Clear Implications for Anthropomorphism in Marketing Campaigns - Literature review Example

?INTRODUCTION DEFINITION Anthropomorphism is imagining nonhumans to be human-like – anthropomorphism means that non-human beings or robots are ascribed the characteristics, intentions, motivations and emotions of a human (Goodrich & Allen, 2007). Anthropomorphism can extend to anything which is non-human, including forces of nature, animals, religious deities and electronic and mechanical devices (Epley et al. 2007). Another kind of anthropomorphism concerns anthropomorphic form. According to DiSalvo et al. (2005), this encompasses either structural, character, aware or gestural anthropomorphic form. In a structural anthropomorphic form, shapes, mechanism, volumes and arrangements are created to approximate the human anatomy. An example of this is a pose-able artist’s model of the female nude. In gestural anthropomorphic form, the form approximates the way that humans communicate with others. DiSalvo et al. (2005) states that the Jean–Paul Gaultier “Le Male” perfume bottle, which captures a male in a certain style of dress and specific traits, is an example of this. Aware anthropomorphic form is the other form cited by DiSalvo et al. (2005), and this means a form which can imitate human’s thought capacities, as well as human’s capacity for inquiry. DiSalvo et al.,(2005) states that a good example of this would be a robot which is designed for this function. THEORIES OF ANTHROPOMORPHISM. Epley et al. (2007) state that there are cognitive and motivational determinants of anthropomorphism. They state that, for instance, one of the cognitive determinants of anthropomorphism is inductive inference. This means that the human who sees a non-human agent interprets the behavior, mentality and verbal reports of the non-human agent within an existing theoretical framework. So, the human’s theoretical framework is what that person feels and thinks, or what other humans feel and think, and these are the feelings and thoughts which are ascribed to the non-human agent. Another factor which would determine the amount of anthropomorphism a given human would attribute to a non-human is the need for cognition, which refers to how individuals structure their world and how engaged they are in intellectual exercises and pursuits. Epley et al. (2007) state that humans who have a high need for cognition would be less likely to anthropomorphize, because they are more likely to look for other reasons why a non-human agent would behave the way that they do. They will discount their own framework in viewing non-human agents, and look at external stimuli and other theories for their anchor on non-human behavior. Those with a low need for cognition, which means that the individual is less likely to intellectualize situations and events, will be more likely to anthropomorphize, because they are more likely to use existing frameworks to ascribe attributes to the non-human (Epley et al., 2007). Another factor about why people anthropomorphise is perceived similarity (Epley et al., 2007). This means that the more that the non-human resembles a human, the more likely the human is to anthropomorphize. There are two subfactors to perceived similarity. One is that the more the non-human resembles the human in motion, the more likely the non-human will be anthropomorphized. For instance, Epley (2007) states that robots which have the gait and speed of a human are more likely to be anthropomorphized than robots who are slower and have dissimilar gaits to humans. The other sub-factor is morphology, which means that the more human-like the non-human observable features, the more likely that non-human will be anthropomorphized. A cartoon character which looks human will anthropomorphized, while a cartoon character that has exaggerated features, or other non-human features will not be (Epley, 2007). Epley et al. (2007) also state that development plays a role in anthropomorphism. A human must have a sense of self and of others and have a developed theory of the mind in order to anthropomorphize. Moreover, young children anthropomorphize more than adults, because adults have a developed sense of alternate theories which are not egocentric. In other words, children do not have the experience with objects, so they do not have alternate explanations for how things work. Barring this, they ascribe human motivations to non-human objects such as clouds. To a child, the cloud is “angry.” To adults, who have experience with clouds, therefore understand that dark clouds mean storms, the clouds are not angry (Epley et al, 2007). The final cognitive theory regards cultural influences. These influences provide the individual with norms and ideologies which explain their environment and world. So, Epley et al. (2007) states that some cultures use non-human animals in their everyday life, therefore they are less likely to anthropomorphize these animals than somebody who is of a culture which does not use animals for work. On the other hand, some cultures interact on a daily basis with sophisticated electronics, such as cars and computers, and are less likely to anthropomorphize these objects than are individuals in cultures that do not interact with these technologies (Epley et al., 2007). Social motivation is another motivation, and this regards man’s need to be socially connected to others (Epley et al., 2007). Caporael & Heyes (1997) state that humans have evolved to have a need for group living, and that the interface between habitat and the individual is a group process (Caporael & Heyes, 1997). If a human does not have this social need fulfilled with other humans, the human will anthropomorphize a pet because this anthropomorphism fulfills that human’s need for social contact (Caporael & Heyes, 1997). An anthropomorphism of God fulfills the same need, as the human connects with God as a substitute for other humans. Anthropomorphism in particular occurs among humans who are somehow socially disconnected from other humans (Epley et al., 2007) Scassellati (2002) offers another theory as to why people anthropomorphize, and he states that humans have abilities which are collectively known as the “theory of the mind.” This means the ability to mentalize, which means that we can understand “the actions and expressions of others within an intentional or goal-directed framework” (Scassellati, 2002). Discussion REASONS WHY PEOPLE ANTHROPOMORPHIZE Motivational determinants are a broad category espoused by Epley et al. (2007). One motivational determinant is the need to reduce uncertainty. If one can anthropomorphize the unexplainable, such as when Hurricane Katrina was ascribed by New Orleans Mayor Ray Nagin as the expression of God’s wrath regarding the United States being in Iraq and regarding African-Americans not taking care of themselves. In this way the hurricane could be explained as something other than a random occurrence (Epley et al., 2007). Related to this is the need for closure and the need for control. With a need for closure, an individual needs a definitive answer on a given topic, and turns to anthropomorphism for that answer. These individuals do not do well with unexplained events or phenomenon, and need to have answers to ambiguities. Epley et al. (2007) states that individuals who need answers often form conclusions based upon information readily available and are slow to correct this perception when different information presents itself. This makes the individual more likely to anthropomorphize, because information received about non-humans are framed in their readily available cognitive framework, which tends towards human traits (Epley et al., 2007). With the need for control, the individual feels the need to predict future interactions with the non-human (Epley et al., 2007). DEHUMANIZATION AND HOW THIS RELATES TO ANTHROPOMORPHISM Dehumanizing is the act of thinking of another individual or group of individuals as being less than human (Castano & Kofta, 2009). This may take the form of denying human qualities in the dehumanized individual or group, or by denying secondary emotions to the individual or group. Secondary emotions are, for example, nostalgia and humiliation (Castano & Kofta, 2009) . Central to dehumanization is the denial that the other person has a mind or awareness (Waytz & Epley, 2012). Loughnan & Haslam (2007) suggest that there are two distinctly different types of dehumanization. One is that individuals may be compared to animals, which means that they are seen as less intelligent than humans, as well as being uncivilized, amoral and coarse. Another type of dehumanization is that humans may be compared to automatons, which means that they are denied attributes which constitute human nature (Loughnan & Haslam, 2007). An example of dehumanization, according to Waytz & Epley (2012) were the Nazis. They had to dehumanize their subjects whom they tortured and starved, or else they would not have been capable of the atrocities that they committed on these people (Waytz & Epley, 2012). This is because dehumanization of others “licenses wrongdoing towards them” (Waytz et al., 2010, (A) p. 60 ). Waytz & Epley (2012) suggest that social connection actually enables dehumanization. The theory is that individuals need social connection. When they have a close social connection with other people, then it reduces their motivation to seek out other connections. Therefore, when people have close social connections with others, these people are more likely to see others outside their close circle as similar to oneself (Waytz & Epley, 2012). Those with close ties are therefore more likely to see others outside their circle as “other” (Demoulin, 2004). Costello & Hodson (2010) concurs with this analysis, stating that people are more empathetic to others who are similar to oneself. Dehumanization may extend to ethnicity or race, in that people who are outside one’s ethnicity or race are seen as not having the same human characteristics as the person who is viewing them (Bastian et al., 2011). Dehumanization may also extend to objectivism, in which individuals are seen as objects, and not humans (Haslam, 2006). Feminist writers state that this is the basis for pornography, as the women in these movies are seen as objects for the men. Disabilities is another category which is subject to dehumanization, as individuals with cognitive difficulties are characterized as “parasites” or animals. Medicine is another area which is subject to dehumanization, as modern medicine is seen as dehumanizing patients and treating them as nothing but a number. Technology is another domain, in which people are expected to have the efficiency and conformity of a machine (Haslam, 2006). Video games is another version of dehumanizing, according to Bastian et al. (2011). This is dehumanizing to real-life antagonists, and this leads to aggressive behavior towards these antagonists in real life. This is because, in the context of the violent video games, the goal is to be aggressive towards the video game target, which takes the form of a human. Once video game players enter the real world, they feel the same sense of aggressiveness towards actual humans, because of the dehumanizing effects of the video games (Bastian et al., 2011). Waytz et al. (2010 B) state that dehumanization and anthropomorphism are inversely related, in that the reasons why one person might dehumanize another is the opposite of why the same person might anthropomorphize another. They state that inverting the theories behind anthropomorphism might explain dehumanization. For instance, they state that individuals who anthropomorphize do so because the object of the anthropomorphism is similar to oneself. Dehumanization is the inverse of that, in that people are dehumanized because of dissimilarity to oneself. Social connection also bridges the gap between the two concepts. Feeling socially isolated leads to anthropomorphism, as people take comfort in God or pets when they are not closely connected with other humans. Feeling socially connected leads to dehumanization, as the people in the in-group are humanized and the people in the outgroup are dehumanized as “other” (Waytz et al., 2010 B). How dehumanization relates to anthropomorphism is that it offers an explanation as to why individuals do not anthropomorphize. At the same time, individuals who dehumanize others might anthropomorphize non-humans, if the individual who dehumanizes is isolated from other humans because of his or her beliefs. For instance, an individual who has hatred for ethnic groups might not be able make friends because of his or her extreme beliefs. In this case, the isolation might lead to anthropomorphism, and pets may be their only companion because of their misanthropy. EMPATHY The opposite of dehumanization is the concept of empathy (Allison et al., 2011). When one is empathetic, the person is able to stand in the shoes of another individual, and this helps make sense of that other person’s behavior, as well as enabling one to make a prediction about that person’s behavior and feelings, and to feel connected with the other person and appropriately respond to that other person (Allison et al., 2011). Riek et al. (2009) suggests that empathy is affected by anthropomorphism with regard to robots. In their study, they showed clips of different robots which increasingly resembled humans (Riek et al., 2009). Harrison (2007) suggests robots may resemble humans if they possess artificial intelligence that allow them to converse. Regarding Riek et al. (2009)'s study, one robot was a vacuum cleaner. Another was an LED robotic lamp which can use lights to convey colors and intensities. Andrew was another robot who was human-like with facial expressivity, a full range of movement and a mechanical sounding voice. Alicia was another robot who was the most human of the robots – she has a fully human appearance and a human-sounding voice. The participants of this study were then showed clips of people being either very cruel to the robots or neutral clips of the robots doing prosaic tasks, such as cleaning or cooking. The participants were then asked how sorry they felt for the robot when the study participant was being cruel to it by shouting at it, making it do embarrassing things and pushing them. The participants were also asked to imagine if there were an earthquake, and which robot would they save? The researchers found that the more human the robot, the more sorry the participants felt for them and the more likely they were to save that robot over all the others (Riek et al., 2009). This suggests that the Respondents in the study are anthropomorphizing the more human-like robots, therefore they felt more empathetic towards these human-like robots. This sense of empathy for robots may be applied to other kinds of technology and computers. For instance, if an individual has a new I-Phone with Siri, who is the human-sounding voice which instructs the I-Phone user, gives directions, takes commands, etc., losing this I-Phone might make the person feel more devastated than if they lost a simple phone. It would be akin to losing a friend, in this case. Gong (2008) also suggests that computers which are human-like will elicit a more social response from the individuals using that computer than other computers who do not have human-like characteristics. Gong (2008) conducted a study in which the participants of the study were shown a series of images from low anthropomorphism to high anthropomorphism. They found that the computer images which are considered to be either high anthropomorphistic or actually portrayed a real human were judged as being more trustworthy, competent and influential. Powers et. al (2005) concurs with this finding that robots may be anthropomorphized, and extended the findings to gendered robots. What they found was that female robots were assumed to have human female characteristics, and vice versa for male robots. In their study, the participants were introduced to different robots who were male or female, depending upon the characterization of the particular robot. Their hypothesis was that the participants in this study would assume that the robots would have the same basic knowledge as they do. The females would assume that the female robot has similar knowledge to them, and the males would assume the same for the male robot. The researchers then asked the subjects to ascertain how much knowledge each of the robots had regarding dating. The males thought that the male robot had more knowledge about dating than the female robot, and vice-versa. This suggests that each of the genders assumed that the same gendered robot had the same amount of knowledge as they themselves did, and vice-versa for the different gendered robot. The researchers state that this is an example of the “common ground” theory, which means that individuals ascribe similar characteristics on those who are seen as occupying common ground with oneself. This relates to anthropomorphism and empathy, in that the robots were assumed human characteristics stereotypically common to males and females, which is why the male and female robots were ascribed the characteristics they were. (Kiesler & Goetz (2002) similarly measured anthropomorphism that people held for robots. In their study, the participants conversed with a toy robot vehicle or a toy robot man, then answered interview questions about the robot in question. They found, perhaps surprisingly, that the appearance of the robot did not affect anthropomorphism. Moreover, Nass & Moon (2000) state that, while individuals understand that computers are not humans, nonetheless, the participants in their study unconsciously subscribed social rules and motivations to these computers. In their study, they first determined if the computers would elicit the same sorts of reactions as real people would. For instance, in the real world, males who are aggressively are often seen as more positively than aggressive females. The males are “assertive” and the females are “pushy.” They then directed the participants to different computers – one with a female voice and one with a male voice. Both computers issued identical commands, and both computers praised the participants. The researchers found that the participants thought that the female computer was less friendly than the male computer, and that the praise was more compelling coming from the male computer than the female computer. This suggests that the participants anthropomorphized the computers, ascribing the female computer the characteristics of a female person, and vice-versa for the male computer. They found similar result for ethnicity, in which the computer had either a Korean or a Caucasian face, and the stereotypes of these ethnicities were ascribed to the computer. Marakas et al. (2000) concurs that computers may be anthropomorphized. They state that computers are seen by some as having human-like characteristics, such as stating that computers read, write, think, catches viruses and is friendly. Moreover, the more human-like the computer seems, the more likely that computer will be seen as productive and helpful. ANTHROPOMORPHISM IN DESIGN Anthropomorphism may also be seen in figurines and designs. DiSalvo & Gemperle (2003) state that anthropomorphism may be seen in many different types of product forms and behaviors. They state that household products, humanoid robot and vehicles are among the items which may take a human form and be anthropomorphized. They state that anthropomorphism fulfills a number of goals for the marketers of these products. For instance, anthropomorphism makes the products seem more familiar and comfortable, and it also makes the object seem like they share our values. Moreover, anthropomorphism of these objects may serve a variety of functions. For instance, a perfume bottles may take the female shape, as well as soap bottles, such as Ivory soap. These shapes are the curvaceous hourglass shape which is often attributed to females. They state that the shape of these bottles serve the purpose of maintaining conventions. Perfume is seen as sexual or sensual – putting the perfume into the shape of a curvaceous woman reinforces this. Another example put forth by DiSalvo & Gemperle (2003) is making a pot scrubber in the shape of a human. This is designed to make scrubbing pots seem like fun. Mac Icon faces, in which either a happy face or a sad face was displayed on the screen serves the function of conveying complex technological information. The sad Mac face means that there is something wrong with the computer. The happy Mac face means that the computer is working well. DiSalvo & Gemperle (2003) also state that vehicles are anthropomorphized. For instance, a VW Beetle has a “face” made up of the headlights and grill, and this face looks cute and harmless. On the other hand, a Dodge Truck has a “face” which is menacing. This is intentional, as the faces on these and other vehicles conveys the characteristics of the vehicle (DiSalvo & Gemperle, 2003). Just as VW Beetles are known to be “fun” girly cars, a Dodge Truck is known to be more intimidating and masculine. WHY PEOPLE MAY NOT ANTHROPOMORPHIZE The above evidence suggests why individuals may anthropomorphize. There are also reasons why they may not, at least when the focus is on technology and machines, such as computers and robots. One reason is that computers are sometimes extremely frustrating. Lazar et al. (2006) studied this factor. They found that users’ frustration with their computer was high, which means that users typically are very frustrated with their computers. In both students and professionals, there was a great deal of frustration with the computers, as well as a great deal of time lost to the computer malfunctions. Frustration with computers was also strongly correlated with the time lost to the malfunction, the time it takes the fix the malfunction, as well as the importance of the task which was being worked on when the malfunction occurred (Lazar et al., 2006). What this suggests is that the inherent frustration in working with computers may result in less anthropomorphizing of the computers. The reason for this is because anthropomorphizing occurs, generally, when the non-human object or animal is seen as similar to oneself. When a computer malfunctions and causes frustration, then that computer may also be seen as dissimilar to oneself. This is assuming that the individual does not see his or herself as being frustrating to others or somehow damaged. Therefore, malfunctioning computers may result in less anthropomorphizing of these particular computers. IMPLICATIONS There are clear implications for anthropomorphism. One of the implications is in marketing and why and how we buy. As stated above, inanimate objects are seen as marketing ploys, such as the perfume bottles which are shaped like a sexy woman, and the car which looks friendly verses another car that looks menacing. These objects subliminally appeal to our senses and values. A man who wants to intimidate on the road is instinctively going to shy away from the cutesy Volkswagon and choose a vehicle which looks like a menacing man. A cutesy girl will do the opposite. A woman may be drawn to a certain perfume bottle because it is suggestive of the shape and sexiness that she sees in herself. She may buy the Ivory soap for the same reason. These may all be subliminal reactions, but that does not make them any less powerful. Waytz et al. (2010 B) also states that anthropomorphism has clear implications in marketing campaigns, as well as in human-computer interactions. They state that anthropomorphism is clearly implicated in human-computer interactions, which includes artificial intelligence, engineering and computer science. They state that anthropomorphizing enhances the connection between the computer and the user by making the computer seem more predictable and understandable. Anthropomorphic computer interfaces also increase engagement with between the user and the computer. Robots are liked more when they seem more human, and they may provide social support for the elderly and those with physical and mental health challenges. That said, Waytz et al. (2010B) state that there can be drawbacks as well. Certain things like the old Microsoft Word paperclip was disliked because it was distracting and seemed like a co-worker who came into one’s office too often. Robots that look too human repulse and discomfort the users (Waytz et al., 2010B). More anthropomorphic interfaces are also more likely to be scapegoated when there is a computer malfunction (Waytz et al., 2010BB). What this suggests is that creators of computers should do market research on what features will be helpful and which would be annoying, such as the annoying Microsoft paperclip, and that robots should not seem too human-like. The degree of anthropomorphism may influence how and why people might buy computers, however, if the computer is made to resemble a human in some way and it conveys that it shares the values of the potential buyer. Beyond this, anthropomorphism may also influence marketing decisions. Waytz et al. (2010 B) states that marketers have long used anthropomorphic designs to sell products. The Koolaid man is one example of this, as are the car parts, as noted above. Each brand also has a “personality,” argues Waytz et al. (2010B), and this personality is effective in marketing because people will buy products which enhance their own self-concepts. For instance, Eddie Bauer has the personality of being rugged and outdoorsy. A person who has a similar concept of his or her self will buy Eddie Bauer clothes and an Eddie Bauer vehicle because the personality of the brand fits the conception that the person has of his or herself. Waytz et al., (2010B) also state that humans are naturally attuned to cues in the environment which look human, such as car grills that look like smiles and a watch that appears to be smiling when set at 10:10, and this natural inclination towards human cues in the environment can enhance a marketing campaign. Anthropomorphism may also be used to improve the interface of computers (Murano & Holt, 2011). Murano and Holt (2011) state that anthropomorphism in computer interfaces are helpful in that they enable cognitive, physical and functional affordances. A cognitive affordance means that the computer helps to give clear and precise feedback, therefore it facilitates the users knowing. Physical affordance means that the computer helps the user physically accomplish a task. Anthropomorphic computers accomplish this by the designer manufacturing a button that can be clicked by the user that will facilitate a task. Sensory affordances helps the user sense (see, feel or hear) something (Murano & Holt, 2011). Anthropomorphism with robots can also be very helpful in assisting autistic children in learning to play. This is important because playing is the norm in children, and not knowing how to play can lead to social isolation (Dautenhahn, 2007). Dautenhahn (2007) suggests that robots can help with this, as shown through the Aurora project. This project features a robot who reacts to the movement of the autistic child, and enhances play by keeping a certain distance from the child, following the child, encouraging the child to run from it in a playful way, or run away from the child. This can lead to interactive games, where the child is chasing the robot and the robot is chasing the child, and this can help autistic children learn how to play. CONCLUSION Anthropomorphizing is the act of ascribing human characteristics to non-human objects. The bulk of the research for this project focuses upon the implications for technology, including computers and robots. Anthropomorphizing may be helpful in that it can serve as an effective marketing tool, as well as assist in interactions with technology. Bibliography Allison, C., Baron-Cohen, S., Wheelwright, S., Stone, J. & Muncer, S. (2011) Psychometric analysis of the empathy quotient (EQ). Personality and Individual Differences, 51, 820-835. Bastian, B., Jetten, J. & Radke, H. (2011) Cyber-dehumanization: Violent video game play diminishes our humanity. Journal of Experimental Social Psychology. Caporael, L.R., Heyes, C.M., (1997). Why anthropomorphize? Folk psychology and other stories. In Mitchell, R.W, Thompson, N.S, Miles, H.L (Eds.), Anthropomorphism, Anecdotes, and Animals (pp. 59 - 73). State University of New York Press, Albany, New York. Castano, E. & Kofta, M. (2009) Dehumanization: Humanity and its denial. Group Processes and Intergroup Relations, 12, 6, 605-607. Costello, K. & Hodson, G. (2010) Exploring the roots of dehumanization: The role of animal-human similarity in promoting immigrant humanization. Group Process and Intergroup Relations, 13, 1-21. doi: 10.1177/1368430209347725. Dautenhahn, K. (2007) Socially intelligent robots: Dimensions of human-robot interactions. Philosophical Transactions of the Royal Society, 362, 679-704. Demoulin, S. (2004) Emotional prejudice can lead to infra-humanisation. European Review of Social Psychology. Available at: http://www.tandf.co.uk/journals /pp/10463283.html DiSalvo, C., Gemperle, F. & Forlizzi, J. (2005) Imitating the human form: Four kinds of anthropomorphic form. Available at: http://www.mendeley.com/research/imitating-the-human-form-four-kinds-of-anthropomorphic-form/ DiSalvo, C. & Gemperle, F. (2003) From seduction to fulfillment: The use of anthropomorphic form in design. Available at: http://anthropomorphism.org/pdf/Seduction.pdf Epley, N., Waytz, A. & Cacioppo, J. (2007) On seeing human: A three-factor theory of anthropomorphism. Psychological Review, 114(4), 864-886. doi: 10.1037/0033-295X.114.4.864 Gong, L. (2008) How social is social responses to computers? The function of the degree of anthropomorphism in computer representations. Computers in Human Behavior, 24, 1494-1509. Goodrich, G. & Allen, C. (2007) Conditioned anti-anthropomorphism. Comparative Cognition and Behavior Reviews, 2, 147-150. Harrison, T. (2007) Different types of intelligence: Bridging the gap between systems design and cognitive theory. Available at: http://www.cs.bath.ac.uk/hiins/papers/07_Harrison.pdf Haslam, N. (2006) Dehumanization: An integrative review. Personality and Social Psychology Review, 10, 252-264. Doi: 10.1207/s15327957pspr1003_4. Kiesler, S. & Goetz, J. (2002) Mental models of robotic assistants. Available at: http://www.cs.cmu.edu/~kiesler/publications/2002pdfs/2002KIESLER.mental_chi_short.pdf Lazar, J., Jones, A., Hackley, M. & Schneiderman, B. (2006) Severity and impact of computer user frustration: A comparison of student and workplace users. Interacting With Computers, 18, 187-207. Loughnan, S. & Haslam, N. (2007) Animals and androids. Psychological Science, 18(2), 116-121. Marakas, G., Johnson, R. & Palmer, J. (2000) A theoretical model of differential social atributions toward computing technology: When the metaphor becomes the model. International Journal of Human-Computer Studies, 52, 719-750. doi: 10.1006/ijhc.1999.0348. Murano, P. & Holt, P. (2011) Evaluation of anthropomorphic feedback for an online auction and affordances. International Journal of Computer Science Issues, 8(2), 13-21. Nass, C. & Moon, Y. (2000) Machines and mindlessness: Social responses to computers. Journal of Social Issues, 56(1), 81-103. Riek, L., Rabinowitch, T., Chakrabarti, B. & Robinson, P. (2009) How anthropomorphism affects empathy towards robots. Available at: http://www.cl.cam.ac.uk/~pr10/publications/hri09.pdf Scassellati, B. (2002) Theory of mind for a humanoid robot. Autonomous Robots, 12, 13-24. Waytz, A. & Epley, N. (2012) Social connection enables dehumanization. Journal of Experimental Social Psychology, 18, 70-76. A. Waytz, A., Epley, N., Cacioppo, J. (2010) Social cognition unbound: Insights into anthropomorphism and dehumanization. Current Directions in Psychological Science, 19(1), 58-62. B. Waytz, A., Cacioppo, J. & Epley, N. (2010) Who sees human? The stability and importance of individual differences in anthropomorphism. Perspectives in Psychological Science, 5(3), 219-232. Read More