Inferring action structure and causal relationships in continuous sequences of human action

In the real world, causal variables do not come pre-identified or occur in isolation, but instead are embedded within a continuous temporal stream of events. A challenge faced by both human learners and machine learning algorithms is identifying subsequences that correspond to the appropriate variables for causal inference. A specific instance of this problem is action segmentation: dividing a sequence of observed behavior into meaningful actions, and determining which of those actions lead to effects in the world. Here we present a Bayesian analysis of how statistical and causal cues to segmentation should optimally be combined, as well as four experiments investigating human action segmentation and causal inference. We find that both people and our model are sensitive to statistical regularities and causal structure in continuous action, and are able to combine these sources of information in order to correctly infer both causal relationships and segmentation boundaries.     

Do the right thing: The assumption of optimality in lay decision theory and causal judgment

Human decision-making is often characterized as irrational and suboptimal. Here we ask whether people nonetheless assume optimal choices from other decision-makers: Are people intuitive classical economists? In seven experiments, we show that an agent’s perceived optimality in choice affects attributions of responsibility and causation for the outcomes of their actions. We use this paradigm to examine several issues in lay decision theory, including how responsibility judgments depend on the efficacy of the agent’s actual and counterfactual choices (Experiments 1–3), individual differences in responsibility assignment strategies (Experiment 4), and how people conceptualize decisions involving trade-offs among multiple goals (Experiments 5–6). We also find similar results using everyday decision problems (Experiment 7). Taken together, these experiments show that attributions of responsibility depend not only on what decision-makers do, but also on the quality of the options they choose not to take.     

Agents and Causes: Dispositional Intuitions As a Guide to Causal Structure

Currently, two frameworks of causal reasoning compete: Whereas dependency theories focus on dependencies between causes and effects, dispositional theories model causation as an interaction between agents and patients endowed with intrinsic dispositions. One important finding providing a bridge between these two frameworks is that failures of causes to generate their effects tend to be differentially attributed to agents and patients regardless of their location on either the cause or the effect side. To model different types of error attribution, we augmented a causal Bayes net model with separate error sources for causes and effects. In several experiments, we tested this new model using the size of Markov violations as the empirical indicator of differential assumptions about the sources of error. As predicted by the model, the size of Markov violations was influenced by the location of the agents and was moderated by the causal structure and the type of causal variables.     

Structural heterogeneity in a binary Pd–Si metallic glass

A Pd₈₁Si₁₉ bulk metallic glassy rod with a diameter of 4.5 mm was produced by water quenching the fluxed alloy. Despite a negative heat of mixing between Pd and Si elements and very simple components constituting the binary Pd–Si glass-forming system, structural heterogeneity was induced either by slow cooling of a liquid or sub-T_g annealing of glassy ribbons. The sub-T_g annealing experiments evidenced that a more ordered amorphous phase emerged from the original glassy matrix. Our work provides an alternative way to tune the microstructure of metallic glasses by subsequent thermal treatment on an as-prepared single glassy phase.     

Causal judgements about two causal candidates: Accounting for occurrences,estimating strength,and the importance of interaction judgements

When two possible causes of an outcome are under consideration, contingency information concerns each possible combination of presence and absence of the two causes with occurrences and nonoccurrences of the outcome. White (2008) proposed that such judgements could be predicted by a weighted averaging model integrating these kinds of contingency information. The weights in the model are derived from the hypothesis that causal judgements seek to meet two main aims, accounting for occurrences of the outcome and estimating the strengths of the causes. Here it is shown that the model can explain many but not all relevant published findings. The remainder can be explained by reasoning about interactions between the two causes, by scenario-specific effects, and by variations in cell weight depending on quantity of available information. An experiment is reported that supports this argument. The review and experimental results support the case for a cognitive model of causal judgement in which different kinds of contingency information are utilised to satisfy particular aims of the judgement process.     

Intuitive understanding of the relation between velocities and masses in simulated collisions

The participants in our experiments were asked to judge whether simulated horizontal collisions appeared to be “natural” or “unnatural”. We manipulated the implied masses and the velocity ratio of two colliding objects. Implied masses were varied through manipulations of the objects' simulated materials (Experiment 1) and sizes (Experiments 2 and 3). For each participant, we determined the interval of velocity ratios that produced “natural” responses most of the time and evaluated how this interval varied as a function of implied masses. The results revealed a fair degree of consistency between predictions of Newtonian mechanics and the participants' responses; this consistency was greater when the implied masses of the colliding objects were varied through manipulations of the simulated materials. Overall, the results support the hypothesis that the cognitive system can integrate multiple sources of stimulus information, enabling individuals to understand multidimensional mechanical events.     

When functions and causes compete

The discounting principle states that ‘the role of a given cause in producing a given effect is discounted if other plausible causes are present’ (Kelley, 1972 Kelley, H. H. 1972. “Attribution theory in social psychology”. In Attribution: Perceiving the causes of behaviour, Edited by: Jones, E., Kanouse, D., Kelley, H., Nisbett, R., Valins, S. and Weiner, B. 1–26. Morristown, NJ: General Learning Press.  [Google Scholar], p. 8). The principle has only been tested with cases where the two explanations are of the same kind (i.e., causal explanations). However, explanations of properties of objects, people, or events often involve functions. Zebras have stripes in order to be better camouflaged. Humans have eyebrows to keep sweat from running into their eyes. Adrenaline is secreted in order to modulate fight and flight responses. Thus, what happens when we are faced with two different kinds of explanation for the same property: one functional and one causal? People evaluated explanations of properties for natural kinds and artefacts. Functional explanations were discounted in favour of causal explanations, however this was only true for properties of artefacts. The presence of an alternative explanation for properties of natural kinds did not affect the plausibility of either kind of explanation.     

Evidence for an illusion of causality when using the Implicit Association Test to measure learning

Our ability to detect causal relations and patterns of covariation is easily biased by a number of well-known factors. For example, people tend to overestimate the strength of the relation between a cue and an outcome if the outcome tends to occur very frequently. During the last years, several accounts have attempted to explain the outcome-density bias. On the one hand, dual-process performance accounts propose that biases are not due to the way associations are encoded, but to the higher-order cognitive processes involved in the retrieval and use of this information. In other words, the outcome-density bias is seen as a performance effect, not a learning effect. From this point of view, it is predicted that the outcome-density bias should be absent in any testing procedure that reduces the motivation or opportunity to engage in higher-order cognitive processes. Contrary to this prediction, but consistent with the most common single-process learning accounts, our results show that the outcome-density effect can be detected when the Implicit Association Test is used to measure the strength of cue–outcome associations.     

Sticking to the Evidence? A Behavioral and Computational Case Study of Micro‐Theory Change in the Domain of Magnetism

Constructing an intuitive theory from data confronts learners with a “chicken‐and‐egg” problem: The laws can only be expressed in terms of the theory's core concepts, but these concepts are only meaningful in terms of the role they play in the theory's laws; how can a learner discover appropriate concepts and laws simultaneously, knowing neither to begin with? We explore how children can solve this chicken‐and‐egg problem in the domain of magnetism, drawing on perspectives from computational modeling and behavioral experiments. We present 4‐ and 5‐year‐olds with two different simplified magnet‐learning tasks. Children appropriately constrain their beliefs to two hypotheses following ambiguous but informative evidence. Following a critical intervention, they learn the correct theory. In the second study, children infer the correct number of categories given no information about the possible causal laws. Children's hypotheses in these tasks are explained as rational inferences within a Bayesian computational framework.