The general framework of my researches is the study of motion perception. When we move around in our environment, our brain continuously processes a set of information, which tells us about our relative motion and position in the world. I have been conducting behavioural and psychophysical experiments on healthy humans, often using virtual reality in order to control the motion simulation. The collected data can eventually be confronted to the predictions from existing models.

Speed perception during self-motion: on structural stability of the world

Optic flow plays a fundamental role in the analysis of our own movements. I currently work, in collaboration with Paolo Pretto from the MPI in Tübingen, on a project in order to characterize how speed is extracted from optic flow during visual self-translations over planar surfaces. In this particular project, I focus on visual information that cannot provide direct knowledge about locations, meaning that I exclude cognitive strategies relying on visual landmarks for motion perception.

The relationship between angular velocity (retinal) and linear velocity (egocentric) is not trivial, and still the world we perceive does not get distorted when we move around our gaze. We are running experiments to study: the underlying compensation mechanisms for varying gaze tilt angle; the contribution of visual areas in speed perception (peripheral, central and full field); and the influence of scene contrast. The latter, while being currently debated in the literature, is of crucial importance for driving simulators in order to assess the dangers of fog while on the road.

Visio-vestibular motion perception and memorization

On the other hand, inertial information is available thanks to the vestibular system, pressure on the internal organs, and proprioception allows positioning our body segments in space. Many previous studies at the LPPA have focused on motion perception with vestibular information alone. In the context of the Moves European funded project, I am investigating how internal information (vestibular and proprioceptive) might combine with optic flow in order to create robust self-motion percepts. I try to assess the respective contributions of each sensory modality, and in particular whether there is a continuous integration with visual inputs or just a punctual qualitative use of vestibular inputs. I am also trying to determine whether the encoding of motion perception is specific to each modality or if we have the capacity to transfer from one to another (for instance, compare a pure visual rotation to a body rotation in darkness).

Spatial updating: a visio-vestibular integration?

In another line of research, I plan to study how visual and body motion cues are integrated in order to update environing spatial locations during self-motion. We found previously that for moving observers the facilitation of mental rotations – this capacity to update spatial relationships – is enhanced by rich sensory contexts. How exactly the visual and vestibular cues are combined to update spatial information?

Time-to-contact and motion prediction

Together with Aurore Capelli, a post-doc from the LPPA, we are conducting a series of experiments on self-motion time-to-contact estimations. We want to assess how visual information is processed in order to estimate the time delay before reaching a visually specified location on our way. More details about these experiments are available on the web page of Aurore Capelli.