Controlling for alternative causes is essential for learning the strength of any one cause on an effect. Several processes have been proposed for how people control for alternative causes, including probabilistic contrasts within focal sets and associative processes. We investigated another mechanism called the informative transitions heuristic; people selectively attend to temporally adjacent observations (informative transitions; IT) in which the state of the target cause changes but the alternative causes remain the same. Within ITs, whether the effect also changes in the same direction, does not change, or changes in the opposite direction implies that the target cause has a positive, neutral, or negative influence on the effect. Participants judged the strength of the relationship between two drugs and a side effect in a trial-by-trial learning task. Causes with more positive as opposed to neutral ITs were judged to have stronger causal relations, consistent with the IT heuristic.

People reliably make two errors when making inferences aboutthree-variable causal structures: they violate what is known asthe Markov assumption (mediation) on causal chains andcommon cause structures, and fail to sufficiently ‘explainaway’ on common effect structures. Our goal for the presentstudy was to quantitatively compare these two errors aftersubjects have learned the statistical relations between threevariables using procedures designed to maximize the accuracyof their learning and inferences. Aligning with prior research,we found that subjects violated the Markov assumption, anddid not sufficiently explain away. We also found judgmentsabout mediation were worse than judgments about explainingaway for one inference, but better for another, suggesting thatpeople are not uniquely worse at reasoning about one structurethan another. We discuss the results in terms of a theory of cueconsistency.

When learning the relation between a cause and effect, howdo people control for all the other factors that influence thesame effect? Two experiments tested a hypothesis that peoplefocus on events in which the target cause changes and allother factors remain stable. In both four-cause (Experiment 1)and eight-cause (Experiment 2) scenarios, participants learnedcausal relations more accurately when they viewed datasets inwhich only one cause changed at a time. However,participants in the comparison condition, in which multiplecauses changed simultaneously, performed fairly well; inaddition to focusing on events when a single cause changed,they also used events in which multiple causes changed forupdating their beliefs about causal strength. These findingshelp explain how people are able to learn causal relations insituations when there are many alternative factors.