Invited Speakers and Symposia

The ESPP 2025 conference will feature four keynote speakers, each delivering a keynote address followed by a related symposium organized by a leading expert in the field. See below for details and dates.

2nd September: Petra Schumacher

Petra B. Schumacher is a Professor of German Linguistics / Empirical Linguistics at the University of Cologne. Her research focuses on experimental pragmatics, information structure, psycho- and neurolinguistics, and reference. Dr. Schumacher investigates language processing, neurocognitive mechanisms, discourse processes, and the syntax-semantics interface, with a keen interest in speaker-hearer interaction and the neurocognitive aspects of speech disorders.

Keynote: Meaning extensions and adaptation in real-time

Abstract

Despite the fact that language users often underspecify certain meaning aspects, communication is highly successful and addressees manage to enrich their interpretation rather effortlessly. In my talk, I will focus on meaning extensions involving metonymic relations (as in The cappuccino asked for the menu., where the cappuccino refers to the person associated with the cappuccino) and investigate the underlying mechanisms in real-time language comprehension using a series of event-related brain potential studies. The data reveal distinct processing patterns for different types of metonymic use (e.g., The cappuccino asked for the menu. vs. Tom is listening to Chopin.), indicating differences in the representation and inferential demands of the respective entities. I argue that these differences are due to discrete stages in meaning evolution and conventionalization. The talk further discusses how meaning evolution can be traced within a communicative situation, suggesting that meaning adaptation eases inferential demands and pointing to the momentary coactivation of different representations. The ability to adapt meaning extensions in real-time is further modulated by an individual’s capacity to handle neural noise.

Accompanying Symposium: 'Individual Differences in Semantic and Pragmatic Processing'

Organiser: Vera Demberg, with Margreet Vogelzang and Maria Spychalska

Experimental studies in semantics and pragmatics often report relatively high variability among participants, on phenomena ranging from scalar implicatures over reference resolution and perspective taking to processing of the scope of quantifiers and negation.

Our symposium will explore which cognitive and other individual differences factors may be underlying this variability, while also discussing methodological challenges related to reliably measuring cognitive constructs and eliciting interpretation from participants.

This session will include 3 speakers each presenting their studies on individual differences in semantic and pragmatic processing.

Margreet Vogelzang (Newcastle University)

Lexical innovation involves the creation and interpretation of new word forms or uses, often requiring listeners to pragmatically infer the word's meaning from context. One source of lexical innovation is to use an existing noun as a novel verb (e.g., “to cocktail guests”), where the new verb meaning relates to the noun’s original sense. I will present a series of experimental studies examining how individuals process such innovative denominal verbs, and to what extent this is influenced by individual differences in cognitive skills, sociolinguistic experiences, meaning generation abilities, and communication difficulties such as those associated with autism spectrum disorder.

Vera Demberg (Saarland University)

In Gricean pragmatics, inference during communication is regarded as a form of rational, domain-general reasoning about the intentions of other agents. However, experimental evidence shows that there are substantial individual differences in behaviour with participants often requiring many rounds of interaction before they exhibit the theoretically expected patterns. Based on experimental evidence and cognitive modelling, I will argue that these patterns emerge from resource-rational strategy selection behaviour subject to individualized parameters for reinforcement learning. Experimental evidence also shows that participants vary with respect to how strongly they adapt their pragmatic inferences to their conversational partner – I will report on results relating this to differences in working memory updating abilities across several pragmatic tasks.

Maria Spychalska (Ruhr University of Bochum)

Predictive processing is a considered integral to language comprehension, whereas negation is an essential part of every human language. However, it has been debated at which stage in the incremental sentence processing negation is integrated into the compositional sentence meaning and, in relation to that, how it modulates predictive comprehension mechanisms. In a series of ERP experiments we investigated how context-based predictability interacts with negation processing. In this talk, I explore the results of these studies that relate to individual differences in the processing that relate to the participants' chosen and declared strategy, where the focus and attention given to objects that are unique in the context model interferes with prediction, facilitating it in certain cases and inhibiting it in others.

3rd September: Emma Borg

Emma Borg is a Professor at the Institute of Philosophy, School of Advanced Studies, University of London. She is widely known for her work in philosophy of language (notably ‘minimal semantics’), philosophy of mind (reasons for action, rationality, and understanding chronic pain), and business ethics (corporate responsibility to society). Dr. Borg has held distinguished positions at the University of Reading and has published extensively on these topics.

Keynote: Do we do what we do for the reasons we have?

Abstract

Commonsense, or Folk, Psychology holds that people’s actions are generally rational responses to their reasons: I pick up the glass because I’m thirsty and think there is something drinkable in it, you book a train because you want to go to a conference and believe the train will get you there. This view has had the status of philosophical orthodoxy. For instance, Fodor 1989: 77 writes: “if it isn’t literally true that my wanting is causally responsible for my reaching,…and my believing is causally responsible for my saying…if none of that is literally true, then practically everything I believe about anything is false”. However, the idea that typically we do what we do for the reasons we have has come under significant pressure recently, being seen as overly-intellectualising and wildly out of touch with experimental evidence. In this talk I focus on a fundamental worry for common-sense psychology: the claim that much action-generation comes about with no appeal to reasons and reasoning. I survey the key evidence for this claim but argue that none of it suffices to show that the common-sense model must be rejected.

Accompanying Symposium: 'The Underintellectualization of Everyday Life'

Organiser: Eric Mandelbaum, with Benedek Kurdi, Liuba Papeo and Barbara Pomiechowska

Liuba Papeo (CNRS & University Claude Bernard Lyon1, France)

Social Relationship in the Eye of Perception

In the street, two people are running one after the other; instantly you know they are not just running, but one is chasing the other, who is in danger. How can we form such immediate, effortless impressions of social interactions? The prevailing idea in cognitive neuroscience has been that perceiving social interactions is a high-level process implemented in association cortices. But what if, instead of inferring chase, we see chase. I will present behavioral and neural phenomena suggesting that social interactions are extracted earlier ¬–and rapidly– in visual perceptual processing. This research may reveal the visual foundations of social cognition and the function of visual perception in unraveling the causal and social structure of the world, thus advancing our understanding of human vision, social cognition –and their bridging.

Benedek Kurdi (University of Illinois)

Implicit attitudes form (and change?) propositionally

A key insight from experimental social psychology over the past decades is that attitudes (i.e., evaluative information stored in long-term memory, such as BLACK PEOPLE–GOOD or “Central Illinois is beautiful”) can be activated automatically. Such automatic (implicit) attitudes are usually juxtaposed with explicit attitudes, which are retrieved deliberately. According to influential early and contemporary dual-process accounts of social cognition, implicit and explicit attitudes are fundamentally different in nature: They respond to different types of information, form and change via different processes, and are represented differently in the mind. In this talk I will review 10+ years of experimental evidence produced by my lab and others that challenge this basic idea by showing that implicit attitudes robustly shift in response to persuasive messages (not just vast numbers of co-occurrences), and they reflect inferential transitions and relational information (not merely the associative structure of language or events experienced in the world). I will conclude by highlighting open questions with respect to the representational format of implicit attitudes and the applicability of these findings to the change of entrenched implicit attitudes, such as implicit attitudes toward familiar social categories.

Barbara Pomiechowska (University of Birmingham)

Babies’ secret cognitive tools: symbols and compositionality

Symbols are one of the basic components of human thought and communication. As adults we constantly use complex sequences of symbols while conversing, consuming news, navigating new cities, learning new skills, and more. What are the origins of our symbolic behavior? How does the ability to understand and manipulate symbols emerge during human development? Classical studies in developmental psychology suggest that symbolic representations take years to develop. In this talk, I will explore the idea that symbolic representation may be operational already in infancy. I will present recent behavioural and neuroscience evidence that, by the end of the first year of life, infants already deploy symbolic thought in service information compression and learning. I will also explore limits of this early symbolic competence compared to that of adults. Finally, I will argue that the infant brain may be hard-wired for compositionality, spontaneously combining familiar symbols to generate new meaning.

4th September: Nora Newcombe

Nora S. Newcombe is the Laura H. Carnell Professor of Psychology at Temple University. A leading figure in spatial cognition, cognitive development, and memory research, she has made groundbreaking contributions to our understanding of cognitive maps and spatial reasoning across the lifespan. Dr. Newcombe’s many accolades include membership in the National Academy of Sciences and the APA Distinguished Scientific Contributions Award.

Keynote: Cognitive Maps & Mental Representation

Abstract

The term “cognitive map” is in common usage, and Nobel Prizes were awarded for the discovery of the cellular basis of “GPS in the Brain”. Nevertheless, whether the mind-brain forms map-like representations has long been controversial. I will offer evidence that cognitive maps can be formed, but that doing so is easier in some environments than others, and depends on the individual’s cognitive capacities, prior navigational experience and motivation. I will then consider what we know about the development of the capacity to form cognitive maps. I will argue against the core knowledge proposal of a geometric module, and then turn to examine an emergentist view, according to which strong starting points exist in the organization of the ventral visual stream, behavioral capacities emerge in an experience-expectant way as the motor milestones are passed and the hippocampus matures, and further fine tuning occurs at least until the start of adolescence.

Accompanying Symposium: 'Maps, Cognition & Representation'

Organiser: Roberto Casati, with Pablo Fernandez Velasco and Misun Kim

Roberto Casati

Cognitive maps and non-cooperative environments

Cognitive maps are theoretical posits that get their support from an inference to the best explanation (IBE). In particular, their postulation straightforwardly explains certain abilities, such as the capacity to envision detours and shortcuts through unperceived and unvisited places that are connected to places one has visited, perceives, or one is at. The main idea is that exploration of an environment automatically (in "system 1" mode) produces representations of that environment that can be piecemeal and scattered, but are nevertheless organized in a maplike fashion, or in a format that is compatible with a type of retrieval of information for navigational purposes that mimics the type of retrieval that we perform on (external) maps. For instance, if on a map one sees that present location A is aligned with both unsurmountable obstacle B and target location C, one can plan a detour around B, as the detour is present in the topology of the map vehicle. Much of the IBE, and the effectiveness of cognitive maps, is predicated on what I propose to term 'environment cooperation'. Some environments are navigationally non-cooperative in the sense that they make it near impossible (or noneffective) to automatically generate a cognitive map of their layout. An example of non-cooperative environment is the open seas. When sailing on the open seas, one must actually inhibit "system 1" cognitive map formation, as the environmental near and distant landmarks are inherently dynamic and ephemeral (waves, clouds). I shall discuss some interpretations of documented non-instrumental techniques of open seas sailing. To, conclude, I shall tentatively suggest that contemporary theoretical focus on cognitive maps may be a consequence of a cartographic bias, i.e. the implicit recognition of the enormous effectiveness of modern cartographic tools.

Pablo Fernandez Velasco

Cross-cultural variation in the use of cognitive maps for navigation

One assumption in most studies of spatial cognition in cognitive science is that expert navigation depends on the use of cognitive maps. This contrasts with anthropology, where there is an active debate between mental map proponents, who argue that wayfinding depended on stored spatial information, and practical mastery proponents, who argue that wayfinding depends on attunement to the environment. What the debate shows is that different cultures vary in the degree to which they rely on allocentric information for navigation. Many groups have expert navigators that seem to find their way in a predominantly egocentric fashion, such as Saami reindeer herders, the Walbiri people or the Ongees people of Little Andaman Island. In this talk, I focus on the case of Evenki reindeer herders and hunters. I show that their wayfinding methods –involving a particular gait, path networks, and vast hydrological and toponymical knowledge– allow the Evenki to navigate without requiring the integration of egocentric and allocentric frames of reference. I finish by exploring the connection between cognitive map use and environmental experience.

Misun Kim

Cognitive maps for a 3D world

Humans can build a mental model of the world, known as a cognitive map. Despite physical space being 3D, much of previous research has focused on spatial navigation in simple 2D environments, leading to the dominance of 2D Euclidean maps in the field. How do humans build cognitive maps for a 3D world where they can move vertically as well as horizontally? How do they navigate on non-flat surfaces embedded in 3D space? Do they rely on dimension-reduced maps or volumetric maps? Can they overcome the influence of Euclidean geometrical prior? In this talk, I will present evidence from virtual reality experiments, where participants completed various spatial memory and path integration tasks on non-flat surfaces, including a spherical surface. These findings suggest that cognitive maps are flexible and adaptive, and are not necessarily veridical representations of the 3D world.

5th September: Cameron Buckner

Cameron Buckner began his career in computer science, focusing on non-monotonic logic approaches to artificial intelligence. His research explores the interplay between learning and meaning, especially how contemporary learning models inform theories of mental content and knowledge representation. Dr. Buckner’s work spans animal cognition, AI, and philosophical methodology, with a current focus on the philosophical implications of large language models.

Keynote: Large Language Models as Models of Human Reasoning

Abstract

Recent advances in large language models that use self-prompting like GPT’s o1/o3, Gemini 2.5, and DeepSeek r1 have begun to encroach on human-level performance on “higher reasoning” problems in mathematics, planning, and problem-solving tasks. OpenAI in particular has made ambitious claims that these models are reasoning and that by scrutinizing their chains of self-prompting, we can “read the minds” of these models, with obvious implications for solving problems of opacity and safety. In this talk, I review four different methodological approaches to evaluate the success of these models as models of human reasoning (“psychometrics”, “signature limits”, “inner speech”, and “textual culture”), focusing especially on comparisons to philosophical and psychological work on “inner speech” in human reasoning. I argue that this work suggests that while the achievements of self-prompting models are impressive and may make their behavior more human-like, we should be skeptical that problems of transparency and safety are solved by scrutinizing chains of self-prompting, and more philosophical and empirical work needs to be done to understand how and why self-prompting improves the performance of these models on reasoning problems.

Accompanying Symposium: 'LLMs and Human Reasoning'

Organiser: Emily Sullivan, with Michael Franke and Bojana Grujičić

We look forward to an exciting week of keynotes and symposia at ESPP 2025!