Many domains involve gathering evidence, from forensic investigations and medical diagnosis, to everyday life. Howshould one order this collection, given the costs involved (e.g. time, financial, information)? Scheduling theory offersoptimal solutions, but requires clear metrics. Evidence can have many influences on it, which affect prioritization, e.g.degradation, contamination, etc. However, to date there has been no clear way to bring this into a unified metric, andthus optimal scheduling has remained out of reach. We propose a new information-based measure, KL, as a way ofencapsulating these information costs, and present maximum KL preservation as a clear rule & metric for scheduling. Wego on to test several heuristic rules for scheduling evidence collection, based on optimally derived algorithms, providingnovel formal backing for a dominant heuristic strategy for scheduling information gathering.

Despite the increase in studies investigating people’s explana- tory preferences in the domains of psychology and philoso- phy, little is known about their preferences in more applied do- mains, such as the criminal justice system. We show that when people evaluate competing legal accounts of the same evidence that vary in complexity, their explanatory preferences are af- fected by: i) whether they are required to draw causal mod- els of the evidence, and ii) the actual structure that is drawn. Although previous research has shown that people can reason correctly about causality, ours is one of the first studies that shows that generating and drawing causal models directly af- fects people’s evaluations of explanations.

Recent research suggests that people do not perform wellon some of the most crucial components of causal reason-ing: probabilistic independence, diagnostic reasoning, and ex-plaining away. Despite this, it remains unclear what con-texts would affect people’s reasoning in these domains. Inthe present study we investigated the influence of manipulatingpriors of causes and structural complexity of Causal BayesianNetworks (CBNs) on the above components. Overall we foundthat participants largely accepted the priors and understoodprobabilistic independence, but engaged in inaccurate diagnos-tic reasoning and insufficient explaining away behavior. More-over, the effect of manipulating priors on participants’ perfor-mance in diagnostic reasoning and explaining away was sig-nificantly larger in a structurally less complex CBN than in astructurally more complex CBN.

In order to make accurate inferences and judgments, one needs to not only be able to aptly evaluate and integrate informa-tion, but be able to seek and acquire the right information in the first place. The present work explored human informationacquisition and evaluation in a novel probability context and utilising a more naturalistic criminal investigation scenario.Focus was placed on exploring the relationship between searching for information, evaluating it and integrating it withinones belief model in order to make a causal judgement. Results indicated that although participants search choices ap-proximated those of informed Bayesian OED models, belief updating accuracy systematically decreased throughout thetask. Findings suggested a dichotomy between information evaluation and belief integration, questioning the descriptiveabilities of OED principles to account for these processes. The implications of these finding in relation to the psychologicalliterature of human inquiry are discussed.

Recent research (Pilditch, Fenton, & Lagnado, 2019) shows that people are susceptible to zero-sum thinking in evidence evaluation, where they dismiss or underweight the probative value of evidence that is equally predicted by multiple independent hypotheses. But such an assumption is only valid when explanations are mutually exclusive and exhaustive. The present work extends these findings by looking at the context of information selection, and the decisional consequences of the zero-sum fallacy. It uses an information metric to quantify the cost of the error in terms of overlooked information.

What makes an explanation satisfying? Much work hasinvestigated explanatory preferences for things like animalsand artifacts, but how do explanation preferences manifest ineveryday life? Here, we focus on the criminal justice system asa case study. In this domain, outcomes critically depend on howmembers of the system (e.g., lawyers, jurors) generate andinterpret explanations. We investigate lay preferences for twodifferent classes of explanations: those that appeal to‘mechanistic’ aspects of a crime (i.e., how the culpritcommitted the crime) vs. ‘teleological’ aspects of that crime(i.e., the purpose of the crime). In two experiments, wedemonstrate that people have a systematic preference for'motive' accounts of crimes (analogous to a teleologypreference) at different stages of the investigative process. Wediscuss these findings in light of a broad literature on thecognitive basis of explanation preferences. We also discussimplications for the criminal justice system.