KEYNOTE ABSTRACTS

---

Gert Westermann (Lancaster University, UK)

When words and objects combine: The transition from preverbal to language-based cognition in infancy

Before they begin to learn language, infants have already acquired considerable knowledge of the world. An important question is how the onset of language affects and shapes this early knowledge. I will describe a range of studies that have addressed this question, for example by asking how labels help infants form categories of objects, if knowledge of the name for an object affects the mental representation of this object, and how infants request object labels for novel objects from social partners and the role of temperament in this process. I will also describe computational models of the transition from preverbal to language-based object processing to provide explanations of the mechanisms underlying the integration of object and linguistic knowledge. This work highlights the benefit of considering the development of language as embedded in the processes of cognitive and social development.

---

Julian Jara-Ettinger (Yale University, US)

The social basis of referential communication

Human communication is an intrinsically social activity that allows us to share our thoughts through sound and movement. Accordingly, theoretical work has long argued that this capacity must rely on Theory of Mind — our ability to understand other people’s behavior in terms of unobservable mental states. Yet, classical work suggests that the interaction between Theory of Mind and communication is surprisingly limited. In this talk I will argue that social reasoning is necessary for successful communication even in the simplest linguistic events. Using referential communication as a case study, I will show that traces of nuanced social reasoning are visible in both production and comprehension in everyday communication, and that complex social inferences underlie seemingly ‘irrational’ communicative behaviors. These findings advance the idea that communication is best understood as a fundamentally cooperative activity grounded in social representations provided by Theory of Mind.

---

Jutta L. Mueller (University of Vienna, AUT)

Sound matters – especially when we are young

Humans have remarkable abilities to learn non-adjacent sequential structure from auditory input from early on. This is true for linguistic as well as for non-linguistic stimuli. To demonstrate this, I will present data from infants younger than 6 months of age and neural network simulations explaining potential underlying mechanisms. On top of a sensitivity to complex sequential structure, infants and adults are different with respect to their ability to learn from mere input. Those differences include different sensitivities to sensory properties in the learning input across age. To show this I will present sequence learning and word learning experiments that varied sensory properties of the associated stimuli.  I will discuss different explanations of the observed effects, including developmental changes in cognitive control and sensory biases and a potential combination of both. It will be argued that auditory sequential learning may contribute to initial steps of language learning, specifically segmentation and identification of basic units and relations and thus forms an important stepping stone towards the later discovery of linguistic grammar. While the processing of full-fledged human grammar requires additional, potentially human-specific, layers of combinatorial processing basic mechanisms of complex auditory sequence processing are shared with our phylogenetic relatives.

---

Olivier Pascalis (Université Grenoble Alpes & CNRS, FR)

Face Processing in Infancy and Beyond: The Case of Social Categories

Arriving in an unfamiliar world, infants need to quickly develop a core knowledge that allow them to organize the various multimodal items/stimuli present in their environment in order to predict and respond appropriately to those events. To achieve this, the infant will require the ability to categorize stimuli and relate this knowledge to previous experiences. Creating a memory representation for previously experienced events enables infants to develop more complex representations of stimuli that they repeatedly encounter and will help them to learn how to respond to other similar stimuli encountered in the future. Infants’ early categorization has been observed both in visual and auditory modalities: Human speech / animal sound, native/non-native language; male/female faces etc. (Quinn et al., 2019). The scientific literature indicates that young infants already possess categories (someone is human, from a specific gender or ethnic group). Current evidence suggests in fact that infants demonstrate early perceptual differences that may be associated with changes in how they represent other humans. I will present a series of studies on infant’s face categorization.

---

Victoria Southgate (University of Copenhagen, DK)

Uniquely infant social intelligence

The classic view of very early cognition is that it is egocentric, and that sufficient cognitive control is required to overcome an egocentric bias. However, this view is difficult to reconcile with data accumulated over the last decade, indicating that infants readily adopt others’ perspectives, and do so despite limited cognitive control. In this talk, I will present a radically different view of infant cognition in which infants are predominantly altercentric and biased to encode information that is the focus of others’ attention, even at the expense of their own perspective. I argue that this is possible, in part, due to an initial absence of self-representation and propose that this bias will constrain and facilitate infant learning by allowing them to exploit others’ information selection at a time when their own ability to act on the world is limited. I will present recent empirical studies from my lab in which we have been testing the various hypotheses derived from this account.