2.2 Technological Acceptance of Social Robots
Based on social psychology, technological acceptance models have been developed that attempt to analyse and relate informational factors that explain how users accept and use a new technology. Although there are many [
51], we are going to introduce those that have become the most relevant in HRI [
22,
23]. One of the pioneers was the Technology Acceptance Model (TAM), developed by Davis [
14], and based on the proposals made by the Theory of Reasoned Action (TRA) [
15] and by the Social Cognitive Theory (SCT) [
16]. Davis [
14] proposed TAM from a utilitarian perspective and considering a work environment. The intention to use the new technology depended on the perceived usefulness (how this technology improves work efficiency) and the perceived ease of use (how easy it is to use this technology), both of which are components of attitude. In addition, he considered that this relationship could be moderated by external variables, such as age, gender, etc. Although the TAM model was studied for more than a decade and underwent numerous modifications, changes in theoretical paradigms (Theory of Planned Behaviour proposed by Ajzen) [
51] and consumers’ greater familiarity with such devices led the model to become the Unified Theory of Acceptance and Use of Technology (UTAUT) [
17,
18]. It is an eclectic model, which combines theoretical foundations and empiricism, and due to its greater predictive capacity, it is considered a more integrative model than TAM and has imposed itself upon the other theories of technology adoption [
19].
Although more advanced models such as the UTAUT-2 [
18] have been proposed, they are better suited for use in an environment of market maturity and widespread familiarity on the part of consumers (computers, mobile phones). The UTAUT can be a good basis for developing models of technological acceptance of social robots for two reasons. First, although the TAM model could be considered the most appropriate at an introductory stage, social robots, due to their embodiment and social skills, are completely different from other technological devices, such as laptops or smartphones [
23,
52]. As noted above, the TAM model considers that the intention to use depends exclusively on its cognitive evaluation (perceived usefulness and perceived ease of use), which is deemed appropriate in the context of industrial robotics. For example, Bröhl et al. [
53] proposed an adaptation of the TAM model to study how industrial workers accept sharing their task with a robot (the acceptance of human–robot collaboration) and, furthermore, analysed differences in the degree of acceptance in four countries (Germany, Japan, China and the United States). Second, when a social robot acts by offering a service, it should not only provide the core of the service, but also add value with socio-emotional and relational elements [
54]. That is, the core of the service generally relates to functional outcomes, while the relational form (a female/male voice, the tone of the conversation) is linked to its social outcomes [
55]. This view is in line with that of Heerink et al. (2010) [
23], who extended the TAM by including a number of socio-emotional and relational variables in the context of elderly care. Ultimately, the UTAUT seems more suitable for service delivery than the TAM.
According to UTAUT the intention to use a new technology, such as an entertainment social robot, is explained by four constructs: Performance Expectancy (a new name assigned to TAM’s Perceived Usefulness), Effort Expectancy (a new name for TAM’s Perceived Ease to Use), Social Influence and Facilitating Conditions. In addition, it is moderated by user characteristics (Gender, Age, Experience and Voluntariness of Use). The UTAUT represented an improvement of the TAM by incorporating two new constructs: the perceived social influence, which represents the degree of acceptance that an individual receives from his/her social environment when using the new technology, and the Facilitating Conditions, that is, the individual’s perception of their degree of control over this new technology. These modifications represented an adaptation to the Theory of Planned Behaviour and an improvement in the theoretical basis.
There are numerous precedents for studying the technological acceptance of social robots, and they have been conducted within a wide range of services and scenarios. These scenarios range from text descriptions, video representations and, more realistically, Wizard-of-Oz scenarios [
51]. The outcomes of front-office services are experiences, and social robots, with their ability to generate social interaction during the customer encounter, contribute to co-creating this experience [
55]. The use of social robots in different services and with different degrees of abilities to interact with customers will result in different experiences [
55]. This study, to our knowledge, is the first to evaluate the technological acceptance of an entertainment robot after undergoing a direct, face-to-face experience and having a conversation, including messages giving advice and encouragement (as recommended by Suchman [
28] and Savela et al. [
51]).
In order to study users’ attitude or intention towards robots, on the one hand, it is necessary to take into account their degree of accessibility. This refers to the degree of intensity of the relationship between the evaluating subject (user) and the object of evaluation (the robot) [
43]. And, one way to achieve greater accessibility is to have a direct experience, rather than to do so indirectly through other users' comments or published information. This direct experience results in more extreme and less ambivalent ratings [
43,
51]. Similar results have been obtained in robotics research [
23,
57]. For example, Savela et al. (2018) [
51] found that in studies of robot technology acceptance, when participants rated hypothetical experiences, they were more likely to report negative attitudes than when exposed to real robots, whose ratings were more positive. When people lack first-hand experience with robots, they must base their ratings on social representations or mental images stored in memory, which affect not only the degree of intensity but even the direction of the attitude [
43,
51].
On the other hand, while a first impression contributes significantly to the formation of a general idea about HRI, it is not sufficient to be able to appreciate all the relevant factors involved, as accessibility also requires more experiences [
43]. Therefore, after a first brief experience, it is better to propose to participants the evaluation of a parsimonious model (with few factors), but with the essential elements, rather than a more complex model with relationships that they will not have had enough time to appreciate and internalise [
43]. But what are the essential elements that the model should include? According to social psychology proposals, they are cognitive, affective and behavioural elements [
43].
Taking into account the above considerations and the context of a still developing social robot technology, a parsimonious model is proposed to evaluate users’ technological acceptance after a real entertainment experience [
58]. Although the proposal is based on the UTAUT, it must take into account the three essential elements proposed by Gerrig [
43], adapted to the technological acceptance of social robots. The three elements are: functional (perceived usefulness and perceived ease of use), socio-emotional (social influence) and relational (perceived enjoyment) [
55]. Thus, given that the service provided by the robot is to entertain, perceived enjoyment has been incorporated into the model and, due to the nascent stage in which social robotics currently find itself, the use of the Facilitating Conditions construct does not make sense until consumers become more familiar with it, so it has been removed. The proposed model is defined by the following hypotheses:
H1. The perceived usefulness of using an entertainment robot is positively related to the intention to use it.
H2. The perceived ease of use of an entertainment robot is positively related to the intention to use it.
H3. The perceived enjoyment of the use of an entertainment robot is positively related to the intention to use it.
H4. The social influence of the use of an entertainment robot is positively related to the intention to use it.
H5. The perceived ease of use of an entertainment robot is positively related to the perceived usefulness.
2.3 User Gender
Consumer gender is frequently used both as a basic criterion for product and service segmentation [
56] and in communication and advertising campaigns [
59]. In fact, gender is one of the most widely used segmentation factors, as it is easy to identify, accessible and the segments generated are broad enough to be cost-effective [
56]. This practice is based on the assumption that men and women have different cognitive structures, personality traits and interests associated with their gender roles, which guide their judgements and decision-making processes [
60]. Under this belief, enterprises such as the Coca-Cola Company have designed products with almost the same ingredients, Coca-Cola Zero and Diet Coke, the former targeting male consumers and the latter targeting female consumers of zero-calorie soft drinks [
61].
The literature on HRI has also addressed the gender issue from different perspectives. Nomura [
11] proposed a classification of these studies into three categories: i) implications of the male, female or neutral gender of robots in users’ acceptance [
26,
62]; ii) acceptance of robots according to the user’s gender [
8]; and iii) interaction effects between the genders of robots and users [
27]. All these perspectives have been criticised both for not differentiating between sex and gender, and for reproducing heteronormative and binary gender stereotypes [
8,
30]. That is, while sex is formed by the biological nature of the human body, gender refers to the stereotyped characteristics related to social expectations about behaviours, attitudes and preferences that are considered adequate or inappropriate [
6,
56].
Furthermore, it has also been pointed out that reducing the analysis of participants to their sex or gender leads to an excessive emphasis on gender differences, neglecting other relevant factors such as age, socioeconomic level, experience, etc. [
6]. For example, Wang and Young [
8] highlighted the under-representation of women in science and engineering and even as participants in HRI studies. They also criticised the oversimplification that the gender dichotomy implies and advocated for greater sensitivity in order to achieve more nuanced results, while also proposing some advice on how to accomplish an inclusive approach [
8]. Later, Rea et al. [
25] carried out a study that replicated other studies that use gender stereotypes with the aim of trying to validate these stereotypes or show them to be false. Mixed results were obtained: while some stereotypes such as male participants being less polite than females towards robots were validated, others such as women's lower commitment or greater concern with robots were disproved. However, the test conducted by Rea et al. [
25] used small samples and only considered gender, ignoring any other factor.
This controversy stems from the ongoing debate among psychologists regarding the different behaviour of men and women, that is, whether these differences are mainly due to genetic-biological inheritance (nature) or environmental influence (nurture) [
56,
63]. A wide range of theories attempt to explain the causes of these differences, ranging from the extreme nativists (those who believe that genetic, hormonal and brain structure influences are responsible for different behaviours) [
32] to the extreme ecologists (who attribute them to the socio-cultural environment, stereotypes and the media) [
16,
33].
According to nativist theories, the basis for behavioural differences between the sexes is due to chromosomal and hormonal differences, as well as differences in brain structure and cognitive processes [
43,
56]. Among the hormonal differences, testosterone plays a key role in determining the sex of a newborn and also affects mood formation and personality. In particular, androgynous differences affect their interests, activities and levels of aggression [
64]. The fact that these differences emerge at a very early age and are found in all cultures suggests the involvement of biological factors [
65]. But, in addition, technical advances in brain scanning have also detected structural differences between males and females. For example, the use of MRI to analyse sex differences in the orbitofrontal cortex revealed that women have a larger volume in regions related to emotional and affective regulation than men [
64]. Evidence has also been collected on the different ways in which men's and women's brains work to perform cognitive and emotional tasks. For example, Canli, Desmond and Gabrieli (2002) [
66] showed that men and women activated different neural circuits to encode stimuli in memory, in this case the stimuli consisting of funny and neutral images. The results showed that, in general, women recalled the funny images stored in memory with more humour than men, and the explanation had to do with the greater number of brain regions activated both during the viewing of the more emotional images and during subsequent recall.
In summary, while decades ago biological differences between the sexes were considered to have little impact on their behaviour [
56], recent research is changing that view by uncovering significant differences in brain activity. In the prefrontal cortex, men, compared to women, produce more neural activity to achieve the same behaviour, in line with the neural efficiency hypothesis. According to this hypothesis, the more intelligent (IQ) produce less neural activity than the less intelligent to perform the same cognitive task. In addition, evidence has been collected showing that in visual processing regions, males report higher activity than females in storing and retrieving realistic visualisations from long-term memory. Finally, gender differences in parahippocampal cortex activity have also been found, for example, greater use of verbal strategies by females than males [
64].
According to the ecological school of thought, gender role is a social construct. First, children identify themselves with a gender (usually their own) and then, through the learning process, try to make their identification congruent with the norms of behaviour, motivations and feelings they perceive as appropriate for that gender [
43,
56]. Within this school of thought, several theories have been proposed, one of the most prominent being social role theory [
67]. According to this theory, differences between men and women in skills and personality traits often reflect traditional gender roles in society. Thus, men are guided by agentic goals that emphasise assertiveness, self-efficacy and dominance. In contrast, women are guided by communal concerns that emphasise interpersonal affiliation and harmonious relationships [
56].
Today, controversy continues to rage over the origins of the behavioural gender difference, as there are: (1) researchers who argue for a decisive socio-cultural influence on the formation of behavioural patterns [
33,
67], (2) those who emphasise biological factors [
64,
65] and (3) those who believe that the truth must lie somewhere between nature and nurture [
68,
69]. One of the theoretical streams that recognises the influence of both biological and socio-cultural factors is evolutionary psychology. As its name suggests, this theory adopts the Darwinian principles of natural and sexual selection that ensure the survival of the species. According to its precepts, behaviour is due to the existence of adaptive programmes, the results of human evolution, which have taken root in the brain and reflect the overcoming of different survival challenges faced by our hunter-gatherer ancestors over thousands of generations [
70]. For example, biology has endowed women with the ability to procreate, and they have ancestrally had to raise their offspring in a hostile environment. Thus, those women who invested more in affection and care for their children and, in addition, developed cognitive skills to detect dangerous situations may have promoted their survival and gained an evolutionary advantage [
71,
72]. In addition, these programmes acquired over thousands of years take time to change, resulting in temporary mismatches between the prevalence of ancestral behaviours and a modern environment, which are called “evolutionary delays” [
73].
Since hunter-gatherer societies still survive, it has been possible to conduct ethnographic studies on their traditions and customs, which have shed light on our ancestors’ behaviours (an example is Hill and Hurtado [
74]). A prominent feature of these societies is labour division between genders, so while men have the main task of hunting and defence, women are engaged in gathering, reproduction and the care of offspring [
75]. This is one of the most controversial and contentious issues among different theoretical currents [
67]. For example, social role theory suggests that labour division between sexes creates gender role expectations that lead to differences in social behaviour and personality. Thus, men are more assertive and aggressive because they have historically been more likely to assume leadership positions, while women do not develop these characteristics because they do not play these roles [
67]. This explanation contrasts with that provided by evolutionary psychology. According to DeVore and Tooby [
76], this labour division that assigns the most dangerous tasks to men is the most efficient for promoting group survival.
This separation of tasks for thousands of generations has led to the formation of differences in physical characteristics (greater physical strength and size in men) and in perceptual and cognitive skills (ways of processing information) between genders. Thus, a study conducted in 40 countries and with seven ethnic groups showed that women, as a result of their adaptation to gathering, have developed a better memory for remembering the spatial location of objects, while men, as an adaptation in hunting animals, have developed better navigation, map-reading and three-dimensional mental rotation skills [
77]. And, although it has been shown that spatial skills can be improved with practice, this does not eliminate the male advantage in these skills [
78].
To analyse how these cognitive gender differences affect the intention to use a social robot, this paper proposes that it is necessary to study their effect on antecedents or explanatory factors from an eclectic theoretical perspective. In other words, we need to know how the different cognitive abilities of men and women can affect the perception of usefulness, ease of use, perception of enjoyment and social influence and their effect on the intention to use. In studies on technological acceptance, significant differences have been detected between men and women, the latter being more reluctant to accept technological innovations than men [
29,
79]. Even in one of the most widespread technologies, such as the use of the Internet, women make less intensive use of it than men [
80]. The literature has compiled several explanations for this behaviour: women perceive themselves as being less skilled, having less control over new technologies [
8,
81] and encountering greater difficulties to understand them than men [
79,
82]. Even in the search for information they believe they are less effective than men [
83]. All this has led, in the most advanced models of technological acceptance, to consider gender as a moderating variable, although without any theoretical support [
17].
Similar findings appear in the acceptance models of social robots. For example, Nomura et al. [
57] showed that women had more negative attitudes than men and, in another experiment, Nomura et al. [
84] also found that women had more negative attitudes towards the social impact of robot implementation. In another study conducted in Taiwan on how parents perceive educational robots, results showed more negative attitudes in mothers than in fathers towards entrusting their children’s education to robot machines [
85]. However, in the second study there may be a problem of confounding factors, since the degree of acceptance of robots can be confused with the difference in parenting styles between men and women [
59]. Regardless of these results, for both researchers and practitioners in social robotics, it is essential to know how men and women differ in their cognitive processes, affective responses and reactions to robot stimuli in order to adjust robots to their preferences and to be able to initiate a process of market segmentation.
Meyers-Levy and Stemthal [
68] established some of the different cognitive abilities between genders. In their relationship with the environment, women capture and process more information, in a more comprehensive and enveloping way than men. This ability is due to the fact that they have a lower perception threshold, which allows them to capture information that is unnoticeable to men, such as the perception of non-verbal emotions and the ability to develop a deeper understanding of emotions [
62]. In other words, women will pick up non-verbal language better than men, and will also be more sensitive to possible risk signals from the robot, whereas men will pick up less non-verbal language and will not be aware of low-risk signals coming from the robot [
86].
On the other hand, men perceive and process information with greater focus, are more selective in stimuli and pay attention to pieces of information that they consider more relevant, while women tend to process information more comprehensively [
59,
79]. These different cognitive abilities have resulted in a greater preference for instrumentality [
87] and the usefulness of objects in men than in women [
88,
89]. Thus, we can predict that in the case of acceptance of entertainment robots:
H6. For women, perceived usefulness will be a less important precedent of the intention to use the robot than for men.
Since women have a greater ability to scan the environment and gather information (for example, they scrutinise with greater eye fixations) than men, this competence has its advantages and disadvantages. On the one hand, this ability allows them to better recognise the environment, to have a global vision and to quickly grasp the robot's possible subtle gestures of danger but, conversely, makes it difficult to establish quick relationships between objects [
90] and, consequently, to elucidate the intrinsic difficulty of robot handling. This proposition is in line with the idea that men pay more attention to specific elements (characteristic attributes of the robot) while women are processors of the relationships between all elements (of the robot with the environment) [
65]. With these precedents, in the case of entertainment robots we can expect:
H7. For women, perceived ease of use will be a less important precedent of intention to use the robot than for men.
On the other hand, since men process information in a less exhaustive way and focus on specific elements, this procedure helps them establish quick relationships with less information [
59] and, consequently, the perception that ease of use will be less relevant to their usefulness. Conversely, women, who see more complex relationships between the elements of the robot, will consider ease of use to be a more relevant factor for their usefulness. This has resulted in women’s self-assessment as having less confidence in their performance with new technologies [
79,
91] and considering that they have greater difficulty than men to understand them [
82].
H8. For women, perceived ease of use will be a more important precedent of perceived usefulness than for men.
Regarding the hedonic precedent of HRI, Wang and Young [
8] pointed out that men show greater enthusiasm for using robots for entertainment and, in addition, women tend to feel more nervous when interacting with robots, since, as noted above, they are more sensitive to any possible risk signals [
86]. Some authors have also noted the greater propensity of men to agency and women to community, which drives men to look for new technologies in order to explore more personal interests, such as entertainment and economic performance, and women to seek greater social participation [
8,
92]. However, as Canli et al. [
66] pointed out, women showed a greater ability than men to store in memory and later recall entertainment experiences. Thus, in the case of entertainment robots:
H9. For women, perceived enjoyment will be a more important precedent of intention to use the robot than for men.
Studies on different motivational factors between the sexes show that women outperform men on socio-emotional, expressive and interpersonal traits, while men outperform women on task-oriented, instrumental and agentic traits [
65]. In addition, differences in the influence of context on emotion expression have also been reported, i.e. women are more emotional in an interpersonal context, while men are more emotional in an achievement context [
59]. In this study, both elements could be present: (1) men being emotional about achieving the goal of the game, and (2) women being emotional about engaging in an interpersonal conversation with the robot. However, the literature reports a greater tendency for women to be socially affiliated and to take others' opinions into account when making their own judgements [
8,
59,
93]. And, unlike men, they use new technologies more for social than for personal purposes. Women use the Internet with greater intensity than men to communicate with other people, for example, by sending emails [
94]. In the case of entertainment robots, we propose:
H10. For women, perceived social influence will be a more important precedent of intention to use the robot than for men.
2.4 Intuitive and Reflexive Processes
Although the participant's reasoning system is a common variable in the literature on strategic management decisions [
95], to our knowledge it has never been used in technology acceptance models of social robots. Nevertheless, it is not a variable that is entirely alien to the world of the new technologies, since it has been used to study the effect of use of smartphones [
96]. According to the dual-process theory, when individuals must make a decision, they can tackle it through a fast heuristic process or through a more elaborate and slower analytical process [
97,
98]. Although dual-process theories began to develop in the 1970s [
99], it was not until the last decade that they have reached greater diffusion in the literature, probably due to the further development of evolutionary psychology [
97,
98], the impact achieved by some of its applications [
36] and their use in the study of strategic decisions [
36,
95].
This theory proposes that individuals use two different evolutionary cognitive systems to process information: a primeval system, similar to that used by the more developed animals, and another that is unique to humans and superior to that of animals [
98]. The first, considered primitive, autonomous, intuitive and which does not require memory work, is called system1 and the second, deliberative, analytical, reflective, which requires memory work, cognitive decoupling and mental simulation, is called system 2 [
34‐
36]. System 1, which is intuitive, is conceptualised as a rapid, non-conscious decision-making mechanism that uses patterns and associations of ideas to derive affective judgements [
100]. As individuals who use heuristics to make their decisions, they later find it difficult to reconstruct and explain thoroughly the decision-making process [
34]. In contrast, in system 2, individuals use an analytical, systematic, rule-based and explicit mechanism to make decisions [
101]. That is, they follow a step-by-step process that includes identifying and formulating the problem, thoroughly evaluating relevant information, generating a set of alternatives, evaluating them, and finally making a logical decision based on conscious deliberation [
34,
95].
Although system 1 (intuitive) and system 2 (rational) are recognised as valuable for decision-making, they are fundamentally different [
97,
98]. It seems that people tend to use intuitive processes more frequently than cognitive decoupling, that is, more heuristic than abstract reasoning and cognitive reflection [
102]. In an attempt to explain this bias, Todd [
103] proposed that human beings, like all other living beings, have adopted the principle of energy efficiency in their behaviour. Because analytical processes require effort and consumption of energy resources, individuals generally try to avoid them by replacing them with quick and economical heuristics. This process has led to qualify human beings as miserly cognitive [
35].
However, the use of quick heuristics does not mean making incorrect decisions. An evolutionary heuristic has been formed over thousands of years as a result of overcoming challenges to survival and, in many situations, can improve decision-making performance [
104] and overcome more complex decisions [
105]. Yet, some evidence shows that when individuals trust in their intuitive response, they are unlikely to rethink or change their decisions after reflection [
106]. Furthermore, heuristic effectiveness depends on the environment. While in routine environments, where people have learned to identify the key elements, the use of system 1 can generate acceptable responses, whereas in hostile or stochastic ones, the responses of system 2 are more efficient [
34].
Not all people exhibit the same behaviour, however, and some are less miserly and make more intense use of rationality than others [
107]. In a study on the use of new technologies, it was found that smartphones were considered more useful and practical by the more miserly cognitive than by those with a greater propensity to rationality [
96]. That is, those who rely more on intuition are more prone to intensive use of smartphones, since they resort to their smartphone more than to their mind to obtain information, thereby showing a greater dependence on that external source than people who use their own mental resources [
106]. In other words, the principle of energy efficiency still applies [
103] and users are getting used to having a portable brain and not consuming energy from their own. We expect these individuals to adopt the same logic with social robots and, by accepting an entertainment robot, we can propose the following hypothesis:
H11. For participants using heuristics (system 1), perceived usefulness will be a more important precedent for the intention to use the robot than for participants using reasoning (system 2).
As noted above, people with system 1 (heuristic) resort to looking for key elements that facilitate their assessment. That is, among all the variables that can affect decision-making, they only focus on a few that they consider relevant (for example, if a consumer is considering the purchase of a product, s/he must use price or brand as a heuristic criterion). But, when these elements are difficult to assess, they are often replaced by easier ones, even if they are less accurate [
107]. This implies that for a consumer applying heuristic logic there is no real perception of difficulty [
34], since, to continue with the example, if the consumer is unfamiliar with the brand, s/he simply resorts to price as a decision criterion. Therefore, for participants who use heuristic logic in their decision process, ease of use will not be relevant for them. In the case of entertainment robots, we hope this predisposition also prevails:
H12. For participants using heuristics (system 1), perceived ease of use will be a less important precedent for the intention to use the robot than for participants using reasoning (system 2).
However, the principle of energy efficiency that prevails in participants applying system 1 [
103] will cause them to consider ease of use as a precedent for determining the usefulness of the robot. Something similar has been detected with the use of smartphones, since their greater use among the miserly cognitive is closely related to proximity, accessibility and ease of use [
96]. With these precedents, in the case of entertainment robots we can expect:
H13. For participants using heuristics (system 1), perceived ease of use will be a more important precedent of perceived usefulness than for participants using reasoning (system 2).
It is difficult to establish a relationship between cognitive systems and entertainment perception. Studies that have analysed entertainment have traditionally focused on the measure of hedonic and pleasurable responses [
108]. However, Oliver and Bartsch [
109] considered that from the feeling of enjoyment two different qualitative experiences must be distinguished: mere enjoyment and appreciation. Thus, while enjoyment generates immediate positive responses of excitement and enthusiasm, appreciation generates more reflective and long-lasting cognitive responses [
109]. Similarly, Vorderer and Ritterfeld [
110] discussed the distinction between enjoyment and appreciation in the context of digital games, considering that, in contrast to entertainment games, experiencing appreciation was more related to educational games (called serious games) that allow for higher-order goals, such as improving competence [
110].
Lewis et al. [
111] associated these experiences derived from enjoyment with the dual-process theories. They proposed that while enjoyment follows a quick and intuitive process to generate a response, appreciation, in contrast, requires a slow and controlled evaluation, capable of weighing up the existence of conflicting needs or emotions [
111]. Although this distinction has not been made in this study, the context in which the study is framed means that we consider perceived enjoyment in an HRI to be closer to appreciation than mere enjoyment. Therefore, given that appreciation requires analytical effort and the consumption of energy resources, people who are more predisposed to the use of heuristics (system 1) will try to avoid their involvement and will therefore value this driver less when accepting this new technology [
35]. This consideration is in line with findings obtained by Barr et al. [
96] in their experiments. Although they have not been able to find significant relationships, their results pointed towards a greater predisposition to entertainment with the use of smartphones by more rational users. In this study we want to determine whether this indication can be verified in the case of entertainment robots.
H14. For participants using heuristics (system 1), perceived enjoyment will be a less important precedent for the intention to use the robot than for participants using reasoning (system 2).
Something similar occurs between the social influence and the cognitive system of individuals. Although we have not found any direct evidence, the work by Pennycook and Rand [
35] suggests that individuals who follow system 1 show greater credulity regarding epistemically suspicious information (fake news), greater religious belief and, conversely, a greater disbelief towards scientific proposals, such as evolution, than those who follow a deliberative reasoning process [
36]. That is, all these studies underline a greater independence on opinion among people who use reasoning rather than intuition and, therefore, a lesser need for social approval to make their decisions. Other evidence has shown that more intuitive users make more intensive use of smartphones and social networks than those who show greater intensity ability [
96]. Associating this idea with the acceptance of an entertainment robot, we can propose that:
H15. For participants using heuristics (system 1), perceived social influence will be a more important precedent for the intention to use the robot than for participants using reasoning (system 2).