Children attribute higher social status to people who have extraordinary capabilities

Throughout history, individuals believed to have extraordinary capabilities were generally highly ranked in their communities; this suggests a universal "extraordinary-dominant expectation" in human minds, which may play a key role in religious thought, even in modern societies. This study shows that 5-6-year-old children, who begin to understand real-world causalities regarding how the body and mind of human beings work, predict that individuals who exhibit extraordinary capabilities have higher social status in interactions with individuals who exhibit ordinary capabilities. In Experiment 1, we showed children two individuals achieving goals using either humanly possible or impossible methods, the latter involving simple forms of violation of intuitive psychology (knowing without seeing), physics (flying), or biology (fire breathing). The children clearly judged the latter as surprising and unusual. More importantly, the children predicted that individuals showing extraordinary capabilities will gain contested resources and play a dominant role in interactions with ordinary individuals, indicating a higher social status. Further investigations suggested that the children specifically linked extraordinary capacities to social status, as they did not attribute dominance to individuals who apply surprising/unusual but possible methods (Experiment 2), and that they did not indiscriminately attribute positive characteristics to extraordinary capabilities despite a strong extraordinary-dominant expectation being replicated (Experiment 3). These findings demonstrate that extraordinary-dominant expectations can be observed in childhood across different intuitive knowledge domains, helping understand the cognitive mechanisms of religious thought and the cognitive foundations of hierarchical social systems.

Throughout history, individuals believed to have extraordinary capabilities were generally highly ranked in their communities; this suggests a universal "extraordinary-dominant expectation" in human minds, which may play a key role in religious thought, even in modern societies. This study shows that 5-6-year-old children, who begin to understand real-world causalities regarding how the body and mind of human beings work, predict that individuals who exhibit extraordinary capabilities have higher social status in interactions with individuals who exhibit ordinary capabilities. In Experiment 1, we showed children two individuals achieving goals using either humanly possible or impossible methods, the latter involving simple forms of violation of intuitive psychology (knowing without seeing), physics (flying), or biology (fire breathing). The children clearly judged the latter as surprising and unusual. More importantly, the children predicted that individuals showing extraordinary capabilities will gain contested resources and play a dominant role in interactions with ordinary individuals, indicating a higher social status. Further investigations suggested that the children specifically linked extraordinary capacities to social status, as they did not attribute dominance to individuals who apply surprising/unusual but possible methods (Experiment 2), and that they did not indiscriminately attribute positive characteristics to extraordinary capabilities despite a strong extraordinary-dominant expectation being replicated (Experiment 3). These findings demonstrate that extraordinary-dominant expectations can be observed in childhood across different intuitive knowledge domains, helping understand the cognitive mechanisms of religious thought and the cognitive foundations of hierarchical social systems.
Hierarchical structure (i.e., some individuals are ranked above other members in a group) is a common feature of group-living animals, favored by evolution, as it largely contributes to the stability and growth of a group (Cheney & Seyfarth, 1990;Meng, Nakawake, Nitta, Hashiya, & Moriguchi, 2019;Sapolsky, 2005;Smith et al., 2016;Tinbergen, 1936). Within hierarchies, higher-ranked individuals generally enjoy privileges in resource control (e.g., food and mates) and influence group decisions (Cillessen & Rose, 2005;Fiske & Berdahl, 2007;Guinote, 2017). Therefore, having a higher social rank than others is a crucial determinant of survival for individuals, which leads to larger reproductive successes (Jiménez & Mesoudi, 2019. However, how we establish social hierarchies is not well known. Throughout human history, the link between extraordinary power and a high social hierarchal position has been observed in cultures globally (Burkert, 1998;Weber, 1947). From diviners, shamans, mystics, cult leaders, Tianzi, and divine kings, people accord high status to individuals believed to have extraordinary powers; these powers commonly entail capacities to efficiently achieve goals by methods that ☆ We thank our research participants for their help in this study and the staff and volunteers at Miraikan for their advice and support. This work was supported by JSPS Grants-in-Aid for Scientific Research (KAKENHI) (20K20156, 19K14370, 17H06344, 22H00090, 23H03702, 21J22803). We have no conflicts of interest to disclose. violate intuitive psychology (e.g., mind reading), physics (e.g., levitation), or biology (e.g., healing) (Barrett, 2000;Bowie, 1999;Boyer, 2003;Mullins et al., 2018;Singh, 2017;Winkelman, 1990). The extraordinary-dominant expectation is considered a key component of religious thought that contributes to cooperation among groups, as well as the hierarchical structure and stability of groups, even in modern societies (Lalich, 2004;Riggio, 1998;Turner, 2003;Wilson & Kwileck, 2003). However, little is known about the psychological foundations of this ubiquitous psychological bias, and less about its developmental origin. From when and to what extent do we expect extraordinary individuals to have a higher social hierarchal position? Answering these questions may help us theoretically understand the nature of the expectation itself, provide direction for studies on the underlying cognitive mechanisms of religious thought, and extend our knowledge of the cognitive foundations of religion.  first approached this research topic through experimental investigations. In their study, 12-to 16-month-old infants watched animated videos in which two agents gain a reward using either a physically intuitive or physically counterintuitive method, the latter involving more efficient actions that violate the constraints of gravity, object coherence and continuity that could be detected intuitively from infancy (Baillargeon & Hanko-Summers, 1990;Baillargeon, Spelke, & Wasserman, 1985;Needham & Baillargeon, 1993;Stahl & Feigenson, 2017). The results showed that infants looked longer when the physically intuitive agent subsequently outcompeted a physically counterintuitive agent in securing a reward. Moreover, their control conditions demonstrated that the differences in looking behavior were absent when both agents obtained the reward by moving in trajectories and speed identical to the previous conditions but in physically intuitive ways . The authors considered the results as evidence indicating that infants expect agents exhibiting counterintuitive capacities to gain access to contested resources, suggesting that preverbal infants link social rank to physically counterintuitive capacities.
Beyond the empirical evidence gained from infants' looking behavior,  further discussed theoretical notions explaining why infants would attribute high social rank to extraordinary agents. They proposed that infants respect, in comparison to ordinary agents, extraordinary agents for their persuasive skills and consequently expect them to have a higher social rank. This view is in line with a competence model that has received much attention from social scientists recently. The competence model posits that social stratification is principally based on persuasive capacities among individuals (Cheng, 2020;Henrich, 2016;Henrich & Gil-White, 2001). It has been argued that because high-ranked individuals can benefit the group by transmitting knowledge, motivating group-wide cooperation, and contributing to collective action, humans may have evolved a stable tolerance for stratification based on competence-based hierarchy (Garfield, Hubbard, & Hagen, 2019;Glowacki & von Rueden, 2015;Henrich, 2016;Henrich, Chudek, & Boyd, 2015;Price & Van Vugt, 2014;Smith et al., 2016). Examples supporting the competence model have been found across different forms of societies; even among more egalitarian huntergatherers, people tend to defer to and attribute higher social rank to people who possess skills and capabilities that lead to valued success (Boyd & Richerson, 1985;Cheng & Tracy, 2014;Garfield & Hagen, 2020;Glowacki & von Rueden, 2015;Henrich & Gil-White, 2001;Jiménez & Mesoudi, 2019).
As counterintuitive physical capacities (e.g., the capacity to cause oneself and/or other matter to float in mid-air; Aarne & Thompson, 1961;Singh, 2017) are generally impossible and, thus, could hardly be observed in everyday human activities, people may perceive such capacities to have the highest rank in the classification of competence. Consequently, according to the competence model, this would lead to an expectation that individuals who have such competence likely also have a high position in the social hierarchy. However, given the limitations that are often observed in infant studies, the past literature still leads to unclear conclusions. First, although  used video stimuli in which geometric agents "fly" and "teleport," as the infants could not give verbal responses regarding how they perceived the stimuli, it was difficult to know whether the infants found these actions as counterintuitive in the real world or whether the infants respected these agents. Second, the study focused only on the domain of intuitive physics, in which the infants have been shown to detect the difference between physically possible events and physically impossible events (Baillargeon et al., 1985;Baillargeon & Hanko-Summers, 1990;Needham & Baillargeon, 1993;Stahl & Feigenson, 2017). Yet, it was unclear whether the extraordinary-dominant expectation was generalizable to other domains, such as psychology (e.g., "Omniscient of a God") or biology (e.g., "Immortality"), which are also common motifs in mythical stories and religions (Barrett, 2008;Karsdorp, van der Meulen, Meder, & van den Bosch, 2015;Thompson, 1995;Uther, 2004).
The current study aimed to draw a more complete picture of the developmental origin and psychological mechanism of the extraordinary-dominant expectation. In a series of experiments, we tested whether 5-to-6-year-old children expect individuals exhibiting extraordinary abilities to have a higher social rank than normal individuals, and whether the possible extraordinary-dominant expectation could be observed across different domains. Furthermore, to evaluate the validity of applying the competence model to the extraordinarydominant expectation, we investigated whether the children respect extraordinary individuals (i.e., as "great").
Studies on cognitive development consistently show that from late in the preschool period (i.e., approximately five to six years of age), children show a coherent and theoretical network of causal beliefs regarding how human bodies and minds work (Gopnik & Wellman, 2012). For instance, the understanding of the relationship between seeing and knowing in other individuals develops sharply after three years of age; by approximately four years of age, most children correctly acknowledge that a person (e.g., their mother, a girl, or a fictional superhero) who has not seen the inside of a container cannot know its contents (psychological domain; Lane, Wellman, & Evans, 2010; evidence could also be found with the "false belief task"; Baron-Cohen, Leslie, & Frith, 1985;Perner & Wimmer, 1985). Other studies have shown that children are aware that certain phenomena associated with the human body follow certain biological principles; children possess three essential components of biology, including "living-nonliving distinction," "mindbody distinction," and "nonintentional causality" (biological domain; Hatano & Inagaki, 1997). For instance, when asked "why do we take in air?" six-year-old children were more likely to apply vitalistic (e.g., because our chest takes in vital power from the air) or mechanistic causalities (e.g., because the lungs take in oxygen and change it into useless carbon dioxide) than intentional causalities (e.g., because we want to feel good; Inagaki & Hatano, 1993). Furthermore, from four years of age, more than half of the participating children judged that no one could float in the air without touching anything, also providing sensible explanations, such as "he didn't have any air (to) float up" and "he didn't have any wings" (physical domain; Schult & Wellman, 1997). Together, although infants show an intuitive understanding of some physical rules (e.g., gravity), from the late preschool period, children develop a theoretical understanding of the environment across physical, biological, and psychological domains, giving appropriate explanations based on their causal structure about the environment (Gopnik & Wellman, 2012). Therefore, the late preschool period would be the earliest stage in which children think about the extraordinary-dominant expectations across different domains based on causal reasoning (Gopnik & Wellman, 2012), and respond to questions through verbal communication with experimenters Meng & Moriguchi, 2021). Although studies have found that by the age of five, children attribute greater cognitive capacities to specific agents (e.g., "God" and "Mr. Smart") than to humans (Heiphetz, Lane, Waytz, & Young, 2016;Makris & Pnevmatikos, 2007), no research has investigated whether children generally expect people exhibiting extraordinary capabilities to be highly ranked in their communities.
Regarding the evaluation of hierarchical relationships, studies have shown that preschool children are capable of representing and evaluating relative hierarchical positions among individuals based on various cues. Specifically, from the second year of life, infants utilize cues, such as spatial location and past achievements, to predict protagonists' hierarchical relationships; they expect highly-ranked individuals to prevail in zero-sum conflict over contested resources and exhibit dominant postures and gestures in dyadic interactions (Bas & Sebastian-Galles, 2021;Meng, Nakawake, et al., 2019). From the preschool age, children use gender, decision power (e.g., imposing one's preferred game on others who want to play different games), and age as cues to predict hierarchical relationships among others (Charafeddine et al., 2015(Charafeddine et al., , 2020Gülgöz & Gelman, 2017). Of particular interest is the fact that, in comparison to dominance-based powerholders (e.g., individuals who excel in physical fights and elicit fear from their peers), 5-6-year-old children are more inclined to perceive respect-based powerholders (e.g., individuals who demonstrate skill in activities such as jumping rope and teach others with kindness, leading to admiration from their peers) as socially preferable (Amakusa, Meng, & Kanakogi, 2022). Furthermore, children predict that these respect-based powerholders will acquire contested resources (e.g., an attractive chair), which subsequently reinforces their elevated social status. This suggests that 5-6 years is an appropriate age range to investigate whether children's extraordinarydominant expectations are based on the competence model (Cheng, 2020;Henrich, 2016;Henrich & Gil-White, 2001).
Although 5-6-year-old children have an understanding of human limitations and hierarchical relationships, no previous research has investigated whether they are biased toward attributing higher social status and respect to people who achieve goals using extraordinary (impossible) methods than those using ordinary (possible) methods. We started the current research with the experiment in which the participants judged the hierarchical relationships between two individuals who had achieved goals in three different events (i.e., a psychological event, a physical event, and a biological event) using either ordinary or extraordinary methods. Following the methodologies used in previous investigations on children's understanding of human capabilities and hierarchical relationships (Amakusa et al., 2022;Enright, Alonso, Lee, & Olson, 2020;Gülgöz & Gelman, 2017;Lane et al., 2010;Lane, Wellman, & Evans, 2012), we used cartoon figures to represent the individuals. Prior studies have shown that children perform equally well on experimental tasks (e.g., false belief task) whether the protagonists are drawings or live performers (Wellman, Cross, & Watson, 2001). More importantly, compared to photographs of real people, cartoon figures might avoid the influence of cues (e.g., appearance and emotion) that are out of the scope of the current research. For instance, experimenters could easily manipulate and control the facial information (and even remove such information) of cartoon characters. Cartoon figures may thus be more appropriate than photographs to access children's pure response toward the experimental manipulations (Koval & Rosette, 2021;Okanda et al., 2022).

Experiment 1
In Experiment 1, the children judged whether the ordinary individual or the extraordinary individual was more highly ranked, more surprising, great, and whether they thought people would do such things in daily life.

Participants
The final sample included 41 5-6-year-old children living in Tokyo, Japan (20 girls; M age = 72.1 months, SD = ±6.03). The sample size, stopping rule, and the experimental procedure were preregistered prior to the experiment (https://osf.io/8nad5/?view_only=bad45405225c4 22aa42c858f8d48ba81, an anonymized OSF link for review). Specifically, a power analysis indicated that at least 39 participants were required to detect any potential effects using a one-sample Wilcoxon signed-rank test with a medium effect size (d = 0.5), significance of α = 0.05, and power of 1-β = 0.80 (Cohen, 1992). Moreover, as we collected the data in a science museum, even when confident that the sample size exceeded the required number, we did not stop the experiment if additional children showed interest in participating. The total number of participants was 57. Based on the preregistered data exclusion rules, 12 children were excluded because they gave at least one incorrect answer for either the chair comprehension question or the posture comprehension question. Four more children were excluded because of experimental errors in voice recording. Written informed consent was obtained from the children's carers before the experiment was conducted. The study was approved by the Behavioral Research Ethics Committee of the Osaka University School of Human Sciences (HB021-020).

Set-up
The children were tested individually by a male experimenter in a separate booth at the Japan National Museum of Emerging Science and Innovation (Miraikan) in Tokyo, Japan. The participating child and the experimenter sat facing each other at the corner of a table (Fig. 1). A MacBook Pro 13-in. notebook computer was placed in front of the experimenter and used to control the stimulus presentation. A MacBook Pro 15-in. notebook computer was placed in front of the child and used to present the stimuli of the Dominance Task. A 60-in. television screen was placed in front of the child across the table to present other stimuli. Two black trays-one for the participating child and the other for the experimenter-served as coin containers. Three sheets of stickers, each containing approximately 20 stickers that are attractive to children, were used as rewards in the games. The experiment was recorded using a voice recorder placed on the table.

Procedure
The task consisted of two phases: the warm-up and the testing phase. In the warm-up phase, the participating children selected their favorite sheet of stickers from the three options, which was then used during the experiment as a reward. The experimenter explained to the child that they would take part in several quiz games in which they could win coins for each correct answer, which could be traded for stickers. All the participating children categorically affirmed that they liked the stickers they had selected and wanted to win more coins to get more stickers. To familiarize the children with the experimental circumstances, the experimenter asked them to answer four easy quiz questions (e.g., "Is an elephant bigger than a mouse or smaller than a mouse?").
In the testing phase, the children were shown a sequence of Power-Point slides on the TV screen and they answered related questions (see Supplementary Information). Each of the two continuous slides described an event that included a specific goal, as well as two animated individuals who applied different goal-directed actions (Fig. 2). The "ordinary" individual applied a normal method, while the other individual applied an extraordinary method. We presented three types of events with different goals. In the psychological event, the goal was to know what was in a sealed box; in the physical event, the goal was to reach a house; and in the biological event, the goal was to light a fire. To achieve these goals, the ordinary individuals opened the box to see what was inside, walked to the house, and lit the fire with a long lighter. In contrast, the extraordinary individuals knew what was inside the box without opening it, flew to the house, and emitted fire from their mouths to light the fire (Hatano & Inagaki, 1997;Lane et al., 2010;Schult & Wellman, 1997).
These events were presented five times. During each presentation sequence, the children answered a specific question regarding the paired individuals (i.e., ordinary individuals and extraordinary individuals). The five main questions are as below.
(1) The surprise question: "Which individual surprised you?" (2) The chair dominance question: The chair dominance question and the posture dominance question aimed to test whether the children judged that the extraordinary individual had more social power than others. The questions were presented on the notebook screen in front of the children. The chair question was a modified version based on the work of Bas and Sebastian-Galles (2021) and Enright, Gweon, and Sommerville (2017). The children were shown a vertical arrangement of two chairs, where the upper chair was bigger and more attractive than the lower chair (Fig. 3). The experimenter instructed each child as follows: "There are two chairs here: one is in the higher position, bigger, and more attractive than the other. Everyone said that they wanted to sit in the upper chair. See the following pictures and tell me who should sit on the upper (or lower) chair?" To prepare the children for the dominance question and confirm that they could correctly link the higherranked character to the upper chair, the chair comprehension question was asked. Specifically, each child was shown images of a lion and a mouse, presented on the left and right sides of the TV screen, respectively, and asked: "Who should sit on the upper (or the lower) chair?" After the child answered the question, they were shown a picture that showed that the lion was sitting on the upper chair and the mouse was on the lower chair. Then, the experimenter showed the children the goal-directed events and asked the child to indicate the individual who should sit on the upper (or lower) chair (i.e., chair dominance question).
(3) The posture dominance question: The posture task was a modified version of that used by Charafeddine et al. (2015Charafeddine et al. ( , 2019. The experimenter showed each child a slide in which two characters displayed postures of dominance and subordination (Fig. 3). The experimenter then instructed the child, "Look! There are two main characters. One character tells the other: 'You must do everything I say! Do whatever I want!' The other character says: 'OK! I will do what you want.'" These instructions were followed by posture comprehension question 1: "Who is saying, 'Do what I want?' and who is saying 'OK! I will do what you want?'" The children were then shown the lion and the mouse again, and posture comprehension question 2 was asked (for same purposes as the chair dominance task): "In fact, this picture represents the relationship between the lion and the mouse. Could you tell me which animal is represented by the character saying, 'Do what I want?' Could you tell me which animal is represented by the character saying, 'OK! I will do what you want?'" After the child answered the question, the experimenter told the child that the lion should be represented by the dominant character. Then, the experimenter showed the children the goal-directed events and asked them to indicate the individual who should be presented by the dominant (or subordinate) character (i.e., posture dominance question).
(4) The "great" question: "Which individual do you think is greater than the other?" (5) The expectation question: "In our world, do you think people would fly/emit fire from their mouth/know what is inside of the box without opening the box?" The expectation question aimed to test whether the children recognize individuals exhibiting extraordinary methods as usual or unusual-whether the children think people would do such things in daily life. This question may provide a more specific explanation regarding why the children feel these individuals are surprising.
The questions were presented in the above order, except that the order of the chair dominance question and the posture dominance question was counterbalanced across the participants. To counterbalance factors-such as the characteristics of the individuals (extraordinary or ordinary), position (upper or lower; e.g., to ensure that the spatial positions of the extraordinary and ordinary characters do not always match with the spatial positions of the high and attractive/low and unattractive chair), and the order of the events (e.g., psychological, Note. The column titled "Ordinary" shows individuals who exhibited normal capabilities and that titled "Extraordinary" shows those who exhibited impossible capabilities, across the psychological, physical, and biological domains. Also shown, side by side, are the sample images used in the tasks. The black dots indicate the mean of the extraordinary-dominant score (scores higher/ lower than 0.5 indicate that the children attribute higher status to extraordinary/ordinary individuals than the other). Error bars are 95% confidence intervals. biological, or physical)-we prepared eight patterns in which the factors were pseudo-randomized. Furthermore, the order of the chair dominance task and the posture dominance questions was counterbalanced; half of the children participated in the chair dominance task first and the other half participated in the posture dominance task first. In the first task, the questions were fixed as "Which individual should sit on the upper chair?" and "Which individual is represented by the character saying, 'Do what I want'?" In the second task, the questions were fixed as "Which individual is represented by the character saying, 'OK! I will do what you want'" and "Which individual should sit on the lower chair?" Regardless of how the children answered the questions, the experimenter gave them one coin for one main question, without indicating whether the answer was correct. The task lasted for approximately 10 min per child.

Analysis plan
To determine whether the children expected the extraordinary individuals to be socially dominant, we calculated an extraordinarydominant score and tested whether its mean was statistically different from chance (i.e., a score of 1). This score indicated the child's total number of extraordinary-dominant answers to the chair dominance question and the posture dominance question (range = 0-2). Children's answers to the chair dominance question were assigned a 1 if they judged that the extraordinary individual would sit on the upper chair. Similarly, children's answers to the posture dominance question were assigned a 1 if they judged that the extraordinary individual is represented by the character who says, "Do what I want." As the data were expected to show non-normal distributions, we used a single-sample Wilcoxon signed-rank test for hypothesis testing. The data for each event type were analyzed separately. We set the p-value for significance at 0.05.

Results and discussion
The children reported being surprised by the extraordinary rather than the ordinary individual in 118/123 trials (reserved judgment was found in two trials for the psychological domain, one trial for the physical domain, and two trials for the biological domain; based on the preregistered data exclusion rules, responses corresponding to these five events were excluded from the following analyses). Our single-sample Wilcoxon signed-rank tests indicated that the extraordinary-dominant score was significantly above chance for all the events: The children judged the individuals who showed extraordinary capabilities in the psychological domain (M = 1.64, SD = ±0.584, V = 405, p < .001, r = 0.743), physical domain (M = 1.6, SD = ±0.632, V = 419, p < .001, r = 0.693), and biological domain (M = 1.39, SD = ±0.815, V = 368, p = .007, r = 0.431) to be more socially dominant than those who showed ordinary capabilities (Fig. 2).
Across the three domains, the children judged the extraordinary individuals as "great" in all of the 118 trials (in 39 trials for the psychological domain, 40 trials for the physical domain, and 39 trials for the biological domain). Furthermore, 37/39 children judged that a person would not know what is inside a box without opening it, 38/40 children judged that a person would not fly, and 39/39 children judged that a person would not emit fire from their mouth.
Our exploratory analysis showed that the extraordinary-dominant score did not vary significantly by the children's gender (Mann-Whitney U tests, adj. ps > 0.123) or age (Spearman correlations, ps > 0.146). Moreover, because children may perceive manipulations of social status in different ways (Kajanus, Afshordi, & Warneken, 2020), we also tested whether the children link ordinary capabilities to high status in response to one question (e.g., the chair dominance question) but not the other (e. g., the posture dominance question). Our single-sample Wilcoxon signed-rank tests indicated that the children were more likely to judge (higher than 0.5) the individuals who showed extraordinary capabilities to be socially dominant than those who showed ordinary capabilities, for both the chair dominance question (M = 0.78, SD = ±0.416, V = 5474, p < .001, r = 0.559) and the posture dominance question (M = 0.763, SD = ±0.427, V = 5355, p < .001, r = 0.525). This tendency was also observed in terms of psychological, physical, and biological factors when we separated the scores by different status questions (singlesample Wilcoxon signed-rank tests, ps < 0.018), with no significant differences among domains within each status question (Kruskal-Wallis tests, ps > 0.277) or significant difference between status questions for each domain (paired-samples Wilcoxon signed-rank tests, ps > 0.565).
The children clearly judged the extraordinary individuals as surprising (the surprise question), great (the "great" question), and unusual (the expectation question), compared to the ordinary individuals. More importantly, the children predicted that individuals showing extraordinary capabilities will gain contested resources and play a dominant role in interactions with ordinary individuals, indicating a higher social status. However, it is not clear whether the children's inferences are specific to the link between extraordinary capabilities and social status. For instance, children may tend to attribute higher social status to any "unusual" individuals who show surprising but humanly possible capabilities (i.e., unusual but not extraordinary). Moreover, children may tend to assume that extraordinary abilities go with any positive characteristic (a "halo effect" whereby children attribute positive qualities to individuals who are known to possess other positive qualities). The following experiments were conducted to address these alternative explanations, which help to evaluate to what extent the extraordinarydominant expectation can be supported.

Experiment 2
Experiment 2 tested whether the children are more likely to attribute higher social status to individuals whose goal-directed methods are surprising and unusual, but not humanly impossible, than ordinary individuals.

Participants
The final sample included 41 5-6-year-old children living in Tokyo, Japan (18 girls; M age = 71.2 months, SD = ±6.81). The sample size, stopping rule, and the experimental procedure were identical to those of Experiment 1. The total number of participants was 47. Based on the data exclusion rules of Experiment 1, six children were excluded because they gave at least one incorrect answer for either the chair comprehension question or the posture comprehension question. Written informed consent was obtained from the children's carers before the experiment was conducted.

Set-up
The set-up was identical to that of Experiment 1.

Procedure
The procedure was identical to that of Experiment 1, except for the following manipulation (Fig. 4). In Experiment 2, the participants made judgments between an "ordinary" individual, who was introduced in Experiment 1, and an "unusual" individual who applied obviously unusual but humanly possible methods. The unusual individuals opened the box from below to see what was inside, crawled to the house, and lighted their shoes that they then used to light the fire.

Analysis plan
The analysis plan was identical to that of Experiment 1, except for the following. In Experiment 2, to determine whether the children expected the unusual individuals to be socially dominant, we calculated an unusual-dominant score and tested whether its mean was statistically different from chance (i.e., a score of 1). This score indicated the child's total number of unusual-dominant answers to the chair dominance question and the posture dominance question (range = 0-2). Children's answers to the chair dominance question were assigned a 1 if they judged that the unusual individual would sit on the upper chair. Similarly, children's answers to the posture dominance question were assigned a 1 if they judged that the unusual individual is represented by the character who says, "Do what I want."

Results and discussion
The children reported being surprised by the unusual rather than the ordinary individual in 109/123 trials (reserved judgment was found in five trials for the event of opening the box, four trials for the event of reaching the house, and five trials for the event of lighting the fire; based on the preregistered data exclusion rules, responses corresponding to these 14 events were excluded). Single-sample Wilcoxon signed-rank tests indicated that the unusual-dominant score was significantly lower than chance for all the events: The children judged the individuals who showed unusual capabilities in the event of opening the box (M = 0.17, SD = ±0.378, V = 0, p < .001, r = − 1), the event of reaching the house (M = 0.22, SD = ±0.417, V = 0, p < .001, r = − 1), and the event of lighting the fire (M = 0.17, SD = ±0.378, V = 0, p < .001, r = − 1) to be less socially dominant than those who showed ordinary capabilities (Fig. 4).
Across the three events, the children judged the unusual individuals as "great" in 41/109 trials (in 14 trials for the event of opening the box, 14 trials for the event of reaching the house, and 13 trials for the event of lighting the fire; binomial tests showed that the children's judgments did not differ from chance, ps > 0.133). Furthermore, 36/36 children judged that a person would not open a box from below, 33/37 children judged that a person would not crawl to a house, and 36/36 children judged that a person would not light their shoes and then uses those shoes to light a fire.
Further analysis showed that the extraordinary-dominant score did not vary significantly by the children's gender (Mann-Whitney U tests, ps > 0.146) or age (Spearman correlations, ps > 0.207). Moreover, single-sample Wilcoxon signed-rank tests indicated that the children were less likely to judge (lower than 0.5) the individuals who showed unusual capabilities to be socially dominant than those who showed ordinary capabilities, for both the chair dominance question (M = 0.03, SD = ±0.164, V = 165, p < .001, r = − 0.945) and the posture dominance question (M = 0.16, SD = ±0.364, V = 953, p < .001, r = − 0.688). This tendency was also observed in terms of event factor when we separated the scores by different status questions (single-sample Wilcoxon signed-rank tests, ps < 0.001), with no significant differences among events within each status question (Kruskal-Wallis tests, ps > 0.792) or significant difference between status questions for each event (paired-samples Wilcoxon signed-rank tests, adj. ps > 0.06).
The children clearly judged the unusual individuals as surprising (the surprise question) and unusual (the expectation question), but not great (the "great" question), compared to the ordinary individuals. More importantly, the children predicted that the unusual individuals have lower social status than the ordinary individuals: The ordinary individuals will gain contested resources and play a dominant role in interactions with unusual individuals. These findings ruled out the possibility that children are biased to attribute higher social status to any "unusual" individuals even if their capabilities are humanly possible (i. e., unusual but not extraordinary).

Experiment 3
The extraordinary-dominant expectation observed in Experiment 1 might be explained as a manifestation of a "halo effect" in children, whereby children attribute positive qualities to individuals who are known to possess other positive qualities. For example, children may be likely to positively evaluate being able to fly, breathe fire, or know the content of containers without looking inside. Then, if children are asked to attribute some other positive quality (e.g., high status) to either these characters or to characters who lack those abilities, a halo effect would manifest as children attributing higher status to individuals who possess the extraordinary abilities.
Experiment 3 aimed to test whether, as predicted by the "halo effect," the children are indiscriminately attributing any positive characteristic to extraordinary capabilities. The participants observed the pairs of ordinary and extraordinary individuals who were introduced in Experiment 1. There were two main questions: a "mighty" question, aimed to replicate the extraordinary-dominant expectation and a "good" question, aimed to test whether the children really perceive extraordinary individuals as positive.

Participants
The final sample included 53 5-6-year-old children living in Osaka, Japan (26 girls; M age = 70.9 months, SD = ±3.35). The sample size, stopping rule, and experimental procedure were identical to those of Experiment 1. All the children gave correct answers to a mighty comprehension question so that no child was excluded. Written informed consent was obtained from the children's carers before the experiment was conducted.

Set-up
Children were tested individually by a male experimenter in a separate booth at a kindergarten in Osaka, Japan. The set-up was identical to that of Experiment 1, except in Experiment 3, all the stimuli were presented by a MacBook Pro 16-in. notebook.

Procedure
As in Experiment 1, the participants made judgments about an "ordinary" individual and an "extraordinary" individual in three domains Note. The column titled "Ordinary" shows individuals who exhibited normal capabilities and that titled "Unusual" shows those who exhibited surprising/ unusual but possible capabilities. Also shown, side by side, are the sample images used in the tasks. The black dots indicate the mean of the unusualdominant score (scores higher/lower than 0.5 indicate that children attribute higher status to unusual/ordinary individuals than the other). Error bars are 95% confidence intervals. (Fig. 5). The stimuli and the procedure were identical to those of Experiment 1, except for the following manipulations.
The ordinary events and the extraordinary events in all the domains were presented three times. During each presentation sequence, the children answered a specific question regarding the paired individuals (i.e., ordinary individuals and extraordinary individuals). The three main questions were: (1) the surprise question: "Which individual surprised you?"; (2) the "mighty" question: "Which individual do you think is mightier than the other?"; and (3) the "good" question: "Which individual do you think is better than the other?" The surprise question was asked first. The order of the "mighty" question and the "good" question was counterbalanced across the participants.
The "mighty" question aimed not only to test whether the extraordinary-dominance expectation observed in Experiment 1 can be conceptually replicated but also to extend the findings by testing whether children show extraordinary-dominance expectations when they are explicitly asked linguistic questions. Although previous studies have used different linguistic expressions to evaluate children's judgments about social status (e.g., "which is the boss?" and "who is in charge?"; Charafeddine et al., 2015;Gülgöz & Gelman, 2017), they are not commonly used in the daily life of Japanese children. "Mighty" ("Tsuyoi" in Japanese) is a very child-friendly linguistic expression in Japanese for depicting the social dominance (e.g., powerful, dominant) of individuals. More importantly, the concept of "mighty" has been widely used in studies on social dominance in human and non-human animals (e.g., Smith et al., 2016;Thomsen, Frankenhuis, Ingold-Smith, & Carey, 2011). Therefore, we deemed this word the most suitable for testing whether the current samples of children consider the extraordinary individuals as being more socially dominant and having higher social status than the ordinary individuals.
The "good" question was used to test whether children attribute positive characteristics other than having high status to the extraordinary individuals. Although many characteristics, such as "smart" and "trustworthy," could be valued as positive, we used "good" because it is one of the best-known words for children to describe positive things, at least in Japanese culture.
According to the procedure of Experiment 1, we also asked a "mighty" comprehension question immediately before the "mighty" question. This was used to prepare children for the "mighty" question and confirm that the participants could correctly link the higher-ranked character to the word "mighty." Specifically, each child was shown images of a lion and a mouse, presented on the left and right sides of the TV screen, respectively, and asked: "Which one do you think is mightier than the other?" The children who did not provide correct answers to the "mighty" comprehension question (i.e., who did not judge the lion as mightier than the mouse) were excluded from the analysis because they may not understand the meaning of the word of "mighty."

Analysis plan
The analysis plan was identical to that of Experiment 1, except for the following. In Experiment 3, we investigated whether the children judged the extraordinary individuals to be more socially dominant than the ordinary individuals by testing whether children were more likely to choose the extraordinary individuals than chance (i.e., 0.5) in the "mighty" question. This was tested using an extraordinary-dominant score, which was coded as 1 if the child judged the extraordinary individuals to be mightier (coded as 0 for the reverse judgments). Similarly, an extraordinary-good score, which was scored as 1 if the child judged the extraordinary individuals to be better than the ordinary individuals (scored as 0 for the reverse judgments), was also compared to chance.

Results and discussion
Children reported being surprised by the extraordinary rather than the ordinary individual in 142/159 trials (reserved judgment was found in eight trials for the psychological domain, five trials for the physical domain, and four trials for the biological domain; according to Experiment 1, responses corresponding to these 17 events were excluded from the following analyses). Single-sample Wilcoxon signed-rank tests indicated that the extraordinary-dominant score was significantly above chance for all the events: The children judged the individuals who showed extraordinary capabilities in the psychological domain (M = 0.96, SD = ±0.208, V = 989, p < .001, r = 0.911), physical domain (M = 0.92, SD = ±0.277, V = 1125, p < .001, r = 0.837), and biological domain (M = 0.84, SD = ±0.370, V = 1071, p < .001, r = 0.680) to be more socially dominant than those who showed ordinary capabilities (Fig. 5). Further analysis showed that the extraordinary-dominant score did not vary significantly by the children's gender (Mann-Whitney U tests, ps > 0.414) or age (Spearman correlations, ps > 0.164). Moreover, the score showed no significant differences among domains (Kruskal-Wallis tests, p = .151).
Single-sample Wilcoxon signed-rank tests also showed that the extraordinary-good score was significantly lower than chance for all the events: The children were less likely to judge the individuals who showed extraordinary capabilities in the psychological domain (M = 0.20, SD = ±0.405, V = 207, p < .001, r = − 0.600), physical domain (M = 0.02, SD = ±0.143, V = 25, p < .001, r = − 0.959), and biological domain (M = 0.06, SD = ±0.240, V = 76.5, p < .001, r = − 0.880) to be better than those who showed ordinary capabilities. Further analysis showed that the extraordinary-good score did not vary significantly by the children's gender (Mann-Whitney U tests, ps > 0.537) or age (Spearman correlations, ps > 0.253). The score showed significant differences among domains (Kruskal-Wallis tests, p = .007): The children were more likely to evaluate the ordinary individuals as better than the extraordinary individuals in the physical domain (Mann-Whitney U tests, adj. p = .005), compared to the psychological domain.
The participants clearly judged the extraordinary individuals as surprising (the surprise question) and mighty (the "mighty" question), compared to the ordinary individuals. These findings conceptually replicated those of Experiment 1. However, the participants were not biased to evaluate extraordinary individuals as "good." On the contrary, Note. The column titled "Ordinary" shows individuals who exhibited normal capabilities and that titled "Extraordinary" shows those who exhibited impossible capabilities, across the psychological, physical, and biological domains. Also shown, side by side, are the sample images used in the tasks. The black and white dots indicate the mean of the extraordinary-dominant score (scores higher/lower than 0.5 indicate that the children attribute higher status to extraordinary/ordinary individuals than the other) and the extraordinary-good score (scores higher/lower than 0.5 indicate that children assume extraordinary/ordinary individuals as better than the other), respectively. Error bars are 95% confidence intervals.
they judged the ordinary individuals as better than the extraordinary individuals. These findings showed that children attribute high social status, but not every positive quality, to extraordinary individuals.

General discussion
In Experiment 1, across different knowledge domains regarding realworld causalities, the children consistently judged extraordinary capabilities as surprising and humanly impossible. More importantly, they predicted that individuals achieving goals using extraordinary methods would prevail and play a dominant role-in other words, have higher social status-in interactions with ordinary individuals. Experiment 2 showed that children do not attribute higher social status to any individuals who show surprising, unusual, but humanly possible capabilities (i.e., unusual but not extraordinary). Experiment 3 conceptually replicated the extraordinary-dominant expectation observed in Experiment 1 and ruled out the possibility that children assume that extraordinary abilities go with any positive characteristic. Together, these findings showed that from the age at which we begin to show a coherent and theoretical network of causal beliefs regarding how human bodies and minds work, we are already biased toward attributing higher social status to people with extraordinary (impossible) capabilities than those with ordinary (possible) capabilities. Human history shows that highranked social status has been bolstered by claims of extraordinary power that overcome human limits (Bowie, 1999;Burkert, 1998;Mullins et al., 2018;Singh, 2017;Weber, 1947;Winkelman, 1990). This is the first time we have observed the phenomenon of early-development extraordinary-dominant expectation during the socialization process in ontogeny.
The participating children judged the extraordinary individuals (Experiment 1), but not the unusual individuals (Experiment 2), as "great." This suggests that children respect extraordinary individuals, possibly because children acknowledge their persuasive competence. These findings support the idea of explaining the extraordinarydominant expectation based on the competence model. Specifically, children perceive extraordinary capacities as high competence and, therefore, expect a highly hierarchical position for the individuals who are believed to have such competence (Cheng, 2020;Henrich, 2016;Henrich & Gil-White, 2001;. A previous study suggests that preverbal infants show "surprise" (look longer) when seeing a "normal" animated geometric character outcompeting a "levitating" one in securing a reward ; this indicates an extraordinary-dominant expectation in very early development. However, given the limitations of infant studies (e. g., it is difficult to know whether infants see a "flying" character as counterintuitive in the real world), the conclusion is still unclear. The current compelling evidence demonstrates that, from the age in which we show a coherent and theoretical network of causal beliefs regarding how human bodies and minds work and their limits (Gopnik & Wellman, 2012), we already attribute higher status to individuals who show extraordinary capabilities across different domains of intuitive knowledge, which suggests that this may be a natural consequence of the way human minds develop.
The current experiments counterbalanced experimental factors such as the position (i.e., upper and lower; in all experiments) of the individuals, the identities (i.e., clothes; in all experiments) of the individuals, the order of the events of different domains (in all experiments), the order of the status questions (in Experiments 1 and 3), the target that the experiment asked to identify in the status questions (in Experiments 1 and 3), and the order of the "mighty" question and the "good" question (in Experiment 2). Therefore, the children's judgments can hardly be explained as by-products of any errors of experimental manipulations. For example, in Experiments 1 and 3, if children mostly selected the extraordinary individuals as the response to every question in which they were asked to pick one of the two characters (i.e., if they are inclined to select the extraordinary individual regardless of what they are asked), then the extraordinary-dominant expectation would not be observed. This is because in half the questions, the children were asked to identify which individual is the low-status target.
Experiment 2 showed that the children tend to attribute higher social status to ordinary individuals rather than unusual individuals who apply surprising, unusual, but humanly possible goal-directed methods. The findings of Experiments 1 and 2 suggest that whether the individuals' capabilities are humanly impossible or humanly possible influences whether children attribute high social status to them. The differentiated evaluations regarding impossible and unusual individuals are in line with previous findings: children judge that impossible events cannot happen while usual events can happen (Lane, Ronfard, Francioli, & Harris, 2016); children are more likely to recall and show curiosity toward impossible concepts (i.e., counterintuitive events violating our ontological expectation) than unusual but possible concepts (Banerjee, Haque, & Spelke, 2013;Lewry, Gorucu, Liquin, & Lombrozo, 2023). The present results suggest that such a distinction not only influenced cognitive domains (e.g., memory, attention) but might also have influenced social evaluations.
A difficulty usually faced by developmental psychologists is that sometimes infants' and children's social judgments seem to be interpretable by a "halo effect" (i.e., children attribute positive/negative qualities to individuals who are known to possess other positive/negative qualities; Bas & Sebastian-Galles, 2021;Brey & Shutts, 2015;Cheng, Wan, An, Gummerum, & Zhu, 2021;Enright et al., 2017;Meng, Ishii, et al., 2019). However, when shown the pictures used in the posture dominance question of Experiment 1, unlike preschoolers in Western culture who were found to value dominant more than subordinate individuals, preschoolers in Japanese culture valued dominant and subordinate individuals similarly (Charafeddine et al., 2019). Given that the extraordinary-dominant expectations for the posture dominance question were similar to those of the chair dominance question, the expectations did not seem to be influenced by an association of positive/ negative qualities. Experiment 3 showed more conclusive evidence regarding the "halo effect." Although children judged the extraordinary individuals as mightier than the ordinary individuals, their evaluation about "good" was reversed: They thought that the ordinary individuals were better than the extraordinary individuals. Indeed, these findings may even suggest that the participants were more likely to negatively evaluate the extraordinary individuals than the ordinary individuals. Therefore, the psychological attribution process of the extraordinary-dominant expectation does not seem to be that children see a person who could breathe fire and think "this person has extraordinary abilities and likely has other good properties, such as having high social status," but rather "this person has extraordinary abilities and is therefore likely to be high in status." The possibility of the "halo effect" should also be evaluated from various angles in future studies. For instance, researchers could conduct a study in which persons have extraordinary but negative qualities: If the extraordinary qualities still lead children to attribute higher status to individuals, those results may lend greater support to the extraordinary-dominant expectation.
The children evaluated the ordinary individuals as better than the extraordinary individuals. This might be because the latter violated the normative approach toward goals (e.g., Rakoczy, Warneken, & Tomasello, 2008;Tomasello & Vaish, 2013). Interestingly, this seems to be consistent with the fact that, historically, individuals who possess extraordinary yet socially undesirable qualities can often be the targets of punishments, such as shunning (e.g., Mace et al., 2018;Winkelman, 1990). It would also be beneficial to investigate how children evaluate extraordinary characters in other ways, such as smart, charismatic, caring, and more trustworthy. This would provide a more comprehensive picture regarding children's social evaluations of these characters and to what extent the inference of the characters is specifically limited to social status.
Cartoon figures have been widely used in studies of children's causal reasoning across different domains and their understanding of hierarchical relationships, partly because their superficial perceptual properties can be easily controlled (Amakusa et al., 2022;Enright et al., 2020;Gülgöz & Gelman, 2017;Koval & Rosette, 2021;Lane et al., 2010Lane et al., , 2012Okanda et al., 2022;Wellman et al., 2001). To increase the likelihood that the children recognize the characters as human beings rather than something like fictional characters, we introduced the characters with human cartoon figures and described them by the term "hito." This word is used daily to represent "human beings" in Japanese. Previous studies have shown that Japanese preschool children discriminate between human and non-human existences (e.g., God, stone, and invisible agent) regarding motivational/emotional and cognitive mental capacities (Moriguchi & Shinohara, 2012;Moriguchi, Takahashi, Nakamata, & Todo, 2019). The notion that the children viewed the characters as human beings could also be supported by the fact that most of the children answered that they were surprised by the extraordinary rather than the ordinary individuals; this would not happen if the children had treated the extraordinary characters as fictional existences. An example based on this logic is that the children would not find the Japanese superhero character Anpanman's ability to fly surprising because they understand that flying is considered a normal action in that fictional world. Furthermore, as all the experiments used cartoon figures, and children's judgments varied by experiments, we believe that what capacities the individuals had (i.e., the experimental manipulation) were more effective than whether the children considered the individuals as real people or fictional existences to influence the children's evaluation of the individuals.
Nevertheless, the current findings did not provide strong evidence regarding to what extent the children believed the characters were meant to represent actual people rather than non-human existences. Even if we introduce the individuals as pictures of real humans, it might still be difficult to completely prohibit children from recognizing the characters as fictional existences (e.g., Superman in fictional movies has human-like appearance). However, at the same time, this issue would raise a series of important future research directions that would contribute to the psychological mechanism of human cognition regarding extraordinariness and the relationships with social status. One of the questions could focus on whether the link between extraordinary power and a high social hierarchal position needs people's recognition of unreal properties in the character. For instance, although many characters who are highly ranked in human societies have human properties, such as physical appearance (e.g., shamans, cult leaders, and Tianzi), this does not mean that people do not see them as non-human or super-human existence. Another question could focus on how people are differentiating between human and non-human existences based on their capacities. Which individual would people treat as human being or non-human existence? -An individual who runs 100 m in 3 s, an individual who floats in the air, or both? These unsolved issues are not limited to child participants but also adults. Future studies are expected to address these issues to draw a more complete picture of human cognition regarding extraordinariness and social status.
This study has several potential limitations, which may lead to further research. First, by using the "mighty" question, Experiment 3 conceptually replicated the extraordinary-dominance expectation observed in Experiment 1 and demonstrated that children show extraordinary-dominance expectations when they are explicitly asked linguistic questions. However, because definitions of the Japanese word "tsuyoi" ("mighty") include "strong" and "powerful," one might argue that the extraordinary-dominance score merely reflects the children's expectations that the extraordinary individuals have greater physical power, not higher social status (which the current study is focusing on), than the ordinary individuals. We acknowledge this possibility but also would like to point out that many studies have shown that children evaluate hierarchical relationships among others based on prestige cues (e.g., children attribute higher social status to individuals who have more followers than individuals who have fewer; Enright et al., 2020;Gülgöz & Gelman, 2017) and dominant cues (e.g., children attribute higher social status to individuals who have more physical power than individuals who have less physical power; Enright et al., 2020;Terrizzi, Brey, Shutts, & Beier, 2019). Therefore, the evaluation of physical power would trigger the expectation of social status in children. Moreover, Terrizzi et al. (2019) showed that 4-5-year-olds attributed both strength ("Who's stronger?") and authority ("Who's in charge?")-multiple notions of power-from a single aspect of a person's physical appearance, suggesting that preschool-aged children's understanding of power weaves together notions of both physical formidability and normative authority. Therefore, even if in Experiment 3 children interpreted "tsuyoi" as "physically strong," the results would still support the argument that children expect extraordinary individuals to have higher social rank than ordinary individuals. Nevertheless, future studies should test in a more detailed manner, with empirical evidence, whether children distinguish between high social status and great physical power and whether children's extraordinary-dominant expectation is derived from the evaluation of the physical power of the individuals.
Second, the expectation question in Experiments 1 and 2 showed that the participants did not expect that people would exhibit the extraordinary actions in daily life. This question provided specific explanation regarding why the participants viewed the extraordinary individuals as surprising (i.e., the surprise question): The children recognized that the possibility of the extraordinary actions occurring is low (i.e., unusual). However, the question phrasing presented in Japanese might be ambiguous. In addition to the intended meaning of "do you think people would XXX," the question could also be interpreted as "do you think people can XXX?" If the participants interpreted the question as the latter, then the participants' "no" answers might suggest that the participants did not think the unusual actions could be achieved (it was not likely that the participants did not understand what the question means because in that case the participants would make judgments randomly). This would be a very interesting phenomenon because obviously people can crawl, open boxes on the wrong end, or light their shoes. However, in Experiment 2, many participants spontaneously argued that although they think even babies can crawl, they do not think that people would crawl to a house because doing that would make their clothes dirty. Therefore, we believe that the participants interpreted the expectation question as "do you think people would XXX?" More importantly, even if the participants thought that the unusual actions cannot be achieved, this still suggests that the participants recognized that the possibility of the extraordinary actions occurring is low. Future studies should use more accurate phrasing to test how children recognize the possibility that extraordinary and unusual actions occur, from both the "would" and "can" perspective.
In the current study, we did not collect information about the children's socioeconomic status (SES; e.g., household income level). This was because we did not have theoretical hypotheses related to the relationship between the children's "extraordinary-dominant expectation" and SES. However, we would like to point out that Experiment 1 was conducted in Tokyo and Experiment 3 in Osaka. Despite the average SES in Osaka being lower than that of Tokyo (Ministry of Health, 2021), Experiment 3 conceptually replicated the findings of Experiment 1 by showing the strong "extraordinary-dominant expectation." Therefore, we think that the "extraordinary-dominant expectation" would be generalizable across different regions without regard for SES.
The current findings might represent a new wave of research aiming to understand how human beings treat religious phenomena and link it to social interactions. What cognitive, psychological, and neural components are involved? To what extent is expectation driven by sociocultural regulation? Is it an evolved adaptation or a by-product of the normal operation of human cognition? Given their more important and broader impacts, acquiring extraordinary-dominant expectations in early childhood would explain why children are often charmed by "supernatural" individuals; this should serve as a warning that young minds might be susceptible to inducements by such agencies.