Controlling Uncertainty: Decision Making and Learning In Complex Worlds (original) (raw)

Everyday we are faced with the problem of adaptive control. How can we reliably generate desirable changes in a dynamic environment if we cannot be sure when the observed changes originated from our own actions or occurred independently of them? After nearly fifty years of research in applied, cognitive, and social psychology, and more recently in neuroscience, this review argues that we have reached a point where we can understand the critical characteristics of decision making and learning that enables adaptive control. Essential to this skill is the ability to set appropriate goals, accurately interpret feedback, develop a coherent causal representation, and monitor and evaluate our decisions. Above all, across the disparate domains of research on adaptive control, there is a consensus that our sense of agency appears to be crucial. Maintaining a sense of agency helps to reduce uncertainty stemming from the ambiguity we experience when we fail to locate the source of changing events we experience in a dynamic environment. This book considers the empirical and theoretical approaches to investigating adaptive control, as well as new directions in research. Timely issue: The charted increase in complexity of the environments that we interact with in our lives (e.g., phones, computers, automated driving systems) makes ever more demands on our ability to impose control over them. Adaptive control behaviors are called upon in various dynamic decision making contexts: ecological, economic, industrial, mechanical, medical and organizational. Broadly, these contexts share similar basic characteristics in the way that they induce uncertainty – that is, the decision maker is not always sure that the effects they are observing are directly related to the actions they are generating. In order to find ways of improving our ability to control events in ever increasing uncertain dynamic contexts, the scientific community needs to approach these issues in a unified way which this book aims to do. The objective of the book is to benefit the research community outside of those studying dynamic decision making and problem solving, and bring to light the many issues that adaptive control is connected to.