BBC News recently reported on a pretty interesting three-year long research project that is not the typical non-human primate-focused research we usually highlight on this site, but I couldn’t resist bringing it up. The project, FEELIX Growing, a multi-national project is aiming to create robots that read and react to humans in an appropriate manner.

Dr. Lola Canamero, coordinator of the project, from the University of Hertfordshire:

“We are most interested programming and developing behavioural capabilities, particularly in social and emotional interactions with humans.”

Here’s the project summary:

“If robots are to be truly integrated in humans’ everyday environment in order to provide services such as company, caregiving, entertainment, patient monitoring, aids in therapy, etc., they cannot be simply designed and taken off the shelf to be directly embedded into a real-life setting. Adaptation to incompletely known and changing environments and personalization to their human users and partners are necessary features to achieve successful long-term integration. This integration would require that, like children (but on a shorter time-scale), robots develop embedded in the social environment in which they will fulfill their roles. The overall goal of this project is the interdisciplinary investigation of socially situated development from an integrated or global perspective, as a key paradigm towards achieving robots that interact with humans in their everyday environments in a rich, flexible, autonomous, and user-centred way. To achieve this general goal we set the following specific objectives:

  1. Identification of scenarios presenting key issues and typologies of problems in the investigation of global socially situated development of autonomous (biologically and robotic) agents.
  2. Investigation of the roles of emotion, interaction, expression, and their interplays in bootstrapping and driving socially situated development, which includes implementation of robotic systems that improve existing work in each of those aspects, and their testing in the key identified scenarios.
  3. Integration of (a) the above capabilities in at least 2 different robotic systems, and (b) feedback across the disciplines involved.
  4. Identification of needs and key steps towards achieving standards in: (a) the design of scenarios and problem typologies, (b) evaluation metrics, (c) the design of robotic platforms and related technology that can be realistically integrated in people’s everyday life.

FEELIX GROWING takes a highly interdisciplinary approach that combines theories, methods, and technology from developmental and comparative psychology, neuroimagery, ethology, and autonomous and developmental robotics, to investigate how socially situated development can be brought to robots that grow up and adapt to humans in everyday environments. We expect to have a significant impact on the scientific community, on two grounds. On the one hand, our research focus poses an important and as-yet largely unexplored scientific question that is increasingly recognized as a keystone in the development of human-oriented social technology and in the understanding of humans, and can contribute to the advancement of entertainment, developmental, service, and rehabilitation robotics. On the other hand, our strongly interdisciplinary effort could make important contributions to a number of disciplines and set the grounds towards long-term collaborations among them.”

The description of the robot itself is quite reminiscent of a family pet (or a service animal) in that it is

“to be truly integrated in humans’ everyday environment in order to provide services such as company, caregiving, entertainment, patient monitoring, aids in therapy, etc.”

With that in mind, I couldn’t help but think about the vastly different perceptions science has on the animal mind (some may say that I’m a bit obsessed about it). It’s fascinating that we as a scientific community are on the verge of creating a machine (or more specifically software) with the capabilities to “learn from humans and respond in a socially and emotionally appropriate manner,” yet frequently deny this ability in non-human animals.

It’s frustrating to work with a scientist that comes into their lab telling stories of how when he got home last night, he could immediately tell that his golden retriever, Harry, did something wrong based upon the look on his face alone… and then proceed to close off all thought about the possibility that his subjects share those emotions. He of course will say that he is simply being objective. But couldn’t one argue that not taking into account all of your subject’s abilities be a hindrance to your objectivity?

This is not to suggest that we should bias our opinions based on what happens at home, but to suggest that we as humans are the only emotional beings appears to be a bit arrogant. I feel that not taking a chance and hiding behind scientific precedents (potentially to save one’s career) is a hindrance upon what we can discover. Sure new theories and thoughts can be intimidating, but exploring those in an objective manner is what we thrive upon… it’s why we do what we do.

Please don’t take this the wrong way, I’m not saying that this is something I’m seeing in the FEELIX Growing project, it’s just some thoughts that have popped up while reading about their interesting project… and it leaves me wondering what will come from this project in terms of the animal mind.

Advertisements