Causal reasoning through intervention

Causal knowledge enables us to predict future events, to choose the right actions to achieve our goals, and to envision what would have happened if things had been different. Thus, it allows us to reason about observations, interventions, and counterfactual possibilities. Philosophers and computer scientists have begun to unravel the relations among these three kinds of reasoning and their common basis in causality (e.g., Pearl, 2000; Spirtes, Glymour, & Scheines, 1993; Woodward, 2003). Observations can provide some information about the statistical relations among events. According to the principle of common cause (Reichenbach, 1956), there are three possible causal explanations for a reliable statistical relation between two events A and B: A causes B, B causes A, or both events are generated by a third event or set of events, their common cause. For example, dieting and obesity are statistically related because obesity causes people to go on a diet, because dieting disturbs regulatory physiological processes that eventually lead to obesity (many obese people went on a diet before they became extremely overweight), or because obesity and dieting may be causal consequences of our modern eating habits. In this last case, we can say that the correlation between obesity and dieting is spurious. Regardless of the underlying causal structure, an observation of one of these events allows us to infer that other events within the underlying causal model will be present or absent as well. Thus, when we have passively observed an event, we can reason backward diagnostically to infer the causes of this event, or we can reason forward and predict future effects. Moreover, we can infer the presence of spuriously related events. Interventions often enable us to differentiate among the different causal structures that are compatible with an observation. If we manipulate an event A and nothing happens, then A cannot be the cause of event B, but if a manipulation of event B leads to a change in A, then we know that B is a cause of A, although there might be other causes of A as well. Forcing some people to go on a diet can tell us

[1]  O. Penrose The Direction of Time , 1962 .

[2]  W. Talbott,et al.  THE NATURE OF RATIONALITY , 1995 .

[3]  Robert Nozick,et al.  Newcomb’s Problem and Two Principles of Choice , 1969 .

[4]  S. Sloman,et al.  The advantage of timely intervention. , 2004, Journal of experimental psychology. Learning, memory, and cognition.

[5]  David M. Sobel,et al.  A theory of causal learning in children: causal maps and Bayes nets. , 2004, Psychological review.

[6]  Michael R. Waldmann,et al.  Seeing versus doing: two modes of accessing causal knowledge. , 2005, Journal of experimental psychology. Learning, memory, and cognition.

[7]  J. Woodward Making Things Happen: A Theory of Causal Explanation , 2003 .

[8]  Joshua B. Tenenbaum,et al.  Inferring causal networks from observations and interventions , 2003, Cogn. Sci..

[9]  D. Dewsbury,et al.  The Principles of Learning and Behavior. , 1982 .

[10]  M. Domjan The principles of learning and behavior , 1982 .

[11]  Steven A. Sloman,et al.  Do We "do"? , 2005, Cogn. Sci..

[12]  Joshua B Tenenbaum,et al.  Theory-based causal induction. , 2009, Psychological review.

[13]  C. Glymour Learning, prediction and causal Bayes nets , 2003, Trends in Cognitive Sciences.

[14]  Laura Martignon,et al.  Bayesian network models of causal learning , 1998 .

[15]  Tom Burr,et al.  Causation, Prediction, and Search , 2003, Technometrics.

[16]  Michael R. Waldmann,et al.  KNOWLEDGE-BASED CAUSAL INDUCTION , 1996 .

[17]  R. Nozick THE NATURE OF RATIONALITY , 1995 .

[18]  Illtyd Trethowan Causality , 1938 .

[19]  Keith A. Markus,et al.  Making Things Happen: A Theory of Causal Explanation , 2007 .

[20]  Anthony Dickinson,et al.  The 28th Bartlett Memorial Lecture. Causal learning: an associative analysis. , 2001 .

[21]  M. Waldmann,et al.  A Bayesian Network Model of Causal Learning , 1999 .

[22]  C. Glymour The Mind's Arrows: Bayes Nets and Graphical Causal Models in Psychology , 2000 .

[23]  J. I The Design of Experiments , 1936, Nature.