Social Robotics and Engineering Students: Do They Match? Does Culture Matter?

This inquiry investigates whether social robotics is an attractive learning and research domain for Electrical Engineering (EE) and Computer Science (CS) Students at Heriot-Watt University (HWU) in the UK and Electrical Engineering Students at Xidian University (XDU) in China? Social robotics is an emergent research domain that brings together engineering and social science researchers and requires knowledge of these two domains. In the core of social robotics lies the question of how we can make the robots more interactive with humans by equipping them with technologies to perceive, interpret, and reason about human activities and at the same time react to these human activities in ways that are similar to and/or acceptable by humans [8, 11]. In other words, social robotics studies the so-called 3C tasks, “tasks that require human–robot communication, coordination, or collaboration” [20].

Due to the rising interest in research, scholars have started thinking about including social robotics also in education: “Physical robotics has been very popular and effective in…education…, but social robotics in education is almost completely unexplored” [20]. Mead and Mataric [20] investigated using social robotics in middle school classrooms to address the needs of students undergoing changes at the stage of biological development. Edwards et al. [6] tested a social robot for delivery of lessons in a college classroom. Bolea et al. [3] shared the idea that “the bridge from research to teaching is not clear in social robotics”. They performed a study towards integrating social robotics with a humanoid robot setup into a taught laboratory in the Industrial Engineering Degree. Ruiz-del-Solar [22] used social robots as keynote speakers in the outreach activities towards students in order to attract them to the electrical, electronic, and computer engineering degrees.

Cultural background is an important factor affecting how people perceive robots, especially considering the acceptability of robots in social life [16] and considering how people interact with them [9]. Lee and Sabanovic [17] conducted a survey in South Korea, Turkey, and the United States in order to investigate the cultural dependencies of preferences for the design and use of robots in daily life. Their study emphasized “the importance of performing cross-cultural studies of robots that…delve into…contextually and socially situated dynamics of sense making and interaction” [17]. Li et al. [19] demonstrated that cultural differences exist in people’s perception of likeability, engagement, trust, and satisfaction in human–robot interaction. Rehm suggested that cultural differences were present even in the development of robotic systems “unconsciously influencing how robots look, what we envision with them to do, and how they are programmed to interact with the user” and demonstrated that culture influences how people design affective body movements of a small humanoid robot [21]. Similarly, Shaw-Garlock [24] presented an in-depth discussion of the differences in the seminal work of Cynthia Breazeal from North America and Hiroshi Ishiguro from Japan on humanoid social robots. Haring et al. [14] observed differences in perception of different types of robots by Japanese and Australian university students. In a review paper on social robotics, Dunstan et al. [5] highlighted “cultural robotics” to stress the role of social robots as participant and also creators of culture.

As a roboticist at HWU, I have been particularly interested in how much our students at the engineering departments are familiar with and interested in social robotics. The answer to this question is not trivial because social robotics is highly interdisciplinary in a somewhat uncommon way that the interdisciplinarity is not between two technical domains but between two different research paradigms, engineering and social science. Breazeal et al. [2] stated that social robotics “requires a multidisciplinary approach where the design of social robot technologies and methodologies are informed by robotics, artificial intelligence, psychology, neuroscience, human factors, design, anthropology, and more”. Eyssel [10] stated that social roboticists’ “interest in human psychology is driven by the motivation to identify principles of human–human interaction in order to extract and model these onto a given platform to mirror psychological processes in a technical system.” In the context of social robotics, she added that “Ph.D. candidates in engineering and related fields, …, are commonly required to provide an empirical ‘‘evaluation’’ study along with the platform they have engineered, designed or refined. …At the same time, it appears that Ph.D. candidates…may lack the academic training in research methods, experimental design, and statistics” [10]. In that respect social robotics requires not only engineering knowledge, but also knowledge and skills of field studies, questionnaires, statistical analyses, and qualitative and interpretive analyses. Therefore, the familiarity and interest of engineering students towards social robotics is linked to their familiarity with and interest in these broad range of studies. In this survey, I aim at investigating the familiarity with and interest in social robotics through questions not only focusing directly on social robotics per se but also with questions on psychology, user studies, questionnaires, and statistics.

The joint undergraduate program on Telecommunications Engineering between HWU and XDU provides a rare opportunity to study the impact of cultural background of engineering students, in contrast to the impact of educational background. This is because Electrical Engineering students at HWU and the joint Telecommunications Engineering program students at XDU follow exactly the same curriculum in their first 3 years: the content and style (lecture slides) of the courses, the coursework, and the exams are the same. Moreover, one-third of the courses of the joint Telecommunications Engineering program at XDU are delivered by the staff of HWU. These two group of students constitute a pool where the impact of difference in university education is largely eliminated and any observed difference could be more easily attributed to the cultural background. On the other hand, the Electrical Engineering and Computer Science Students in HWU constitute a pool of subjects where the cultural background can be assumed to be similar whereas the educational background differs. A pool of second year students in the two disciplines of HWU and in the joint program of XDU with HWU allow me to study isolated impacts of cultural background and educational background on the question of engineering students’ familiarity with and interest in social robotics. To my knowledge, there is no literature focusing specifically on this question, particularly through a controlled comparison where educational and cultural background are largely isolated from each other. Therefore, the results of this study is important to provide evidence on the impact of cultural background in perception of engineering students about social robotics and showing how they differ across the two cultures studied.

This study was motivated by two reasons that pertain to our specific circumstances at HWU. First, we as the academics in robotics at HWU are increasingly including social robotics topics in our research and therefore we would like to know whether this is appealing to our students in their undergraduate projects and post-graduate studies, both in Computer Science and Electrical Engineering. Besides that, traditionally social robotics is associated with Computer Science rather than Electrical Engineering and therefore there is more social robotics activity going on in the Computer Science department than in the Electrical Engineering department at HWU. We are curious whether social robotics is as familiar and interesting to the Electrical Engineering Students as to the Computer Science Students. This comparison has been motivated by my observation that different educational programs might result in different world-views with students [7]. Accordingly I hypothesised that:

Second, at HWU, we have a joint engineering program with the XDU in Xi’an, China. Starting from 2018, each year a cohort of electrical engineering students from XDU have been taught in HWU in Edinburgh in their fourth years. They are conducting student projects some of which are part of the running research projects and they will potentially like to pursue post-graduate studies at HWU. For these purposes, we would like to know the level of interest and familiarity of our Chinese students towards social robotics in comparison to our local students and would like to develop and propose projects according to their interests. The results of this survey will inform us about whether our students are familiar with and interested in social robotics and whether there is any difference in the level of interest between the students from Edinburgh and Xi’an. Accordingly I hypothesised that:

I believe that the results of this study are interesting also for the social robotics community at large. The results related to the impact of cultural background in perception of social robotics might be of interest to even a larger community including social scientists who focus on ethnicity and life-style factors in perception of technology, and to market analysts of industry commercializing interactive robotic products.

In this study, I aim at understanding the students' level of familiarity with and interest in social robotics. The study does not aim at measuring their actual level of knowledge of social robotics concepts; rather the interest here is to understand the level of their general familiarity with and interest in social robotics based on what they have heard and got to know from their surroundings and regardless of their having learned of any technical details. The survey was conducted on students with the same level of technical education in the two universities. This approach makes the assumption that the social and cultural environment is effective in order to shape the motivation and interest of students towards an emergent field like social robotics, and this environment makes a difference in that respect regardless of the students having previously acquired the same level and similar education. The familiarity and interest that result from the environmental impacts can be captured by simply asking the students whether they would agree or not agree with the related statements and letting them express themselves in relation to those.

Different societal causes might impact students to have high or low interest in an emergent domain of study. Lee and Sabanovic [17] investigated the impact of religion and media exposure on the preferences towards social robots (though did not reach conclusive results) and they referenced to the reports [15, 25] that suggest that the active efforts of the government, industry, and academia have been influential on social acceptance of robotics in Japan. Walden et al. [26] reported that the mental model of seniors about robots were “shaped by their recent experiences with advanced communications technologies and mediated representations of robots in popular culture”. Destephe et al. [4] demonstrated with a survey that the attitude of the participants towards robots in general affected their perception of the specific humanoid robot they experimented with: “Participants who had positive views toward robots rated our robot and its motions less eerie and more attractive than those with negative views”. Bartneck et al. [1] conducted a cross-cultural survey in seven countries which suggested that people from different cultures had different attitudes towards robots. Lee et al. [18] identified variable cultural models of robotic technology and variable cultural meaning of home context, which shaped the different expectations and preferences of participants of different culture from future domestic robots.

In the current study I aim at showing that the cultural background—which I consider as a compound of all the societal causes surrounding the individual in their daily-life—has an impact in shaping the preferences of undergraduate students. My assumption is that as a result of the impact of such various societal causes, ontologically [12], a cultural background exists and surrounds the students to shape their preferences in choosing the study matters. Epistemologically [12], we can measure the impact of this cultural background by investigating the level of interest in and familiarity with a specific topic.

The qualitative analysis was based on the responses of the students to the open-ended questions in the survey. After a first reading of all the responses, I realized that many of the XDU Students used words and phrases that referred to an understanding and willingness of using robots as companions. The HWU students, on the other hand, did not use similar words and phrases; rather, they mostly used words and phrases that referred to an understanding and willingness of using robots as utility equipment, such as using in the factories, for repetitive and dull tasks, for cleaning, and in dangerous situations. These latter were shared also by some of the XDU students, but the interesting point was that none of the HWU students expressed any inclination towards an understanding and wiliness of using robots as companions. After having this first impression, I extracted the phrases used by XDU students given below. 17 of the 46 XDU students expressed at least one or more of these words and phrases in their responses. None of the HWU students expressed any of these or any other words or phrases similar to these.

The findings presented here point out to the following conclusions.

This observation implies for our local context at Heriot-Watt University that we can pursue running social robotics projects and encourage our students to take social robotics topics in their projects and post-graduate studies.

This observation is important in order to eliminate the hesitations for introducing social robotics in Electrical Engineering Department in comparison to Computer Science Department at HWU. The results of this survey show that Electrical Engineering Students are as familiar with and interested in social robotics as the Computer Science Students.

This observation has both local and general implications. The local implication for the Electrical Engineering Department at Heriot-Watt University is that we can perhaps attract more students to social robotics studies from among the Xidian University students when they spend their fourth year of the joint program in the Heriot-Watt University Campus in Edinburgh. This might be an opportunity to increase the number and depth of social robotics research at Heriot-Watt University. The more general implication of this observation is related to the following observation.

Both the qualitative analysis presented above and the difference in the quantified responses to the questions P1-P2 support this observation. Therefore, this study provides combined qualitative and quantitative features which “contribute complementary form of evidence in addressing the same issue” [13].

This observation provides evidence to the more general argument that robots are more easily perceived as companions in some cultural environments compared to some others. This implies that social robotics research, particularly the ones that perform user studies, should take into account the cultural background of the context the research is being conducted. The results obtained in one particular cultural context might not necessarily generalize to another, or a research or technology output that shows success in one cultural context, might not necessarily achieve the same success in another context. This observation also implies that, as argued by Lee and Sabanovic [17], the development and marketing strategies for socially interactive robot technologies should seriously take into account the cultural background of their target end-users.

This study investigated the level of familiarity and interest of students towards social robotics through a survey conducted with the Electrical Engineering and Computer Science students at Heriot-Watt University and Electrical Engineering students at Xidian University. The results indicated that whereas there is no significant difference in the level of familiarity across the three groups of students and no significant difference in the level of interest between the Electrical Engineering and Computer Science students at Heriot Watt University, there is a statistically significant difference in the level of interest towards social robotics between the Heriot-Watt University and Xidian University students. Xidian University students demonstrate a higher level of interest towards social robotics. The qualitative analysis conducted over the responses of the students revealed that many of the Xidian University students were motivated to perceive and have robots as companions whereas none of the Heriot-Watt University students showed such or similar tendency. This observation strongly supports the claim that cultural background plays a significant role in determining the interests and preferences of students towards an emergent research and study field such as social robotics. This observation also supports that cultural background should be taken into account while conducting research, development, and marketing activities related to robots that interact with people.