Person:

Warneken, Felix

Humans constantly generate and solve ambiguities while interacting with each other in their every day activities. Hence, having a robot that is able to solve ambiguous situations is essential if we aim at achieving a fluent and acceptable human-robot interaction. We propose a strategy that combines three mechanisms to clarify ambiguous situations generated by the human partner. We implemented our approach and successfully performed validation tests in several different situations both, in simulation and with the HRP-2 robot.

In human-robot interaction, a robot must be prepared to handle possible ambiguities generated by a human partner. In this work we propose a set of strategies that allow a robot to identify the referent when the human partner refers to an object giving incomplete information, i.e. an ambiguous description. Moreover, we propose the use of an ontology to store and reason on the robot's knowledge to ease clarification, and therefore, improve interaction. We validate our work through both simulation and two real robotic platforms performing two tasks: a daily-life situation and a game.

Young children begin helping others with simple instrumental problems from soon after their first birthdays. In previous observations of this phenomenon, both naturalistic and experimental, children’s parents were in the room and could potentially have influenced their behavior. In the two current studies, we gave 24-month-old children the opportunity to help an unfamiliar adult obtain an out-of-reach object when the parent (or a friendly female adult) (i) was present but passive, (ii) was present and highlighted the problem for the child, (iii) was present and actively encouraged the child to help, (iv) was present and ordered the child to help, or (v) was absent from the room. The children helped at relatively high levels and equally under all these treatment conditions. There was also no differential effect of treatment condition on children’s helping in a subsequent test phase in which no parent was present, and children had to disengage from a fun activity to help. Young children’s helping behavior is not potentiated or facilitated by parental behavior in the immediate situation, suggesting that it is spontaneous and intrinsically motivated.

As many studies of cognition and behavior involve captive animals, assessing any psychological impact of captive conditions is an important goal for comparative researchers. Ferdowsian and colleagues (2011) sought to address whether captive chimpanzees show elevated signs of psychopathology relative to wild apes. They modified a checklist of diagnostic criteria for major depression and posttraumatic stress disorder in humans, and applied these criteria to various captive and wild chimpanzee populations. We argue that measures derived from human diagnostic criteria are not a powerful tool for assessing the psychological health of nonverbal animals. In addition, we highlight certain methodological drawbacks of the specific approach used by Ferdowsian and colleagues (2011). We propose that research should (1) focus on objective behavioral criteria that account for species-typical behaviors and can be reliably identified across populations; (2) account for population differences in rearing history when comparing how current environment impacts psychological health in animals; and (3) focus on how changes in current human practices can improve the well-being of both captive and wild animals.

This study investigated young children’s commitment to a joint goal by assessing whether peers in collaborative activities continue to collaborate until all received their rewards. Forty-eight 2.5- and 3.5-year-old children worked on an apparatus dyadically. One child got access to her reward early. For the partner to benefit as well, this child had to continue to collaborate even though there was no further reward available to her. The study found that 3.5-year-olds, but not 2.5-year-olds, eagerly assisted their unlucky partner. They did this less readily in a noncollaborative control condition. A second study confirmed that 2.5-year-old children understood the task structure. These results suggest that children begin to appreciate the normative dimensions of collaborative activities during the 3rd year of life.

Some children’s social activities are structured by joint goals. In previous research, the criterion used to determine this was relatively weak: if the partner stopped interacting, did the child attempt to re-engage her? But re-engagement attempts could easily result from the child simply realizing that she needs the partner to reach her own goal in the activity (social tool explanation). In two experiments, 21- and 27-month-old children interacted with an adult in games in which they either did or did not physically need the partner to reach a concrete goal. Moreover, when the partner stopped interacting, she did so because she was either unwilling to continue (breaking off from the joint goal) or unable to continue (presumably still maintaining the joint goal). Children of both age groups encouraged the recalcitrant partner equally often whether she was or was not physically needed for goal attainment. In addition, they did so more often when the partner was unable to continue than when she was unwilling to continue. These findings suggest that young children do not just view their collaborative partners as mindless social tools, but rather as intentional, cooperative agents with whom they must coordinate intentional states.

Human adults will sometimes help without being asked to help, including in situations in which the helpee is oblivious to the problem and thus provides no communicative or behavioral cues that intervention is necessary. Some theoretical models argue that these acts of ‘proactive helping’ are an important and possibly human-specific form of prosociality. Two experiments examined whether young children proactively help in a situation where an adult did not provide any concurrent behavioral cues that help was needed. Specifically, in Experiment 1 an experimenter either dropped an object without noticing (experimental condition) or on purpose (control). Even though children were bystanders engaged in their own task, they spontaneously intervened by helping instrumentally in the experimental condition in the absence of concurrent behavioral cues from the actor (significantly more often than in the control condition). These acts increased significantly from 21 to 31 months of age, probably reflecting children’s emerging social-cognitive capacities to represent goal-directed action. Experiment 2 replicated proactive helping in 2-year-olds in a more closely matched comparison in which in both experimental and control conditions the actor did not notice the accident, and children thus had to infer whether help was needed from the actor’s previous responses alone. This result shows that children are able to infer a need for intervention on concurrent situational cues, without behavioral or communicative cues by the helpee. These results indicate that proactive prosociality might be a characteristic of early human ontogeny, emerging in children as young as two years of age.

Robots should be capable of interacting in a cooperative and adaptive manner with their human counterparts in open-ended tasks that can change in real-time. An important aspect of the robot behavior will be the ability to acquire new knowledge of the cooperative tasks by observing and interacting with humans. The current research addresses this challenge. We present results from a cooperative human–robot interaction system that has been specifically developed for portability between different humanoid platforms, by abstraction layers at the perceptual and motor interfaces. In the perceptual domain, the resulting system is demonstrated to learn to recognize objects and to recognize actions as sequences of perceptual primitives, and to transfer this learning, and recognition, between different robotic platforms. For execution, composite actions and plans are shown to be learnt on one robot and executed successfully on a different one. Most importantly, the system provides the ability to link actions into shared plans, that form the basis of human–robot cooperation, applying principles from human cognitive development to the domain of robot cognitive systems.

This chapter discusses motivational factors and the contributors responsible for the empathic and prosocial behavior of young children. The reasons that people engage in prosocial behaviors, including self-benefit and society’s approval, are discussed. Empathy as an underlying prosocial behavior, along with its associated process sympathy, is studied to determine whether these are the factors responsible for young children’s prosocial behavior. The chapter discusses young children’s understanding of the situational cues and responding empathically to them. Instrumental helping acts of children aimed at helping other people in attaining their goals rather than looking for self-benefit are discussed, as is the role of culture and experience in influencing young children’s prosocial behavior.

We compared the performance of 3- and 5-year-old children with that of chimpanzees in two tasks requiring collaboration via complementary roles. In both tasks, children and chimpanzees were able to coordinate two complementary roles with peers and solve the problem cooperatively. This is the first experimental demonstration of the coordination of complementary roles in chimpanzees. In the second task, neither species was skillful at waiting for a partner to be positioned appropriately before beginning (although children did hesitate significantly longer when the partner was absent). The main difference between species in both tasks was in children's, but not chimpanzees’, ability to profit from experience as a collaborator in one role when later reversing roles. This difference suggests that as they participate in a collaboration, young children integrate both roles into a single “birds-eye-view” representational format in a way that chimpanzees do not.