Children’s knowledge of human behavior and mental functioning changes greatly from early childhood through adolescence. These changes are manifested by increased sophistication and subtlety in children’s inferences concerning other persons’ mental states and children’s explanations of others’ actions. At the same time, children’s ability to monitor their own cognitive activities improves as well. My research examines metacognitive and social cognitive development from early childhood to adolescence. I have investigated the development of children’s ability to infer another person’s perspective, children's understanding of cognitive activities, and children’s explanations of another person’s actions. I also am interested in the development of cognitive monitoring, especially the contributions of children’s monitoring ability to the acquisition of concepts of cognitive processes. Thus, I have investigated (a) children’s ability to infer another person’s knowledge, beliefs, or visual experience, (b) children’s understanding of cognitive processes such as attention, inference, comprehension, and memory, (c) children’s explanations of interpersonal events, and (d) children’s monitoring of their own cognitive activities. In addition, I also investigate children’s use of gender when reasoning about the characteristics of other people.
Conceptions of Cognitive Processes
Understanding the nature of cognitive functioning is important for children’s academic and social lives. Intuitive knowledge or awareness of cognitive functioning contributes importantly to (a) perspective-taking during social interactions, (b) understanding science, (c) understanding perspectives in narrative text, and (d) monitoring reading and comprehension. Thus, the ability to infer the psychological states of others is crucial for successful social interaction, effective communication, and social reasoning. Furthermore, to understand science, students need to be aware of the reasoning processes involved in testing hypotheses, drawing inferences, and evaluating evidence. Children’s understanding of cognitive activities lays the groundwork for further epistemological development during adolescence and adulthood. Thus, an important goal of my research is to bridge the literature on young children’s theory of mind and adolescents’ naive epistemology by examining developmental changes and processes during middle and late childhood.
My research has investigated the children’s understanding of specific psychological processes, such as perception, attention, memorization, and inference, and children’s knowledge of the properties of cognitive activities. Around 6- or 7-years of age children begin to progress from understanding the direct perceptual origins of mental states to understanding the role of cognitive activity in the construction of knowledge and beliefs (Pillow, 1988, 1995, 2008). For example, preschool children appear to understand perceptual experience as a direct source of knowledge, but during the early elementary school years, children begin to understand how processes of interpretation and inference contribute to beliefs. Furthermore, children increasingly differentiate the properties of specific cognitive activities between 6- and 10-years of age. Studies of children’s understanding of inference, interpretation, attention, and perception are described below.
In a series of studies, I have investigated children's understanding that (a) knowledge can be acquired through inference, (b) valid deductive inferences are distinguished from guesses in terms of both the certainty of the knowledge that they yield and the underlying process of integrating premises, and c) deductive and inductive inference are distinct forms of reasoning that differ in their informational basis and the certainty of their conclusions. An initial study indicated that 4-year-olds do not attribute knowledge of an easily deduced fact to others, but by 6-years of age children do attribute inferential knowledge to others (Pillow, 1999). Furthermore, from 6- to 9-years of age, children increasingly recognize deductive inference, direct perception, and guessing as distinct cognitive activities. That is, they recognize that direct perception and deductive inference yield more certain knowledge than does guessing and they explain deductively acquired knowledge by referring to premise information (Pillow, Hill, Boyce, & Stein, 2000).
Logically necessary deductive inferences also differ from inductive generalizations. Inductive inferences are generalizations that go beyond experience. They often involve inferring a general pattern from experience with a few instances. Conclusions reached through inductive inference vary in likelihood or plausibility, but are not logically necessary. Deductive inferences involve integrating information from two or more premises to reach a novel conclusion based on logical relations among the premises. A deductive inference is valid when the conclusion must be true if the premises are true. In contrast to inductive and deductive inferences, guesses are conjectures that may be arbitrary and are not necessarily based on knowledge. Distinguishing among deduction, induction, and guessing, and also recognizing variations in inductive strength, would represent an important advance beyond the basic understanding that inferences occur. Appreciation of these distinctions could facilitate understanding of knowledge as constructed through diverse processes and the ability to critically evaluate conclusions based on different patterns of evidence and reasoning.
My studies of children’s understanding of guesses have included children from kindergarten through fifth-grade, as well as adults. One set of experiments examined children’s evaluation and explanation of their own reasoning (Pillow & Pearson, 2009; Pillow, Pearson, Hecht, & Bremer, 2010), and another set of explanations examined children’s understanding of another person’s reasoning (Pillow & Pearson, 2012). Beginning in kindergarten, children rate their own deductions as more certain than weak inductions or guesses. Children rated deductions as more certain than strong inductions beginning in third-grade, and fourth-grade children and adults differentiated strong inductions, weak inductions, and informed guesses from pure guesses. Fourth-grade children’s and adults’ evaluations of inferences were influenced by the sample of evidence provided and by population base-rate information. Evaluating another person’s reasoning appears to be more challenging. In a series of experiments, children from first to fifth-grade observed a puppet make strong inductions, weak inductions, and guesses. Children either had no information about the correctness of the puppet’s conclusion, knew that the puppet was correct, or knew that the puppet was incorrect. Children ages 5- to 10-years rated the puppet as more certain about correct than incorrect statements. Thus, rather than evaluate the puppet’s reasoning, they appeared to evaluate the puppet’s conclusion. Children were more likely to differentiate between inductions and guesses when the puppet made an incorrect statement, but even the oldest children did not differentiate consistently. When assessing another observer’s certainty, children have difficulty inhibiting their own knowledge and focusing on the observer’s evidence and reasoning (Pillow & Pearson, 2012).
Recognizing the fallibility of one’s own beliefs often requires understanding that one may have misinterpreted information. Likewise, accurately inferring another person's belief often requires consideration of both what information is perceptually available to the person and how the person is likely to interpret that information. Two people may witness the same event but interpret it differently due to differences in their prior expectations. Understanding such differences in interpretation is another instance of understanding a mediated relation in the origin of beliefs. Understanding how differences in interpretation occur is an important social cognitive achievement because such differences may result in misunderstandings and conflicts. Recognizing the interpretive origins of beliefs also facilitates appreciation that different conclusions may be drawn from the same evidence.
In an initial study, I found that 6- and 8-year-olds understood that an observer's prior expectations may bias the observer's interpretation of another person's actions, but younger children did not seem to appreciate the relevance of prior expectations for determining the observer's belief (Pillow, 1991; Pillow & Weed, 1995). For example, in one experiment children ages 4 to 8 years were told stories in which one character, the actor, performed an action that could be interpreted in either of two ways (e.g., as taking a toy out of a donation box or putting a toy into the box). Two other characters, the observers, held contrasting biases concerning the actor (one liked the actor, the other did not). When asked what action each observer thought the actor was performing, both 6- and 8-year-olds attributed negative interpretations to negatively biased observers and positive interpretations to positively biased observers. Four-year-olds responded randomly, despite remembering the information in the stories. Because preschool children's errors might have reflected story comprehension difficulties, in subsequent research I have sought converging evidence with tasks that did not require story comprehension (Pillow & Henrichon, 1996; Pillow & Mash, 1999). Children were asked about an observer's interpretation of a simple drawing. Again, 4- and 5-year-olds failed to ascribe to the observer an interpretation consistent with the observer's expectation, but 6-year-olds answered correctly.
As suggested above, to arrive at an accurate assessment of another's mental state, psychological activities often need to be considered. A person's focus of attention is one factor that mediates between perceptual exposure to information and knowledge of that information. Learning about attention might include discovering limitations on the ability to perform certain combinations of tasks concurrently, learning that because attention is selective, one's knowledge of an event or scene may be incomplete or inaccurate, and realizing that different people may attend to different aspects of the same situation. In a study of children's knowledge of attentional capacity limits (Pillow, 1988), I found that 4-year-olds, but not 3-year-olds, expected messages spoken simultaneously to be difficult to understand. By 4 years, children appear to appreciate the difficulty of dividing attention. Nevertheless, they do not yet fully understand the significance of attention for comprehension and knowledge acquisition. A further series of experiments investigated children's understanding of selective attention (Pillow, 1989). For example, when two messages were presented simultaneously, children and adults were asked to predict whether or not they would understand an incidental message while focusing attention on a target message. The results indicated that 6-year-olds, 8-year-olds, and adults all believe that attending to one source reduces the reception of information from other source. In contrast, 4-year-olds, and some 5-year-olds, predicted they would understand the incidental message while focusing attention on the target message. Four-year-olds also overlooked the importance of attention when asked to judge another person's knowledge and understanding.
Understanding perception as a source of knowledge.
Although preschool children appear to have difficulty understanding the mediating role of cognitive activities, they have some understanding of more direct sources of information. Thus, my research on children's understanding of perception as a source of knowledge has demonstrated that children as young as 3-years of age understand that knowledge is related to perceptual experience (Pillow, 1989). When asked whether another observer knew the color of an object hidden in a container, 3- and 4-year-olds attributed knowledge of the object's color to an observer who had looked at into the container and attributed ignorance to an observer who had not looked. In addition, the results of a recent study (Pillow & Weed, 1997) indicate that 3- and 4-year-olds attribute knowledge on the basis of perceptual experience rather than on the basis of a person's age. That is, they realize that a child who has received perceptual experience possesses the knowledge, but an adult who has not received perceptual experience lacks that knowledge. However, further investigation has revealed an interesting limitation on children's early competence at inferring another's knowledge (Pillow, 1993). Preschool children have difficulty understanding that different perceptual modalities provide knowledge of different aspects of the world. Thus, prior to age 4-1/2 or 5 years, children do not realize that a person who has listened to an object knows what sound it makes but not its color or that to discover an object's color one should look at it rather than listen to it. Although preschool children appear to understand that knowledge is acquired through perception, they do not seem to fully understand what type of information is gained through different kinds of perceptual experience.
Understanding cognitive control.
In a recent set of experiments I investigated children’s and adults’ recognition of the distinction between controlled and automatic cognitive processes (Pillow & Pearson, 2015). Controlled processes are deliberate, subject to conscious control, and require effort, whereas automatic processes are spontaneous, outside of conscious control, and do not require deliberate effort. My research has examined 1st-, 3rd-, and 5th-grade children’s and adults’ intuitions about whether cognitive activities involved in the reception of information or the construction of knowledge are controlled or automatic. In one experiment I investigated whether children and adults regard some activities as more effortful and purposeful than others, and in another experiment I investigated whether children and adults regard some activities as having more optional alternative outcomes and being more influenced by wanting than others. Participants engaged in simple cognitive or perceptual tasks, including one activity that should be judged to occur automatically (object recognition), one activity that should be seen as deliberate and effortful (either counting or pretending), and examples of inferential activity, for which intuitions about controllability might vary by age and across individuals. In one experiment, 5th-grade children and adults rated transitive inference and interpretation of ambiguous pictures as more effortful and purposeful than object recognition or deduction by elimination, and first-grade children rated transitive inference as more effortful than deduction by elimination. In another experiment, , third- and fifth-grade children rated it more difficult to arrive at alternative outcomes for object recognition than for pretending, and adults distinguished object recognition and deduction from pretending and interpreting ambiguous pictures. From first-grade to adulthood there were increases in both the consistency of judgments regarding controllability and the number of distinctions made among cognitive activities in terms of level of controllability.
In addition to documenting age-related changes in children’s understanding of cognitive activities, I am interested on children’s ability to monitor their own cognitive enterprises, and use the first-person experience of their own cognitive monitoring as a source learning that can inform understanding of others’ perspectives (Pillow, 2012). Thus, I have examined children’s monitoring of their own interpretative and inferential activities, as well as children’s monitoring comprehension and memory.
Monitoring and perspective-taking.
Recently I have investigated how children use cognitive monitoring to acquire information about cognitive activities. In one set of studies, I have investigated how children's capacity to reflect on their own interpretative activities contributes to children’s ability to infer another person's misinterpretation (Pillow & Mash, 1998; Pillow, Mash, Aloian, & Hill, 2002). When asked to explain why they themselves misinterpreted an ambiguous drawing, even 4-year-olds provide reasonable explanations of their own mistake. However, the type of explanations children give changes between ages 4- and 6-years. Moreover, there appears to be a relationship between the types of explanation children provide for their own misinterpretation and the ability to infer another observer's misinterpretation. Six-year-olds are more likely to mention previous experiences to explain why they misinterpreted a picture, and children who mention previous experiences to explain their own mistakes are more likely to predict that another observer will misinterpret an ambiguous picture under similar circumstances (Pillow & Mash, 1998). The results of a recent training study suggest that practice explaining their own errors of interpretation improves children’s performance on a perspective-taking tasks requiring them to infer another person’s interpretation of an ambiguous picture (Pillow, Mash, Aloian, & Hill, 2002). In a study exploring children’s ability to learn about inferential activities through monitoring their own certainty, I have found that improvement in children’s memory for their own feelings of certainty is associated with the development of children’s recognition that another person’s deductions are more certain than guesses (Pillow & Anderson, 2006). This result suggests that children’s monitoring of their own experience contributes to children’s understanding of other perspectives.
Monitoring and Study Strategies: Memorization and Comprehension.
Children's beliefs about mental functioning may be an important form of self-knowledge. For instance, to select effective learning strategies, children need to identify specific cognitive goals and monitor their progress toward those goals. In a recent series of experiments, I have been investigating the development of children's ability to distinguish between memorizing material and comprehending it. Recognizing comprehension and memorization as distinct goals, learning which strategies are effective for each goal, and being able to monitor one's own comprehension and memory may contribute to skilled learning. In collaboration with Suzanne Lovett of Bowdoin College, I have investigated children’s sensitivity to the distinction between memorization and comprehension. We have assessed this ability in two ways (Lovett & Pillow, 1995, 1996). In some experiments children were given the goal of either memorizing material verbatim or comprehending it (i.e. correctly following block-building instructions) and then were asked to choose a strategy for achieving that goal. Third-graders were more likely than first-graders to select an effective study strategy for either goal. In other experiments, children were given practice with a strategy that allows them to memorize instructions that they cannot comprehend (due to the inclusion of unfamiliar words) or allows them to understand instructions that they have not memorized. Then children were asked to evaluate how easy or difficult it would be to recall the instruction and how easy or difficult it would be to carry it out. Although both first- and third-graders showed some ability to evaluate their own state of memorization or comprehension after using a strategy, often children failed to appreciate that a strategy that facilitated comprehension, would not equally facilitate memorization. Thus, our results suggest a gradual development of the ability to distinguish comprehension and memory.
Explanations of Action: Inferring Psychological Causes of Behavior
Inferring the reasons for other persons’ actions is one of the most important social cognitive tasks facing children and adults. Although causal attribution has been studied extensively in adults, developmental changes in the content of children’s explanations of other persons’ actions have not been investigated thoroughly. Despite evidence that preschool children explain actions in mentalistic terms, there is little evidence concerning subsequent age-related changes in children's explanatory frameworks. The sophistication of children's explanations (e.g. the complexity of the psychological phenomena they refer to in order to explain an action) may increase with age. In an initial study, I investigated 4- to 6-year-olds use of mentalistic concepts such as ignorance, false belief, forgetting, and inattentiveness to explain another person's mistaken actions and mistaken beliefs. Results suggested that whereas 4-year-olds tend to rely on concepts of ignorance, false belief, or lack of perceptual experience to explain mistaken actions, by 5-years of age children begin to refer to cognitive activities, such as attention and forgetting, to explain people's actions (Pillow & Lovett, 1998).
In subsequent studies, I have investigated the development of causal explanations from early childhood to adolescence (Lovett & Pillow, 2010; Pillow, Lovett, & Hill, 2008). These studies focus on children’s explanations of interpersonal actions (e.g., helping, sharing, rejecting a request to play, etc.). Between kindergarten and fourth-grade children develop an increasingly sophisticated understanding of the goals motivating interpersonal actions. For example, older children were more likely to recognize that actions may be intended to affect someone other than the direct recipient of the action, and may be intended to affect one person’s thoughts or feelings about another person. However, use of these more sophisticated types of psychological explanations continued to increase between fourth grade and adulthood. I plan to build on this research by examining variations in the content and complexity of children’s attributions across different relationship contexts, such as parent-child, teacher-child, and peer and friendship relationships
Social Categories and Social Inferences: Reasoning about Gender
Children begin to categorize people according to gender at an early age, and children also learning stereotypes concerning attributes associated with being male or female. Previous studies indicate that preschool children expect two individuals who belong to the same category (such as having the same gender or belonging to the same species) to share many features in common, including characteristics that have not been directly observed, but children do not necessarily expect two individuals who look similar, but belong to different categories, to share common features. In a sequence of experiments I have investigated children’s reasoning about the biological and behavioral characteristics of others. My results indicate that when presented with gender neutral information preschool children do not necessarily expect two individuals of the same gender to possess the same characteristics (Pillow, Pearson, & Allen, 2015). However, when presented with information concerning gender-stereotype behavior, young children often do use gender as a basis for inferring an individual’s biological characteristics. Currently, I am examining the impact of gender stereotypes and gender salience on children’s reasoning. These studies examine whether inducing children to think about gender stereotypes will increase children’s tendency to infer that two individuals of the same gender share both biological and behavioral characteristics. These studies also examine the question of whether making gender salient in the absence of stereotyping will produce the same effect as stereotyping, i.e., increase children’s tendency to focus on gender when making inferences about another person’s characteristics.
Bradford H. Pillow
Inquiries about the graduate Developmental Psychology Program are welcome.
To request additional information about my research or copies of publications, please contact me at firstname.lastname@example.org.
For information about the admission process, please see the Psychology Department homepage.