David Lagnado

University College London | UCL · Department of Experimental Psychology

Activities of Daily Living (ADLs) are often disrupted in patients suffering from dementia due to a well-known taxonomy of errors. Wearable technologies have increasingly been used to monitor, diagnose, and assist these patients. The present paper argues that the benefits current and future wearable devices provide to dementia patients could be enha...

Mechanisms play a central role in how we think about causality, yet not all causal explanations describe mechanisms. Across five experiments, we find that people evaluate explanations differently depending on whether or not they include mechanisms. Despite common wisdom suggesting that explanations ought to be simple in the sense of appealing to as...

We present three experiments using a novel problem in which participants update their estimates of propensities when faced with an uncertain new instance. We examine this using two different causal structures (common cause/common effect) and two different scenarios (agent-based/mechanical). In the first, participants must update their estimate of t...

How critical are individual members perceived to be for their group's performance? In this paper, we show that judgments of criticality are intimately linked to considering responsibility. Prospective responsibility attributions in groups are relevant across many domains and situations, and have the potential to influence motivation, performance, a...

Do people hold robots responsible for their actions? While Clark and Fischer present a useful framework for interpreting social robots, we argue that they fail to account for people's willingness to assign responsibility to robots in certain contexts, such as when a robot performs actions not predictable by its user or programmer.

Despite the increase in studies investigating people's explanatory preferences in the domains of psychology and philosophy, little is known about their preferences in more applied domains, such as the criminal justice system. We show that when people evaluate competing legal accounts of the same evidence, their explanatory preferences are affected...

Over-flexibility in the definition of Friston blankets obscures a key distinction between observational and interventional inference. The latter requires cognizers form not just a causal representation of the world but also of their own boundary and relationship with it, in order to diagnose the consequences of their actions. We suggest this locate...

We study statistical aspects of the case of the British nurse Ben Geen, convicted of 2 counts of murder and 15 of grievous bodily harm following events at Horton General Hospital (in the town of Banbury, Oxfordshire, UK) during December 2013–February 2014. We draw attention to parallels with the cases of nurses Lucia de Berk (the Netherlands) and D...

This article seeks to propose a framework and corresponding paradigm for evaluating explanations provided by explainable artificial intelligence (XAI). The article argues for the need for evaluation paradigms – different people performing different tasks in different contexts will react differently to different explanations. It reviews previous res...

The goal of perception is to infer the most plausible source of sensory stimulation. Unisensory perception of temporal order, however, appears to require no inference, because the order of events can be uniquely determined from the order in which sensory signals arrive. Here, we demonstrate a novel perceptual illusion that casts doubt on this intui...

Much work has investigated explanatory preferences for things like animals and artifacts, but how do explanation preferences manifest in everyday life? Here, we focus on the criminal justice system as a case study. In this domain, outcomes critically depend on how actors in the system (e.g., lawyers, jurors) generate and interpret explanations. We...

Did Tom's use of nuts in the dish cause Billy's allergic reaction? According to counterfactual theories of causation, an agent is judged a cause to the extent that their action made a difference to the outcome (Gerstenberg, Goodman, Lagnado, & Tenenbaum, 2020; Gerstenberg, Halpern, & Tenen-baum, 2015; Halpern, 2016; Hitchcock & Knobe, 2009). In thi...

How do we make sense of complex evidence? What are the cognitive principles that allow detectives to solve crimes, and lay people to puzzle out everyday problems? To address these questions, David Lagnado presents a novel perspective on human reasoning. At heart, we are causal thinkers driven to explain the myriad ways in which people behave and in...

How do we make sense of complex evidence? What are the cognitive principles that allow detectives to solve crimes, and lay people to puzzle out everyday problems? To address these questions, David Lagnado presents a novel perspective on human reasoning. At heart, we are causal thinkers driven to explain the myriad ways in which people behave and in...

In Chapter 2 I argue that we reason by building and manipulating mental models. These causal models allow us to predict, control and explain the world around us. They are idealized and schematic, which reduces complexity but can lead to reasoning biases. I also present an account of how we generate these models, by using intuitive theories of how t...

How do we make sense of complex evidence? What are the cognitive principles that allow detectives to solve crimes, and lay people to puzzle out everyday problems? To address these questions, David Lagnado presents a novel perspective on human reasoning. At heart, we are causal thinkers driven to explain the myriad ways in which people behave and in...

How do we make sense of complex evidence? What are the cognitive principles that allow detectives to solve crimes, and lay people to puzzle out everyday problems? To address these questions, David Lagnado presents a novel perspective on human reasoning. At heart, we are causal thinkers driven to explain the myriad ways in which people behave and in...

Chapter 1 introduces the main questions to be tackled in the book, tracing a crime case from the discovery of a body to the verdict of a court. How do we generate initial hypotheses from sparse information? How do we develop our hypotheses as we gather new evidence? How do we make sense of a large body of conflicting evidence to reach a final decis...

Chapter 5 examines expert reasoning, with a focus on detectives solving murder cases. I introduce the data-frame theory of sensemaking, which argues that experts and novices share the same modes of reasoning, both relying heavily on causal models and simulation, but with experts using richer models informed by their experience and training. We also...

How do we make sense of complex evidence? What are the cognitive principles that allow detectives to solve crimes, and lay people to puzzle out everyday problems? To address these questions, David Lagnado presents a novel perspective on human reasoning. At heart, we are causal thinkers driven to explain the myriad ways in which people behave and in...

How do we make sense of complex evidence? What are the cognitive principles that allow detectives to solve crimes, and lay people to puzzle out everyday problems? To address these questions, David Lagnado presents a novel perspective on human reasoning. At heart, we are causal thinkers driven to explain the myriad ways in which people behave and in...

How do we make sense of complex evidence? What are the cognitive principles that allow detectives to solve crimes, and lay people to puzzle out everyday problems? To address these questions, David Lagnado presents a novel perspective on human reasoning. At heart, we are causal thinkers driven to explain the myriad ways in which people behave and in...

In Chapter 8 I look at the strategies that people use to gather and interpret evidence, focusing on the classic confirmation bias. I argue that this ‘bias’ covers various different strategies, some of which are reasonable, whereas others are genuine biases. One danger is when investigators misinterpret evidence and these errors cascade undetected t...

Chapter 6 examines the legal concept of evidence and the key dimensions of relevance, strength and reliability. I argue that a purely probabilistic conception of evidence is insufficient and present a causal perspective on evidence. The causal approach provides a unified framework to capture the interrelations between hypotheses and evidence and he...

How do we make sense of complex evidence? What are the cognitive principles that allow detectives to solve crimes, and lay people to puzzle out everyday problems? To address these questions, David Lagnado presents a novel perspective on human reasoning. At heart, we are causal thinkers driven to explain the myriad ways in which people behave and in...

While the laws of probability are rarely disputed, the question of how we should interpret probability judgments is less straightforward. Broadly, there are two ways to conceive of probability—either as an objective feature of the world, or as a subjective measure of our uncertainty. Both notions have their place in science, but it is the latter su...

A prominent finding in causal cognition research is people’s tendency to attribute increased causality to atypical actions. If two agents jointly cause an outcome (conjunctive causation), but differ in how frequently they have performed the causal action before, people judge the atypically acting agent to have caused the outcome to a greater extent...

Epistemic states, what an agent knows or beliefs, play a crucial role in people's moral evaluations of the agent's actions. Whether and to what extent epistemic states also influence an agent's perceived causal contribution to an outcome remains the subject of debate. In three experiments, we investigate people's causal and counterfactual judgments...

The need to update our estimates of probabilities (e.g., the accuracy of a test) given new information is commonplace. Ideally, a new instance (e.g., a correct report) would just be added to the tally, but we are often uncertain whether a new instance has occurred. We present an experiment where participants receive conflicting reports from two ear...

How do people make causal judgments about physical events? We introduce the counterfactual simulation model (CSM) which predicts causal judgments in physical settings by comparing what actually happened with what would have happened in relevant counterfactual situations. The CSM postulates different aspects of causation that capture the extent to w...

In many complex, real‐world situations, problem solving and decision making require effective reasoning about causation and uncertainty. However, human reasoning in these cases is prone to confusion and error. Bayesian networks (BNs) are an artificial intelligence technology that models uncertain situations, supporting better probabilistic and caus...

A significant challenge for recommender systems (RSs), and in fact for AI systems in general, is the systematic definition of explanations for outputs in such a way that both the explanations and the systems themselves are able to adapt to their human users' needs. In this paper we propose an RS hosting a vast repertoire of explanations, which are...

Automated Vehicles (AVs) have made huge strides towards large scale deployment. Despite this progress, AVs continue to make mistakes, some resulting in death. While some mistakes are avoidable, others are hard to avoid even by highly-skilled drivers. As these mistakes continue to shape attitudes towards AVs, we need to understand whether people dif...

In criminal trials, evidence often involves a degree of uncertainty and decision-making includes moving from the initial presumption of innocence to inference about guilt based on that evidence. The jurors’ ability to combine evidence and make accurate intuitive probabilistic judgments underpins this process. Previous research has shown that errors...

Selon le story-model , théorie psychologique du raisonnement et de la prise de décision des jurés proposée par Nancy Pennington et Reid Hastie dans les années 1980, les jurés d’un procès criminel interprètent et évaluent les preuves qui leur sont présentées au moyen d’une représentation mentale des événements – un récit –, plutôt que par un calcul...

Despite the increase in studies investigating people’s explanatory preferences in the domains of psychology and philosophy, little is known about their preferences in more applied domains, 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 p...

The need to update our estimates of probabilities (e.g., the accuracy of a test) given new information is commonplace. Ideally, a new instance (e.g., a correct report) would just be added to the tally, but we are often uncertain whether a new instance has occurred. We present an experiment where participants receive conflicting reports from two ear...

We discuss statistical issues in cases of serial killer nurses, focussing on the Dutch case of the nurse Lucia de Berk, arrested under suspicion of murder in 2001, convicted to life imprisonment, but declared innocent in 2010; and the case of the English nurse Ben Geen, arrested in 2004, also given a life sentence. At the trial of Ben Geen, a stati...

Whether assessing the accuracy of expert forecasting, the pros and cons of group communication, or the value of evidence in diagnostic or predictive reasoning, dependencies between experts, group members, or evidence have traditionally been seen as a form of redundancy. We demonstrate that this conception of dependence conflates the structure of a...

Within the domain of psychology, Optimal Experimental Design (OED) principles have been used to model how people seek and evaluate information. Despite proving valuable as computational-level methods to account for people's behaviour, their descriptive and explanatory powers remain largely unexplored. In a series of experiments, we used a naturalis...

The study of people's ability to engage in causal probabilistic reasoning has typically used fixed-point estimates for key figures. For example, in the classic taxi-cab problem, where a witness provides evidence on which of two cab companies (the more common 'green' / less common 'blue') were responsible for a hit and run incident, solvers are told...

In reasoning about situations in which several causes lead to a common effect, a much studied and yet still not well-understood inference is that of explaining away. Assuming that the causes contribute independently to the effect, if we learn that the effect is present, then this increases the probability that one or more of the causes are present....

The study of people's ability to engage in causal probabilistic reasoning has typically used fixed-point estimates for key figures. For example, in the classic taxi-cab problem, where a witness provides evidence on which of two cab companies (the more common 'green' / less common 'blue') were responsible for a hit and run incident, solvers are told...

A prominent finding in causal cognition research is people's tendency to attribute increased causality to atypical actions. If two agents jointly cause an outcome ("conjunctive causation"), but differ in how frequently they have performed the causal action before, people judge the atypically acting agent to have caused the outcome to a greater exte...

How do people judge the degree of causal responsibility that an agent has for the outcomes of her actions? We show that a relatively unexplored factor – the robustness (or stability) of the causal chain linking the agent’s action and the outcome – influences judgments of causal responsibility of the agent. In three experiments, we vary robustness b...

Temporal binding refers to a phenomenon whereby the time interval between a cause and its effect is perceived as shorter than the same interval separating two unrelated events. We examined the developmental profile of this phenomenon by comparing the performance of groups of children (aged 6-7, 7-8, and 9-10 years) and adults on a novel interval es...

In temporal binding, the temporal interval between one event and another, occurring some time later, is subjectively compressed. We discuss two ways in which temporal binding has been conceptualized. In studies showing temporal binding between a voluntary action and its causal consequences, such binding is typically interpreted as providing a measu...

Bayesian reasoning and decision making is widely considered normative because it minimizes prediction error in a coherent way. However, it is often difficult to apply Bayesian principles to complex real world problems, which typically have many unknowns and interconnected variables. Bayesian network modeling techniques make it possible to model suc...

How do people make causal judgments? We propose the counterfactual simulation model (CSM) of causal judgment which predicts causal judgments by comparing what actually happened with counterfactual simulations of what would have happened in relevant contingencies. It postulates different aspects of causation that capture whether the cause made a dif...

The present crisis demands an all-out response if it is to be mastered with minimal damage. This means we, as the behavioural science community, need to think about how we can adapt to best support evidence-based policy in a rapidly changing, high-stakes environment. This piece is an attempt to initiate this process. The ‘recommendations’ made are...

Temporal binding refers to a phenomenon whereby the time interval between a cause and its effect is perceived as shorter than the same interval separating two unrelated events. We examined the developmental profile of this phenomenon by comparing the performance of groups of children (aged 6-7-, 7-8-, and 9-10- years) and adults on a novel interval...

In many complex, real-world situations, problem solving and decision making require effective reasoning about causation and uncertainty. However, human reasoning in these cases is prone to confusion and error. Bayesian networks (BNs) are an artificial intelligence technology that models uncertain situations, supporting probabilistic and causal reas...

In many complex, real-world situations, problem solving and decision making require effective reasoning about causation and uncertainty. However, human reasoning in these cases is prone to confusion and error. Bayesian networks (BNs) are an artificial intelligence technology that models uncertain situations, supporting probabilistic and causal reas...

How do we deal with unlikely witness testimonies? Whether in legal or everyday reasoning, corroborative evidence is generally considered a strong marker of support for the reported hypothesis. However, questions remain regarding how the prior probability, or base rate, of that hypothesis interacts with corroboration. Using a Bayesian network model,...

Mechanisms play a central role in how we think about causality, yet not all causal explanations describe mechanisms. Across four experiments, we find that people evaluate explanations differently depending on whether or not they include mechanisms. Despite common wisdom suggesting that explanations ought to be simple (appealing to as few causes as...

What are the criteria that people use to evaluate everyday explanations? We focus on simplicity, coherence, and unification. We consider various operationalizations of each construct within the context of explanations to measure how people respond to them. With regard to simplicity, some of the psychological literature suggests that people do have...

Although it has long been known that time is a cue to causation, recent work with adults has demonstrated that causality can also influence the experience of time. In causal reordering (Bechlivanidis & Lagnado, 2013, 2016) adults tend to report the causally consistent order of events, rather than the correct temporal order. However, the effect has...

Bayesian models of legal arguments generally aim to produce a single integrated model, combining each of the legal arguments under consideration. This combined approach implicitly assumes that variables and their relationships can be represented without any contradiction or misalignment, and in a way that makes sense with respect to the competing a...

What are the criteria that people use to evaluate everyday explanations? This chapter focuses on simplicity, coherence, and unification. It considers various operationalizations of each construct within the context of explanations to measure how people respond to them. With regard to simplicity, some of the psychological literature suggests that pe...

According to the “story-model” of jurors’ decision-making, as advocated by Pennington and Hastie (1986, 1988, 1992, 1993), jurors in criminal trials make sense of the evidence through the construction of a mental representation of the events, rather than through the estimation and combination of probabilities. This ‘story’ consists in a causal expl...

Is the nested sets approach to improving accuracy on Bayesian word problems simply a way of prompting a natural frequencies solution, as its critics claim? Conversely, is it in fact, as its advocates claim, a more fundamental explanation of why the natural frequency approach itself works? Following recent calls, we use a process-focused approach to...

Knowledge of intention and outcome is integral to making judgments of responsibility, blame, and causality. Yet, little is known about the effect of conflicting intentions and outcomes on these judgments. In a series of four experiments, we combine good and bad intentions with positive and negative outcomes, presenting these through everyday moral...

One of the greatest challenges to the use of probabilistic reasoning in the assessment of criminal evidence is the ‘problem of the prior’, i.e. the difficulty in establishing an acceptable prior probability of guilt. Even strong supporters of a Bayesian approach have often preferred to ignore priors and focus on the likelihood ratio (LR) of the evi...

The idea of uncertainty analyses, which typically involves quantification, is to protect practitioners and consumers from drawing unsubstantiated conclusions from scientific assessments of risk. The importance of causal modelling in this process – along with the inference methods associated with such modelling – is now increasingly widely recognize...

Current theories of causality from visual input predict causal impressions only in the presence of realistic interactions, sequences of events that have been frequently encountered in the past of the individual or of the species. This strong requirement limits the capacity for 1-shot induction and, thus, does not sit well with our abilities for rap...

This paper is one in a series of analyses of the Dutch Simonshaven murder case, each using a different modeling approach. We adopted a Bayesian network (BN)–based approach which requires us to determine the relevant hypotheses and evidence in the case and their relationships (captured as a directed acyclic graph) along with explicit prior condition...

Bayesian models of legal arguments generally aim to produce a single integrated model, combining each of the legal arguments under consideration. This combined approach implicitly assumes that variables and their relationships can be represented without any contradiction or misalignment, and in a way that makes sense with respect to the competing a...

Examples of reasoning problems such as the twins problem and poison paradox have been proposed by legal scholars to demonstrate the limitations of probability theory in legal reasoning. Specifically, such problems are intended to show that use of probability theory results in legal paradoxes. As such, these problems have been a powerful detriment t...

While it has long been known that time is a cue to causation, recent work with adults has demonstrated that causality can also influence the experience of time. In causal reordering (Bechlivanidis & Lagnado, 2013, 2016) adults tend to report the causally consistent order of events, rather than the correct temporal order. Across four experiments, 4-...

There are many instances, both in professional domains such as law, forensics, and medicine and in everyday life, in which an effect (e.g., a piece of evidence or event) has multiple possible causes. In three experiments, we demonstrated that individuals erroneously assume that evidence that is equally predicted by two competing hypotheses offers n...

Testing of evidence in criminal cases can be limited by temporal or financial constraints or by the fact that certain tests may be mutually exclusive, so choosing the tests that will have maximal impact on the final result is essential. In this paper, we assume that a main hypothesis, evidence for it and possible tests for existence of this evidenc...

Human rights are claimed to be innate and based on moral principles. Human rights attitudes have been shown to be related to political ideology, but there have been few studies investigating their relationship with morality. Using moral foundations theory, we examine whether morals can predict human rights attitudes across two studies. The first st...

It is well‐established that the temporal proximity of two events is a fundamental cue to causality. Recent research with adults has shown that this relation is bidirectional: events that are believed to be causally related are perceived as occurring closer together in time—the so‐called temporal binding effect. Here we examined the developmental or...

Current causal theories argue that the statistical normality or abnormality of an action makes a difference to people's causal judgements. In this paper, we present two experiments that explore the role of statistical norms in causal cognition. In our first experiment, we provide a preliminary test of two competing theories that aim to explain the...

This paper investigates a problem where the solver must firstly determine which of two possible causes are the source of an effect where one cause has a historically higher propensity to cause that effect. Secondly, they must update the propensity of the two causes to produce the effect in light of the observation. Firstly, we find an error commens...