«A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Psychology) in the University of Michigan ...»
Developing a Concept of Social Power Relationships
A dissertation submitted in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
in the University of Michigan
Professor Susan A. Gelman, Chair
Associate Professor Ramaswami Mahalingam
Professor Robin M. Queen
Professor Henry M. Wellman
To my parents, Berna and Sami, for their endless love, support, and humor.
ii Acknowledgements I would like to thank my adviser and committee chair Susan Gelman. Without her dedication, her approach to new ideas, and her encouragement to develop my own path, I believe none of this would have been possible. I would also like to thank Henry Wellman, who has been a great mentor to me, as well as a committee member. I am grateful for all the feedback Susan and Henry have given me over the years. My other committee members, Ram Mahalingam, Robin Queen, and Margaret Evans, have also been incredibly encouraging and passionate when discussing my research. I thank all of them for the diverse viewpoints they have brought to this work.
This research would not have been possible without the enthusiastic research assistants of the Conceptual Development Lab who helped collect and code the data: Jennifer Ashley Asong, Amanda Balakirsky, Aaron Chuey, Kristina Ljukovic, Donald Alfred Lyons, Pragya Mathur, Rebecca Sheinman, and Anna Wendorf. I am also grateful to participating families, as well as the Living Lab Initiative, the Ann Arbor Hands-on Museum, the University of Michigan Museum of Natural History, and the Ann Arbor District Library for allowing me to become a part of each of their teams in collecting my data. I would like to thank Defne Civelekoglu for designing the stimuli used in the tasks, and Craig Smith for setting up the Living Lab in Ann Arbor.
I would also like to extend my thanks to the many intelligent people that have made Ann Arbor (and East Hall) my second home: my incredible friends and lab mates, Maria Arredondo, Margaret Echelbarger, Meredith Meyer, Steven Othello Roberts, and Craig Smith; my amazing cohort members, Lauren Reed, Ju-Hyun Song, and Sarah Trinh; and other wonderful people who iii have shared this journey with me, Tissyana Camacho, Soraya and Jaron Giaccardi, Meghan Martz, Ryota Nishiyori, Paige Safyer, and Irene Wu. I am especially indebted to Maria Arredondo, Paige Safyer, and Steven Othello Roberts for their friendship.
I am grateful for the years of support that my best friends Beyza Boyacioglu, Defne Civelekoglu, and Ceren Serifoglu have given me, despite the distances between us. Finally, I would like to thank my partner, Francisco Velasquez, who brings immense joy, love, and laughter to my every day. His unwavering positivity and belief in me have made these past four years the best years of my life.
5 Frequencies for Level 2 coding of 3- to 4-year-old participants’ responses to the 115 open-ended questions (Study 2b) 6 Frequencies for Level 2 coding of open-ended responses to permission (Study 2b) 116
6 Comparison of mean correct scores on choice questions of Study 2b and Study 3 122 7 (Study 4) Mean number of times participants judged the powerful character to be 123
Power differences organize social relations across species. They emerge early in development, and are observed in children’s early relationships with peers and adults. Despite the ubiquity of social power relations, little is known about how children conceptualize them.
This dissertation provides an experimental examination of children’s developing understanding of social power relationships between individuals, and among members of social groups.
In Part I, Studies 1, 2, and 3 provide an extensive investigation of 3- to 9-year-old children’s and adults’ sensitivity to interpersonal social power relations across five manifestations of power: resource control, goal achievement, permission, giving orders, and setting norms. These studies examine children’s understanding of power both in situations where the powerful individual may be perceived as unkind (Studies 1 and 2), and in situations where the powerful individual may be perceived as benevolent (Study 3). Findings reveal that children as young as 3 or 4 years old represent social power relations between individuals across several dimensions of power, when presented with powerful individuals who were malevolent as well as benevolent. As predicted, sensitivity to social power in resource control, goal achievement, and permission situations emerges earlier in development. With age, children’s sensitivity extends over all five of the dimensions tested, becoming almost adult-like by age 7 to 9.
Part II of the dissertation examines children’s sensitivity to power relations between members of social categories. Participants are shown vignettes depicting two individuals contrasting in power, and are asked to identify the relative age (Study 4) or gender (Study 5) of
gender based on power differentials, and that even adults do not consistently map power onto these social categories.
Overall, this dissertation provides one of the first in-depth experimental examinations of children’s developing concepts of social power. The findings show that children are sensitive to social power relations early on, and even use these power relations to make inferences about people’s social group memberships.
“[T]he fundamental concept in social science is Power, in the same sense that Energy is the fundamental concept in physics…The laws of social dynamics are laws which can only be stated
Children are born into systems of social relations characterized by power hierarchies. In the family, power dynamics shape relationships between two parents, between parents and children, and among siblings. At school, children are exposed to rules and power hierarchies, with teachers and principals formally identified as those who are in charge. Children’s peer relationships involve intricate power relations, where even play is structured by dynamics between those who lead and those who follow. Power also characterizes dynamics between and within social groups, including those based on gender, race, and social status - concepts that children develop and use from an early age. The ubiquity of power in social relationships, both at the individual and group level, implies that developing an understanding of social power is crucial for children to successfully navigate the social world. Having a concept of power early on would allow children to recognize relational dynamics, gain understanding of behaviors culturally appropriate for different situations, predict others’ behaviors, and know which social alliances will benefit them. Yet surprisingly little is known about the developmental origins of a concept of social power, its predictors, and its consequences.
∗ Cited in Guinote & Vescio (2010).
perspectives. From philosophers to historians and literary theorists, from natural scientists to social scientists, power has been associated with numerous aspects of social life at the level of the individual (e.g., parent-child relationships), the level of social groups (e.g., relationships between majority and minority groups), and the level of institutions (e.g., relationships between the government and the people). Social psychologists have shown that perceptions of social power influence cognitive and social psychological outcomes such as attention, emotion, selfregulation, and social cognition (for a review, Keltner, Gruenfeld, & Anderson, 2003). In addition, social power has been used to explain processes of social discrimination based on factors including gender, race, and political ideology (e.g., Glick & Fiske, 1999; Guinote, Willis, & Martelotta, 2010; Ho, Sidanius, Cuddy, & Banaji, 2013; Jost & Banaji, 1994; Sidanius & Pratto, 1999; Vescio, Gervais, Heidenreich, & Snyder, 2006;). However, there is often little communication between these various fields of study, and therefore there is no unified definition of social power. Moreover, there is a dearth of understanding in developmental psychology as to how humans develop a concept of power. The studies in this dissertation provide an in-depth, experimental examination of children’s and adults’ concepts of social power relations between individuals, as well as between members of different social categories.
Origins of conceptualizing social power An individual’s social world is vast, complex, and dynamic. Being able to accurately represent and successfully navigate their large number of interrelated and changing social ties is an important and seemingly difficult task for developing children. Evolutionary psychologists suggest that humans have developed cognitive adaptations that allow them to conceptualize social relations (Cosmides & Tooby, 1992). Studies of nonhuman primates indicate that the
tracking of grooming networks, kinship relations, coalitions, transitive and third-party relations, and quality of relationships with allies and enemies (for a review see Silk, 2007). The adaptive advantage of such social mappings is to maximize access to limited resources, and thus aid survival (Hawley, 1999; Silk, 2007). Across species, recognizing social power allows individuals to build alliances with powerful others, increasing cooperation and in turn, dominance, leading to superior reproductive success (Barton & Whiten, 1993; de Waal & de Waal, 2007; Moll & Tomasello, 2007). Thus, the early-emerging ability for humans to track their social network and the relationships within it carries vital importance.
Similarly, human infants may be born with cognitive biases allowing them to recognize social power differentials early on. Such an expectation is in line with core cognition theories of conceptual development, which suggest that understanding of social relations emerges from innate primitives (Thomsen & Carey, 2013). Recent research in preverbal infants’ concepts of social dominance provides support for this argument. Thomsen, Frankenhuis, Ingold-Smith, and Carey (2011) presented 8- to 13-month-old infants an animated scene where two agentic blocks differing in size were depicted as having conflicting goals (i.e., moving towards each other from opposite directions). Based on findings suggesting that larger physical size (Carney, Cuddy, & Yap, 2010; Stulp, Buunk, Verhulst, & Pollet, 2012; Marsh, Henry, Schechter, & Blair, 2009; Re et al., 2012; Schubert, Waldzus, & Seibt, 2008; Smith & Galinsky, 2010; Yap, Mason, & Ames,
2013) and the ability to achieve goals (Guinote, 2007; Hawley, 1999; Slabu & Guinote, 2010) are associated with dominance, in the expected-outcome condition of the study, the smaller block bent over and moved aside to allow the larger block to continue on its path and achieve its goal, whereas in the unexpected-outcome condition, the larger block bent over and moved aside to
that 10- to 13-month-olds, but not 8-month-olds, looked longer in the unexpected outcome condition, indicating that their expectation of the outcome was violated and that they were surprised by this situation. The authors interpreted this finding as an indication of children’s early disposition to represent social dominance, where infants expect a correlation between an agent’s size and dominance.
This expectation is consistent with findings showing that physical size is tightly linked to adult humans', as well as adult nonhuman primates’, perceptions of social power (Keating, 1985).
Studies with adults have shown not only that taller people are perceived as having more authority (Stulp et al., 2012; Re et al., 2012), but also that powerful individuals are perceived by others as larger in size (Marsh et al., 2009), and that powerful individuals overestimate their own height (Yap et al., 2013). People respond to dominant, larger postures by assuming more submissive postures (Schubert et al., 2008), and adults who are primed to assume a more open and upright posture feel more confident, are willing to take more risks, and experience increased testosterone (Carney et al., 2010). According to evolutionary theories and the core cognition framework of development (Thomsen & Carey, 2013), these findings suggest that the relationship between size and power may be a literal expectation and not simply a cultural metaphor, and further that this association may be a readily accessible cognitive representation.
A recent study by Brey and Shutts (2015) found that preschool- and kindergarten-age children may show sensitivity to other nonverbal cues to social dominance as well. Three- to sixyear-old participants were shown brief videos of two people sitting across from each other, with one displaying several nonverbal bodily cues of higher power (e.g., erect and open posture, upper head orientation) and the other displaying corresponding bodily cues of low power. When asked
identified the person in charge based on the nonverbal cues. Similarly, when shown static photos of two people standing facing each other, 5- and 6-year-olds, but not 3- and 4-year-olds, selected the more upright vs. slouching person, the person with downward head orientation vs. upward head orientation, and the person with direct vs. averted gaze orientation as the person in charge.
Finally, in a study by Charafeddine et al. (in press), French 3- to 5-year-olds were presented with two puppets engaged in a physical fight over two episodes, with the same puppet winning on both instances. When asked “Who is the boss?/C’est qui le chef?”, participants selected the puppet that prevailed.