Are Robots Autistic?

Are Robots Autistic?

Neha Khetrapal
Copyright: © 2010 |Pages: 8
DOI: 10.4018/jse.2010070104
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

This paper discusses the implications of the embodied approach for understanding emotional processing in autism and the consequent application of this approach for robotics. In this pursuit, author contrasts the embodied approach with the traditional amodal approach in cognitive science and highlights the gaps in understanding. Other important issues on intentionality, intelligence and autonomy are also raised. The paper also advocates a better integration of disciplines for advancing the understanding of emotional processing in autism and deploying cognitive robotics for the purpose of developing the embodied approach further.
Article Preview
Top

Embodied Theories Vs Amodal Approach

The traditional symbolic approaches for understanding information processing maintained that information is initially encoded in basic sense modalities but in order to interact with higher level cognitive processes these basic codes need to be “transduced” or changed into amodal language like symbols. Later the embodied theories were propounded as an alternative to the traditional approach where the basic sensory modalities play an important role for both lower and higher order information processing (Barsalou, 1999, 2008; Wilson, 2002). According, to the embodied view any cognitive activity does not require a separate set of representations except for what is contained within the modality specific systems. Within the framework of amodal approaches, the mind is usually described as a highly abstract entity with little connections to the outside world which stands in stalk contrast to the embodied view where the very interaction of the body with the environment is seen as important. In the latter, cognitive activity becomes situated as it takes place in a real world context3.

More recently, the amodal approach has been described as inadequate to explain the workings of social cognition (see Niedenthal, Barsalou, Winkielman, Krauth-Gruber, & Ric, 2005). Traditionally these theories advocated a representation of the outside world as the basis for symbol manipulation. For instance, repeated encounters with a “similar” kind of social situation leads to re-description of the experienced events into an abstract code that permanently gets stored in memory (see Collins & Quillian, 1969) supporting social inferences, reasoning and decision making (examples include semantic networks, schemata and propositions). The most popular reason to adhere to the amodal approach is its ease of implementation in computer simulation (e.g., intelligent systems) even though it has been hard to obtain solid neural and cognitive empirical support for the re-description process.

More specifically, the embodied framework has also been applied for understanding emotional information processing (Niedenthal, 2007). The amodal view would explain the perceptual and conceptual processing of emotion in a similar manner as the processing of other neutral objects (Ortony, Clore, & Collins, 1988) but the embodied theories emphasize the importance of both central and peripheral resources that support emotional information processing for perceptual and conceptual processing of emotion (see Winkielman, McIntosh, & Oberman, 2009). For instance, thinking about an emotional event will recruit the same bodily reactions and brain regions that are helpful for processing an actual/real emotional event and on parallel lines, understanding the emotional reactions of others is dependent upon similar mechanisms that support the same emotional reactions in one’s own self. An important stance of the embodied approaches for emotional processing is the close relation between the peripheral and central processes (Damasio, 1994), where similar modal representations could be generated peripherally or centrally (Barsalou, 1999, 2008).

Complete Article List

Search this Journal:
Reset
Volume 11: 2 Issues (2020)
Volume 10: 2 Issues (2019)
Volume 9: 2 Issues (2018)
Volume 8: 2 Issues (2017)
Volume 7: 2 Issues (2016)
Volume 6: 2 Issues (2015)
Volume 5: 2 Issues (2014)
Volume 4: 2 Issues (2013)
Volume 3: 2 Issues (2012)
Volume 2: 2 Issues (2011)
Volume 1: 2 Issues (2010)
View Complete Journal Contents Listing