Two studies investigated how political attitudes affect causal learning. Participants were tasked with testing economicpolicies to maximize the economic output of an imaginary country. Based on their political attitudes, participants wereeither strongly in favor or strongly against the policies (Study 1), or could also have neutral attitudes (Study 2). Somepolicies had fairly clear positive or negative effects. But some were more ambiguous; they initially had positive effects buteventually had negative effects on the economy, or vice versa. After testing the policies, participants falsely believed thatthe policies that fit with their political attitudes were more effective, and this bias was exacerbated for the policies that haddifferent short vs. long-term effects. This research shows the power of motivated reasoning and provides a well-controlledmethod to study the effects of motivated reasoning on causal learning in explore-exploit situations.

The current study seeks to extend research on motivated reasoning by examining how prior beliefs influence the interpretation of objective graphs displaying quantitative information. The day before the 2018 midterm election, conservatives and liberals made judgments about four economic indicators displaying real-world data of the US economy. Half of the participants were placed in an 'alien cover story' condition where prior beliefs were reduced under the guise of evaluating a fictional society. The other half of participants in the 'authentic condition' were aware they were being shown real-world data. Despite being shown identical data, participants in the Authentic condition differed in their judgments of the graphs along party lines. The participants in the Alien condition interpreted the data similarly, regardless of politics. There was no evidence of a „backfire‟ effect, and there was some evidence of belief updating when shown objective data.

Issues of transfer and of human-computer systems are central, but largely separate questions in cognitive science.We take the human-computer system as the unit of analysis and explore how well a human-computer system transfers to tasksoutside the scope of the humans’ training and the software’s design. In two experiments, participants used the procedure au-tomation software (PRIDE) to control simulations of International Space Station habitat systems. Both the software design andthe user training addressed routine procedure execution. In the transfer problems the conditions assumed for routine procedureexecution were not met, requiring novel problem solving. We report on our methods for complex behavior analysis and ourresults showing high though imperfect transfer, noteworthy given the widespread difficulty of transfer. Further investigationof transfer at the level of human-computer system is important for understanding what combinations of technology design anduser experience enable effectively dealing with the unexpected.