Perceptual and production underpinnings of phonological processes

Languages of the world have different inventories of sounds and different ways in which these sounds may combine to convey meaning. Linguists have long been interested in describing this variability and in discovering its limits to single out those universal properties that hold of human languages as a whole. Phonologies of all languages are affected by aerodynamic and articulatory constraints in the production of sounds.

Issues in cognitive control

The purpose of these bibliographical internships is to survey the empirical work on cognitive control (or specific control functions such as working memory, attention, task switching, and so on), and to identify both empirical and philosophical issues that this work raises. These internships may lead to subsequent collaborative, experimental internships.

The representational space of consciousness

The aim of the internship is to better characterize the representational space of consciousness - why does conscious content appear unified, whereas sensory and cognitive neural maps use very different reference frames? Depending on the intern's background, this question can be approached by a novel experimental paradigm with brain imaging, and/or the analysis of already collected experimental data, and/or the development of a formal model. All approaches require programming skills, with either previous experience or a willingness to develop such skills.

Graphic communication: How minds shape culture

Our research group studies the evolution of graphic communication, the exchange of information through images — letters, symbols, emblems. We study the cultural history of these visual signs, with quantitative tools informed by evolutionary theory and by cognitive science. Using large-scale datasets assembled from publicly available collections or collected by us, we test predictions concerning changes in the shape of letters, on the information carried by emblems, or on the complexity of symbols.

Robustness to environmental noise in early speech perception

When the environment is noisy, speech perception takes place in under-optimal listening conditions. Young adults with normal hearing are efficient at processing such a degraded speech signal thanks to efficient auditory and linguistic perceptual mechanisms. However, this perceptual robustness is only attained at the end of childhood. How do infants cope with noise in the very early phases of language acquisition? Do they rely on any of the factors that are at play in adulthood? Very little is known on these points.

From the mouth to the ear: the role of sensorimotor maturation in phonological development

Traditionally, speech processing has mainly been regarded as an auditory-based phenomenon. However, to date, a vast amount of evidence has shown that both speech actions and speech sounds yield perceptual correlates. Indeed, in healthy adults, the brain network recruited by speech processing includes a regular contribution of the sensorimotor areas. This fits well with neuroscientific frameworks positing shared representations for actions and their perception.

Invariant cortical representations of sounds in ambient noise

In our everyday lives, sounds of interest like speech, alarms and other foreground sounds are effortlessly listened to in noisy background environments. This perceptual capability implies that internal representations of sounds are robust and unaffected by the acoustic background. This process, called noise invariance, emerges throughout cortical stages. In particular previous works highlighted noise invariant representation of sounds throughout human auditory cortex (Kell & McDermott. 2019. Nat. Comm.).

Task-relevant sound encoding in the primary auditory cortex

Any sensory stimulus consists of a superposition of myriad different features. A simple visual object can for instance be characterized by its shape and colour, while an auditory tune is a combination of pitch and rhythm. The brain is able to flexibly focus on and extract the single feature that is behaviourally relevant, and for example segregate objects first by color, and then by shape. How do neural circuits achieve such flexible categorization?