reflective reasoning requires considerably more than pattern recognition. It depends upon causal understanding, contextual interpretation, uncertainty management and the capacity to evaluate competing explanations before reaching conclusions. Research into neuro-symbolic architectures, causal inference, probabilistic reasoning and knowledge representation increasingly suggests that future intelligent systems will combine statistical learning with structured reasoning. Such developments are likely to move Artificial Intelligence beyond prediction towards explanation, thereby strengthening its contribution to reflective decision-making.

Metacognitive Capability

Closely associated with this evolution is the growing importance of metacognitive capability. Human expertise is distinguished not merely by the ability to solve problems but by the capacity to recognise the limits of existing knowledge, identify uncertainty and revise previous assumptions when confronted by new evidence. Future developments in Reflective Intelligence may therefore involve computational systems capable of monitoring their own confidence, identifying ambiguous situations and recommending additional investigation where evidence remains incomplete. Rather than presenting deterministic conclusions, reflective systems may increasingly communicate alternative interpretations together with the degree of confidence associated with each possibility. Such capabilities would substantially enhance trust because users would gain a clearer understanding of both the strengths and limitations of computational recommendations.

Collective Human–Machine Intelligence

Another important trajectory concerns the emergence of collective intelligence. Increasingly complex societal challenges seldom fall within the expertise of a single individual or discipline. Climate resilience, healthcare transformation, financial stability, cyber security and national infrastructure each require collaboration among specialists possessing different forms of knowledge. Future Reflective Intelligence systems are therefore likely to facilitate cooperation between multiple human experts and multiple intelligent agents operating within shared analytical environments. Instead of replacing collaborative decision-making, Artificial Intelligence may strengthen it by synthesising diverse evidence, identifying relationships across disciplines and supporting collective reflection upon complex problems. Such developments may fundamentally alter organisational decision-making by creating distributed knowledge ecosystems capable of continual adaptation.

Adaptive Organisational Knowledge

The future of Reflective Intelligence is also likely to be influenced by advances in organisational science. Contemporary institutions increasingly recognise that knowledge constitutes one of their most valuable strategic assets. Yet organisational knowledge often remains fragmented across departments, information systems and individual experience. Reflective Intelligence offers the possibility of integrating these diverse forms of knowledge into coherent organisational memory capable of supporting continuous learning. Intelligent systems may increasingly preserve institutional expertise, identify emerging trends across historical experience and assist organisations in adapting more rapidly to changing external conditions. This capacity for organisational reflection may become an essential characteristic of resilient institutions operating within environments characterised by continual technological, economic and geopolitical change.

Human-Centred Design and Professional Autonomy

Human-centred Artificial Intelligence will almost certainly remain another defining influence upon future development. International research increasingly emphasises that technological progress should strengthen human capability rather than diminish it. Reflective Intelligence aligns naturally with this perspective because it places professional judgement, ethical reasoning and contextual understanding at the centre of intelligent decision-making. Future systems are therefore likely to become increasingly collaborative, functioning less as autonomous machines and more as cognitive partners capable of enhancing creativity, strategic thinking and evidence-based analysis. Such collaboration may prove particularly valuable within healthcare, scientific research, engineering, law, education and public administration, where successful outcomes depend upon nuanced interpretation rather than computational optimisation alone.

Adaptive and Lifecycle Governance

Governance will similarly assume increasing significance. As Artificial Intelligence becomes embedded within critical infrastructure and public services, regulatory frameworks will continue to evolve in response to societal expectations concerning transparency, fairness and accountability. Reflective Intelligence provides an important conceptual contribution because it assumes that governance should not be regarded as an external constraint imposed upon technological innovation but as an intrinsic component of intelligent system design. Future governance models are therefore likely to incorporate continuous auditing, explainability, ethical evaluation and adaptive oversight throughout the operational lifecycle of intelligent systems. Such approaches may ultimately prove more effective than static compliance models because they recognise that Artificial Intelligence evolves continuously through interaction with changing operational environments.

Quantum, Neuromorphic and Embodied Systems

The relationship between Reflective Intelligence and emerging computational technologies also warrants consideration. Quantum computing, neuromorphic engineering, advanced robotics and distributed intelligent networks may substantially expand computational capability during the coming decades. Yet the significance of these developments will depend less upon their raw processing power than upon their integration within reflective organisational processes. Increasing computational sophistication alone cannot guarantee wiser decisions. Indeed, greater complexity may increase the importance of reflective frameworks capable of ensuring that technological capability remains aligned with human objectives and institutional values.

Reflective Education and Professional Learning

Education and professional development are also likely to undergo significant transformation. Traditional educational models have frequently concentrated upon the acquisition of information. Future professional competence may instead depend increasingly upon critical thinking, interdisciplinary reasoning, ethical judgement and the ability to collaborate effectively with intelligent systems. Reflective Intelligence therefore has implications extending beyond technology into the philosophy of education itself. Universities may increasingly emphasise reflective capability alongside technical competence, preparing graduates not simply to use Artificial Intelligence but to evaluate, question and govern its application responsibly.

Augmentation, Inclusion and Social Wellbeing

At the societal level, Reflective Intelligence offers a vision of technological development that differs markedly from narratives emphasising human replacement through automation. Instead, it proposes that the greatest long-term value of Artificial Intelligence lies in strengthening collective human capability. Democratic institutions, scientific communities, healthcare systems and global research networks all depend fundamentally upon informed judgement, open dialogue and continual learning. Reflective Intelligence supports these objectives by encouraging intelligent systems that facilitate understanding rather than merely accelerating information processing.

Redefining Success Beyond Human Imitation

This perspective also has important philosophical implications. Throughout much of the history of Artificial Intelligence, success has frequently been measured according to whether machines could perform tasks previously regarded as uniquely human. Reflective Intelligence suggests an alternative criterion. The future significance of Artificial Intelligence may ultimately depend less upon whether machines become increasingly human-like and more upon whether they enable human beings to become more thoughtful, better informed and more capable of addressing the complex challenges confronting contemporary civilisation. Such a transition represents a profound shift from technological imitation towards intellectual augmentation.

A New Phase of Artificial Intelligence Evolution

Viewed historically, Reflective Intelligence may therefore represent the beginning of a new phase within the broader evolution of Artificial Intelligence. The first phase sought to mechanise calculation; the second formalised symbolic reasoning; the third introduced learning through data; the fourth transformed perception and language through deep learning and foundation models. The emerging phase seeks to integrate these achievements within systems that enhance reflection, responsibility and adaptive organisational intelligence. This progression should not be interpreted as replacing earlier developments but as building upon them, expanding the objectives of Artificial Intelligence from computational performance towards societal value.

Reflective Intelligence as an Evolution in Human–Machine Collaboration

The historical evolution of Reflective Intelligence demonstrates that its origins lie not within a single technological breakthrough but within the gradual convergence of philosophy, cognitive science, cybernetics, organisational learning and Artificial Intelligence. From Aristotle's conception of practical wisdom and John Dewey's theory of reflective inquiry to the cybernetic principles of Norbert Wiener and the organisational learning theories of Chris Argyris and Peter Senge, each intellectual tradition has contributed to an increasingly sophisticated understanding of intelligence as an adaptive, ethical and socially situated phenomenon. Contemporary advances in Artificial Intelligence have accelerated this convergence by exposing the limitations of purely computational models and highlighting the continuing importance of human judgement, contextual understanding and responsible governance.

Human Judgement and Institutional Learning

Reflective Intelligence therefore represents an important conceptual evolution rather than simply another technological innovation. Its central proposition is that Artificial Intelligence achieves its greatest value when designed to augment reflective human capability instead of replacing it. By integrating computational analysis with ethical reasoning, organisational learning and adaptive decision-making, Reflective Intelligence provides a framework capable of supporting more resilient institutions and more informed societies.

Increasing Significance in Critical Institutions

The future trajectories explored throughout this paper suggest that Reflective Intelligence is likely to assume increasing significance as Artificial Intelligence becomes embedded within critical aspects of economic, governmental and social life. Advances in reasoning systems, collective intelligence, adaptive governance and human-centred design will almost certainly reinforce the need for approaches that balance innovation with responsibility. In this respect, Reflective Intelligence offers not merely a description of emerging technological developments but a vision of how those developments might be directed towards enhancing human wisdom, institutional resilience and societal wellbeing.

Ultimately, the enduring contribution of Reflective Intelligence may not be measured by the sophistication of the technologies it inspires, but by the extent to which those technologies enable individuals, organisations and societies to think more carefully, learn more continuously and govern more responsibly in an increasingly complex world.

Bibliography

• Argyris, Chris and Schön, Donald A., Organisational Learning II: Theory, Method and Practice. Reading, Massachusetts: Addison-Wesley, 1996.
• Dewey, John, How We Think. Boston: D.C. Heath and Co., 1910.
• Floridi, Luciano, The Ethics of Artificial Intelligence: Principles, Challenges and Opportunities. Oxford: Oxford University Press, 2022.
• Hinton, Geoffrey E., Deep Learning. Cambridge, Massachusetts: MIT Press, 2016.
• Kahneman, Daniel, Thinking, Fast and Slow. London: Penguin Books, 2012.
• McCarthy, John, 'Programs with Common Sense', Proceedings of the Teddington Conference on the Mechanisation of Thought Processes, 1959.
• Pearl, Judea, The Book of Why: The New Science of Cause and Effect. London: Allen Lane, 2018.
• Russell, Stuart and Norvig, Peter, Artificial Intelligence: A Modern Approach. Fourth Edition. Harlow: Pearson, 2021.
• Schön, Donald A., The Reflective Practitioner: How Professionals Think in Action. Aldershot: Ashgate, 1983.
• Senge, Peter M., The Fifth Discipline: The Art and Practice of the Learning Organisation. London: Random House Business Books, 2006.
• Turing, Alan M., 'Computing Machinery and Intelligence', Mind, Vol. 59, No. 236, 1950.
• Wiener, Norbert, Cybernetics: Or Control and Communication in the Animal and the Machine. Cambridge, Massachusetts: MIT Press, 1948.