This item has the following downloads:
1 INDIVIDUAL DIFFERENCES IN THEORY OF MIND AND NARRATIVE IN EARLY ADULTHOOD By MARY ELLEN MCKAY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Mary Ellen McKay
3 To Jane
4 ACKNOWLEDGMENTS First, I would like to acknowledge the endless support and invaluable input of my advisor, Dr. Jeff Farrar know I would also like to thank my committee, Dr. James Algina, Dr. Julia Graber, and Dr. Greg Neimeyer for their time, effort, advice, and guidance. I would also lik e to acknowledge Dr. Susan Bluck, Dr. Scott Miller, Dr. Shari Ellis, and Dr. Walter Cunningham for their input and encouragement over the years. Next, I would like to thank my family. My parents, Daniel and Beverly McKay, my brother Andy, and my sister Lo urdes have been there for me through everything and have always let me know that I can accomplish anything I set my mind to. Special thanks go to Jane, for being my sunshine. Finally, I would like to thank all of the women who have motivated me, uplifted m e, and helped me in every imaginable way. In addition to my mother, my sister, my daughter and my professional icons mentioned earlier I have been so lucky to have the most brilliant, inspiring and devoted friends who have surrounded me throughout graduat e school Virginia Tompkins, Sarah Lynne Landsman, Teri De Lucca, and Erin Hastings Monari my gratitude is beyond measure, personally, professionally spiritually, emotionally, and beyond. I love you all.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATI ONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 2 LITERATURE REVIEW ................................ ................................ .......................... 16 Theory of Mind in Childhood ................................ ................................ ................... 16 Theoretical Accounts of Early Theory of Mind Development ............................ 20 Cognitive Contributors to Theory of Mind ................................ ......................... 25 Role of Theory of Mind in Childhood ................................ ................................ 29 Theory of Mind Beyond Childhood ................................ ................................ .......... 34 Study Aims ................................ ................................ ................................ .............. 44 3 METHODS ................................ ................................ ................................ .............. 47 Participants ................................ ................................ ................................ ............. 47 Procedure ................................ ................................ ................................ ............... 48 Measures ................................ ................................ ................................ ................ 49 Social Cognition ................................ ................................ ............................... 49 General Cognitiv e Abilities ................................ ................................ ............... 53 Autobiographical Memory and Story Narratives ................................ ............... 55 4 RESULTS ................................ ................................ ................................ ............... 59 Descriptive results ................................ ................................ ................................ .. 59 Study Aim 1 ................................ ................................ ................................ ............ 60 Study Aim 2 ................................ ................................ ................................ ............ 61 Study Aim 3 ................................ ................................ ................................ ............ 62 Study Aim 4 ................................ ................................ ................................ ............ 63 5 DISCUSSION ................................ ................................ ................................ ......... 69 Study Aim 1 ................................ ................................ ................................ ............ 69 Study Aim 2 ................................ ................................ ................................ ............ 74
6 Study Aim 3 ................................ ................................ ................................ ............ 80 Study Aim 4 ................................ ................................ ................................ ............ 83 Implications and Con clusion ................................ ................................ ................... 84 LIST OF REFERENCES ................................ ................................ ............................... 87 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 94
7 LIST OF TABLES Table page 4 1 Descriptive information for key measures. ................................ .......................... 65 4 2 knowledge (Cu rse), hindsight bias (HB), language (PPVT 4^) and inhibitory control (Stroop) and story quality. ................................ ................................ ....... 66 4 3 Multiple regression analyses of the effects of emotion recognition (Eyes), inhibitory control (Stroop), and language (PPVT 4) on story quality. .................. 67 4 4 Summary of mediation results for Language (PPVT 4) and Inhibitory Control (Stroop) mediated by Emotion Recognition (Eyes) on story quality using bootstrapping method (5000 bootstrap resamples). OSQ=Overall Story Quality, SB=Storybook only, AM= Autobiographical mem ory only. .................... 68
8 LIST OF FIGURES Figure page 2 1 Mediation model showing an indirect effect of language and executive functioning through Theory of Mind. ................................ ................................ ... 46 3 1 Curse of knowledge task stimulus. ................................ ................................ ..... 58
9 LIST OF ABBREVIATIONS AM Autobiographical Memory EF Executive Functioning FB False Belief(s) IC Inhibitory Control PPVT 4 Peabody Picture Vocabulary Test Version 4 T O M Theory of Mind
10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INDIVIDUAL DIFFERENCES IN THEORY OF MIND AND NARRATIVE IN EARLY ADULTHOOD By Mary Ellen McKay August 2013 Chair: M. Jeffrey Farrar Major: Psychology Theory of Mind (ToM ) can be understood as the uniquely human social cognitive ability to reason about the complex mental states of the self and of others. ToM has typically been studied as a skill that develops across early childhood. Current research is directed at further understanding the development and the processes ToM through the study of adults. Historically, social cognition in adulthood has been studied separately from child social cogni tion. This study attempted to bridge the gap between the child and adult literat ure on social cognition by exploring relationships among a variety of tasks use d in both of these literatures Specific aims of the present study were as follows: (1) to examine whether there is an underlying ToM factor comprised of Reading the Mind in the Eyes, hindsight bias and curse of knowledge in adulthood ; (2) to examine whether language and inhibitory control (IC) are related to ToM; (3) to examine the role of these factors for story quality; and (4) to test a mediation model of the indirect effect of language and IC on story quality through ToM. Results indicated that the three factors included as indices of ToM did not correlate with one another. Regression analyses support the ongoing role of inhibitory control for ToM in everyday life, but did not provide strong evidence of the ongoing role
11 of language for adult ToM. ToM was found to be significantly related to the quality of socially shared stories, fictional and autobiographical in nature. Inhibitory control may have an indirect effect on stor y quality through ToM. Essentially, adults apply cognitive effort in inhibiting their own perspectives when communicating with others who have different perspectives. Results are discussed in light of the most current theoretical viewpoints emerging in th e growing literature describing ToM in adulthood as well as their implications in socio cultural theory.
12 CHAPTER 1 INTRODUCTION More than t hirty years ago, Miller, Kessel, and Flavell (1970) brought to our attention the relative lack of research regarding The ways and means by which children become adult like in their ability to understand and participate in social situations has subsequently become one of the most fruitful areas of research in developmental psychology. Theory of Mind (ToM) as this type of social cognition is commonly referred to in developmental psychology, includes the uniquely human ( T omasello 1998 (Tomasello and Carpenter, 2007) to take the perspective of others alongside our own, and to mentally represent the var ious mental states of the self and others. This special representational ability allows human s to see the mind as changeable, fallible, and unique to every individual; we are able to mentally represent mental representations. ToM includes the understandin g that individual mental states are the starting point for individual behavior. ToM is a sophisticated understanding of mental states where physical perception leads to knowledge, which leads to behavior. Baron Cohen (2000) highlights two essential distinctions that individuals with ToM are able to make. First, the mental physical distin ction is the ability to distinguish between the mental world and the physical world. A person may think or believe in accordance or discordance with the true state of the physical world, and a person may think about something that does not exist in the phy sical world. Second, the self other distinction is the ability to distinguish are not visible or otherwise readily available to others, and others possess mental state s
13 commonplace for typical adults, are difficult for children, and to varying degrees in individuals with autism Typically developing individuals may understand that these disti nctions exist, but are able to employ this understanding in their behavior to varying degrees. For example, we may understand that there is a distinction between self and other, but nonetheless assume that we know what others are thinking, or that others s hare our viewpoints. An important milestone in the development of a mature ToM is passing tests of false beliefs (FB) and representational change, for both the self and for other people. An understanding of FB includes the ability to understand that any pe rson may hold a belief that is in contrast to reality (based on incorrect experiential information), and that this person will behave in accordance with this untrue information. Representational change refers to the ability to simultaneously represent the thoughts or beliefs one had prior to a specific experience, and the new thoughts or beliefs which resulted from that experience. This includes knowing that one currently has information not held before, were different, perhaps false. According to the abundance of research evidence accumulated over the past 30 years, children reach this level of social cognition around the age of 4.5 years (Wellman, Cross and Watson, 2001). Indeed, some researchers view th e passing of FB tasks around the debate whether this is a qualitative, discontinuous change or a more quantitative, continuous change (Wellman, Cross & Watson, 2001, provid e an overview)
14 Specifically, is child performance on standard FB tasks equal to an all or nothing ToM competence, or indicative of the emergence of a nascent form of a highly individualized, and potentially continuously developing trait? That is, even tho ugh preschool aged children pass basic FB tasks, does the ability to utilize, or incorporate into daily functioning, ToM skill continue to develop or differ across individuals? Individual differences in these skills have been relatively unexplored in typic ally developing populations, particularly in typically developing adults. The present paper takes the position that ToM is not an all or nothing, discontinuous phenomenon but varies between individuals and is subject to development not only through childho od but into adulthood as well. As will be discussed below, it is possible that an ultimate in other skills, such as language or mathematics. Thus, individuals vary in terms of how competent they become at employing ToM skills. The present research explores individual differences in social cognition, or ToM in early adulthood. Additionally, this research is intended to bridge the gap that exists in the empirical literature b etween social psychology and developmental psychology in describing and understanding the processes of social cognition like ToM Similar social cognitive skills have been viewed as being separate in these disciplines, though similarities in the processes of these skills are suggestive that they may suitably be viewed as one and the same. Finally, th e current study explores other cognitive abilities that may be associated with the social cognitive abilities of primary interest, based on the literature describing ToM in childhood. Specifically, language, executive functioning, and autobiographical memory have been stu died as factors related to ToM. Here, I will
15 explore the relationships between these factors and social cognition through a different lens. These relationships have not previously been explored in early adulthood, and these relationships have not previousl y been explored using the proposed methods of assessing social cognition, or ToM discussed later.
16 CHAPTER 2 LITERATURE REVIEW Theory of Mind in Childhood FB task, which is often operationalized using a change of locations task (commonly known as the Sally Anne task Baron Cohen, Leslie, & Frith, 1985 ). In the traditional Sally Anne change of locations task, the child is told a brief story in which one pupp et (Sally) hides a toy, and then leaves the room. While the first puppet is away, another puppet (Anne) moves the toy to a location unknown to the first puppet. The child is asked to predict where the Sally will look for the toy upon returning to the room. suppressed for the child to understand (and r eport to the experimenter) that Sally would look in the old location, where the toy no longer resides. The first puppet will act according to a FB. Before age 4.5 years, children will reliably and with great certainty dren seeing behaviors. This includes the ability to predict behavior that will lead to an undesired outcome, based on a lack of knowledge or a FB. Another common method for assess appearance reality distinction (Flavell, 1986) To determine if a child can distinguish
17 between two conflicting identities of one object, that is, the appearance and the reality of the object, children are often sho wn a piece of sponge painted to resemble a rock, and asked if the object is really a rock or a sponge, and asked if the object looks like a rock think it was, when yo representation of the object, that it looks like a rock, had subsequently changed with experience to represent the object as really a sponge (representational change). Questions about FB can also be added to this type of task; the child can be asked what a nave puppet will think the object is, having never seen or touched the object before (Gopnik & Astington, 1998) Each of these skills, appearance reality distinction, representational change, and FB understanding require the ability to coordinate Evidence that children pass these tasks between 4 and 5 years of ag e is well of debate. Some (Wellman, Cross & Watson, 2001) argue that the passing of standard FB tasks is evidence of a developmental discontinuity, or a genuine conceptual change. Wellman and colleagues argue that based on the preponderance of evidence, it is clear that prior to the average age of 4.5 years young children cannot reason about FBs and after this point children have achieved a ToM competence resembling that of the adult. Passing FB tasks at this age may reflect a gradually emerging way of thinking about social interactions with roots in earlier skills. Newborn infants show a strong preference for human faces, and early in development engage in social smiling,
18 imitation of other humans and social referencing. Around 9 12 months, children who do not yet speak can nonetheless engage in a social exchange with others known as joint attention the ability to interact with another person about an object in the externa l world. Joint attention can be viewed as a developmental precursor to later ToM achievements. Longitudinal studies (Charman, et al., 2000; Nelson, Adamson, & Bakeman, 2008) have indicated that joint attentional engagement with mother s predict ater ToM. Joint attention brings together a several important social cognitions. First, the act of holding in mind a representation of an entity in the external world, and understanding that another person is capable of mentally representing the same entit y clearly necessitates at least an implicit understanding of other minds. Second, the interaction which occurs provides a stage for the child to experience another person experiencing similar, as well as different ideas about the external word. The partner in a joint attentional episode may use a word unfamiliar to the child about the object, may manipulate the object in a new way, or may ask the child to manipulate the object. Tomasello discusses this uniquely human ability to share mindspace as perhaps t he foundation of human society and responsible for the human ability to build upon prior knowledge (known as the ratche t effect, Tomasello, Kruger, & Ratner, 1993). Recent studies have also shown children younger than age 4.5 engaging in reasoning that may be considered FB, or at least an implicit, earlier emerging form of FB reasoning than would be suggested by a genuine conceptual change account of ToM (Onishi & Baillargeon, 2005; Buttelman, Carpenter, & Tomasello, 2009). Onishi and Baillargeon (2005) fou nd that 15 month olds looked longer when a character looked for, and found, an object in a location that the character could not have known was
19 co rrect. Buttelman, Carpenter, and Tomasello (2009) found that 18 month old children are capable of taking into belief when attempting to help the adult find an object. These studies may indicate that younger children are capable of some nascent form of FB reasoning which leads to later, more explicit abilities like those evidenced by standa rd change of location tasks. These studies of children younger tha n age 4.5 years suggest that FB may not emerge suddenly, but that it is a skill, like language, that is gradually strengthened and honed through time and social experience. Thus, the abilit y to entertain multiple representations of the same event is an element of ToM that develops. There are other social cognitive skills that also reflect ToM understanding. In particular, e motion recognition (the ability to recognize and identify emotions in the self This skill involves understanding and interpreting the internal mental states of others by derstanding of conflicting or multiple representations as FB tasks do. Emotion recognition is often Cohen, Jolliffe, Mortimore, & Robertson, 1997 ; Baron Cohen et al., 2001 ), in which a participa nt chooses emotion words thought to best describe each of a series of photographs of the eyes region of the face. Performance on this task has been shown to improve from chil dhood to adulthood (Baron Cohen et al., 1997). However, the relationship between e motion recognition and other ToM tasks is less clearly defined than, for example, the relationships among joint attention, FB, and representational change. Sabbagh and
20 (namely, the Reading the Mind in the Eyes task, Baron Cohen et al., 2001), was a ppearance r eality, suggesting that whether through genetics or socialization, the two distinct skills are relat ed. A study comparing high functioning autism (HFA) and control individuals found that only for control individuals was emotion recognition related to performance on advanced tests of ToM (Kaland et al., 2007). For HFA individuals, performance on the Readi ng the Mind in the Eyes task was not correlated with other ToM tasks. Despite this, theorists such as Baron Cohen (2000) assert that emotion recognition is part of the constellation of skills and behaviors comprising ToM. Importantly, the findings of Baro n Cohen and colleagues (1997) provide evidence that a skill which falls under the heading of ToM is subject to development beyond childhood. However, no study to date has determined whether performance on the Reading the Mind in the Eyes task is related to performance on other ToM based skills within the individual This question will be addressed in the present study. Theoretical Accounts of Early Theory of Mind Development Across the period of infancy and early childhood, and before the emergence of an adult like concept of ToM (including FB reasoning ability and mentalistic metarepresentation) social behaviors of the child can be viewed as the roots or foundations of this later achievement (Carpenter, Nagel & Tomasello, 1998; Charman et al., 2000 ). A number of theories exist which attempt to explain the origins and development of ToM. Each theory takes into account that some basic, or innate, human capability lays the groun dwork for ToM, but strong differences exist in how these theories explain the development of ToM beyond that biological starting point. The following is an overview of the prevailing theories of ToM driving research in the field.
21 Modularity. According to the modularity viewpoint (Baron Cohen, 2000), all of the achievements leading up to the ability to pass FB (and indeed, beyond into social interactions in general) are connected within an innate Theory of Mind M odule (ToMM) Thus, face preferences, joint a ttention, and the physical mental distinction are specialized brain based skills Notably, the ToMM is proposed to be involved specifically with the metarepresentation on mentalistic concepts, and is separate from other, non mentalistic representational ab ilities. This view is intricately linked to the ToM hypothesis of autism. In autism, individuals have deficits that appear to be isolated within this ToM M These deficits include difficulties with the pragmatics of language and with communication in genera l, difficulty with the appearance reality distinction, lack of pretend play, difficulties with imagination, and inability to interpret facial expressions, to name a few. According to Baron utism involves a specific defici t in ToM Theory of Mind, in Baron and cognitions that hinge on a fundamental distinction the mental physical distinction which is difficult or impossible for individuals with autism To illustrate, individu als with autism have difficulty distinguishing between a person who is holding a cupcake and a person who is only thinking about a cupcake in determining who can actually eat the cupcake. Behaviors and skills that are impacted by the mental physical distin ction include the understanding that seeing is knowing, the ability to distinguish mental state verbs from other verbs, the tendency to pay attention to mental states, comprehension of issues of the mind such as false belief (FB) and representational chang e, pretend play, real life social skills, pragmatic speech, understanding the link between mental
22 states and emotions, and the ability to interpret the eyes mentalistically. If this range of skills is founded on one underlying concept of mental physical di stinction, we might expect that tasks which assess these skills would be interrelated within an individual. ability to use FB reasoning Social c onstructivist. Accordin g to the social construc tivist viewpoint (Carpendale & Lewis, 2004; Fernyhough,2008), more advanced ToM abilities are built upon earlier, related, achievements as described above. Carpendale and Lewis (2004) to socially interact using the skills of infancy leads to the construction of more advanced describing how joint attention is the cornerstone of the development of all social understanding. Here, the child, adult, and external world is grounded in the exte rnal world, and the two converging perspectives of the child and adult reflect upon the world/object. This leads the child to mentally represent the physical mental distinction, as well as the self other distinction. To this, Fernyhough (2008) adds the cri tical aspect of child language as mediating the advancement from earlier (less advanced) and later (more advanced) T o M abilities. When children internalize social dialogues they are able to engage in dialogic thinking. This is evidenced by self regulatory speech (which itself becomes internalized with development). This dialogic thinking is critical to the development of the mind which is
23 not only representational of duality in general, but which represents perspectives of the self and other. Shared intenti onality and cultural l earning Tomasello and colleagues ( Tomasello, Kruger & Ratner, 1993 ) claim that human beings are evolutionarily endowed with the uniquely human skill of shared intentionality, and draw many conclusions from comparative psychology. It is shared intentionality or the innate desire to share mindspace with others that takes skills of the social world and transforms them into uniquely human skills of ToM The infant enters the world with the desire to share attention with others. The chim panzee is able to follow the gaze of another, but only the human child is able to engage in joint attention about an external object. The chimp communicates to make requests or manipulate another, but only the human communicates with the intent to cooperat e. Social skills of the chimpanzee are raised to another level in humans because humans understand the minds of others. Indeed, under the rubric of cultural learning, ToM here the person has internal states, and behaviors result from perceptions and desire. The child who engages in joint attention and learns the vocabulary of the language unders tands the person as intentional, and can engage in imitative learning. Next, the child begins to see the person as a mental agent. Behaviors are seen as the result of more complex mental states such as thoughts and beliefs. This child can benefit from inst ructed learning and internalize the perspective of other persons. Adults have the desire to pass on cultural information and the child is now able to understand and benefit from this cultural transmission of knowledge. This child can
24 pass FB, because of th e perception of the person as a mental agent. Finally, children come to see individuals as reflective agents. Not only does the person have representational thoughts and beliefs, but the person has their own, subjective inner world. That inner world can co me to affect the external world, and minds can influence other minds. Here, cooperative learning becomes possible. Existing knowledge is built upon further, and minds converge. Theory theory ToM is just that: an organized, rule governed set of statements intended to describe, equivalent to any scientific theory. ToM is seen as based on evidence collected from ex ToM changes with the acquisition of new evidence thus undergoes paradigm shifts. Specifically, 3 year old children have a desire psychology theory of human behavior. They understand others as being driven to behave by the internal state of desire, but this desire is non representational. A person desires an apple because they perceive and apple, and because the apple is real and available. The mind is an exact copy of the real world. Thus, according to this c theory, the person behaves based on a mental state, but this is a rudimentary and non representational theory. T with experience and under the weig ht of an abundance of evidence The 4 year old child is able to develop a bel ief desire psychology. This new theory is representational, and based on a working knowledge that the mind is not simply a copy of the external world, but is representational. This conceptualization of ToM development is largely based on the ability to pas s false belief tasks.
25 The theoretical perspectives described above represent a summary of the more prominent perspectives on the development of ToM in childhood. These perspectives vary in the precise mechanism by which children are able to express and ut ilize their ToM abilities, but all incorporate certain evidence from the empirical literature in one way or another. Specifically, each perspective takes into account the role of some innate, brain based, or distinctly human characteristic, and further, ta kes into account the relative roles of other related cognitive functions. L anguage and executive f unctioning (spe cifically inhibitory c ontrol ) for the development of ToM are discussed below The various theories previously reviewed incorporate the precise roles of language and executive functioning differently in childhood The p resent study takes the position that language and executive functioning can be incorporated into these perspectives when considering ToM in adulthood as well. That is, the present s tudy does not intend to test a specific theory, but rather examines the roles of language and executive functioning in adult ToM. Cognitive Contributors to Theory of Mind Executive f unctioning Executive functioning (EF) abilities, including self regulatio n, planning and inhibitory control ( IC, among other related skills associated with the development of the frontal lobes) have been shown to play a role in the development of ToM. Considerable research has shown links between of ToM and EF skill in young ch ildren ( Perner, Lang & Kloo, 2002; also see Perner and Lang, 2000, for a review). Carlson, Moses and Breton (2002) describe the specific role played by IC and FB tasks. IC is defined as the ability to inhibit a p repotent response in favor of a desired response, block interfering information, IC is
26 required as the participant must inhibit a prepotent response of reading a color name in f avor of the desired response of naming the conflicting color of ink the word is printed in. FB tasks, in this view, require the participant to inhibit a response that is in accordance with reality (where the object really is) in favor of the desired respon se (where the character in the story falsely believe s the object to be). Carlson, Moses and Breton (2002) describe two accounts of the possible role that IC might play in the development of ToM. First, IC may play a role in the actual emergence of FB EF ability must be adequately developed to allow for the suppression of more salient states of reality in favor of less tangible mental representations when reasoning about the world. EF may be necessary for competence in du al representation. Dual representation refers to the ability to hold two conflicting aspects of one situation in mind simultaneously, and is viewed as essential in distinguishing mental states from the physical reality of a situation. EF thus provides the child with the cognitive tools to represent mental versus physical simultaneously, or true and false simultaneously. Thus ToM skills are not present until sufficient EF has developed. Alternatively, EF development may play the role of allowing for the exp ression of an already present ToM understanding. Before children are able to correctly answer target questions on FB tasks, that is, when they provide an answer that is consistent with reality rather than with a FB some researchers have noted that the chi FB Clements & Perner, 1994 reality even though a latent understanding exists that there is an alternative answer
27 consistent with reality is sufficiently developed, he or she can finally provide a verbal IC simply allows the child to show what he or she already knows. Supporters of the important role of EF do not exclude the importance of other factors in the development of ToM, either emerge nce accounts or expression accounts, specifically the role of language in the development of ToM. Indeed, some theorists contend that EF belies both language and FB and is the essential factor in ToM development. Language In addition to EF, s pecial atte ntion has been paid to the influence of language as a correlate, and most specifically as a predictor of the development of FB understanding. A metaanalysis of the extant literature on the relationship between language and FB by Milligan, Astington, and Da ck (2007) provides a concise illustration that both general language and specific syntactic abilities are associated with performance on FB tasks. Why this relationship exists is a matter of debate for theorists. Astington and Jenkins (1999) propose that language is an important cognitive tool of representation. Specifically, syntax, the aspect of language concerned with logical and rule based word order, is viewed as critical for FB. This position of linguistic determinism holds that syntax is a necessary scaffold in the mind which holds and supports the representations required for a child to comprehend, remember, process, and understand FB tasks. This also applies to the real world, in that syntax is a cognitive tool that allows children to organize even ts in mind, and keep track of sequences of social situations. Some theorists argue for a strong position of linguistic
28 determinism that specifies the syntactic structure of embedded complements as the central linguistic ability that determines FB ability i n the individual (de Villiers & d e Villiers, 2000 ) Ruffman and colleagues (Ruffman, Slade, Rowlandson, Ru msey, & Garnham, 2003; Slade & Ruffman, 2005) argue for the importance of general language (semantic and syntax) for FB. Ruffman and colleagues suppo rt the notion that language and FB are bi directionally related. That is, language contributes to the understanding of ToM (and the ability to pass FB) but that having a ToM also contributes to the development of language. The two are social skills that in teract across development, and the relationship between the two is not necessarily deterministic. Ruffman claims that language likely plays the role of allowing implicit understandings of FB to be made explicit both internally, in the sense that the child is able to think about and reflect upon issues of FB and conflicting representations and externally, in the sense that the child can express their ToM competence. An important aspect of the position of Ruffman and colleagues is the role of social interac tion. Thus, this research group places emphasis on naturalistic conversations between mothers and their children, and the types of linguistic interactions that take place. Nelson (2005) proposes a social constructivist view of the role of language in the development of ToM referred to as the encompassing one on one interactions with individuals as well as the ability to interact in and understand a world of social situations across time and cultures. Language plays the role of bring ing the child into this community of minds Mental states are, by definition, hidden.
29 Language is the way that mental states are revealed to others. By talking to children, we reveal that others have internal states that are unique, variable, and interpret ive. Special populations provide further insight into the intricate relationship between language and ToM. Individuals with autism show difficulties with the representational aspect of language, particularly the pragmatics of language (Baron Cohen, 2000). Alongside ToM deficits, individuals with autism have difficulty tailoring their speech to the listener, staying on topic, and employing social norms of conversational manners, turn taking, an d appropriateness (Colle & Baron Cohen 2008). Children with sp ecific language impairment (SLI) show a similar correspondence between language and ToM difficulties ( Farrar et al., 2009). Role of Theory of Mind in Childhood When considering the development of ToM in children, much empirical research is devoted to understanding what factors lead to the development of this complex set of social cognitive skills, such as language and EF discussed above. Another perspective on the development of ToM concerns what deve lops as a result of ToM. Research has fantasy ideation a child is capable of (Taylor & Carlson, 1997), peer a cceptance (Slaughter, Dennis & Pritchard, 2002), memory suggest ibility (Welch Ross, 1999), and even acting ability in early adolescence (Goldstein, Wu & Winner, 2009). Of particular interest is the role of ToM in autobiographical memory development. Autobiographical Memory (AM) can be simply described as a remembered event linked with other social cognitive abilities. Theoretically, AM and ToM can be linked in several important ways. First, AM and ToM both require that an individual entertain
30 multiple representations of a single event. AM requires that the individual understand that the event is situated in the past as well as in memory. Further, when sharing AMs, individuals must entertain their own representation of the event as w representation of the event (Welch Ross, 1997). For ToM, individuals must represent understanding of the origins of knowledge (see Welch Ross, 1995, fo r a review). To relate an AM to another person, one must understand that having experienced the event personally endows the individual with privileged knowledge, and individuals not present are not likely to share that knowledge. Thus, in sharing an accoun t of the AM and ToM require that an individual have an understanding of the important causal links between events in time. For ToM, this includes that seeing leads to knowing, for example, and not the other way around. For AM, this includes that one event led to the next, or that the past led to the future. Story structure, or the idea that a coherent narrative includes important links between pieces of information, i s knowledge that develops around the same time as children pass traditional FB tests. In addition, individuals with autism have trouble understanding and employing narrative story structure and also have trouble with ToM ta sks. Adler and colleagues found a direct relationship between ToM (as measured by the Reading the Mind in the Eyes task) and AM in individuals with h igh f unctioning autism (Adler, Nadler, Aviatar & Shamay Tsoory, 2010) As we have seen, ToM is linked both theoretically and practically to AM. Welch Rosss (1997) directly tested the connection between ToM and AM in normally
31 developing children. To do so, Welch Ross (1997) focused on the role of the adult in scaffolding AM reminiscence with children. When conversing with their children about AMs, mothers may adopt different styles of interaction. Research evidence has shown that children of Fivush, 1993), characterized by sharing extended amounts of information and rich detail, asking open ended questions, and in cluding evaluative content, experience several advantages over children whose mothers do not use this style of conversation. These advantages include improved story structure understanding, more elaborate AMs, and enhanced self understanding ( Reese, Haden & Fivush, 1993; also see Reese, 2002, for a review). Welch Ross style of conversation contributes to ToM. The findings of Welch Ross (1997) are suggestive that parental el aborative style may contribute to ToM development. However, a question that has been relatively unexplored in the literature concerns the individual characteristics that mothers employing this elaborative style may possess. It is possible that mothers with superior understanding of the world of mental states, or ToM, are able to discuss memories with their children in this way because they understand differing perspectives to a greater degree than non elaborative mothers (Welch Ross, 1997). Sabbagh and Seam ans
32 of ToM occurs is underexplored. Determining if a link exists between adult ToM and storyt elling style is an important step in uncovering how ToM may be transmitted from parent to child. It is important to note that the direction of relationship between ToM and autobiographical memory may be bidirectional. Whereas some view ToM as a precursor t o autobiographical memory (Welch Ross, 1995), others note that autobiographical memory deficits may lead to increasing ToM deficits in cl inical populations (Corcoran & Frith, 2003). That an understanding of story structure is relevant to telling a coherent narrative is not limited to the case of AMs. Tompkins and Farrar (2010) found that wordless storybooks afforded more opportunities for elaborations between mothers and their children than simple AM sharing, thus the wordless storybook paradigm is employed in this study. Narrating a fictional story from a book clearly utilizes the ability to link episodes in time and causality. Good story telling also entails taking the perspective of the listener in terms of what needs to be shared, or told, in order to re late a good story. Wordless storybook narratives by individuals with autism have been found to be lacking in objective measur es of story structure (Colle & Baron Cohen, 2008), tend to include many more off topic utterances, unnecessary or unrelated informa tion, and evidence a general lack of understanding of the perspective of the listener. Their stories may be a good story, who also happen to be younger than the av erage age of passing FB tests. In their key paper depicting the development of autobiographical memory, Nelson and Fivush (2004) provide a comprehensive account of the role that language plays in the emergence of AM in children. Notably, these authors dis cuss the importance of
33 language in allowing the young child to organize information in mind for the purpose of coherence of memory as well as the ability to recall events from memory. In addition, Nelson and Fivush cite the social nature of language as pla ying a role in allowing the child to mentally represent personal experiences as separate from fictional events or the experiences of others. A study by Haden, Ornstein, Eckerman, and Didow (2001) showed that very young children, with even very limited verb al recall of an event, largely remembered and verbalized information that was spoken and discussed by their mothers during the event. Thus, experiencing an event that is spoken about in a social situation plays a role in what and how much a young child can recall. Language also is inextricably linked to narrative beyond that narratives require language at a basic level. Skilled language allows for coherent, interesting, evaluative, and meaningful narration of remembered events. AM storytelling, language a nd ToM have been shown to be related in a number of ways. The representational abilities that are part and parcel to ToM may help children develop AM and storytelling skills, and storytelling interactions may lead children to understanding the processes of the mind. Further, having an understanding of minds contributes to the quality of the story told. In these ways, ToM skill is linked to the ability to tell a good story, whether autobiographical or fictional. Thus, it well established that FB understandin g becomes evident in childhood, around 4 or 5 years of age. Further, both language and EF contribute to its memory. Of interest in the current research is how ToM develops beyond childhood and whether there are meaningful individual differences in adult ToM If so, are these
34 differences linked to EF and language, the same as they are in childhood? Finally, is adult ToM similarly related to autobiographical memory? Theory of Mind Beyond Childhood Research is beginning to highlight the differences between children and adults in their social cognitive abilities but, more surprisingly their many similarities. Whereas we can be sure that the adult is quite savvier at parties and in job interviews to the nuances of social interaction than a 4.5 year old child, a wealth of evidence suggests that humans of all ages make some mistakes in their social cognition. Children and adults alike are surprisingly apt to attribute their own, how ever private, knowledge to others. Equally so, children and adults are likely to assume that others share personal likes, dislikes and viewpoints, and to ascribe these to others especially strongly when even the most trivial of similarities are told to the m to exist. In general, adult social cognition and perspective taking has been viewed through a different theoretical lens than has ToM in children, despite apparent similarities. There exists a large amount of evidence in the social psychology literature that adults commit a variety of errors and are prone to certain biases in social situations which closely resemble the errors that young children make prior to age 4.5, with some important differences. These biases are founded in the same skills that under lie T oM, but are operationalized differently. Whereas tests of ToM for young children are relatively simple for adults, the tasks that have been used in social psychology to elucidate these biases are more nuanced, and can reveal more subtly that adults so not simply employ their ToM skills in an all or nothing way. In fact, some research (Apperly, Riggs, Simpson, Chiavarino & Samson, 2006) provides evidence that correctly answering basic questions about FB does not happen automatically for adults, but requ ires extra time or cognitive effort when compared to
35 questions not involving FB. Four specific biases, the curse of knowledge hindsight bias the f alse c onsensus e ffect, and visual perspective taking are discussed next. As we have seen, theorists claim that ToM develops in childhood either 4 to 5 years. Others (Birch & Bloom, 200 7; Dumonthiel, Apperly, & Blakemore, 2009) take an altogether differen pass FB tasks prior to age 4.5 years reflects a cognitive bias that exists across the ccording to their perspective, children fail tests of FB because they lack the ability to suppress their current knowledge when inferring the knowledge of another (nave) person. The evidence of Birch and Bloom suggests that adults are likely to make this very mistake, but in a more discreet manner. That is, rather than unequivocally adults who know the true location of a displaced object are surprisingly more likely tha n those who do not know the true location to assign a higher search probability to the true location than other possible new locations. This, despite the fact that both groups are told that the character has no knowledge of the new location. For Birch and Bloom, thereby have a mistaken impression that their knowledge is shared by others. In addition to the curse of knowledge, adults are found to have a strong propensity t o c ommit a hind s i ght bias (Pohl & Hell, 1996). Easily understood as the
36 1975, Fischhoff found that participants were likely to claim that they could have predicted the outcome of certain historical events and that others would be likely to have correctly predicted the outcome of those events (also, Fischhoff & Beyth, 1975) These fi ndings have been replicated in a variety of ways. For example, if two groups of adults are asked to guess the true height of the Eiffel Tower they will provide a wide range of responses. Later, when asked to recall those guesses the group that is told the t c ommitting this bias (Pohl & Hell, 1996). Thus, adults are not perfect when considering their own representational changes, nor are they perfect when taking into account the possibility of another person havi ng a FB (Birch & Bloom, 2007). The belief, and th e Both of these mistakes are highly similar to the mistakes made by young children who do not pass traditional representational change and FB tasks. Bernstein, Atance, L oftus, and Meltzoff (2004) demonstrated that adults and children are prone to making this mistake using visual tasks. When armed with the knowledge of what an ambiguous picture really is, individuals report seeing the object sooner as the picture resolutio n is made clearer than those who are not told what the picture really is. Interestingly, in two studies these researchers found two different pictures of the developmental trajectory of child to adult visual hindsight bias In one study, visual hindsight b ias was greater for children than for adults, suggesting a developmental decline in hindsight bias In another study, children and adults were equally likely to commit visual hindsight bias
37 Thus, despite being much better at traditional FB and representat ional change tasks, adults may not be much better than children when the same skills are required of them in a different type of task. Another bias, or social error, reliably committed by adults is the f alse c onsensus e ffect. The f alse c onsensus e ffect oc curs when adults expect that others share their beliefs, or that their personal preferences are common to the population as a whole. Early work by Ross, Greene and House (1977) showed that regardless of actual commonality, individuals are likely to conside r their own personal preferences or showed that college students attributed their own level of drug use as common to the general public as well. Considering that ToM compe tence includes the conscious awareness that minds differ, it stands to reason that ToM competence should, ideally, prevent this error. That ToM may not be a discontinuous achievement of early childhood, but may in fact be a skill that continues to develop across adolescence and adulthood is highlighted in a study by Dumontheil, Apperly, and Blakemore (2010), who studied visual perspective taking. These researchers showed that the ability to take into consideration the viewpoint of another person increased a cross five age groups ranging from approximately age 7 through 27. The viewpoint in question was a literal viewpoint. Participants were able to see certain objects on a set of shelves that were occluded cts were visible to both parties. The director asked the participant to move certain objects around the set of shelves. When
38 consideration that the smallest truck visible to the participant may be occluded from the view of, therefore unknown to the director. The tendency to move objects that were invisible to the director decreased (but did not disappear) across age groups. Even the propensity to stereotype based upon outward app earances flies in the face of the notion that adults with ToM require sensory knowledge of an event before or character based on the minimal information of type of clothing or color of skin. Further, adults are by no means perfect when it comes to effectively taking the reaction to directions given by another person are often egocentric, but adults simply correct for this egocentrism more quickly than do children. As we have seen, adults commit several errors that appear to reveal anomalies explore whether th e processes of ToM in early adulthood reflect those of the development of ToM in early childhood. This includes multiple related questions. First, this study will determine whether multiple social cognitive biases are interrelated. No study has attempted t o draw a parallel between social cognitive biases of adulthood and ToM Following this, the present study is intended to show individual differences in these biases. The expectation that individuals will vary meaningfully in the degree to which they fall p rey to social cognitive biases is rooted in the perspective that ToM is not an all or nothing change that happens at age 4.5 years, but rather is a gradually developing skill affected by a wide range of experiences across the lifespan.
39 Meaningful individu al differences in ToM are expected to be associated with individual differences in other skills conceptually related to the development of ToM. Specifically, language and EF have been shown to be important factors contributing to ited research has examined similar relations in adulthood. Botting and Conti Ramsden (2008) found that language skill in adolescents was associated with functional social outcomes such as friendships and social engagements, and that adolescents with a hist ory of SLI were at particular risk for reduced social functioning. This study suggests that language may remain an important social tool beyond early childhood as ToM is emerging. However, for the most part research evidence for the role of language for a dult ToM is lacking. Apperly, Samson, and Humphreys (2009) outline evidence that in the case of brain injury, ToM skill is found to remain intact if language areas of the brain are tients are able to pass modified versions of standard FB tasks despite deficits in language skill. Apperly and colleagues propose that language is an important tool supporting the emergence of ToM, but that once ToM is in place, language may no longer be n ecessary to employ ToM skill. In contrast, Newton and de Villiers (2007) showed that verbal interference, but not non verbal interference had the effect of reducing adult accuracy on simple non verbal FB tasks. These researchers suggest that the impact of verbal interference is indicative of the continued importance of language for ToM and FB reasoning in adulthood. Dungan and Saxe (2012) however, showed that the most likely cause for this reduction in adult accuracy on FB tasks was due to increased demand s of working memory and EF Thus, using the interference paradigm has not
40 provided conclusive evidence regarding the role of language for adult FB. T o date no study has directly examined the relationship between ToM and language in early adulthood. It is possible that language skill remains associated with ToM throughout life. T hree theoretical viewpoints provide a basis for this assertion. First, the findings of Botting and Conti Ramsden (2008) that adolescents with a history of SLI experience conti nued social deficits support the continued role of language for ToM beyond early childhood. An individual experiencing early difficulties interacting with others through language may carry those difficulties in both language and social skill into adolescen ce and adulthood. However, this evidence is not a direct test of whether language remains an important predictor of social skill (in general) or ToM (specifically) throughout development. Second, Fernyhough (2008) proposes a Vygotskian account of the role of language for ToM, in which language based interactions, or dialogues, are central to by which children come to understand and internalize the menta l states of others, and even of the self. The dialogic thinking framework describes how the experience of dialogues highlights the alternating and often conflicting representations held within different minds (that is, the different participants in the dia logue). Multiple aspects of time and experience these dialogues become internalized as a general way of thinking. For the present purposes, this dialogic thinking frame work is extended to encompass the continuing dialogues that a person engages in beyond early childhood, and more
41 (and higher quality) social and linguistic interaction may be related to continued improvement in ToM skill as well as use. Third the strong linguistic determinism perspective tested by Newton and de Villiers ( 2007 ) holds that language is crucial for though t specifically language is critical for thinking about FB. Thus, without language the child cannot think about FB, and the same applies in adulthood. If language and ToM are found to be related in early adulthood this would provide further support for a continuity account of ToM development, as opposed to the genuine conceptual change account. The proposition is, in essence, that factors an d experiences which lead to the enhancement of ToM understanding in children continue to be effective in enhancing related ToM skills throughout development. Adults who have been consistently engaged in high quality language based interactions, in this vie w, would be more likely to have honed ToM skill; they would have a better understanding of the minds of others, and be more skilled at social cognition in general. Here, language serves as a constant support of ToM use and development. If language is not f ound to be relevant to performance on ToM tasks for adults, this would be suggestive that language is a tool which supports the initial emergence of ToM, but which ceases to be necessary once the foundations of ToM are established. EF may also play an imp ortant role in adult TOM. Recent research (Apperly, Samson & Humphreys, 2009) has questioned whether EF is important simply for emergence of ToM in childhood, or whether EF remains integral to ToM reasoning throughout development.
42 Apperly et al. (2009) inv estigated the role of EF for adult performance on ToM tasks in their analysis of the existing literature, and found that performance on ToM tasks was affected by EF ability in cases of adults with brain injury as well as in normally developing adults. Sams on, Apperly, Kathirgamanathan and Humphreys (2005) showed that a brain injury patient with impaired EF performed significantly below chance on standard FB tasks used with 4 year old children. Qureshi, Apperly and Samson, (2007) found a relationship between EF and performance on a visual perspective taking task in this case, higher EF was associated with fewer errors in perspective taking. In addition, the increased time it sometimes takes adults to answer questions involving FB as compared to non FB questions ( Apperly Riggs Simpson, Chiavarino, & Samson, 2006 ) hints at the executive processing involved in reasoning about FB. These findings are suggest ive that EF plays a role in ToM beyond childhood. However, the role of EF in measures of adult social cognitive biases is relatively unexplored. Also unexplored in early adulthood is the role of language in the use of ToM. The present study considers the roles of both langua ge and EF for performance on measures of ToM in adulthood. In addition, ToM has been shown to play a role in behaviors considered to be more of an outcome, or a result of ToM, such as effectiveness of communication with others Apperly and coll eagues (201 0) showed that adult participants were limited in their ability to use or apply their ToM skill in communicating with others, resulting in egocentric errors in interpretation. Keysar, Lin and Barr (2003) and Apperly et al. ( 2010) showed that adults have di fficulty applying their ToM understanding using a creative grid paradigm described earlier In this paradigm, participants are instructed to move
43 objects around a shadow opposite side of the grid (Apperly et al., 2010 used a computer based, cartoon version of the original live action task). Participants are clearly instructed to take into account that there are some objects that ONLY the pa rticipant can see. The director has no knowledge of and will not make reference to the items that the director cannot see. However, when considering the discrepancy between what the director and participant see, some directions are ambiguous. For example, the grid may contain three different sized trucks, all three of which are visible to the participant, but only two (the medium and large) are visible to s perspective the smallest truck is not visible to the director, thus the medium sized truck must be this smallest truck that the director refers to. Adults consistently that they frequently will respond egocentrically, by moving the smallest truck. If the tive of a nave other person is not automatic, consistent, or reliably done by participants on this task. In the present study, the role of adult use of ToM in communication is explored in the circumstance of autobiographical memory and fictional storytel ling in a social context. The present study will explore the role of ToM and social cognition, language, and EF in AM as well as fictional storytelling in early adulthood
44 Study A ims Th is study explores ToM in Early Adulthood and its relationship to language, EF and narrative. The first s pecific aim of the proposed study is t o determine if individual differences in measures of ToM as typically used in the developmental psychology literature (emotion recogni tion and FB) are associated with individual differences in hindsight bias as typically studied in the social psychology literature. It is e xpected that performance on emotion recognition and FB task s will be related to one another and will be related to performance on the h indsight bias measure comprising an overall TOM factor made up of e motion r ecognition, curse of knowledge and hindsight bias The second specific aim of the proposed study is t o determine if individual differences in these various mea sures of adult ToM are associated with individual differences in language ability and IC It is expected that individual differences in language and IC will be associated with individual differences in performance on ToM tasks in adults. The third specifi c aim of the proposed study is t o determine if individual differences in measures of ToM, language and IC are associated with individual differences in storytelling ability, including the quality of socially shared narratives of AMs and a fictional story. It is expected that individual differences in adult ToM will be associated with individual differences in story quality. Although conceptualized as an point study, the direction of influ ence of ToM on storytelling cannot be specified. Finally, an overarching fourth specific aim is to elucidate a possible pathway from language and EF (specifically IC ) to AM quality mediated by T o M. In other words, language and IC have an indirect e ffect on storytelling quality through ToM (Figure 2 1)
45 Based in the theoretical perspectives described here, and included as a comprehensive model incorporating the previous specific aims of the study, an overarching model of the interconnections among l anguage, EF (IC), ToM and storytelling is proposed. As seen in Figure 2 1(A) the basic cognitive skills of language quality story. Figure 2 1(B) illustrates the propositi on that language and EF (IC) may be found to affect storytelling indirectly, through their influence on ToM in the individual. Thus, language and EF (IC) will impact storytelling not simply because they are basic cognitive tools that are used when engaging in storytelling, but because these skills are necessary for ToM, and the use of ToM is essential in communicating a good story to another person.
46 A. B. Figure 2 1. Mediation model showing an indirect effect of language and executive functioning through Theory of Mind. Theory of Mind Language (PPVT 4) Storytelling Quality a b c Executive Functioning Language (PPVT 4) Storytelling Quality c Executive Functioning
47 CHAPTER 3 METHODS Participants The age range chosen for the present study represents early adulth ood, ranging from 18 25 years of age. This age range was chosen because it is easily accessible in a university setting, represents relative diversity of individuals, and is a period of ongoing development across a variety of developmental domains. Individ uals in early adulthood, especially those in college, have been studied because they are experiencing continued brain development, social development, emotional development, and cognitive development. Adults in this age range experience a wide range of lif e changes, face a wide range of choices, and have ample opportunity for development through social interactions with peers and other individuals. Additionally, this developmental period is an ideal time to collect autobiographical memories as the life stor y has been found to Participants were recruited through the Psychology Participant Pool at the Ohio State University. Specifica lly, participants attended a satellite campus of the Ohio State University, which promotes open enrollment and samples from this campus include a variety of non traditional and diverse university students. Students who completed the study were given course credit commensurate with university guidelines. Although the 25 years, p articipation was not restricted based on age, and upon examination of the final sampl e, the age range represented was 18 35 .5 years. All individuals in this extended age range were university students, and examination of the data showed no outliers on any measure
48 within this age range. Individuals between the ages of 18 35.5 were included in analyses as this allowed for a larger sample while not otherwise changing the characteristics of the sample. Individuals over the age of 35.5 years were excluded from analyses, as these represented outliers not consistent with the remainder of the sampl e. A total of 10 4 students were tested. Students over the age of 35.5 years were excluded from analyses resulting in a final sample of 9 6 students. Participant age ranged from 18.5 years to 35.2 years (M = 21. 2 9 years, SD = 3. 41 years) and included 59 fres hmen, 19 sophomores, 4 juniors, 9 seniors and 5 post graduates. The sample included 49 males and 47 females (no gender differences were found on any measure) and was largely Caucasian (8 3 Caucasian, 6 African American, 3 Asian, 2 Hispanic, and 2 multi rac ial). Procedure Participants w ere tested in the Psychology Building by graduate and undergraduate student researchers. During each session, participants were given a paper and pen packet, after which they were interviewed by a researcher. Interviews were audio recorded. Participants participated in two sessions. Session 1 last ed one hour and include d an Adult Theory of Mind battery (including hindsight bias Day 1, curse of knowledge and Reading the Min d in the Eyes), self relevant autobiographical memory, two cue word elicited autobiographical memories, and language assessment (PPVT 4 ). Day 1 also included a personal information sheet to gather information about age, gender, race/ethnicity, SAT/ACT scores, and socio economic status. Session 2 occur ed one w eek after session 1 and last ed 30 minutes. Session 2 include d hindsight bias follow up, the Stroop Color Word test and storybook narrative.
49 Measures Social Cognition The Reading the Mind in the Eyes Task, Revised Version (Baron Cohen, et al., 2001). This paper and pencil test consists of 36 photos of the eye region of the faces of a variety of human actors and actresses. Participants are required to choose the word, out of 4 answer choices (one correct, three carefully chosen foils), that best describes t he mental state expressed in the eyes. Participants are provided with a glossary of the mental state terms included in the test for reference. Participants were not timed, but generally this task t ook about 10 minutes to complete. This revised version of t s as expressed through the eyes The previous version (Baron Cohen, Jolliffe, Mortimore, & Robertson, 1997) of the Eyes Task resulted in near ceiling level performance for normally developing adults, and served to distinguish normally developing adults from those on the autism s pectrum. In order to serve as an individual difference measure for normally developi ng adults, t he revised version includes more items (previously there were 25), four answer choices rather than the previous two and answer choices which require a nuanced understanding of the mental state presented, as opposed to a forced choice between 2 opposites as in the original version. Participant scores can range from 0 36. Mean scores for the general populations in the report of this task were 26.2 (SD=3.6) and for student populations the mean score was 28.0 (SD=3.5). This task has been used prev iously with normally developing adu lts (Kemmis, Hall, Kingston & Morgan, 2007) and special populations including High Functioning autism ( Kalan d et al 2008)
50 Curse of k nowledge t ask. Participants are presented with one paper based scenario based on the s eries of experimental versions of a hidden displacement task developed by Birch and Bloom (2007). This scenario resembles the classic Sally Anne FB competence, but is modified for use with adults ( Figure 3 1) Essentially, this is a FB task with responses in the form outside playing, her sister Denise moves the violin to the red container. Then, Denise with an illustration of the same containers in a different configuration. Final ly, She first looks for the violin in the blue container where she left it. What are the chances that Vicki will first look for her violin in each of the remaining cont ainers? Write your answers in percentages in the spaces provided under each container. Your responses This task takes less than 5 minutes to complete. In order to score this assessment, it is important to take into consideration what a expect a participant not cursed with the knowledge of the true location to assign equal probability to each of the three remaining containers (pilot testing suggests that man y, though not all, non cursed participants do this). Thus, individual curse of knowledge scores were calculated by subtracting 33.33 from the probability assigned to the red
51 container, resulting in a raw score range of 0 66.66, with 66.66 indicating the hi ghest The scenario used is a modification of the method used by Birch and Bloom (2007) and is structured to be as sensitive as possible in detecting whether an individual ting the behavior of a nave other person. The participant is told that the violin is hidden in the red container without the knowledge of the character who will search for the violin (Vicki). Thus, Vicki has no reason to search the red container preferen tially over the other containers. By telling the participant that Vicki first searches in the blue/original container we eliminate the possibility that all participants will assign 100% probability to this container (perform at ceiling for standard FB ). In addition, we are maximizing our ability to determine each be assigned anywhere from 0 100% probability for being the first search location. Ideally, an individual who sets aside their privileged knowledge of the true location would assign equal probability to each of the remaining containers (that is, 33.33%), applying their understanding that Vicki is nave to the true location of the violin. Pilot testing of this task has indicated that when individuals are told the true location of the violin, as a group they assign a higher probability to the red container than to the other containers when compared to a control group (in which participants are not told the true location of the violin). In addition, pilot data indicates that those who are told the true location of the violin assign a first search probability ranging from 0 100 % to the red container, with some participants assigning equal probability to each contai ner. This suggests that the proposed procedure will elucidate individual
52 differences in ability to set aside privileged knowledge when predicting the behavior of another person. This specific task has not been used previously as an indicator of individual differences in ToM for adults. Reliability was not calculated for this measure because of the nature of the responses being a division of 100% probability and a limit of 3 responses. Hindsight bias. A battery of 5 e is administered on day 1 of testing using a paper and pencil format. According to previous work, ( i.e., Pohl & Hell 1996 ) the questions are Almanac based questions of facts of which the average person would have no specific knowledge, each with a numeri cal answer. An e What is the height of the Eiffel Tower, in feet? Participants w ere asked to write their answer (or estimate ) in the space provided. On Day 2 of testing (one week later) participants w er e presented with the same l ist of questions. For 2 5 of the 5 0 questions, participants w ere asked to recall the answer they gave on Day 1 of testing after being told the correct answer. For the remaining 2 5 questions, participants w ere asked to recall the answer they gave on Day 1 of testin g without being told the correct answer. Hindsight b ias scores w ere calculated using the procedure of Pohl and Hell (1996). For each question, a bias score was calculated in the following way: B = (E R) / (E S), where B = bias, E = original esti mated response, R = the recalled response, and S = the solution, or actual answer to the question. Using this formula, 5 types of score are possible, indicating (1) that the remembered response was the same as the original estimated response (B = 0), (2) t hat the recalled response moved farther away from the solution (B < 0), (3) that the recalled response moved beyond the solution (B
53 > 1), (4) that the recalled response moved toward the solution (0 < B < 1), or (5) that participants recalled their origin al response as the same as the solution (B = 1). Notably, Pohl and Hell (1996) found this fifth type of response only 3 times out of 900. For each participant, frequency of the four types of bias scores were calculated separately for the the participant is told the solution after recall) lists of questions. Of particular interest here is the number of scores indicating a drift toward the correct response (i.e., 0 < B < 2). The difference between frequencies of drift toward the solution for each list will indicate overall bias for each participant. Scores on this measure can range from 25 to 25. A score of 25 would indicate complete hindsight bias, in which all 25 responses in the biased condition drifted toward the correct answer while no response in the unbiased condition drifted toward the correct answer. General Cognitive Abilities Language. P articipants completed the Peabody Picture Vocabulary Test, Version 4 ( PPVT 4, Dunn & Dunn, 2007). This standardized language assessment was developed for use with participants aged 2 years 6 months to over 90 years, and is often used as a measure of general language ability or verbal intelligence. This test was chosen becau se it is commonly used in studying language development, is prevalent in the literature examining the relationship between language and ToM, and because it is standardized for the widest age range of all language measures, making it useful for comparing ac ross studies and applying to a diverse sample of adults. To administer this test, the researcher states the target word for each item and participants must choose the correct illustration (out of four choices) that matches the target word. Target words
54 bec ome increasingly difficult and administration of the task is ended when 6 of 8 consecutive words are answered incorrectly. Participants were also asked to report their SAT and ACT scores for verbal sections of the test. However, due to a response rate of l ess than 40%, this information was excluded from analyses Executive functioning/Inhibitory control. Participants complete d the St r oop Color Word Test (SCWT), (Victoria Version, Spreen & Strauss, 1998). In this task, est cards were timed individually. The first card consists of 6 rows of 4 dots, colored red, green, blue, and yellow in varying order. Participants are required to name the color of the dots as quickly as possible. The second card consists of 6 rows of 4 w ords (when, hard, over, and) printed in red, green, blue and yellow ink (word color and order varies across rows). Participants are required to name the color of the ink for each word as quickly as possible. The third card consists of 6 rows of 4 color wor ink, for example). Participants are required to name the color of the ink for each color word as quickly as possible. M ean time (in seconds) to complete the first (color reading) and seco nd (ink color naming) cards will be subtracted from the time required to complete the third (IC) card. The resulting score is considered to be an indicator of individual differences in IC, as a participant who answers the third card more quickly (or more s imilarly to the first two cards) is better able to inhibit the more automatic responses of color word reading in favor of the secondary response of ink color naming. The third card maximizes this IC demand by presenting color words in mismatched colored in k.
55 Autobiographical Memory and Story Narratives Participants were interviewed to elicit three AM narrative s and one narrative about a wordless story book. Time to complete these narratives varied among individuals. Autobiographical Memory. For the first au tobiographical memory narrative, p articipants were asked to relate to the interviewer a memory of an event, occurring prompt ed the participant by telling them that often we tell stories to give another person insight into our own sense of who we are, or we remember these events privately to were given time to think about the prompt, and then asked to relate the memory to the b est of their ability to the interviewer. For the remaining two autobiographical memory narratives, participants were memories (car, water, play, fear, animal). Participants w ere then asked to remember events from their lives which were cued to memory by individual words from the list. Participants were then asked to choose a word and to narrate a memory related to the cue word chosen to the best of their ability so that the in terviewer understood. This procedure was repeated with a different cue word. The interviewer was instructed to act as an active and interested listener without speaking, responding or asking questions. Wordless Story Book Narrative. Participants were prov ided with a wordless Breakfast for Jack sense of the story line. Once satisfied with their understanding of the story, participants were asked to narrate the story aloud, keeping in mind that the narrative w ould be intended for child around the age of 4 who has not seen the book or heard the story
56 before, but who will be able to look at the pictures while listening to the recorded narrative. This story was chosen for its potentially engaging storyline, colorful illustrations, as well as its potential for participants to include information regarding mental state, emotional, and perceptual details. Narrative Quality Ratings N arrative s were audio recorded, then rated for overall story (PSQI). This six question scale asks a lay rater to indicate how entertaining, emotional, memorable, original, rich in imagery, and engaging a narrative is using a 5 point likert type sca le. Ratings for each individual scale were averaged to represent an overall story rating ranging from 1 5. Baron and Bluck showed that individual raters tend to rate the same story similarly, reducing the need for a large number of ratings of quality for e ach individual narrative. Thus, each AM narrative, as well as each wordless storybook narrative, was rated using this scale by 3 independent coders and the average rating of these three was used in analyses. Internal validity of the scale was determined us ing validity. Additionally, when considering the average score across three diffe rent AM overall average of all AM narratives and the storybook narrative was examine d, it was shown that Alpha was higher when the storybook narrative was not included in the
57 overall average. Therefore the overall story quality average as well as AM and storybook narratives were examined in separate analyses in order to determine if rela tionships among language, IC, ToM and narrative were different depending on whether the story was autobiographical or fictional. Three narrative scores were used in analyses. Average rating for all AM narratives (AM only), average storybook rating (Storybo ok Only), and an overall story quality average which included both AM and storybook ratings.
58 Figure 3 1. C urse of knowledge task stimulus.
59 CHAPTER 4 R ESULTS Descriptive results Descriptive results are displayed in Table 4 1 for each of the measures of interest. All missing data was handled using listwise deletion, as the missing data was considered to be missing completely at random, and loss of power was not found to be an issue using listwise deletion in this study. Age w as correlated with raw scores on the PPVT 4 in preliminary analyses, so it was deemed appropriate to use standardized scores (consistent with the intended use of the test) in analyses, thus controlling for age in analyses involving the PPVT 4. The mean sta ndard score on the PPVT 4 for this sample (M = 96.74) was close to the population mean, and performance on this test was normally distributed. Also of note, maximum ratings of story quality for the storybook and for AMs remained below 4, out of a maximum o f 5 as a rating of a very good story, with means just below average. Raters consistently found that the stories told in this study were average or below average. Performance on the Reading the Mind in the Eyes task was as expected, with a mean score of 24. 74 items correct out of 36 (SD = 4.19, range 15 33 correct). All participants performed better than chance and there were no ceiling effects, indicating that the test was understood by participants but posed a challenge and highlighted individual differenc es in emotion recognition effectively. Performance on the remaining two proposed ToM measures ( curse of knowledge and hindsight bias ) was slightly different than expected based on prior work. Responses on the curse of knowledge task were not normally distr ibuted. More than half of the responses on this task fell in the
60 30%, indicating that most participants were not biased by the knowledge of the true location of the object in the scenario. Only one participant assigned 100% probabilit y to the red container, indicating being fully cursed by the knowledge of the true location, and only five participants assigned a probability higher than 50% to the red container. Two participants assigned 0% probability to the red container. For the hin dsight bias measure, a mean score of 1.13 indicates that participants, overall, were subject to hindsight bias, but only slightly. That is, their responses drifted closer to the correct answer when the correct answer was given to them roughly one item more frequently than when the correct answer was not given. Hindsight bias scores drifted closer to the correct answer more frequently when the correct answer was not given or dr ifted closer to the correct answer equally in the two conditions (n = 13) resulting in 39.7% of the sample failing to show hindsight bias. Study A im 1 The first aim of the present study was to examine associations among a variety of ToM measures in early adulthood, namely the Reading the Mind in the Eyes task, FB hindsight bias First, these relationships were explored using simple correlations (Table 4 2) No relationships were found to exist among the Reading the Mind in the Eyes task, curse of knowledge task, or hindsight bias Thus, the confirmatory factor analysis planned to show a unitary The Eyes task henceforth will be used as the standard measure of ToM for analyses. This is based on the fact that it is the most established, most frequently used
61 measure in the ToM literature, whereas the curse of knowledge task and the hindsight bias task h ave not previously been used as individual difference measures. Attempting this application of these measures was intended to extend the study of individual differences in ToM in adulthood through the addition of tasks used in the adult social cognition li terature However, because the Eyes task was the only of the proposed ToM measures to be correlated with more than one other measure of theoretical or practical interest (described below) the Eyes task was selected to operationalize ToM in adulthood Stu dy A im 2 The second aim of the present study was to reveal relationships among ToM, language, and IC. These relationships were first explored using simple correlations, shown in Table 4 2. Notably, age was not found to be correlated (Table 4 2) with any of the proposed measures of ToM, indicating that age did not suffice in explaining individual differences in performance on these measures. Higher scores on the Eyes task were found to be significantly associated with higher language scores ( PPVT 4 ) ( r (94) = .20). In contrast, l anguage was not found to be related to either of the remaining proposed ToM measures ( curse of knowledge or hindsight bias ). IC as measured by the traditional Stroop task was found to be related to both the Reading the Mind in the Eye s task ( r (82) = .32, p = .004) and the curse of knowledge task ( r (82)= .25, p = .023). Quicker response times on the incongruent (test) card of the Stroop task were associated with higher scores on the Eyes task and lower likelihood of being cursed by the knowledge of the true location of the object in predicting the behavior of an individual with a false belief.
62 To explore the relative contributions of language (PPVT 4) and inhibitory control (Stroop) for ToM, a regression model using the forced entry me thod was conducted. O nly the Stroop task significantly predicted performance on the Reading the Mind in the Eyes task = 26 t ( 79 ) = 2.65 p < .05 and explained a significant proportion of variance in Reading the Mind in the Eyes scores, R 2 = .13 F (2, 79 ) = 5.74 p < .01, however the effect was small Because of the non normal distribution of scores on the curse of knowledge task, responses were recoded to indicate whether the participants were not cursed (assigned a probablitiy of 30 34% to the red container n = 45 ), cursed in favor of the red container (assigned a probability higher than 34% n = 30 ) or cursed against the red container (assigned a probability less than 30% n = 21 ). A one way ANOVA comparing these three groups on all other measure s was not significant. The groups did not differ in performance on language or IC test indicated no difference on any measure between gr oups. Thus, it is difficult to interpret the small but significant correlation found between the curse of knowledge task and the Stroop task mentioned above. Study A im 3 The third aim of the present study was to explore the relative roles of ToM, language and IC for story quality reflected in the PSQI scores for AM and storybook narratives These relationships were first explored using simple c orrelations (Table 4 2). Higher scores on the Eyes task were found to be related to higher story quality when con sidering storybook narratives (r(83) = .36, p < .01) and the Overall (combined
63 storybook and AM) story quality rating (r(82) = .38, p < .01). Higher scores on the PPVT 4 were found to be related to story quality for AM stories only (r(94) = .23, p .05). No other measure was significantly correlated with story quality. Multiple regression analyses are displayed in Table 4 3. Separate multiple regression models (using the forced entry method) with each of the three narrative scores used as the outcome vari able (AM Only, Storybook Only, and Overall Story Quality) were examined with the Eyes task, the PPVT 4, and the Stroop as predictors. Regression models predicted a significant portion of the variance in Storybook Quality F (3,80) = 3.766, p = .014, and Overall Story Quality, F (3,81) = 3.671, p = .016. The model predicting AM quality alone was not significant. Upon examination of individual predictors within the regression models, for both Storybook only and Overall Story Quality only the Eye s task was a significant predictor of va .3 2 R 2 = .12, R 2 = .13 respectively). Study A im 4 The fourth aim of the study was to show an indirect effect of language and IC on storytelling quality through ToM as measu red by the Eyes task, using a mediation model (Figure 2 1). Hayes (2009) argued for the use of a bootstrapping method of testing for bootstrap resampling method allows f or nonnormal distributions and tests for mediation using a 95% confidence interval. Results are summarized in Table 4. First, when Overall Story Quality (OSQ) was entered as the dependent variable (DV), the omnibus test of the indirect effect indicated tha t the model supported an indirect pathway from the IVs (language and IC) through the mediator (ToM) to the DV (story quality) Examination of the individual IVs indicated that IC but not language, was responsible
64 fo r the significance of the model. However the effect was very small. That is, the 95% confidence interval of the indirect effect of the Stroop through the Eyes task on story quality did not include zero, suggesting the presence of an indirect effect, indirect effect = .01, R 2 = .13, adjusted R 2 = .1, 95% CI [ .0137, .0013]. In order to determine relative effect size for the model, more work is needed in this area. Standardized coefficients were not calculated, as the creators of the method employed do not recommend the interpretation of standard ized coefficients. Thus, the relative size of this effect is not clear. The pattern of results was the same when Storybook Only ratings were entered as DV. When Autobiographical Memory Only ratings were entered as DV, no indirect effect was found.
65 Table 4 1. Descriptive information for key measures. Measure (possible score/metric) N Mean SD Minimum Maximum Age (years) 96 21.29 3.41 18.47 35.16 PPVT 4 (standard) 94 96.74 11.65 75 130 Eyes (0 36) 96 24.74 4.19 15 33 FB Curse ( .33 .67) 96 .02 .14 .33 .67 Hindsight Bias ( 25 25) 78 1.13 2.89 8.0 7.0 Stroop (seconds) 81 10.28 4.63 3.0 25.5 AM Only (1 5) 94 2.55 .35 1.86 3.47 Storybook Only (1 5) 83 2.06 .50 1.13 3.50 Overall Story Quality (1 5) 82 2.30 .36 1.56 3.27
66 Table 4 2. yes ), curse of knowledge (Curse) hindsight bias (HB), language (PPVT 4^) and inhibitory control ( Stroop ) and story quality. Age in years Eyes Curse HB PPVT 4 Stroop Eyes (ER) .05 Curse .17 .09 HB .07 .17 .05 PPVT 4 .02 .18 .04 .09 Stroop .07 ** .31 *.25 .07 .16 AM Quality .02 .20 .08 .03 *.23 .14 Storybook Quality .20 **.36 .09 .02 .09 .18 Overall Story Quality .11 **.38 .19 .01 .06 .19 ^PPVT 4 scores are standardized p < .05, two tailed, ** p < .01, two tailed
67 Table 4 3 Multiple regression analyses of the effects of emotion recognition (Eyes), inhibitory control (Stroop), and language (PPVT 4) on story quality. AM only Storybook only^ Overall Story Quality^ Variable B SE B B SE B B SE B Constant 1.753 .39 1.072 .61 1.341 .41 Eyes .01 .01 .17 .04 .01 **.3 2 .03 .01 **.31 Stroop .002 .01 .03 .01 .01 .07 .01 .01 .06 PPVT 4 .01 .003 .17 .001 .01 .01 .003 .003 .09 R 2 .08 .1 2 .1 3 ^model predicted a significant proportion of variance in the DV; *p < .05. **p<.01
68 Table 4 4. Summary of mediation results for Language (PPVT 4) an d Inhibitory Control (Stroop) mediated by Emotion Recognition (Eyes) on story quality using bootstrapping method (5000 bootstrap resamples). OSQ=Overall Story Quality, SB=Storybook only, AM= Autobiographical memory only. DV Mediator IV Direct Effect (SE) Indirect Effect (SE) 95% CI for Indirect Effect Model R 2 LL UL OSQ Eyes .03 (.01) PPVT 4 .00 (.00) .00 (.00) .0004 .0046 Stroop .00 (.01) .01 (.00) .0137 .0013 Omnibus .00 (.00) .0003 .0095 *.13 SB Eyes .04 (.01) PPVT 4 .00 (.00) .00 (.00) .0007 .0072 Stroop .01 (.01) .01 (.00) .0216 .0029 Omnibus .00 (.00) .0004 .0143 *.14 AM Eyes .01 (.01) PPVT 4 .00 (.00) .00 (.00) .0005 .0027 Stroop .00 (.01) .00 (.00) .0091 .0009 Omnibus .00 (.00) .0006 .0057 .07 p = .01
69 CHAPTER 5 D ISCUSSION The study described was intended to explore social cognition in early adulthood, with a focus on ToM intended to bridge a gap between child ToM research in the d evelopmental p sychology literature and adult social cognition research in the s ocial p sychology literature. These results add to the gr owing literature describing adult ToM. The findings of this study will be discussed by each of the four primary study aims in turn, followed by a discussion of the overall theoretical implications for the development of ToM. Study Aim 1 The first aim of th e present study was to examine associations among a variety of ToM measures in early adulthood, namely the Reading the Mind in the Eyes task, FB hindsight bias These tasks were chosen because they represent th assessing ToM and tasks used in adult social cognition literature. Of primary interest is whether ToM in adulthood represents a unitary construct and whether there are meaningful individual dif ferences in ToM performance. The Reading the Mind in the Eyes task used in this study is the adult version of a child version of the same task used to assess emotion recognition (Baron Cohen, Wheelwright, Scahill, Lawson & Spong, 2001). The curse of knowl edge task is a nuanced version of the traditional FB change of locations task designed to tap into more subtle errors in adult FB reasoning (Birch & Bloom, 2007). The participant is required to predict the probability that a nave character will search for a secretly displaced object in a location
70 referred t o is simply the knowledge of the true location of the object, and the fact that adults do not perform at ceiling on this task suggests that the ability to reconcile conflicting representations (reality and a FB ) is imperfect in adults. The hindsight bias task, because it highlights t tasks. For example, if a young child is shown that a crayon box contains ribbon, the child is likely to claim they always thought the box contained ribbon. Similarly, if an adult is told that the Eiffel Tower is 986 feet tall, the adult is highly likely to claim their guess of the height would have been very similar to 986 feet. Because these tasks fit into the scheme of child ToM research which shows consistent correlations between these types of tasks, and because well respected and well researched theories of ToM and autism (Baron Cohen, 2000) propose that these skills are based on an underlying and unitary ToM trait, results of the pre sent research were expected to show interrelations among individual performance on these adult tasks. This expectation was not met in this study as the three ToM measures were not correlated This result was unexpected, as multiple tasks measuring differe nt aspects of ToM (such as FB and representational change) are often found to be correlated in the child ToM literature (i.e., Gopnik & Astington, 1988). There are multiple explanations for why these tasks were not related in the current study. First, thes e results may indicate that the measures of ToM do not represent a unitary construct, and are not organized as such within the individual. This evidence may be construed as an indication that ToM, as conceptualized in many prominent theories of ToM is not truly a unitary cognitive construct, but more of a constellation of skills that share common skills, such
71 as thinking about mental states, the self other distinction, the physical mental distinction, and meta representation, but which are not based on a co mmon biological or cognitive underpinning. It is important to recognize that ToM in adulthood is less be replicated in order to further apply the finding that multiple ToM skills did not represent an underlying ToM construct. However, in light of the present findings, it is important to realize that ToM as discussed in the present paper from this point forward is operationalized using only the Reading the Mind in the Eye s task, an emotion recognition task that does not require the coordination of multiple conflicting representations. Thus, some may view this as not a true or complete representation of ToM. However, as the Reading the Mind in the Eyes task has previously b een used as a measure of ToM (i.e.,Baron Cohen, Jolliffe, Mortimore, & Robertson, 1997 ), I consider it appropriate to use this task as an indicator of some form of ToM for the present study. A point worth considering is that the individual differences expr essed through the hindsight bias and curse of knowledge tasks were not meaningful as a result of methodological difficulties or sample issues. Further, though the measures of ToM used in this study were not interrelated, it may be that a different selectio n of ToM tasks for adults may show a different pattern of results An interesting distinction has recently received attention in the ToM literature delineating two separate types of ToM reasoning: affective ToM and cognitive ToM (Leopold et al., 2012). Af fective ToM includes the ability to understand, interpret, and empathize with the affective or emotional mental states of others. Cognitive ToM includes the ability to reason about epistemic mental states such as beliefs and
72 knowledge. Evidence from brain injury patients (Leopold et al., 2012) suggests that these two types of ToM are grounded in different regions of the brain. As a specific example of this, Leopold and colleagues showed that damage to the v entromedial p refrontal c ortex and bilateral v entrom edial p refrontal c ortex was associated with significantly impaired affective but not cognitive ToM ability. Results of the present study may be viewed as supporting this brain evidence, as the affective Reading the Mind in the Eyes test was not associated with the cognitive curse of knowledge and hindsight bias test s Some methodological considerations, or limitations, may also provide an explanation of the finding that the ToM measures did not re present an underlying construct Specifically, the ToM metho ds used in the present study, with the exception of the Reading the Mind in the Eyes task, have not previously been used to assess individual differences in ToM reasoning ability. Thus, the use of measures not well established or supported in the existing literature makes strong conclusions about the implications of a lack of association among these measures difficult. These measures were used in the present study as an exploratory tool, and this use was always considered a bit of a gamble in terms of what the findings might be. Further research and development of individual difference measures would be ideal in moving toward a stronger methodology for studying curse of knowledge and hindsight bias as adult correlates of FB and representational change. It is possible that the curse of knowledge and hindsight bias measures employed are not valid measures of ToM but there may be other explanations for individual performance on ToM measures observed in this study First, the underwhelming
73 evidence of both curse of knowledge and hindsight bias in this sample may be an important addition to the hindsight bias and curse of knowledge literature as the robust findings of bias previously reported for these tasks was not replicated here. Next, f actors such as atte ntion, motivation to perform, and time to respond were not measured in this study. Questions arise during test administration using a participant pool (in which students are required to participate in research, or alternatively write a paper, for course cr edit) about the possibility that these factors will affect outcomes. Thus, whether the participant is motivated to apply their full cognitive resources to the measures, or whether they are simply filling in blanks as quickly as possible may be a way of det ermining whether scores on these measures are reliable. For the hindsight bias measure, even though response times were not measured, it was clear during task administration that individual differences existed in attention, motivation, and time to respond both to the Day 1 and Day 2 lists of uncommon knowledge. During the curse of knowledge task, participants may have merely skimmed to the end of the vignette and reasoned that there was an equal probability that the character would look in the three contain their reasoning (that the object was really in the red container). In contrast, participants appeared to enjoy and relate with the Reading the Mind in the Eyes task, potentially leading to more reliable responses on this measure. entering numbers indicating probability on the curse of knowledge task to delve deeper responses. Additionally, more attention grabbing methods of administering these tasks may be employed. For example, a video vignette
74 may be more likely to transmit the essential information to all participants. Other methods for assessing adult reasoning about FB s may be added to this type of study as well. Apperly (2006) recorded time to respond to basic FB questions. This method may be useful in exploring individual differences in FB skill in future research. Study Aim 2 The second aim of the present study was to reveal relationships among ToM, as measured using the Reading the Mind in the Eyes task, language, and IC. The expectation that individual differences in language, as measured by the PPVT 4, would be associated wi th individual differences ToM was supported when considering simple correlations with the Reading the Mind in the Eyes task. However, the effect of language for this task was eliminated when examined within a regression model including the IC task (Stroop) Despite the fact that the Stroop and the PPVT 4 we re not correlated it seems that an association between language and the Eyes task is explained by IC in this sample. This finding is inconsistent with a large amount of research evidence from the litera demonstrating a strong and consistent relationship with language (i.e., Milligan, Astington & Dack, 2007 ) It is possible that, in adulthood, language takes on a different role in the development or use of ToM. The findings are consistent with research evidence from the childhood literature showing the importance of IC for ToM skill (i.e., Carlson, Moses & Breton, 2002 ). Further, the fact that the best indicator of ToM in this study is an emotion recognition tas k, not a task that requires reasoning about conflicting representations, may explain why language is not a good predictor of t his form of ToM. Some may argue that emotion
75 recognition is a separate skill, not entirely representative of ToM as a whole. Further research should address the role of language in both affective and cognitive ToM. Apperly, Samson and Humphreys (2009) provide a clear explanation of the the dif ferent roles of language and EF They explain that language is an essential tool in the initial construction of ToM in childhood, but once ToM is constructed, language ability is no longer associated with the support and use of ToM in adulthood. Applying N it stands to reason that language interactions in childhood do indeed play a role in escorting the child into the understanding of other minds. Through language, children are able to access the contents o f the minds of others, understand that perspectives differ, and realize that the contents of the mind are not visible and must be revealed through language. The findings presented here, supported by the work of Apperly and colleagues, suggest that once the child is escorted into the community of minds through language, language has done its part and is no longer necessary for reasoning about ToM. However, measures of the indirect effect of language, using a cognitive interference paradigm during administrat ion of FB tasks suggest that there may be some role of language (Newton & de Villiers, 2007). Newton and deVilliers showed that requiring adult participants to engage in a language based interference task reduced accuracy on a non verbal test of FB, more s o than when using an interference task that did not require language (for an argument to the contrary, see Dungan & Saxe, 2012). In contrast, EF (specifically, IC) appears to be a skill inherently involved in ToM reasoning itself, and not simply a tool fo
76 improving ability to inhibit prepotent responses, avoid the pull of the real, and suppress an egocentric perspective is clearly linked to their ability to perform on FB tasks (Carlson Moses & Breton, 2002 ). Fi ndings from the current study support the necessary role of IC for ToM reasoning, as opposed to the weaker claim that IC may be important simply for the initial emergence of ToM. Further, these findings suggest that IC is critical for the expression of ToM skill s that exist within the individual. Even adults must actively engage in IC to suppress their own perspective in order to take the perspective of another. In the current study, better IC was associated with better performance on the Reading the Mind in the Eyes task. This finding is particularly strong evidence for the essential and ongoing role of IC for ToM throughout development. This emotion recognition task does not require inhibition of a prepotent response, or suppression of ive in favor of the perspective of another or indeed reasoning about conflicting representations The association found may suggest that IC is critical to the development of the emotion recognition skill itself and not only for performance on a specific test of ToM skill which overtly requires IC. Thus, processing, understanding, and relating to the information conveyed through the facial expressions of others may require the individual to inhibit automatic responses. These automatic responses may be base d on preconceived notions about what a person should feel at a particular over time and across development may enhance emotion recognition, thereby enhancing a ToM skil l.
77 Zelazo and Carlson (2012) describe an emerging way of differentiating EF skills EF under emotional or highly motivated circumstances. Cold EF refers to EF that is less affective and more cognitive. Interestingly, the Stroop task used in the current study be associated with the more emotional ToM task used, the Reading the Mind in the Eyes task. It is possible that even stronger effects of EF for ToM may be found if a more A Stroop like task which requires inhibition of the actual emotion expressed on a face in favor of a response of the opposite emotion, called an Emotion Face Stroop task may be used for this purpose (Avram, Baltes, Miclea & Miu, 2010). Current neuropsychological evidence supports the finding that EF (IC) continue s to play a role in ToM in adulthood a s well. D evelopment of EF regions of the brain in critical in ToM reasoning in adulthood. In their cross sectional study of children and adults, Apperly, Warren, Andrews, G rant and Todd (2011) showed that across age groups, FB reasoning took longer and required more cognitive effort than did reasoning about true beliefs. Processing ToM information clearly requires effort and is not automatic, even for adults. There are many reasons for this supported by research (Apperly, Samson & Humphreys, 2009 provide a review). Briefly, reasoning about mental states sometimes requires inhibiting a known reality to understand the behavior of one with an inexperienced mind. Often a person m ust inhibit their own knowledge, their own opinion, and their own experience in order to understand the knowledge,
78 opinion and experience of others. Adults are capable of these things, but require cognitive effort in the employment of these behaviors. The cognitive effort involved seems to be located largely in the areas of the brain associated with EF Importantly, research is ongoing attempting to determine specific areas of the brain devoted to ToM itself (Apperly, Samson, Chiavarino, Bickerton, & Humphr eys, 2007). The present finding that IC, and not language, is associated with emotion recognition performance in adulthood is congruent with findings from brain injury patients who experience intact ToM reasoning ability despite damage to language areas of the brain, and impaired ToM associated with damage to EF areas of the brain (Apperly, Samson, Carroll, Hussain & Humphreys, 2006; Apperly, Samson & Humphreys, 2009). No previous study has directly examined individual language ability in participants witho ut brain injury and its association with ToM in adulthood (see Newton & de Villiers 2007, for an indirect examination of language and ToM.) The findings presented here dovetail nicely with the findings of Apperly and colleagues demonstrating this lack o f association. Apperly and colleagues (2006) describe a detailed case study of a patient with compromised general language, general grammar, and specific grammatical structures closely scrutinized in the ToM literature as being significant for ToM reasoni ng (embedded complements). This patient was able to perform accurately on first and second order FB tasks. Further, brain injury patients with damage to regions of the brain associated with EF and IC show significant decreases in ToM reasoning ability. App erly and colleagues (2005) describe a case study of a patient with damage to the right frontal and temporal lobe as a result of a stroke. This patient was unable to inhibit his
79 own perspective when confronted with FB tasks and when asked to make judgments about the desires, perceptions, and emotions of other people when he had his own perspective on the same situation. That is, he was unable to inhibit his own personal judgments when considering the perspectives of others. Studies of participants without br ain injury have revealed a similar association between IC and ToM (Apperly, Samson & Humphreys, 2009). Beyond this explanation, other conclusions may be considered when reviewing the findings of the present study. First, it is important to recognize that language was measured using only one indicator of ability, the PPVT 4. This test is widely used and highly reliable, however drawing conclusions about the lack of importance of language for ToM in adulthood should be done with caution. The PPVT 4 is highly correlated with IQ, and assesses receptive vocabulary. To achieve this in adults, the items on the PPVT 4 represent words used with low frequency in speech, conversation, and even in printed text. Advanced a cademic study and high general intelligence are more likely to be predictors of performance on the PPVT 4 for adults than amount and quality of social interaction, for example. Thus, the notion that high quality social interactions through language may contribute to better ToM ability in adults is not n ecessarily addressed when language is measured using the PPVT 4. Similarly, it could be that it is not language, per se, that plays a role in enhancing adult ToM, but ongoing participation in social interaction in general. More research in this area is nee ded. Assessing multiple aspects of language, and obtaining objective measurements of amount and quality of social interaction may elucidate the role of language and speak to the social construction of ToM in adulthood.
80 In the present study, ToM was operati onalized mainly using the Reading the Mind in the Eyes task, a test of emotion recognition. Although other measures were used in this study, intending to provide information about a variety of different ToM reasoning skills, it is unclear why the curse of knowledge and hindsight bias tasks were not found to be associated with language, as measured by the PPVT 4. This may be due to the potential limitations of these tasks discussed earlier. An examination of other indices of ToM reasoning ability is also called for when trying to answer questions about the role of language for ToM in adulthood. Other methods used for assessing ToM in adulthood include the grid paradigm developed by Keysar, Lin and Barr (2003), the Faux Pas test ( Stone, Baron Cohen & Knight, 1998), and the Strange Stories task (Happe, 1994) among others. A study utilizing a variety of ToM tests, representing both affective and cognitive ToM, and their associations with language ability and IC in typically developin g adults is needed. Study Aim 3 The third aim of the present study was to explore the relative roles of ToM as measured by the Reading the Mind in the Eyes task language and IC for story quality. Previous research on autobiographical and fictional storyte lling led to the expectation that language would be a predictor of story quality, as storytelling is essentially an exercise in language use. Importantly, p revious research has also shown that ToM is related to autobiographical memory and in story quality in general (i.e., Kleinknecht & Beike, 2004). As discussed earlier, ToM was expected to allow the storyteller to take the o wn perspective, a storyteller is able to pro vide the most relevant and appropriate information which the listener needs to follow and understand the story.
81 A good story, as perceived by the listener, will necessarily require that the speaker communicate effectively. Effective communication is depen dent upon the 2012). This is the essence of ToM. Take the example of the role of imagery in telling a story. The presence of imagery in a story is an indicator of a g ood story (Baron & Bluck, 2011), and is one of the six indices of story quality assessed in the PSQI. To summon an image to the mind of the listener using only words to tell a story, the speaker must fundamentally understand that it is possible to manipula te or exercise influence over the perceives, and the speaker is required to reveal his or her own inner thoughts in order to affect the listener in this way. Results of th is study support the role of ToM at least as measured by an emotion recognition task, in story quality with higher scores on the Reading the Mind in the Eyes task predicting better perceived story quality but did not support the role of language or of IC (included because of its high level of interconnectedness with ToM, language, and social cognition in general). The present results are in concordance with the finding that individuals with autism have difficulty telling objectively high quality stories, and that in these clinical populations, higher scores on the Reading the Mind in the Eyes task are associated with higher story quality (Colle & Baron Cohen, 2008). That emotion recognition, as measured by the Reading the Mind in the Eyes task, is associa ted with story quality in the present study may have been partially influenced by the procedure used for eliciting narratives from participants. Here, we used a social sharing paradigm in which participants told their stories to an
82 experimenter, face to fa ce. Despite the fact that experimenters were instructed not to interject verbally or ask questions, no instructions were given to control for the facial expressions of the experimenter as instructing experimenters to hold a steady, impassive expression ma y have created a potentially uncomfortable environment not conducive to participant storytelling (based on typical response to the still face paradigm, Mesman, van IJzendoorn & Bakermans Kranenburg, 2009). Not all participants told their stories to the sam e experimenter, but experimenters shared certain key characteristics, including a specific interest in autobiographical memory story sharing research. Experimenters are likely to have shown interest, and responded to participant stories as they were being shared using facial expressions. Those participants more skilled at reading facial expressions may have been more likely to add information when a quizzical look was observed, elaborated on a point when an interested expression was observed, and so forth. Different results may be found if stories are elicited without an experimenter present, either using a written narrative paradigm or solitary verbal narration in front of a video camera or tape recorder, for example. That language as measured by the PPVT 4 was not associated with story quality in this study may, again, be the result of the type of language assessed by the PPVT 4. Knowledge of low frequency vocabulary words may not be the best indicator of the type of language skill that might be needed to t ell a good story. Grammatical skill, as well as pragmatic language and knowledge of story structure might be language indices more likely to show relationships with story quality. Further, it may be useful to consider that the stories told in this study we re overwhelmingly rated as low to average in quality.
83 Clearly, no participant in the study was telling stories to rival the imagery, entertainment, emotion, or originality of Mark Twain. It is possible that personal investment in participation in the study was lacking, and even those participants c apable of telling a high quality story through the use of advanced language skill did not choose to do so. Study Aim 4 The fourth aim of the study was to show an indirect effect of language and IC on storytelling quality through ToM as measured by the Eyes task, using a mediation model (Figure 2 1). Results of the mediation analysis were not straightforward. The results suggest an indirect effect of IC on story quality through the Eyes task but the effect size wa s very small sizes, m ore research in the area of adult ToM and storytelling is needed to truly understand the size of the effect found) Thus, it is possible that IC affects storytelling quality through its est ablished role in ToM, but strong conclusions about the implications of this finding cannot be made. If this result is interpreted as reflecting a true effect, it may be explained by revisiting the role of IC in ToM and the role of ToM in storytelling. This constellation of factors tells an interesting story. Employing ToM, as we have seen, requires cognitive effort, apparently in the form of IC IC may have an indirect effect on story quality because the storyteller must use executive cognitive resources to inhibit his or her own perspective, knowledge, and experience with the event told in the story in order to communicate effectively with the listener. That is, IC is needed to utilize the ToM necessary for good communication with a listener and ultimately good storytelling. This is especially interesting considering that ToM is operationalized using an emotion recognition task, not a more cognitive FB or representational change task, yet the effect was still present.
84 However, because this effect was small a nd only a partial one, more information is needed to make confident conclusions. That language did not emerge as having an indirect effect on story quality through ToM (emotion recognition) can be understood in light of the findings discussed earlier. Lan guage itself was correlated with AM story quality when examining simple correlations. However, that ToM does not explain this relationship can be expected, as we have seen that a good amount of evidence shows that language in adulthood is not a predictor o f ToM use or understanding. Thus, language plays a small role in storytelling because stories are essentially made up of language, but IC is the primary cognitive factor (measured here) which underlies the use of ToM in adulthood. Implications and Conclusi on The present study focused on individual differences in ToM in early adulthood and its relationships with language, IC, and storytelling ability. Results did not support a unitary ToM construct that spanned FB, representational change, and emotion recogn ition in the developmental and social psychology literatures. However, the present f indings concerning the role of ToM as measured using an emotion recognition task, in good communication may promote further explorations of the role of ToM in psycho therapy or counseling, teaching, parent child conversations, and parent child storytelling. In addition, the role of IC in these domains may be considered to be of equal interest. Effective communication of ideas appears to be related not only t o a egocentric focus on his or her own knowledge or perspective. The findings presented here also add to a growing literature that speaks to the issue of the c ontinuity of development of ToM across the lifespan. Evidence from
85 infancy, through childhood and adolescence and extending into adulthood supports a ultimate competence. The ability to use and apply ToM reasoning in social situations and in social reasoning continues to change and vary across individuals even after the typical child gains a new ability to reason about conflicting representations in the preschool years. Th e ability to reason about conflicting representations, to understand the representational nature of the mind and specifically to pass traditional FB tasks is acquired fairly uniformly by ty pically developing individuals. However, applying that reasoning c ontinues to be subject to change, individual differences, and development driven by brain maturation, continued social interaction, or possibly improvements in language, intelligence, and a host of other experiences not yet discovered. Results presented he re support the importance of IC in the application of ToM to communication with others in adulthood. The ways and means by which adults become proficient at applying their uniquely human ToM skill is currently a fascinating and fruitful topic of research. At the most basic level, the utilization of ToM for communication may be applied (1986) socio cultural theory, specifically when considering the role of the cognitive functioning. While simply being in the presence of a more skilled peer or adult may play a role in the cognitive development of the child, active scaffolding may suitably be viewed as being impacted by the use of ToM skills. An adult attempting to scaffold or teach will necessarily have to take into consideration the mental state of the child. By
86 functioning or knowledge, and creating a zone of proximal development within which the child is capable of achieving success in new knowledge and cogniti ve skill. That development are founded in the human ability to understand and utilize the concepts of ToM. In conclusion, the present research adds to a rapidly expanding fiel d exploring ToM in adulthood. These explorations are elucidating the need to view ToM as more than just a skill that emerges in childhood, but as one that must be used and applied to everyday functioning. This use is effortful, is by no means applied perfe ctly automatically, or at a ceiling level by children older than age 5 or even by adults, can come in many different forms, and can be impacted by a variety of other cognitive processes or skills (Samson & Apperly, 2010).
87 LIST OF REFERENCES A dler, N., Nadler, B., Eviatar, Z., & Shamay Tsoory, S. G. (2010). The relationship between theory of mind and autobiographical memory in high functioning autism and Asperger syndrome. Psychiatry Research 178 (1), 214 216. doi:10.1016/j.psychres.2009.11.015 Apperly, I. A., Carroll, D. J., Samson, D., Humphreys, G. W., Qureshi, A., & Moffitt, G. (2010). Why are there limits on theory of mind use? Evidence from adults' ability to follow instructions from an ignorant speaker. The Quarterly Journal Of Experiment al Psychology 63 (6), 1201 1217. doi:10.1080/17470210903281582 Apperly, I. A., Riggs, K. J., Simpson, A., Chiavarino, C., & Samson, D. (2006). Is belief reasoning automatic? Psychological Science, 17, 841 844. Apperly, I. A., Samson, D., Carroll, N., Hussa in, S., & Humphreys, G. (2006). Intact first and second order false belief reasoning in a patient with severely impaired grammar. Social Neuroscience 1 (3 4), 334 348. doi:10.1080/17470910601038693 Apperly, I. A., Samson, D., Chiavarino, C., Bickerton, W. & Humphreys, G. W. (2007). Testing the domain specificity of a theory of mind deficit in brain injured patients: Evidence for consistent performance on non verbal, 'reality unknown' false belief and false photograph tasks. Cognition 103 (2), 300 321. App erly, I. A., Samson, D., & Humphreys, G. W. (2009). Studies of adults can inform accounts of theory of mind development. Developmental Psychology, 45, 190 201. Astington, J., & Jenkins, (1999). A longitudinal study of the relation between language and theo ry of mind development. Developmental Psychology, 35 1311 1320. reflects cognitive biases in anxiety: Evidence from an emotional face stroop task. Applied Psychophysiology And Biofeedback 35 (4), 285 292. doi:10.1007/s10484 010 9138 6 Baron, J. M., & Bluck, S. (2011). That was a good story! Preliminary construction of the Perceived Story Quality Index. Discourse Processes 48 (2), 93 118. doi:10.1080/ 01638531003702109 Baron Cohen, S. (2000). Theory of mind and autism: A fifteen year review. In S. Baron Cohen, H. Tager Flusberg, & D. J. Cohen (Eds.), Understanding other minds: Perspectives from developmental cognitive neuroscience (2 nd ed., pp. 3 20). N ew York: Oxford University Press.
88 Baron Cohen, S., Jolliffe, T., Mortimore, C., & Robertson, M. (1997). Another advanced test of theory of mind: Evidence from very high functioning adults with autism or Asperger Syndrome. Child Psychology & Psychiatry & Al lied Disciplines 38 (7), 813 822. doi:10.1111/j.1469 7610.1997.tb01599.x Baron Cohen, S., Leslie, A., & Frith, U. (1985). Does the autistic child have a "theory of mind"? Cognition, 21 37 46. Baron Cohen, S., Wheelwright, S., Hill, J., Raste, Y., & Plumb, with Asperger Syndrome or high functioning autism. Journal of Child Psychology & Psychiatry, 42 241 251. Baron Cohen, S. Wheelwright, S. Scahill, V. Lawson, J. and Spong, A. (2001). Are intuitive physics and intuitive psychology independent? A test with children with Asperger Syndrome. Journal of Developmental and Learning Disorders, 5, 47 78. Bernstien, D. M. Atance, C., Loftus,G. R., & Meltzoff A. N. (2004). We saw it all along: Visual hindsight bias in children and adults. Psychological Science, 15 264 267. mental state reasoning. Trends in Cognitive Scien ces, 8 255 260. Birch, S. J., & Bloom, P. (2007). The curse of knowledge in reasoning about false beliefs. Psychological Science 18 (5), 382 386. doi:10.1111/j.1467 9280.2007.01909.x Botting, N., & Conti Ramsden, G. (2008). The role of language, social cognition, and social skill in the functional social outcomes of young adolescents with and without a history of SLI. British Journal of Developmental Psychology, 26 281 300. Buttelmann, D., Carpenter, M., & Tomasello, M. (2009). Eighteen month old infant s show false belief understanding in an active helping paradigm. Cognition 112 (2), 337 342. doi:10.1016/j.cognition.2009.05.006 Carlson, S. M., Moses, L. J., & Breton, C. (2002). How specific is the relation between executive function and theory of mind? Contributions of inhibitory control and working memory. Infant And Child Development 11 (2), 73 92. doi:10.1002/icd.298 Carpendale, J. L. M. & Lewis, C. (2004). Constructing an understanding of mind: The social interaction. Behavioral and Brain Sciences, 27 79 151. Carpenter, M., Nagel, K., & Tomasello, M. (1998). Social Cognition, Joint Attention, and Communicative Competence from 9 to 15 Months of Age. Monographs of the Society for Research in Child De velopment, Vol. 63.
89 Charman, T., & Baron Cohen, S. (1995). Understanding photos, models, and beliefs: A test of the modularity thesis of theory of mind. Cognitive Development 10 (2), 287 298. doi:10.1016/0885 2014(95)90013 6 Charman, T., Baron Cohen, S., S wettenham, J., Baird, G; Cox, A. & Drew, A. (2000). Testing joint attention, imitation, and play as infancy precursors to language and theory of mind. Cognitive Development, 15 481 498. Clements, W.A., & Perner, J. (1994). Implicit understanding of false belief. Cognitive Development, 9, 377 395. Colle, L., & Baron Cohen, S. (2008). Narrative discourse in adults with high functioning autism or Asperger Syndrome. Journal of A utism Developmental Disorders, 38 28 40. Corcoran, R., & Frith, C. D. (2003 ). Autobiographical memory and theory of mind: Evidence of a relationship in schizophrenia. Psychological Medicine 33 (5), 897 905. doi:10.1017/S0033291703007529 de Villiers, J., & de Villiers, P. (2000). Linguistic determinism and the understanding of fal se beliefs. In P. Mitchell & K. Riggs (Eds.), Children s reasoning and the mind (pp. 189 226). Hove, UK: Psychology Press. Dumonthiel, I, Apperly, I, & Blakemore, S. (2009). Online usage of theory of mind continues to develop in late adolescence. Developme ntal Science 1 8. Dungan, J., & Saxe, R. (2012). Matched false belief performance during verbal and nonverbal interference. Cognitive Science 36 (6), 1148 1156. doi:10.1111/j.1551 6709.2012.01248.x Epley, N., Morewedge, C. K., & Keysar, B. (2004). Perspec tive taking in children and adults: Equivalent egocentrism but differential correction. Journal of Experimental Social Psychology, 40 760 768. Farrar, M., Johnson, B., Tompkins, V., Easters, M., Zilisi Medus, A., & Benigno, J. P. (2009). Language and theo ry of mind in preschool children with specific language impairment. Journal Of Communication Disorders 42 (6), 428 441. doi:10.1016/j.jcomdis.2009.07.001 Fernyhough, C. (2008). Getting Vygotskian about theory of mind: Mediation, dialogue, and the developme nt of social understanding. Developmental Review, 28, 225 262. Fischhoff, B. (1975). Hindsight is not equal to foresight: The effect of outcome knowledge on judgment under uncertainty. Journal Of Experimental Psychology: Human Perception And Performance 1 (3), 288 299. doi:10.1037/0096 15184.108.40.2068
90 Fischhoff, B., & Beyth, R. (1975). 'I knew it would happen': Remembered probabilities of once future things. Organizational Behavior & Human Performance 13 (1), 1 16. doi:10.1016/0030 5073(75)90002 1 Flavell, J. reality distinction. American Psychologist, 41 418 425. Goldstein, T. R., Wu, K., & Winner, E. (2009). Actors are skilled in theory of mind but not empathy. Imagination, Cognition a nd Personality 29 (2), 115 133. doi:10.2190/IC.29.2.c change and its relation to the understanding of false belief and the appearance reality distinction. Child Development, 5 9 26 37. Mind & Language, 7 145 171. Habermas, T., & Bluck, S. (2000). Getting a life: The emergence of the life story in adolescence. Psychological Bulletin, 126 748 769. Haden, C. A., Ornstein, P. A, Eckerman, C. O., & Didow, S. M. (2001). Mother child conversational interactions as events unfold: Linkages to subsequent remembering. Child Development, 72, 1016 1031. Happe F. (1994). An advanced test of theory of mind: Understanding of story characters thoughts and feelings by able autistic, mentally handicapped and normal children and adults. Journal of autism and Developmental Disorders, 24, 129 154. Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical media tion analysis in the new millennium. Communication Monographs 76 (4), 408 420. doi:10.1080/03637750903310360 Kaland, N., Callesen, K., Mller Nielsen, A., Mortensen, E., & Smith, L. (2008). Performance of children and adolescents with Asperger syndrome or high functioning autism on advanced theory of mind tasks. Journal Of autism And Developmental Disorders 38 (6), 1112 1123. doi:10.1007/s10803 007 0496 8 Kemmis, L. L., Hall, J. K., Kingston, R. R., & Morgan, M. J. (2007). Impaired fear recognition in regul ar recreational cocaine users. Psychopharmacology 194 (2), 151 159. doi:10.1007/s00213 007 0829 5 Keysar, B., Lin, S., & Barr, D. J. (2003). Limits on theory of mind use in adults. Cognition, 89 25 41.
91 Kleinknecht, E., & Beike, D. R. (2004). How Knowing and Doing Inform an Autobiography: Relations Among Preschoolers' Theory of Mind, Narrative, and Event Memory Skills. Applied Cognitive Psychology 18 (6), 745 764. doi:10.1002/acp.1030 Leopold, A., Krueger, F., dal Monte, O., Pardini, M., Pulas ki, S. J., Solomon, J., & Grafman, J. (2012). Damage to the left ventromedial prefrontal cortex impacts affective theory of mind. Social Cognitive And Affective Neuroscience 7 (8), 871 880. doi:10.1093/scan/nsr071 Mesman, J., van IJzendoorn, M. H., & Baker mans Kranenburg, M. J. (2009). The many faces of the Still Face Paradigm: A review and meta analysis. Developmental Review 29 (2), 120 162. doi:10.1016/j.dr.2009.02.001 Miller, C. (2006). Developmental relationships between language and Theory of Mind. Am erican Journal of Speech Language Pathology 15 142 154. Miller, P. H., Kessel, F. S., & Flavell, J. H. (1970). Thinking about people thinking about people thinking about . : A study of social cognitive development. Child Development 41 (3), 613 623. Milligan, K., Astington, J, & Dack, L. (2007). Language and theory of mind: Meta Analysis of the relation between language ability and false belief understanding. Child Development, 78 622 646. Nelson, K. (2005). Language pathways into the community of m inds. In J. Astington & J. Baird (Eds.), Why Language Matters for Theory of Mind (pp. 26 49). New York: Oxford University Press. Nelson, P., Adamson, L. B., & Bakeman, R. (2008). Toddlers' joint engagement experience facilitates preschoolers' acquisition of theory of mind. Developmental Science 11 (6), 847 852. doi:10.1111/j.1467 7687.2008.00733.x Nelson, K., & Fivush, R. (2004). The Emergence of Autobiographical Memory: A Social Cultural Developmental Theory. Psychological Review 111 (2), 486 511. doi:10. 1037/0033 295X.111.2.486 Newton, A. M., & de Villiers, J. G. (2007). Thinking while talking: Adults fail nonverbal false belief reasoning. Psychological Science 18 (7), 574 579. doi:10.1111/j.1467 9280.2007.01942.x Onishi, K. H., & Baillargeon, R. (2005). Do 15 month old infants understand false beliefs? Science, 308, 255 258. Perner, J., & Lang, B. (2000). Theory of mind and executive function: Is there a developmental relationship?. In S. Baron Cohen, H. Tager Flusberg, D. J. Cohen (Eds.), Understanding o ther minds: Perspectives from developmental cognitive neuroscience (2nd ed.) (pp. 150 181). New York, NY US: Oxford University Press.
92 Pohl, R. F., & Hell, W. (1996). No reduction in hindsight bias after complete information and repeated testing. Organizat ional Behavior And Human Decision Processes 67 (1), 49 58. doi:10.1006/obhd.1996.0064 Qureshi, A. W., Apperly, I. A., & Samson, D. (2010). Executive function is necessary for perspective selection, not level 1 visual perspective calculation: Evidence from a dual task study of adults. Cognition 117 (2), 230 236. doi:10.1016/j.cognition.2010.08.003 Reese, E. (2002). Social factors in the development of autobiographical memory: The state of the art. Social Development 11 (1), 124 142. doi:10.1111/1467 9507.001 90 Reese, E., & Fivush, R. (1993). Parental styles of talking about the past. Developmental Psychology 29 (3), 596 606. doi:10.1037/0012 16220.127.116.116 Reese, E., Haden, C. A., & Fivush, R. (1993). Mother child conversations about the past: Relationships of style and memory over time. Cognitive Development 8 (4), 403 430. doi:10.1016/S0885 2014(05)80002 4 bias in social perception and attribution processes. Journal of Exper imental Social Psychology, 13 279 301. Ruffman, T., Slade, L., Rowlandson, K., Rumsey, C., & Garnham, A. (2003). How language relates to belief, desire, and emotion understanding. Cognitive Development, 18 139 158. Sabbagh, M. A., & Seamans, E. L. (2008 ). Intergenerational transmission of theory of mind. Developmental Science, 11 354 360. Samson, D., & Apperly, I. A. (2010). There is more to mind reading than having theory of mind concepts: New directions in theory of mind research. Infant And Child Dev elopment 19 (5), 443 454. Samson, D., Apperly, I. A., Kathirgamanathan, U., & Humphreys, G. W. (2005). Seeing it my way: A case of a selective deficit in inhibiting self perspective. Brain: A Journal Of Neurology 128 (5), 1102 1111. doi:10.1093/brain/awh4 64 Slade, L., & Ruffman, T. (2005). How language does (and does not) relate to theory of mind: A longitudinal study of syntax, semantics, working memory, and false belief. British Journal of Developmental Psychology, 23 141. Slaughter, V., Dennis, M. J., & Pritchard, M. (2002). Theory of mind and peer acceptance in preschool children. British Journal Of Developmental Psychology 20 (4), 545 564. doi:10.1348/026151002760390945
93 Spreen, O., & Strauss, E. (1998). A compendium of neuropsychological tests: Admini stration, norms, and commentary (2nd ed.) New York, NY US: Oxford University Press. Stone, V. E., Baron Cohen, S., & Knight, R. T. (1998). Frontal lobe contributions to theory of mind. Journal of Cognitive Neuroscience, 10, 640 656. Taylor, M., & Carlson, S. M. (1997). The relation between individual differences in fantasy and theory of mind. Child Development 68 (3), 436 455. doi:10.2307/1131670 Tomasello, M. (1998). Uniquely primate, uniquely human. Developmental Science 1 (1), 1 16. doi:10.1111/1467 7687.00002 Tomasello, M., & Carpenter, M. (2007). Shared intentionality. Developmental Science 10 (1), 121 125. doi:10.1111/j.1467 7687.2007.00573.x Tomasello, M., Kruger, A., & Ratner, H. (1993). Cultural learning. Behavioral and Brain Sc iences, 16, 495 522. narratives with children with specific language impairment. Journal Of Communication Disorders 44 (1), 1 22. doi:10.1016/j.jcomdis.2010.06.002 Vygotsky, L. (1 986). Thought and Language. Cambridge: MIT Press. Welch Ross, M. (1995). An integrative model of the development of autobiographical memory, Developmental Review, 15 338 365. Welch Ross, M. K. (1997). Mother child participation in conversation about the p ast: Relationships to preschoolers' theory of mind. Developmental Psychology 33 (4), 618 629. doi:10.1037/0012 1618.104.22.1688 Welch Ross, M. K. (1999). Preschoolers' understanding of mind: Implications for suggestibility. Cognitive Development 14 (1), 101 1 31. doi:10.1016/S0885 2014(99)80020 3 Wellman, H. M., Cross, D., & Watson, J., (2001). Meta analysis of theory of mind development: The truth about false belief. Child Development 72, 655 684. Wolfson, S. (2000). Students' estimates of the prevalence of d rug use: Evidence for a false consensus effect. Psychology Of Addictive Behaviors 14 (3), 295 298. doi:10.1037/0893 164X.14.3.295
94 BIOGRAPHICAL SKETCH Mary Ellen McKay, called Molly, received a Bachelor of Science degree in p sychology from the University of Florida in May 2000. Molly worked in a variety of applied psychological fields, including counseling inpatient child and adolescent psychiatric clients and helping young people in an early delinquency intervention program. Following these exp eriences, Molly returned to graduate school in 2002 to pursue an academic career in d evelopmental p sychology. During her tenure in the d evelopmental p sychology program at the University of Florida, Molly devoted her time to teaching, research, and gaining experience in applied educational and intervention setti ngs. Molly received her Master of Science degree in 2009. Molly worked in the Center for Informal Science Education implementing a grant funded science and literacy curriculum, and worked for Nemours Jacksonville in research and evaluation of an early literacy intervention program. This rounded understanding of the field of d evelopmental p sychology. Molly received her Ph.D from the University of Florida in the s ummer of 2013.