Online research in philosophy

Understanding causal structure is a central task of human cognition. Causal learning underpins the development of our concepts and categories, our intuitive theories, and our capacities for planning, imagination and inference. During the last few years, there has been an interdisciplinary revolution in our understanding of learning and reasoning: Researchers in philosophy, psychology, and computation have discovered new mechanisms for learning the causal structure of the world. This new work provides a rigorous, formal basis for theory theories of concepts and cognitive development, and moreover, the causal learning mechanisms it has uncovered go dramatically beyond the traditional mechanisms of both nativist theories, such as modularity theories, and empiricist ones, such as association or connectionism.

Causal conditional reasoning means reasoning from a conditional statement that refers to causal content. We argue that data from causal conditional reasoning tasks tell us something not only about how people interpret conditionals, but also about how they interpret causal relations. In particular, three basic principles of people's causal understanding emerge from previous studies: the modal principle, the exhaustive principle, and the equivalence principle. Restricted to the four classic conditional inferences—Modus Ponens, Modus Tollens, Denial of the Antecedent, and Affirmation of the Consequent—causal conditional reasoning data are only partially able to support these principles. We present three experiments that use concrete and abstract causal scenarios and combine inference tasks with a new type of task in which people reformulate a given causal situation. The results provide evidence for the proposed representational principles. Implications for theories of the na ve understanding of causality are discussed.

We further develop the mathematical theory of causal interventions, extending earlier results of Korb, Twardy, Handfield, & Oppy, (2005) and Spirtes, Glymour, Scheines (2000). Some of the skepticism surrounding causal discovery has concerned the fact that using only observational data can radically underdetermine the best explanatory causal model, with the true causal model appearing inferior to a simpler, faithful model (cf. Cartwright, (2001). Our results show that experimental data, together with some plausible assumptions, can reduce the space of viable explanatory causal models to one.

Causal structure learning algorithms have focused on learning in ”batch-mode”: i.e., when a full dataset is presented. In many domains, however, it is important to learn in an online fashion from sequential or ordered data, whether because of memory storage constraints or because of potential changes in the underlying causal structure over the course of learning. In this paper, we present TDSL, a novel causal structure learning algorithm that processes data sequentially. This algorithm can track changes in the generating causal structure or parameters, and requires significantly less memory in realistic settings. We show by simulation that the algorithm performs comparably to batch-mode learning when the causal structure is stationary, and significantly better in non-stationary environments.

Learning to understand a single causal system can be an achievement, but humans must learn about multiple causal systems over the course of a lifetime. We present a hierarchical Bayesian framework that helps to explain how learning about several causal systems can accelerate learning about systems that are subsequently encountered. Given experience with a set of objects, our framework learns a causal model for each object and a causal schema that captures commonalities among these causal models. The schema organizes the objects into categories and specifies the causal powers and characteristic features of these categories and the characteristic causal interactions between categories. A schema of this kind allows causal models for subsequent objects to be rapidly learned, and we explore this accelerated learning in four experiments. Our results confirm that humans learn rapidly about the causal powers of novel objects, and we show that our framework accounts better for our data than alternative models of causal learning.

This paper develops the view that in arguing informally individuals construct a dual representation in which there is a coupling of arguments and the structure of the qualitative (mental) causal model to which these refer. Invited to consider a future possibility, individuals generate a causal model and mentally simulate the consequences of certain actions. Their arguments refer to the causal paths in the model. Correspondingly, faced with specific arguments about a policy option they generate a model with particular causal paths and mentally simulate the outcomes. The results of Experiment 1 are consistent with this notion. Decisions on the percentage of funds to be allocated to genetically modified (GM) crop research depended on the structure of the arguments elicited in response to imagining a future state of affairs. Specifically, the presence of a dissuasive argument eliminated the impact of any persuasive argument. The non-monotonic properties of everyday informal argument can then be seen as a corollary of change to causal structure in the model. The dual representation view predicts that the impact of a dissuasive argument will depend on the structure of the causal model. Experiment 2 tested and confirmed this prediction by requiring individuals to judge the relative persuasiveness of two cases referring either to a model with two independent causal paths or to a model in which one causal path depended on the other. In contrast to Experiment 1, prior opinion on GM crop research did not affect allocation decisions. An advisory role in contrast to a participant role may encourage a more decontextualised mode of thinking. According to the dual representation view, ease of mental simulation should exert wide-ranging effects on judgements and the rhetoric of arguments should also be important. The paper concludes with a discussion of some of these expectations.

The application of the formal framework of causal Bayesian Networks to children’s causal learning provides the motivation to examine the link between judgments about the causal structure of a system, and the ability to make inferences about interventions on components of the system. Three experiments examined whether children are able to make correct inferences about interventions on different causal structures. The first two experiments examined whether children’s causal structure and intervention judgments were consistent with one another. In Experiment 1, children aged between 4 and 8 years made causal structure judgments on a three-component causal system followed by counterfactual intervention judgments. In Experiment 2, children’s causal structure judgments were followed by intervention judgments phrased as future hypotheticals. In Experiment 3, we explicitly told children what the correct causal structure was and asked them to make intervention judgments. The results of the three experiments suggest that the representations that support causal structure judgments do not easily support simple judgments about interventions in children. We discuss our findings in light of strong interventionist claims that the two types of judgments should be closely linked.

"Blocking" refers to judgments of a moderate contingency being lowered when contrasted with a strong contingency. The Rescorla-Wagner model and causal model theory account for blocking through different mechanisms. To examine the predictions from these two models, seven experiments tested the extent to which "causal scenario" and "causal order" would influence whether blocking was observed in human contingency learning tasks. "Causal scenario" was manipulated by contrasting responses to two causes of one effect or to one cause of two effects; "causal order" was defined as causes preceding effects or effects preceding causes. The four conjunctions of these two factors were investigated separately in Experiments 1 to 5. In Experiments 1 and 2, two causes preceded one effect and two effects preceded one cause, respectively. Blocking was observed regardless of whether the predictors were causes or effects. In Experiments 3, 4 and 5, participants were presented with one antecedent cue and made separate predictions about each of the trial's two outcomes. Blocking was not observed, irrespective of whether the antecedent cue was a cause or an effect. These initial results were consistent with the Rescorla-Wagner model. An alternative explanation was that blocking failed to occur in Experiments 3 to 5 because participants were asked questions between the predictor and two outcomes. Predicting the outcomes might have implicitly led participants to monitor them separately and to report on subsets of the data at the time of judgment. To address this issue, the volunteers in Experiment 6 observed the events on each trial but did not make any predictions about the outcomes. Blocking was observed, signifying that the intervening questions between the antecedent and consequent cues constitute an important variable influencing cue competition effects. In Experiment 7, all four conjunctions of causal scenario and causal order were tested simultaneously. Furthermore, participants w.

There is now substantial agreement about the representational component of a normative theory of causal reasoning: Causal Bayes Nets. There is less agreement about a normative theory of causal discovery from data, either computationally or cognitively, and almost no work investigating how teaching the Causal Bayes Nets representational apparatus might help individuals faced with a causal learning task. Psychologists working to describe how naïve participants represent and learn causal structure from data have focused primarily on learning from single trials under a variety of conditions. In contrast, one component of the normative theory focuses on learning from a sample drawn from a population under some experimental or observational study regime. Through a virtual Causality Lab that embodies the normative theory of causal reasoning and which allows us to record student behavior, we have begun to systematically explore how best to teach the normative theory. In this paper we explain the overall project and report on pilot studies which suggest that students can quickly be taught to (appear to) be quite rational.

The dynamics model, which is based on Talmy’s (1988) theory of force dynamics, characterizes causation as a pattern of forces and a position vector. In contrast to counterfactual and probabilistic models, the dynamics model naturally distinguishes between different cause-related concepts and explains the induction of causal relationships from single observations. Support for the model is provided in experiments in which participants categorized 3D animations of realistically rendered objects with trajectories that were wholly determined by the force vectors entered into a physics simulator. Experiments 1-3 showed that causal judgments are based on several forces, not just one. Experiment 4 demonstrated that people compute the resultant of forces using a qualitative decision rule. Experiments 5 and 6 showed that a dynamics approach extends to the representation of social causation. Implications for the relationship between causation and time are discussed.