IEEE International Conference on Development and Learning

Somogyi, E. (Invited speaker)

Department of Psychology

Activity: Talk or presentation types › Invited talk

Description of Activity

Workshop: Body Image & Cross-sensory Correspondences: What about Infants and Robots.

Period	12 Sept 2022
Held at	Queen Mary University of London, United Kingdom
Degree of Recognition	International

Documents & Links

Our bodies constitute the interface between our mind and the surrounding world. To interact effectively with the world, babies and robots need to discover the relations between the different parts of their bodies, and between the different senses; i.e., cross-sensory correspondence. For body representation, they need to know where their limbs are in space, how far they can extend, how they work, and what range of actions they can achieve with them. Such body representations are essential for developing a sense of the self and for defining the limits of the peripersonal space. These are important milestones in human development to construct efficient interactions with others and the world. For this construction to be complete, individuals need to extract from the different sense modalities a coherent representation of their surroundings. This is possible thanks to cross-sensory correspondences. Cross-sensory correspondences occur very early on in development, but it is still unknown how they develop, whether some appear before others and why for example. Indeed, many events of the world, such as social interactions, involve perceiving common properties in different sense modalities. The size and shape of an object can, for example, be processed both visually and haptically, which leads to a unitary percept. The intensity of either light or sound can change over time and thus be perceived as a similar phenomenon. In fact, in everyday situations, contingent associations of different attributes simultaneously changing over time are experienced on a regular basis. A bouncing ball, for example, is perceived as regular up and down motion and as regular sound-silence alternation. Its movement generates a multimodal representation based on synchronous and analogous dynamic time variations in the visual and auditory experience.

IEEE International Conference on Development and Learning

Description of Activity

Documents & Links

Related content

Research outputs

Action-outcome contingencies as the engine of open-ended learning: computational models and developmental experiments

Know your body through intrinsic goals

Which limb is it? Responses to vibrotactile stimulation in early infancy

Editorial: Body representations, peripersonal space, and the self: humans, animals, robots