scholarly journals Sensorimotor Representation Learning for an “Active Self” in Robots: A Model Survey

2021 ◽  
Vol 35 (1) ◽  
pp. 9-35 ◽  
Author(s):  
Phuong D. H. Nguyen ◽  
Yasmin Kim Georgie ◽  
Ezgi Kayhan ◽  
Manfred Eppe ◽  
Verena Vanessa Hafner ◽  
...  

AbstractSafe human-robot interactions require robots to be able to learn how to behave appropriately in spaces populated by people and thus to cope with the challenges posed by our dynamic and unstructured environment, rather than being provided a rigid set of rules for operations. In humans, these capabilities are thought to be related to our ability to perceive our body in space, sensing the location of our limbs during movement, being aware of other objects and agents, and controlling our body parts to interact with them intentionally. Toward the next generation of robots with bio-inspired capacities, in this paper, we first review the developmental processes of underlying mechanisms of these abilities: The sensory representations of body schema, peripersonal space, and the active self in humans. Second, we provide a survey of robotics models of these sensory representations and robotics models of the self; and we compare these models with the human counterparts. Finally, we analyze what is missing from these robotics models and propose a theoretical computational framework, which aims to allow the emergence of the sense of self in artificial agents by developing sensory representations through self-exploration.

Author(s):  
Minoru Asada

Proprioception is our ability to sense the position of our own limbs and other body parts in space, and body schema is a body representation that allows both biological and artificial agents to execute their actions based on proprioception. The proprioceptive information used by current artificial agents (robots) is mainly related to posture (and its change) and consists of joint angles (joint velocities) given a linked structure. However, the counterpart in biological agents (humans and other animals) includes more complicated components with associated controversies concerning the relationship between the body schema and the body image. A new trend of constructive approaches has been attacking this topic using computational models and robots. This chapter provides an overview of the biology of proprioception and body representation, summarizes the classical use of body schema in robotics, and describes a series of constructive approaches that address some of the mysteries of body representation.


Perception ◽  
10.1068/p5853 ◽  
2007 ◽  
Vol 36 (10) ◽  
pp. 1547-1554 ◽  
Author(s):  
Francesco Pavani ◽  
Massimiliano Zampini

When a hand (either real or fake) is stimulated in synchrony with our own hand concealed from view, the felt position of our own hand can be biased toward the location of the seen hand. This intriguing phenomenon relies on the brain's ability to detect statistical correlations in the multisensory inputs (ie visual, tactile, and proprioceptive), but it is also modulated by the pre-existing representation of one's own body. Nonetheless, researchers appear to have accepted the assumption that the size of the seen hand does not matter for this illusion to occur. Here we used a real-time video image of the participant's own hand to elicit the illusion, but we varied the hand size in the video image so that the seen hand was either reduced, veridical, or enlarged in comparison to the participant's own hand. The results showed that visible-hand size modulated the illusion, which was present for veridical and enlarged images of the hand, but absent when the visible hand was reduced. These findings indicate that very specific aspects of our own body image (ie hand size) can constrain the multisensory modulation of the body schema highlighted by the fake-hand illusion paradigm. In addition, they suggest an asymmetric tendency to acknowledge enlarged (but not reduced) images of body parts within our body representation.


2012 ◽  
Vol 4 (2) ◽  
pp. 157-165
Author(s):  
Sally Gardner

Drawing on experiences that have entailed watching and learning forms of so-called ‘Indian dance’ (Bharata Natyam and Odissi), and watching Odissi dancers performing in various locations in Orissa’s ‘sacred triangle’ (Puri, Konark, Bhubaneswar), and against my own background in contemporary dance, I propose that the difference of the Odissi body is that the dancer dances with his or her feet in more than one kingdom – that is, he or she maintains a link between human bodies and the bodies of plants. Such a perception can help to displace questions of the dancer’s spatiality and representations, challenging western or westernized visions of the industrial or mechanical body, assumed hierarchies of body parts and their signifying powers, and assumptions about the role of the joints. The sense of a botanical imaginary or specific cultural body-schema at work in Odissi dance is supported by discussion of historical and ethnographic literature pertaining to the (former) female dancers of the Jagannath Temple in Puri; the temple’s links with Oriyan tribal cultures; the dancers’ traditional importance according to an axis of social auspiciousness/inauspiciousness as opposed to social purity/impurity; and the particular processes of the reconstruction of Odissi dance (separate from that of Bharata natyam) after independence.


Body schema refers to the system of sensory-motor functions that enables control of the position of body parts in space, without conscious awareness of those parts. Body image refers to a conscious representation of the way the body appears—a set of conscious perceptions, affective attitudes, and beliefs pertaining to one’s own bodily image. In 2005, Shaun Gallagher published an influential book entitled ‘How the Body Shapes the Mind’. This book not only defined both body schema (BS) and body image (BI), but also explored the complicated relationship between the two. The book also established the idea that there is a double dissociation, whereby body schema and body image refer to two different, but closely related, systems. Given that many kinds of pathological cases can be described in terms of body schema and body image (phantom limbs, asomatognosia, apraxia, schizophrenia, anorexia, depersonalization, and body dysmorphic disorder, among others), we might expect to find a growing consensus about these concepts and the relevant neural activities connected to these systems. Instead, an examination of the scientific literature reveals continued ambiguity and disagreement. This volume brings together leading experts from the fields of philosophy, neuroscience, psychology, and psychiatry in a lively and productive dialogue. It explores fundamental questions about the relationship between body schema and body image, and addresses ongoing debates about the role of the brain and the role of social and cultural factors in our understanding of embodiment.


2021 ◽  
Author(s):  
Lennart Wittkuhn ◽  
Samson Chien ◽  
Sam Hall-McMaster ◽  
Nicolas W. Schuck

Experience-related brain activity patterns have been found to reactivate during sleep, wakeful rest, and brief pauses from active behavior. In parallel, machine learning research has found that experience replay can lead to substantial performance improvements in artificial agents. Together, these lines of research have significantly expanded our understanding of the potential computational benefits replay may provide to biological and artificial agents alike. We provide an overview of findings in replay research from neuroscience and machine learning and summarize the computational benefits an agent can gain from replay that cannot be achieved through direct interactions with the world itself. These benefits include faster learning and data efficiency, less forgetting, prioritizing important experiences, as well as improved planning and generalization. In addition to the benefits of replay for improving an agent's decision-making policy, we highlight the less-well studied aspect of replay in representation learning, wherein replay could provide a mechanism to learn the structure and relevant aspects of the environment. Thus, replay might help the agent to build task-appropriate state representations.


2021 ◽  
pp. 229-243
Author(s):  
Jonathan Cole

In neurological illnesses, the body may present itself to perception in ways which allows insights into the concepts of body image and body schema. Three such conditions are explored. From those who live with spinal cord injury, paralysed and insentient from the neck down, aspects of the importance of the body in one’s sense of self are revealed. Some also describe a coming to terms with their altered bodies. When considering the body image, its adaptability and this reconciliation to a new normal should be considered. Studies on acquired severe sensory loss explore how conscious control, at the body image level, may partially replace the deafferented body schema. There is little evidence, however, for these subjects extending access to previously non-conscious motor schema. Lastly, some narratives from those with congenital absence of movement of facial muscles describe reduced emotional experience and felt embodiment as children. These can be developed as young adults, through shared social interactions. The importance of the social in elaboration of the body image is further implicit in a consideration of the stigma associated with facial disfigurement. Others’ responses to one’s body are crucial in developing our body image and sense of self.


1974 ◽  
Vol 44 (3) ◽  
pp. 346-362 ◽  
Author(s):  
Yigal Gross ◽  
Ross Webb ◽  
Ronald Melzack
Keyword(s):  

2009 ◽  
Vol 21 (3-4) ◽  
pp. 252-260 ◽  
Author(s):  
Lucilla Cardinali ◽  
Claudio Brozzoli ◽  
Alessandro Farnè

2019 ◽  
Author(s):  
Aubrieann Schettler ◽  
Ian Holstead ◽  
John Turri ◽  
Michael Barnett-Cowan

AbstractWe assessed how self-motion affects the visual representation of the self. We constructed a novel virtual reality experiment that systematically varied an avatar’s motion and also biological sex. Participants were presented with pairs of avatars that visually represented the participant (“self avatar”), or another person (“opposite avatar”). Avatar motion either corresponded with the participant’s motion, or was decoupled from the participant’s motion. The results show that participants identified with i) “self avatars” over “opposite avatars”, ii) avatars moving congruently with self-motion over incongruent motion, and importantly iii) identification with the “opposite avatar” over the “self avatar” when the opposite avatar’s motion was congruent with self-motion. Our results suggest that both self-motion and biological sex are relevant to the body schema and body image and that congruent bottom-up visual feedback of self-motion is particularly important for the sense of self and capable of overriding top-down self-identification factors such as biological sex.


Sign in / Sign up

Export Citation Format

Share Document