What Makes a Social Robot Good at Interacting with Humans?

Onyeulo, Eva Blessing; Gandhi, Vaibhav

doi:10.3390/info11010043

Cited by 56 publications

(24 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The type of robot or its social role (teacher, peer, friend) did not matter, either (cf. [43]): A more human-like machine did not improve performance, a teacher role was no better than a peer, and the level of emotional bonding of the child with the tutoring machine (e.g., as it is new and beautiful) had no significant effect on their learning outcomes. This is good news for teaching practice (cf.…”

Section: Discussionmentioning

confidence: 99%

Robot Tutoring of Multiplication: Over One-Third Learning Gain for Most, Learning Loss for Some

et al. 2021

View full text Add to dashboard Cite

In the design of educational robots, it appears to be undecided as to whether robots should show social behaviors and look human-like or whether such cues are insignificant for learning. We conducted an experiment with different designs of social robots built from the same materials, which is unique in robotics research. The robots rehearsed multiplication tables with primary school children in Hong Kong, which is a user group not easily or often accessed. The results show that affective bonding tendencies may occur but did not significantly contribute to the learning progress of these children, which was perhaps due to the short interaction period. Nonetheless, 5 min of robot tutoring improved their scores by about 30%, while performance dropped only for a few challenged children. We discuss topics, such as teaching language skills, which may be fostered by human likeness in appearance and behaviors; however, for Science, Technology, Engineering, and Mathematics (STEM)-related subjects, the social aspects of robots hardly seem to matter.

show abstract

Section: Discussionmentioning

confidence: 99%

Robot Tutoring of Multiplication: Over One-Third Learning Gain for Most, Learning Loss for Some

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The type of robot or its social role (teacher, peer, friend) also did not matter (cf. Onyeulo & Gandhi, 2020): A more human-like machine did not improve performance, a teacher role was no better than a peer, and the level of emotional bonding of the child with the tutoring machine (because it is new and beautiful) made no difference for their learning outcomes. This is good news for teaching practice (cf.…”

Section: Discussionmentioning

confidence: 99%

“…Robots are not human beings (cf. Onyeulo & Gandhi, 2020). It may be that a warm relationship with a human teacher makes a child want to work harder and may improve its social-emotional development (e.g., Frymier & Houser, 2000;Hattie, 2015;Skinner & Belmont, 1993;Hamre & Pianta, 2001).…”

Section: Discussionmentioning

confidence: 99%

Robot Tutoring Multiplications: Over One-Third Learning Gain for Most, Learning Loss for Some

Hoorn¹,

Huang²,

Konijn³

et al. 2020

Preprint

View full text Add to dashboard Cite

In the design of educational robots, it seems undecided whether robots should show social behaviors and look human-like or that such cues are indifferent to learning. We conducted an experiment with different designs of social robots, rehearsing the multiplication tables with primary school children in Hong Kong. Results show that affective bonding tendencies may occur but did not significantly contribute to the learning progress of these children, perhaps due to the short interaction period. Nonetheless, 5 minutes of robot tutoring improved their scores with about 30% and only for a few challenged children, performance dropped. We discuss that topics such as teaching language skills may be fostered by human likeness in appearance and behaviors but that for STEM-related subjects, the social aspects of robots hardly matter.

show abstract

“…Hence, the use of a fully transparent visor is beneficial if those are considered as protective, as it will not only provide protection against infectious disease but will at the same time help in improving the interaction between artificial systems and humans since people like to interact with social robots which can identify emotion (Breazeal et al, 2008;Onyeulo and Gandhi, 2020) than with robots that cannot identify emotions.…”

Section: Discussionmentioning

confidence: 99%

Emotion Categorization from Faces of People with Sunglasses and Facemasks

Shehu

Browne

Eisenbarth

2021

Preprint

View full text Add to dashboard Cite

Emotional information is considered to convey much meaning in communication. Hence, artificial emotion categorization methods are being developed to meet the increasing demand to introduce intelligent systems, such as robots, into shared workspaces. Deep learning algorithms have demonstrated limited competency in categorizing images from posed datasets with the main features of the face being visible. However, the use of sunglasses and facemasks is common in our daily lives, especially with the outbreak of communicable diseases such as the recent coronavirus. Anecdotally, partial coverings of the face reduces the effectiveness of human communication, so would this have hampering effects on computer vision, and if so, would the different emotion categories be affected equally? Here, we use a modern deep learning algorithm (i.e. VGG19) to categorize emotion from faces of people obscured with simulated sunglasses and facemasks. We found that face coverings obscure emotion categorization by up to 74%, whereby emotion categories are affected differently by different coverings, e.g. clear mouth coverings have little effect in categorizing happiness, but sadness is affected badly. While an overall accuracy of up to 97% has been achieved with nothing added to the face, the achieved accuracy decreases in all other cases when the face is obscured. Notably, clear visors have only a small effect across all emotions, where the classifier achieved an accuracy of up to 89.0% compared to other types of facemasks in which the achieved accuracy is less than 36%.

show abstract

What Makes a Social Robot Good at Interacting with Humans?

Cited by 56 publications

References 54 publications

Robot Tutoring of Multiplication: Over One-Third Learning Gain for Most, Learning Loss for Some

Robot Tutoring of Multiplication: Over One-Third Learning Gain for Most, Learning Loss for Some

Robot Tutoring Multiplications: Over One-Third Learning Gain for Most, Learning Loss for Some

Emotion Categorization from Faces of People with Sunglasses and Facemasks

Contact Info

Product

Resources

About