In this paper, we test people’s causal judgments when the graphs have inhibitory causal relations. We find evidence that a particularly important class of errors known as Markov vio- lations extend to these settings. These Markov violations are important because they are incompatible with causal graphical models, a theoretical framework that is often used as a com- putational level account of causal cognition. In contrast, the systematic pattern of errors are in line with the predictions of a recently proposed rational process model that models peo- ple as reasoning about concrete cases (Davis & Rehder, 2020). These findings demonstrate that errors in causal reasoning ex- tend across a range of settings, and do so in line with the pre- dictions of a model that describes the process by which causal judgments are drawn.

Although the causal graphical model framework has achievedsuccess accounting for numerous causal-based judgments, akey property of these models, the Markov condition, is con-sistently violated (Rehder, 2014; Rehder & Davis, 2016). Anew process model—the causal sampler—accounts for theseeffects in a psychologically plausible manner by assumingthat people construct their causal representations using theMetropolis-Hastings sampling algorithm constrained to onlya small number of samples (e.g., < 20). Because it assumesthat Markov violations are built into people’s causal represen-tations, the causal sampler accounts for the fact that those vio-lations manifest themselves in multiple tasks (both causal rea-soning and learning). This prediction was corroborated by anew experiment that directly measured people’s causal repre-sentations.

Recent work has shown that people use temporal informationincluding order, delay, and variability to infer causality be-tween events. In this study we build on this work by investi-gating the role of time in dynamic systems, where causes takecontinuous values and also continually influence their effects.Recent studies of learning in these systems explored short in-teractions in a setting with comparatively rapidly evolving dy-namics and modeled people as relying on simpler, resource-limited strategies to grapple with the stream of information(Davis et al., 2020). A natural question that arises from such anaccount is whether interacting with systems that unfold moreslowly might reduce the systematic errors that result from thesestrategies. Paradoxically, we find that slowing the task indeedreduced the frequency of one type of error, but increased the er-ror rate overall. To capture the differences between conditions,we introduce a novel Causal Event Segmentation model basedon the notion that people compress the continuous scenes intoevents and use these to drive structure inference.

Interventions, time, and continuous-valued variables are all potentially powerful cues to causation. Furthermore, when observed over time, causal processes can contain feedback and oscillatory dynamics that make inference hard. We present a generative model and framework for causal infer- ence over continuous variables in continuous time based on Ornstein-Uhlenbeck processes. Our generative model pro- duces a stochastic sequence of evolving variable values that manifest many dynamical properties depending on the nature of the causal relationships, and a learner’s interventions (man- ual changes to the values of variables during a trial). Our model is also invertible, allowing us to benchmark participant judgments against an optimal model. We find that when in- teracting with systems acting according to this formalism peo- ple directly compare relationships between individual variable pairs rather than considering the full space of possible models, in accordance with a local computations model of causal learn- ing (e.g., Fernbach & Sloman, 2009). The formalism presented here provides researchers in causal cognition with a powerful framework for studying dynamic systems and presents oppor- tunities for other areas in cognitive psychology such as control problems.

Acting effectively in the world requires learning and control- ling dynamic systems, that is, systems involving feedback re- lations among continuous variables that vary in real time. We introduce a novel class of dynamic control environments us- ing Ornstein-Uhlenbeck processes connected in causal Markov graphs that allow us to systematically test people’s ability to learn and control various dynamic systems. We find that per- formance varied across a range of test environments, roughly matching with complexity defined by a set of models trained on the task (an optimal model, a deep Reinforcement Learning agent, and a PID controller). The testbed of dynamic envi- ronments and class of models introduced in this paper lay the groundwork for the systematic study of people’s ability to con- trol complex dynamic systems.

People’s causal judgments exhibit substantial variability, but the processes that lead to this variability are not currently understood. In this paper, we use a repeated-measures design to study the within-participant variability of conditional probability judgments in common-cause networks. We establish that these judgments indeed exhibit substantial within-participant variability. This variability differs by inference type and is related to the extent to which participants commit Markov violations. The consistency and systematicity of this variability suggest that it may be an important source of evidence for the cognitive processes that lead to causal judgments. The systematic study of both within- and between-person variability broadens the scope of behavior that can be studied in causal cognition and promotes the evaluation of formal models of the underlying process. The data and methods provided in this paper provide tools to enable the further study of within-participant variability in causal judgment.

Intervening on causal systems can illuminate their underlyingstructures. Past work has shown that, relative to adults, youngchildren often make intervention decisions that confirm sin-gle hypotheses rather than those that discriminate alternativehypotheses. Here, we investigated how the ability to make in-formative intervention decisions changes across development.Ninety participants between the ages of 7 and 25 completed40 different puzzles in which they had to intervene on vari-ous causal systems to determine their underlying structures.We found that the use of discriminatory strategies increasedthrough adolescence and plateaued into adulthood. Our resultsidentify a clear developmental trend in causal reasoning, andhighlight the need to expand research on causal learning mech-anisms in adolescence.