How Do People Generalize Causal Relations over Objects? A Non-parametric Bayesian Account

How do people decide how general a causal relationship is, in terms of the entities or situations it applies to? What features do people use to decide whether a new situation is governed by a new causal law or an old one? How can people make these difficult judgments in a fast, efficient way? We address these questions in two experiments that ask participants to generalize from one (Experiment 1) or several (Experiment 2) causal interactions between pairs of objects. In each case, participants see an agent object act on a recipient object, causing some changes to the recipient. In line with the human capacity for few-shot concept learning, we find systematic patterns of causal generalizations favoring simpler causal laws that extend over categories of similar objects. In Experiment 1, we find that participants’ inferences are shaped by the order of the generalization questions they are asked. In both experiments, we find an asymmetry in the formation of causal categories: participants preferentially identify causal laws with features of the agent objects rather than recipients. To explain this, we develop a computational model that combines program induction (about the hidden causal laws) with non-parametric category inference (about their domains of influence). We demonstrate that our modeling approach can both explain the order effect in Experiment 1 and the causal asymmetry, and outperforms a naïve Bayesian account while providing a computationally plausible mechanism for real-world causal generalization.

How do people generalize causal relations over objects? A non-parametric Bayesian account

How do people decide how general a causal relationship is, in terms of the entities or situations it applies to? What features do people use to decide whether a new situation is governed by a new causal law or an old one? How can people make these difficult judgments in a fast, efficient way? We address these questions in two experiments that ask participants to generalize from one (Experiment 1) or several (Experiment 2) causal interactions between pairs of objects. In each case, participants see an agent object act on a recipient object, causing some changes to the recipient. In line with the human capacity for few-shot concept learning, we find systematic patterns of causal generalizations favoring simpler causal laws that extend over categories of similar objects. In Experiment 1, we find that participants’ inferences are shaped by the order of the generalization questions they are asked. In both experiments, we find an asymmetry in the formation of causal categories: participants preferentially identify causal laws with features of the agent objects rather than recipients. To explain this, we develop a computational model that combines program induction (about the hidden causal laws) with non-parametric category inference (about their domains of influence). We demonstrate that our modeling approach can both explain the order effect in Experiment 1 and the causal asymmetry, and outperforms a naive Bayesian account while providing a computationally plausible mechanism for real world causal generalization.

From Causal Relations to Causal Constraints

This book explores the notion of causal constraint in science. It examines a family of general constraints encountered in fundamental science, along with the conceptual relations between them: determinism, locality, stability, symmetries, conservation laws, and variation principles. The book shifts the focus away from causal relations between individual events (or properties of events) to the more general causal constraints found in science. This chapter discusses the main contending philosophical accounts of causation and the key issues they raise, including regularity theories (also known as Humean theories), the counterfactual account advanced by David Lewis, process accounts, probabilistic accounts, and interventionist or manipulation accounts. It also provides an overview of determinism, locality, stability, symmetries, conservation laws, and variation principles, along with causal asymmetry and the mutuality of causal relations.

Explaining the Temporal Value Asymmetry

An important feature of life is the past/future value asymmetry. Not to be confused with proximal/distant discounting, the past/future value asymmetry is the fact that we prefer future rather than past preferences be satisfied. Misfortunes are better in the past, where they are “over and done,” than in the future. Some philosophers take this value asymmetry to warrant positing a radical metaphysical asymmetry between the past and future. By contrast, others contend that the value asymmetry is due to the causal asymmetry. Thanks to the causal asymmetry, there is a mechanism between future desires and future fulfilment, but no such mechanism between past desires and past fulfilment. Opponents of this view deride it as a piece of “socio-biological mythology.” Here, appealing to recent work in cognitive science, neuroscience, and evolution, a rich and powerful version of the “causal asymmetry” explanation of the value asymmetry is built.

The Influence of School Leadership on Classroom Participation: Examining Configurations of Organizational Supports

Background In this paper we call for studying school leadership and its relationship to instruction and learning through approaches that highlight the role of configurations of multiple organizational supports. A configuration-focused approach to studying leadership and other essential supports provides a valuable addition to existing tools in school organizational analysis and is particularly useful in examining equifinality and causal asymmetry. Equifinality is the idea that more than one pathway can result in a desired outcome whereas causal asymmetry suggests that the set of conditions that lead to the presence of an outcome need not be the same as the conditions that lead to its absence. Focus of Study This study uses a configurational approach to examine how school leadership and other organizational supports are related to an important aspect of instruction—students’ classroom participation. Research Design We apply fuzzy set qualitative comparative analysis (QCA) to administrative and survey data of high schools from a large urban school district to examine combinations of organizational supports that are associated with classroom participation. Conclusions The study draws attention to the utility of applying configurational approaches to investigate the influence of complex combinations of organizational supports on school outcomes. We compare this approach to more traditional methods that focus on the effects of isolated factors, controlling for each other. Our results show that leadership is associated with students’ classroom participation via multiple configurations of organizational supports. These configurations are different from the set of organizational supports that are related to an absence of classroom participation.

Asymmetry

This chapter shows that the quantitative and qualitative cultures differ on the issue of symmetry. Whereas quantitative research tends to analyze relationships that are symmetric, qualitative research focuses on relationships that have asymmetric qualities. Causal models and explanations can be asymmetric in a variety of ways. This chapter deals mainly (though not exclusively) on the so-called “static causal asymmetry,” in which the explanation of occurrence is not the mirror image of that of nonoccurrence. After comparing symmetric and asymmetric models, the chapter looks at examples of asymmetric explanations using set-theoretic causal models. It highlights the difficulty of translating the fundamental symmetry of standard statistical models into the basic asymmetry of set-theoretic models, as well as the difficulty of capturing the asymmetry of set-theoretic models with the standard symmetric tools of statistics.