Center for Interdisziplinary Research

How Universal is Causal Cognition?

Bielefeld, May 18 - 19, 2012

Convenors: Michael Waldmann | Göttingen | DEU, York Hagmayer | London | GBR, Andrea Bender | Freiburg i.Br. | DEU und Sieghard Beller | Freiburg i.Br. | DEU


The workshop 'How universal is causal cognition' took place as part of a series of workshops organized by the research group 'Cultural constitution of causal cognition'. The main goal of the workshop was to bring together leading researchers from various fields in psychology who study causal cognition. The ability to learn and reason about cause-effect relations is one of our central cognitive capacities. Causal cognition allows us to make predictions, explain surprising events, and plan actions, and therefore is crucial for survival. However, there is still no agreement in psychology about the precise nature of the cognitive processes underlying this important competency. The first day of the workshop focused on basic research on cognitive processes underlying causal cognition. Most of the researchers in this area have worked under the implicit assumption that causal cognition is driven by universal mechanisms, although this assumption has rarely been empirically tested. One of the goals of the workshop was to take a closer look at these theories, and discuss with the presenters how the universality assumption can actually be tested.
The meeting started with the presentation of Marc Buehner (University of Cardiff) who took a closer look at one of the main postulates of Hume's theory of causation, namely that causes temporally precede their effects. Many studies have shown that it is easier for us to learn about cause-effect relations when effects immediately temporarily follow their causes, but Marc could additionally show that this pattern can be moderated by prior knowledge. We know that not every cause is followed by its effects right away (e.g., effects of drugs), and use this knowledge when learning about causal relations. In the second part of his talk Marc presented fascinating new results showing that causal understanding can also influence our perception of events: We are more likely to perceive two events as closer than they were actually presented when we see them as causally related. The next talk shifted the focus to moral judgments. Typically moral judgments are viewed as separate from causal cognition, but Rumen Iliev (Northwestern University) convinced the group that physical and causal cognition can also influence moral judgments in harm-based moral dilemmas. A famous example is the trolley problem in which a fictitious runaway trolley is about to kill several people standing on the tracks. These people could be saved by re-directing the trolley, which, would lead to the death of a different person. Rumen presented evidence showing that physical features of the trolley scenario (and related situations) affect whether people find it acceptable to kill one person to save five others.
Anne Schlottmann, a developmental psychologist (University College, London), presented her research on causal perception in children. One of the visual displays she presented to the children is the famous Michotte situation in which a square moves towards a second square that moves away after contact. Observers typically report that the first square launched the second, even when they see the animation the first time. This observation has led developmental psychologists to believe that causal perception may be innate. Anne presented her own research on mechanical (as in the Michotte task) and social causation casting doubt on the claim that causal perception is fully innate and impervious to differences in learning experiences.
Tom Beckers (University of Leuven) followed with a presentation of his work on causal learning in children and rats. His work takes issue with the popular claim in psychology that causal cognition may be driven by elementary associative learning processes. Tom defended the contrasting claim that both children and rats recruit processes that are more similar to higher-level cognitive processes, such as logical reasoning, than to lower-level processes.
The next speaker was José Perales (University of Granada) who investigated which empirical evidence people use when making causal inferences. Like Tom Beckers, he doubts that low-level covariation assessment is the only cue people use. To support his view that people are more flexible, José presented studies showing that people use various heuristics which are triggered by characteristics of the learning data.
The last speaker of the day was Jean Bonnefon (University Toulouse). His work focuses on reasoning with conditionals ('if p then q') in which the conditional describes situations in which p or q are valued by the agent (e.g., 'if you take the medicine, then your symptoms will disappear'). Bonnefon showed that reasoning with these so-called utility conditionals is also influenced by the causal structure underlying the described situation.
The second day of the workshop was mainly devoted to research in which the possible influence of cultural features on causal reasoning was directly assessed. The first speaker, Denis Hilton (University of Toulouse), focused on causal explanation and showed how important the conversational context is in which queries for causal explanations are embedded. In discourse situations, people typically form hypotheses about the intentions of the person querying them. These hypotheses strongly influence the type of answer that is being offered. The talk made it clear that mutual expectations in the test situation will prove an important factor when testing theories of causal cognition in different cultural contexts.
Michael Morris (Columbia University, New York) is one of the few researchers who has directly tested hypotheses about cross-cultural differences in causal cognition. In his talk he first reviewed his seminal research on differences in causal attributions in 'Western' and 'Eastern' cultures. The key difference in his studies was that participants from Western cultures tended to explain behavior more by emphasizing properties of the agent, whereas participants from Eastern cultures focused more on context, especially the social context of the roles, relationships and groups in which the agent is embedded. In recent work Michael has specified more precisely the boundary conditions of these effects, and has demonstrated the variability of causal attributions within cultures.
The last speaker of the workshop was Larry Hirschfeld (New School for Social Research, New York). Larry is also interested in answering the question how people in different cultures explain behavior. Such explanations often contain references to beliefs and desires of the agent, which in the psychological literature has been interpreted as evidence for a possibly innate Theory of Mind. Larry's important contribution was to show that people also hold a Theory of Society. When explaining behavior we also activate our knowledge about social groups and related norms to which the agent belongs. Larry presented studies showing how implicit assumptions about race, for example, influence the explanations of young children.
The workshop concluded with a plenary discussion in which we re-visited the main question of the workshop: How universal is causal cognition? One of the main conclusions was that we do not really know yet which components of causal understanding are universal, and which can be moderated by cultural influences. Most research has been conducted under the implicit premise of universality, but no empirical evidence supports this premise. One possible option is that causal learning and reasoning mechanisms are universal, but different learning experiences in diverse cultures shape the sometimes varying outcomes of learning. Another possibility is that folk theories held in different cultures are not only the result of causal learning; they also shape the learning processes underlying further learning. If there was a consensual conclusion arising from this workshop, then it is that empirical studies on these questions are desperately needed.


Christoph Antweiler | Bonn | DEU, Tom Beckers | Leuven | BEL, Bettina Beer | Luzern | CHE, Jean-François Bonnefon | Toulouse | FRAU, Marc Buehner | Cardiff | GBR, Thomas Friedrich | Köln | DEU, Don Gardner | Luzern | CHE, Daniel Hanus | Leipzig | DEU, Denis Hilton | Toulouse | FRAU, Lawrence A. Hirschfeld | New York, NY | USA, Rumen Iliev | Evanston, IL | USA, Maria Kronfeldner | Bielefeld | DEU, Gregory Kuhnmünch | Freiburg i.Br. | DEU, Cristine Legare | Austin, TX | USA, Manuela Lenzen | Bielefeld | DEU, Michael W. Morris | New York, NY | USA, Jonas Nagel | Göttingen | DEU, Hansjörg Neth | Göttingen | DEU, Albert Newen | Bochum | DEU, José César Perales López | Granada | ESP, Jana Samland | Nettingen | DEU, Anne Schlottmann | London | GBR, Keith Stenning | Edinburgh | GBR, Christopher Topp | Bochum | DEU, Markus Werning | Bochum | DEU, Thomas Widlok | Nimwegen | NLD, Alex Wiegmann | Göttingen | DEU

Program (PDF)

ZiF - Center for Interdisciplinary Research - Homepage > ZiF Research Groups > The Cultural Constitution of Causal Cognition > events >