Metric biases in body and object representations

The dimensions and proportion of our body parts are typically misestimated. For instance, the hand is perceived as distorted, with its width overrepresented compared to its length. Even though we misperceive its shape and dimensions, our hand is the protagonist of extremely accurate fine movements, as well as the means by which we sense the world. This thesis is organised into two chapters. The first one describes two studies aimed at investigating the role of the biased representation of the hand in motor planning and execution. In Study 1, we provided evidence in support of the hypothesis that our motor system makes use of the distorted hand representation when movements are performed in absence of visual guidance: we observed that the pattern of errors in a proprioceptive matching task was compatible with biases affecting the hand representation. However, we also found that the errors magnitude was reduced compared to what predicted by that hypothesis. Results of Study 2 suggest that the motor system refines the movement trajectories, partially overcoming the misestimation of the hand dimensions, by integrating current somatosensory inflow and motor outflow. Our results highlight the role of these systematic biases, as an important source of error, in movement driven by proprioception only, and prompt to shift the focus from the body as an isolated system, to the body as integrated and active into the environment. In this vein, the second chapter enters the debate regarding the specificity of the metric biases affecting body representations, by testing whether these biases extend to the surrounding environment, and which sensorial information and higher-order factors modulate them. Study 3 addresses the role of visual and somatosensory information in estimating the size of our body, by comparing the perceived dimensions of body parts affording different degrees of tactile acuity and visual accessibility. We found that both visual and somatosensory factors are likely to shape biases affecting body representations, and that somatosensory factors come into play mostly when visual cues are poor, ambiguous or unavailable. In Study 4, we investigated whether the metric biases are specific to the size estimation of the body, or whether they generalise to the size estimation of objects, too. We reasoned that, from an ecological perspective, the selective misestimation of our body dimensions may not be functional to an efficient interaction with the environment. An extensive investigation of the perceived dimensions of the hand and several objects showed that metric biases indeed extended to objects, were stable over time and were unrelated to the degree of familiarity or sense of ownership for the object. Yet, the pattern of the distortions might depend, at least to a degree, on the manipulability of the object, since objects which do not afford manipulation and interaction were differently represented. Finally, Study 5 sought to elucidate the neural underpinnings associated with these last results. We recruited six patients with refractory epilepsy undergoing stereo-EEG recordings for diagnostic reasons, to study the electrophysiological responses elicited during the size estimation of the participants hand and of a highly familiar and manipulable object (participants mobile phone). The similar behavioural pattern of distortions affecting those two targets was reflected by similar activity in the high-γ band, spreading over occipito-temporal, posterior parietal and frontal areas, consistent with the involvement of the visual imagery network. In two patients, we also registered a higher activity over the precentral area during the size estimation of the hand compared to that of the mobile phone, possibly supporting the additional role of the sensorimotor cortex in hand metric representation.