Herbert Alexander Simon (June 15, 1916–February 9, 2001) was an American social scientist whose research spanned economics, public administration, psychology, and computer science. He investigated how people and organizations make decisions when information and computational capacity are limited. His account of bounded rationality challenged models that assumed decision-makers could identify and select an optimal alternative. He received the 1978 Nobel Memorial Prize in Economic Sciences for research on decision-making within economic organizations and shared the 1975 Turing Award with Allen Newell. (nobelprize.org)
Education and academic career
Simon was born in Milwaukee, Wisconsin. He entered the University of Chicago in 1933, earned a bachelor's degree in political science in 1936, and completed his doctorate in 1943. His training combined social science with mathematics, symbolic logic, and statistics. His teachers included Henry Schultz, Rudolf Carnap, Harold Lasswell, and Charles Merriam. He sought to make the study of human behavior more empirically rigorous without confining it to a single disciplinary framework. (nobelprize.org)
Early investigations of municipal administration directed his attention toward organizational choice. He led a research group at the University of California, Berkeley, from 1939 to 1942, then taught at Illinois Institute of Technology. During his Chicago years, participation in Cowles Commission seminars strengthened his engagement with mathematical economics. His dissertation became the basis of Administrative Behavior, published in revised form in 1947. His subsequent career at Carnegie Institute of Technology, later Carnegie Mellon University, brought organizational research into sustained contact with computing and psychology. (nobelprize.org)
Organizations and administrative decisions
In *Administrative Behavior*, Simon treated decision-making as the central activity of administration. Rather than explaining organizations solely through formal structures or general management principles, he examined how members select actions. Organizational arrangements affect the information, goals, and alternatives available to individuals, thereby shaping the premises on which their decisions rest. (digitalcollections.library.cmu.edu)
This approach displaced the picture of an organization as a single, fully informed decision-maker. Managers and employees possess different knowledge, responsibilities, and capacities. Rules, communication, authority, and specialization help coordinate their choices, but do not eliminate uncertainty or conflicting objectives. Organizational behavior therefore requires an account of actual decision processes, not merely an assumption that the organization maximizes profit or efficiency. (nobelprize.org)
Simon also contributed to mathematical analysis of causal relations and the identification of economic equation systems. The Nobel award announcement explicitly recognized this work alongside his organizational research. His economics thus included both behavioral explanations of choice and formal investigations of how economic relationships can be represented and analyzed. (digitalcollections.library.cmu.edu)
Bounded rationality and satisficing
Bounded rationality concerns purposeful choice under limits of knowledge, time, attention, and computational ability. In conventional optimizing models, agents are often assumed to know the relevant alternatives and evaluate their consequences consistently. Simon instead asked how alternatives are discovered, how consequences are estimated, and how a decision-maker determines when to stop searching. His 1955 paper “A Behavioral Model of Rational Choice” formalized this alternative approach to decision theory. (digitalcollections.library.cmu.edu)
A central mechanism is satisficing: searching for an alternative that meets an acceptable aspiration level rather than establishing that it is the best available option. The aspiration level can change with experience and the difficulty of finding acceptable outcomes. Satisficing is not simply careless choice; it describes a decision procedure suited to circumstances in which exhaustive comparison is impracticable. (nobelprize.org)
Simon distinguished the evaluation of outcomes from the examination of the procedures producing them. This shifted research toward search, problem representation, and the interaction between cognitive capacities and environments. His work helped establish the intellectual basis of behavioral economics. He also emphasized that bounded rationality becomes explanatory only when supplemented by empirical knowledge of particular decision processes; acknowledging limits alone does not predict behavior. (nobelprize.org)
Artificial intelligence and human problem-solving
Simon was an early contributor to artificial intelligence. With Newell and Cliff Shaw, he developed Logic Theorist in 1956, a program that proved theorems from Principia Mathematica. It demonstrated how symbolic representations and selective heuristic search could support tasks previously associated with human reasoning. (cmu.edu)
Their subsequent General Problem Solver, developed beginning in 1957, employed means–ends analysis. This procedure compares a current situation with a goal, identifies differences, and selects actions intended to reduce them, sometimes creating intermediate subgoals. The program was intended both as an artificial problem-solver and as a model of human cognitive processes. (iiif.library.cmu.edu)
Simon and Newell's *Human Problem Solving* (1972) consolidated their information-processing approach to cognitive science. They compared computational models with verbal records of people thinking aloud while solving problems. Such evidence helped connect the structure of a program's search with observed human strategies, rather than judging models solely by whether they reached a correct answer. (findingaids.library.cmu.edu)
In their 1976 Turing Award lecture publication, they formulated the physical symbol system hypothesis: appropriately organized physical symbol systems possess the necessary and sufficient means for general intelligent action. This became a central proposition of symbolic artificial intelligence. They presented it as an empirical hypothesis open to testing, not as a mathematical theorem establishing how every intelligent system must operate. (elearning.di.unipi.it)
Design and the sciences of the artificial
In *The Sciences of the Artificial* (1969), Simon examined systems shaped by goals and adaptation. He characterized an artifact through the relationship between its internal organization and external environment. Its performance depends on both: an explanation cannot rest entirely on either internal mechanisms or surrounding conditions. (mitpress.mit.edu)
The book extended this perspective to design, including engineering, management, and architecture. Whereas explanation examines existing phenomena, design considers arrangements capable of fulfilling specified purposes. Simon connected this activity with search, representation, and optimization, treating the creation of artifacts and institutions as a subject for systematic investigation. (mitpress.mit.edu)