In two studies, we examined 99 3- to 5-year-old American children’s concepts of age, aging, and birthdays. Previous research has shown that preschool-age children understand that all living beings grow, and that growth is a biological process. Humans, however, are distinct from other living things in that we attach great significance to the aging process. Specifically, in Western cultures, we have an annual ritual to celebrate the day we were born—the birthday party. Thus, although the biological mechanism of aging is continuous and invisible, it is marked by discrete yearly celebrations. We examine the proposal that, in part due to its salience as a cultural ritual, children interpret the birthday party as playing a causal role in the aging process. Results indicate that young children understand certain important biological aspects of the aging process but exhibit confusion regarding others, including the causal role of the annual birthday party.

“My mother doesn’t have no more birthdays,” Fredrick tells me in school one day. “Do you mean she doesn’t have a birthday party?” “No. She really doesn’t have a birthday. How I know is no one comes to her birthday and also she doesn’t make the cake.” (p. 3) Paley, V. G. (1991). Bad guys don’t have birthdays

Consider the following situation relayed by a colleague: The colleague’s nephew, Gino, was to turn six on August 6. On the day of his birthday, he and his family were out of the country, so instead of a full-fledged party they simply had a cake, he blew out the candles, and they promised him a party when they returned home. Upon their return home, the family began planning the party and discussing the details. At one point in the conversation, Jino turned to his parents and announced that, since he had already turned 6 on August 6, he would be turning 7 at his party. As with the child in the book by Paley (1991) quoted earlier, Jino clearly seemed to believe that birthday parties play a causal role in the aging process. The present set of studies seeks to examine the presence of this potential misconception.

The concept of a birthday, essentially defined as the anniversary of one’s birth, is intimately linked to an understanding of age, aging, and growth. These concepts are considered central components of children’s and adults’ naïve biological theories (Carey, 1985; Inagaki & Hatano, 1987; Keil, 1989, 1992; Wellman & Gelman, 1992). Researchers have documented clear development in children’s understanding of various facets of naïve biological understanding during the preschool years. For example, Rosengren, Gelman, Kalish, and McCormick (1991) investigated young children’s understanding of biological growth. They report that even children as young as 3 understand that animals, but not artifacts, grow, and that they grow bigger but not smaller. By age 4, children know that both plants and animals grow and that different factors affect their growth (Hickling & Gelman, 1995). Inagaki and Hatano (1987) have shown specifically that 5- and 6-year-old children understand that living things grow because they take in food (a biological cause) and not because they want to grow (a psychological cause).

The practice of celebrating the anniversary of one’s birth, however, is a uniquely cultural one. Cultural rituals such as baptisms, weddings, and birthday parties mark important transitions in our lives. Thus, the birthday party presents a unique opportunity to examine the potential interaction of biological and cultural knowledge. At a birthday party, there are various markings of the child’s change in age, including the number of candles on the birthday cake, the birthday cards reading, “now you are 5!,” and the verbal congratulations. Although many biological processes that result in growth and aging are continuous and invisible, these cultural markers are discrete and experientially salient. Research has shown that both children and adults are especially compelled to seek causal explanations for events that are personally meaningful (e.g., Hickling & Wellman, 2001; Legare, Evans, Rosengren, & Harris, 2012; Woolley, Cornelius, & Lacy, 2011). Might the more tangible birthday party event be inferred as playing a causal role in aging, especially given the intangible nature of the underlying biological processes? Specifically, might children erroneously assume that the tangible observables of the birthday party are what result in them suddenly (or so it may seem to them) being a year older on that special day? To the extent that children exhibit this misconception, research on this question will inform our understanding of children’s causal reasoning.

Research by Springer and Keil (1989, 1991) provides evidence that even preschool children differentiate between causal mechanisms appropriate to biological and nonbiological kinds, suggesting that this sort of misconception would be unlikely. For example, preschoolers in their study preferred natural mechanisms when reasoning about the color of a plant but understood that human intention determines the color of artifacts. However, research by Klavir and Leiser (2002) suggests that indeed children may view birthday parties themselves as the underlying cause of the aging process. To probe children’s beliefs about the potentially causal role of parties in aging, Klavir and Leiser (2002) posed two scenarios to 4- to 9-year-old Israeli children. In one, investigating whether age could be increased, the mother of a boy who did not have a party tried to make up for it by having multiple parties “again and again until he is older than anyone else” and then asked children to indicate if that was a “good idea.” Among 4-year-olds, 83% responded affirmatively, whereas approximately half of 5-year-olds and a third of the 7-year-olds did. In a second scenario, they investigated whether age could be decreased. Here, they presented a story of an 80-year-old man who hoped to get younger by next celebrating his 79th birthday, the next year his 78th, and so on, and asked whether children thought that was a good idea. Children seemed more reluctant to agree that this was a “good idea,” with only 38% of 4-year-olds and 36% of 5-year-olds making such a claim. They conclude from these findings that not until age 7 do most children believe that it is impossible to increase age by having additional birthday parties.

The understanding that aging occurs independently of birthday parties is predicated on basic biological knowledge that aging is an irreversible process and occurs independently of other forces more generally. For example, one cannot get younger or even remain the same age simply by desiring it to be so. Although research on biological understanding indicates that older preschool-age children understand that the forces that cause growth are primarily biological (Inagaki & Hatano, 1987; Hatano & Inagaki, 1994), it is unclear whether children believe that desire or other forces could have an effect on the aging process. The children in Klavir and Leiser’s (2002) study appeared to believe that having parties could affect the aging process, but it is not clear how widespread this belief is or whether children also think other nonbiological factors might play a role.

In the present research, we probed children’s understanding of the inevitability of aging and the potential role of birthday parties in the aging process. Our methods primarily involved presenting children with stories about other children who had one party, no party, or multiple birthday parties, and asking them to indicate how old the children would be. With this method, we sought to circumvent two possible concerns in the methods of Klavir and Leiser (2002): (a) By telling children that the parent in the story had multiple parties in order to make the child older, these researchers may have suggested to children that parties cause aging, and (b) in responding that it was a “good (or bad) idea,” children may have been evaluating the idea of having multiple parties, the idea of getting older, or the intended relation between having parties and getting older. Our methods seek to disambiguate these possibilities.

A secondary interest was children’s knowledge of their own age and birth date. When people are asked to describe themselves, age is one of the most common attributes volunteered, in addition to sex and race. Often, when an adult meets a child for the first time, a query about the child’s name is followed immediately by one about his or her age. By age 3, children are able to label themselves as a boy or a girl (Thompson, 1975), and although there is great variability in children’s labeling of their own race, it is one of the earliest emerging social categories (Hirshfeld, 2008). Arguably, a child’s age would also be an important component of self-identity, yet we know very little about children’s awareness of how old they are. Thus, children’s awareness of their own age and birth date is an additional focus of our research.

In summary, the goal of our research is to probe children’s understanding of the nature of the aging process, generally, and their concept of the birthday party more specifically. Research on children’s naïve biology suggests that even 3-year-olds understand some of the basic concepts underlying age, aging, and birthdays, for example, that living things grow bigger not smaller. However, other potentially relevant aspects of biological growth, such as the role of intention, continue to develop between the ages of three and five. Thus, an understanding of the role of intention and other factors in aging may also develop significantly during this age range. A primary focus of our research is to explore young children’s beliefs about the birthday party: Do they understand it similarly to adults as a cultural marker of aging, or do they erroneously attribute to it a causal role in the aging process? We sought to extend previous research on these concepts (Klavir & Leiser, 2002) in three primary ways: (a) by using a more refined and expanded methodology, (b) by probing children’s understanding of the inevitability of aging more broadly, and (c) by examining the role of naïve biological knowledge in concepts of birthdays and aging.

Study 1 Section: Choose Top of page Abstract Study 1 << Study 2 General Discussion References Methods Participants Seventeen 3-year-olds (M = 3.7, range 3.4–3.9, nine girls), sixteen 4-year-olds (M = 4.6; range 3.11–4.10; six girls), and fifteen 5-year olds (M = 5.4, range 4.11–5.8, seven girls) participated. Participants were primarily from middle- or upper-middle class families; 63% of the children were Caucasian, 8% were Hispanic, 19% were Asian, and 10% were African American. Children were recruited from a local child development center. Each child was seen individually for one 15-minute session. Informed consent in the form of a signed parental permission letter was obtained for all children in both studies prior to their participation. Materials Materials were the set of questions and tasks listed in Appendix A, along with various laminated photographs and hand drawn pictures of children and adults. Procedure Children with parental permission were asked if they would like to participate in the study. Participants who consented were tested by a trained experimenter in a quiet hallway adjacent to their classroom. Children were first asked a set of warm-up questions to increase their comfort with the experimenter and with answering questions; these questions included asking the children their age and birth date. Next, children were given a short task to practice holding up their fingers to indicate age and a warm-up story to practice holding up their fingers to answer test questions (Appendix A). Then, children were presented with the Birthday Party Task to assess their beliefs about the causal role of birthday parties. Here, children were told three stories about children who were celebrating their third birthday: The first story was about a child who had no party on his or her birthday, the second was about a child who had two parties, and the third was simply about a child who was turning 3. Children were asked to indicate how old each child was by using their fingers or by responding verbally. Children also received the Aging Process Task, in which their beliefs about the inevitability of aging were probed. Children were given three scenarios in which someone wanted to either stay the same age or get younger and were then asked to indicate whether that was possible or not. Next, they were presented with the ideas that magic and discontinuing birthday parties could affect the aging process and were asked to indicate whether they thought those ideas would work or not work. Finally, we asked children two naïve biology questions focusing on the role of intention in biological processes, specifically whether, if they wanted to, they could stop breathing “for a couple of days” and whether they could stop their heart from beating (a more in-depth set of questions was not possible due to time constraints). Appendix A contains the exact wording of the tasks and questions. Data for all tasks were coded and entered by the authors and were checked by trained research assistants. Results and Discussion Knowledge of their own age and birth date Most children (44 out of 48) answered correctly when asked their age. One 3-year-old reported that he did not know his age and three 3-year-olds said they were 4. As shown in Table 1, children’s knowledge of the month of their birthday increased with age, as did the number of children who knew the exact date of their birthday. A Kruskal–Wallis (KW) test revealed an effect of age on knowledge of the birthday month, χ2(2) = 10.47, p = .005. A post hoc test using Mann–Whitney (MW) tests with Bonferroni correction indicated a significant difference between 3- and 5-year-olds (p = .004). A KW test also revealed a significant effect of age on knowledge of the exact birth date, χ2(2) = 9.23, p = .01, with significant differences between 3- and 4-year-olds (p = .042) and between 3- and 5-year-olds (p = .018). These findings indicate that knowledge of one’s age is present early in the preschool years, and also that knowledge of one’s birth date develops during these years. The high level of awareness of the month and even the exact date of their birthdays testifies to the salience of this concept for young children. Table 1. Performance Accuracy by Age and Task: Study 1. View larger version Beliefs about birthday parties We assessed children’s understanding of the potentially causal role of birthday parties with two tasks, the Birthday Party Task and the Aging Process task. With regard to children’s actual numerical responses on the former control story, 72% of the participants correctly responded that the child would be 3 years old; 16% claimed the child would be 2, and 9% claimed she would be 1. For the no party story, 54% correctly responded that the child would still be 3 years old, with the most common error being to say that the child would remain 2 years old (28%) and 9% claiming she or he would be 4. For the two parties story, 62% responded correctly that the child would be 3 years old; the most common error was to say that the child would in fact be 2 years older (i.e., that she or he would be 4; 18%) and with 10% responding that she would be five. To assess effects of age and story-type, children were given a 1 for each correct response and a 0 for any incorrect response. To check that children were proficient at indicating age by showing the correct number of fingers, the first and second question in each story set simply asked how old the target child was on his or her first and second birthday. Overall, children of all three ages performed close to ceiling on these questions (M = 91% correct). To assess statistically children’s understanding of the influence of parties on aging, we first summed children’s responses to the test questions on the three stories and conducted a univariate analysis of variance (ANOVA) to assess effects of age on correct responses. This analysis revealed a significant effect of age, F(2, 47) = 7.58, p = .001, with both 4- (M = 2.13 out of 3, SD = .81) and 5-year-olds (M = 2.27, SD = .70) performing better than 3-year-olds (M = 1.29) p’s < .009, by post hoc multiple comparisons with Bonferroni correction. Figure 1 indicates that 4- and 5-year-olds appeared to outperform 3-year-olds on both the control story and the no party story, but that all age-groups performed similarly on the two party story. We used a logistic regression to compare statistically children’s responses on the three types of story by age. This analysis revealed significant effects of both age, χ2(2) = 14.82, p = .001 (as reflected in the ANOVA earlier), and story type, χ2(2) = 12.11, p < .001, but the expected interaction between age and story type was not significant. The story type effect reflected more children overall performing correctly on the control story (81% correct) than on either the no party (54% correct) or the two-party (52% correct) stories, p = .007 and p = .006, respectively, by Wilcoxon signed rank tests. Download Open in new tab Download in PowerPoint We also assessed beliefs about the potentially causal role of parties as part of the Aging Process Task. Here, an adult was presented as wanting to get younger and children were asked if ceasing to have birthday parties could help in achieving that goal (see “Mrs. Johnson” in Appendix A for exact wording). Results indicated development in the understanding that ceasing to have birthday parties would not cause someone to get younger; 63% of 3-year-olds demonstrated this understanding, 87% of 4-year-olds, and 100% of 5-year-olds. KW, χ2(2) = 6.14, p = .046, and post hoc (MW) tests indicated that 5-year-olds were significantly different from 3-year-olds, p = .046. Scores of both 4- and 5-year-olds were significantly different from chance (p < .001 by binomial test). We also asked whether this same adult could remain the same age by ceasing to have parties. Interestingly, it was more difficult for children to understand that stopping birthday parties would not allow someone to remain the same age; here, children were only 63% correct overall and there were no age differences. No age-group’s responses differed significantly from chance. Beliefs about other potential causes of aging Finally, also as part of the Aging Process Task, we addressed children’s beliefs about the role of two other potential causes of aging—desire and magic. Children were first asked whether a 3-year-old child who wanted to stay 3 forever could do so. Overall, 88% of children were correct on this question, with a KW test indicating no significant age differences (responses of all age-groups were significantly different from chance, p < .05). When asked whether the child had to turn 4 “even if she doesn’t want to,” children were 94% correct, again with no age differences (all age-groups were significantly different from chance, p < .004). Thus, children demonstrated an understanding that children could not affect the aging process via their desires. Performance when asked about adults changing their age was more variable. When asked whether an adult’s desires could cause her to remain the same age (“Can Ms. Jamison stay the same age and not get older?”), children overall were 81% correct, with no significant age differences (responses of 3- and 5-year-olds were significantly different from chance). When asked this question using a slightly different phrasing (“Does Mrs. Jamison have to get older even if she doesn’t want to?”), children were 90% correct, with no age differences and all responses significantly different from chance. However, children did not perform as well when asked if an adult could get younger if she wanted to. Here, 3-year-olds were only 29% correct, whereas 4-year-olds (69%) and 5-year-olds (87%) performed fairly well, KW χ2(2) = 11.38, p = .003. Post hoc tests (Mann–Whitney with Bonferroni correction) revealed a significant difference between 3- and 5-year-olds (p = .003) and a trend toward a significant difference between 3- and 4-year-olds (p = .067). Only 5-year-olds responded correctly at a rate that was significantly different from chance, p = .007. Three-year-olds’ explanations were examined to explore why they seemed to better understand that adults cannot stay the same age than that they cannot get younger. Many 3-year-olds performed correctly on the former task and explained their answer by reference to change, food, desire, and growth. These same children, however, when asked why they thought a person could get younger offered virtually no codable explanations, with the majority either responding with “I don’t know,” or “because,” or “she can.” One possible explanation is that children have considerable familiarity with the concept of growing older and so understand that age change is inevitable, that is, that one generally cannot remain the same age forever. It may be that they had never before considered the process of getting younger, but extrapolated from their understanding that age can change to the possibility of it changing in either direction. A full explanation of this pattern must await further study. When asked whether someone could use magic to get younger, 83% of the children denied that it would work, although there was significant development in this understanding between the ages of 3 and 5, KW χ2(2) = 7.01, p = .03, with 5-year-olds (100%) performing significantly better than 3-year-olds (71%), p = .039. Responding at all ages was significantly different from chance, p < .001. It is interesting that the majority of 3-year-olds denied that this would work; despite erroneously believing that adults can get younger, most children clearly do not see magic as an appropriate mechanism. As discussed earlier, a significant number of 3-year-olds and some 4-year-olds appear to believe that ceasing to have birthday parties may affect this process. Results are summarized in Table 1. Naïve biological knowledge As in previous research, basic naïve biological knowledge was correlated with age, r(48) = .47, p < .002. We assessed the effects of naïve biological knowledge on understanding of the inevitability of aging, both with respect to the effect of desire on aging and with respect to the effect of parties on aging. To do this, we first categorized each child as being high in biological knowledge (both questions correct) or as being low in biological knowledge (0 or 1 question correct), and we summed scores on the four questions assessing the potential effects of desire on the aging process (i.e., stories involving children or adults who either wanted to stay the same age or wanted to get younger). We then conducted an analysis of covariance with age as a covariate. Results revealed a significant effect of naïve biological knowledge on understanding of the potential effects of desire on aging, F(1, 47) = 7.82, p = .008, with children with higher biological knowledge (M = 3.67, SD = .70) performing better than children with lower biological knowledge (M = 2.67, SD = .91). We also summed responses on the two questions about whether stopping birthday parties could cause one to get younger or stay the same age. This analysis also revealed a significant effect of biological knowledge, F(1, 47) = 7.92, p = .007, χ2 = .15; children with higher biological knowledge (M = 1.71, SD = .73) performed better than children with lower biological knowledge (M = 1.05, SD = .73). Overall, results of this study revealed some degree of confusion in preschoolers about the role of birthday parties in aging, with many children appearing to afford them a causal role. Children of all ages seemed to believe that not having a birthday party can halt, or possibly even reverse, the aging process and that having multiple parties can speed it up. Children of all ages demonstrated a clear understanding that magic cannot cause someone to remain the same age. They also exhibited understanding that desires cannot affect aging; however, this understanding was much more apparent when reasoning about children versus when reasoning about adults; a significant number of children seemed confused about whether the aging process can be halted or reversed in adults, through desire or through ceasing to have birthday parties. The study also revealed significant awareness in young children of their own age, and significant development in children’s knowledge of their birth date. A second study was conducted to probe further some of the misconceptions found in Study 1 and to explore the role of children’s birthday party experiences in their developing concepts.

Study 2 Section: Choose Top of page Abstract Study 1 Study 2 << General Discussion References Methods Participants Participants were sixteen 3-year-olds (M = 3.5, range 2.10–3.11, 4 girls), nineteen 4-year-olds (M = 4.5, range 4.1–4.11, 6 girls), and sixteen 5-year-olds (M = 5.5, range 5.0–5.11, 9 girls). Participants were primarily from middle- or upper-middle-class families and were recruited from both a local child development center and through a database maintained by the Department of Psychology Children’s Research Center; 78% of the children were Caucasian, 14% were Hispanic, 4% were Asian, 2% were African American, and 2% were of multiple ethnicities. Each child was seen individually for one 15-minute session. Materials and procedure There were three primary changes from Study 1. First, we removed the finger component of the parties task and instead allowed children to simply respond verbally, as they often did spontaneously in Study 1. Second, as part of the initial set of warm-up questions, we asked children why they thought people had birthday parties. Finally, we added a parent questionnaire to explore potential environmental effects on children’s conceptions. We were particularly interested in each child’s personal experience with missed or skipped birthday parties, multiple parties, and attendance at other’s parties. The questionnaire can be found in Appendix B. As in Study 1, participants were first asked a series of warm-up questions to increase their comfort in talking to the experimenter. We then presented the Birthday Party Task, which included a control story in which a child had one party, a story in which a child had no party, and one in which a child had two parties. Children were asked to indicate how old each child would be. Next, children received the Aging Process Task, in which they were told about an adult who wanted to get younger and asked whether she could do that. We included this question in an attempt to replicate the difficulties children had with it in Study 1. Because there was significant development both in children’s understanding that people cannot get younger simply via desire, and in their understanding of the role of parties, we also added a question about whether an adult could get younger if she stopped having parties. Finally, children were given two questions aimed at tapping their understanding of the role of intention in biological processes: (a) whether, if they wanted to, they could stop breathing for a couple days and (b) whether, if they wanted to, they could stop their heart from beating. Results and Discussion Children’s knowledge of their own age and birth date Results were similar to those in Study 1. All children (n = 51) correctly answered when asked their age. Children’s knowledge of the month of their birthdays increased with age from 44% in 3-year-olds to 79% and 73% in 4- and 5-year-olds. A KW test revealed an effect of age on knowledge of the birthday month, χ2(2) = 10.47, p = .005. A post hoc test using MW tests with Bonferroni correction indicated a significant difference between 3- and 5-year-olds (p = .004). The number of children who knew the exact date of their birthdays also increased with age, with 18% of 3-year-olds, 47% of 4-year-olds, and 56% of 5-year-olds able to report the correct date. A KW test also revealed a significant effect of age on knowledge of the exact birthday date, χ2(2) = 9.23, p = .01, with significant differences between 3- and 4-year-olds (p = .042) and between 3- and 5-year-olds (p = .018). Role of birthday parties One of the warm-up questions assessed children’s beliefs about why people have birthday parties. Children’s explanations were first examined for the purpose of developing coding categories. Responses fell into five main categories: (a) fun or enjoyment (e.g., “because they like them”), (b) norm or obligation (e.g., “because they have to”), (c) to celebrate age or birth (e.g., “to celebrate the day they were born”), (d) because they cause aging (e.g., “because that’s how they get older”), and (e) do not know or uncodable. Table 2 shows three trends in children’s responding: (a) Explanations that people have birthday parties simply for fun appeared to decrease with age, (b) claims that people have birthday parties to celebrate their age or the day of their birth appeared to increase with age, and (c) claims that birthday parties are held to cause a change in one’s age also appeared to increase with age. Because such a large number of 3- and 4-year-olds offered no codable responses to this question, we did not analyze these patterns further, but the data do indicate the presence of the misconception that parties cause aging. Table 2. Children’s Explanations for Why People Have Birthday Parties. View larger version With regard to children’s actual numerical responses on the Birthday Party task, on the control story, 77% of the participants correctly responded that the child would be 3 years old, 13% claimed she would be 4, and 6% claimed the child would be 2. For the no party story, 57% correctly responded that the child would still be 3 years old, with the most common error being to say that the child would remain 2 years old (22%), 8% claimed she or he would be 4, and 8% claimed she or he would be 1. For the two parties story, 60% responded correctly that the child would be 3 years old; the most common error was to say that the child would in fact be 2 years older (i.e., that she or he would be 4; 20%); 8% responded that she or he would be 2 and 4% responded that she would be 5. To analyze responding on the Birthday Party task, children were assigned a score of 1 for each correct response (3 was the correct answer for all three tasks) and 0 for incorrect responses (any other answer). In our first analysis, we summed responses to the three party questions and conducted a univariate ANOVA assessing effects of age. This analysis revealed a significant main effect of age, F(2, 48) = 7.58, p = .001, η p 2 = .26. Post hoc tests (Bonferroni, p < .009) revealed that both 5-year-olds (M = 2.67, SD = .70) and 4-year-olds (M = 2.13, SD = .80) performed better than 3-year-olds (M = 1.13, SD = .77). As shown in Figure 2, mean response patterns appear to differ for the three types of story. Comparisons to chance indicated that on the control (one party) story, both the 4- (p = .001) and 5-year-olds (p = .007) performed at rates significantly different from chance. In contrast, on the no party story, no age-group performed significantly differently from chance. On the two party story, only the responses of 5-year-olds (p = .035) were significantly different from chance. We used a logistic regression to compare statistically children’s responses on the control (one party), no party, and two party stories by age. This analysis revealed a significant effect of age, χ2(2) = 30.11, p < .001, and a marginally significant effect of story type, χ2(2) = 5.25, p = .072, but no interaction. Download Open in new tab Download in PowerPoint As in Study 1, we further addressed children’s beliefs about the inevitability of aging with the Aging Process Task (see Appendix A). However, this time, we limited our probes to the two questions: (a) Whether an adult could get younger if she or he wanted to, and (b) whether an adult could stay the same age by ceasing to have birthday parties. Regarding whether an adult could get younger if she wanted to, responses were similar to Study 1 with 25% of 3-year-olds, 58% of 4-year-olds, and 75% of 5-year-olds responding correctly that desires cannot affect aging, χ2(2) = 8.16, p = .017, with 3-year-olds differing significantly from 5-year-olds, p = .015. No age-group’s responses differed significantly from chance, however. Likewise responses to the question about whether an adult could stay the same age if she or he stopped having birthday parties were similar to those of Study 1, with 38%, 53%, and 69% of 3-, 4-, and 5-year-olds responding correctly that ceasing to have parties does not cause one to remain the same age; age differences were not significant and no age-group’s responses differed significantly from chance. Naïve biological knowledge Unlike in Study 1, there was no significant age effect with regard to biological knowledge. Thus, we conducted a univariate ANOVA to assess effects of biological knowledge (high vs. low) on children’s understanding of the inevitability of aging. In this study, the two inevitability questions concerned an adult who simply wanted to get younger and an adult who wanted to stay the same age by not having any more birthday parties. Responses to these two questions were scored as 1 for correct and 0 for incorrect and then summed. This analysis revealed a significant main effect of biological knowledge, F(1, 50) = 14.15, p < .001, with children with higher biological knowledge (M = 1.38, SD = .73) exhibiting more knowledge of the inevitability of aging than children with lower biological knowledge (M = .64, SE = .66). Parent questionnaire There were no relations between responses on the parent questionnaire and children’s responses to the questions on our tasks.

General Discussion Section: Choose Top of page Abstract Study 1 Study 2 General Discussion << References The present studies yield new information about a concept located at the intersection of three distinct bodies of knowledge: (a) biology, (b) cultural conventions, and (c) social cognition. The concept of the birthday, and associated knowledge about age and aging, is one that has received little attention from researchers in developmental psychology. The research here thus yields novel information that should be of interest to researchers in both cognitive and social development. We will begin by discussing what our research indicates about children’s understanding of the aging process generally and then narrow our focus to what appears to be an intriguing misconception regarding the causal role of parties. We will end with an attempt to present a comprehensive picture of the nature of knowledge in 3-, 4-, and 5-year-old children and some speculations about how this knowledge develops during the preschool years. Previous research on children’s understanding of growth indicated that, by age 5, children understand that things grow because of biological processes, not because of psychological ones (Inagaki & Hatano, 1987). That is, by this age, children understand that living things do not grow simply because they want to. Our findings extend evidence of this understanding to the process of aging. Our research revealed that even 3-year-olds, when reasoning about children, understood that desires do not affect aging. However, when reasoning about adults, children of all ages exhibited some confusion. In particular, most 3-, a significant number of 4-, and even some 5-year-olds seemed to think that an adult could get younger if he or she wanted to. It is not clear what is responsible for this misconception. One possibility is that children might hear adults saying that they want to be younger and mistakenly infer that they are being realistic in their desires. Alternatively, the results from both studies showed that basic naïve biological knowledge is significantly related to understanding the inevitability of aging. Children who understood that they could not stop breathing or stop their heart from beating also understood that one could not reverse the aging process. Thus, it seems more likely that understanding of the irreversibility of the aging process is part of a general biological theory. Because of their children’s greater experience with their own and other children’s birthday parties, this understanding may manifest itself earlier in their reasoning about children versus about adults. These findings thus build upon our knowledge of children’s understanding of biological growth. It is interesting to consider how growth and aging differ: Signs of growth may be more apparent to children, for example, when they see their puppy grow to become an adult dog or when their shoes become too small. The subtleties inherent in this comparison can be made evident by considering the fact that even after the puppy has grown to full adult size, he or she will continue to age. Parents often link eating to growth, and children seem to understand eating as a potential mechanism for growth (Inagaki & Hatano, 1987). Aging, however, as indicated earlier, is much less salient from a physical perspective and also has a numerical concomitant that is independent of growth. Because the biological mechanisms of aging are not well understood even by scientists, most parents probably do not attempt to provide explanations for aging as often as they do for growth. In the absence of easily clear physical, biological, or verbal cues to what causes changes in age, children appear to turn to an event that is directly correlated with cultural reference to the aging process—the birthday party. Previous research on children’s concept of the birthday party by Klavir and Leiser (2002) indicated that until the age of 6 or 7, a significant number of children believe that birthday parties cause aging. Our studies also documented considerable confusion in preschool-age children regarding this concept; many of the young children that we tested demonstrated this misconception. Given children’s propensity to seek unobservable causal mechanisms for events, this attribution of a causal role to a cultural event might be unexpected (Carey, 1985; Gelman, Coley, & Gottfried, 1994; Gopnik & Meltzoff, 1997; Shultz, 1982). On the other hand, we know from other research (e.g., Hickling & Wellman, 2001; Legare et al., 2012; Woolley et al., 2011) that children, like adults, are especially driven to seek explanations for personal, meaningful events. We suggest that the annual experience of the seemingly sudden change from one age to another is an event of great significance and meaning to preschool-age children in particular, and that this salience drives children to seek an explanation for the apparent change. This drive, combined with the ubiquitous human tendency to misattribute causation in the presence of simple co-occurrence (e.g., Hood, 2009; Shermer, 2002), results in the belief that the birthday party causes the age change. Although we documented significant evidence of this misconception, some of our tasks also captured development in our preschool sample. There was significant development between 3 and 4 years of age in both studies in understanding that missing a party does not delay aging. There was also significant development between the ages of 4 and 5, as shown in Study 2, in understanding that having multiple parties does not speed up aging. By the age of 5, a significant number of children understood both of these facets of the relation between birthday parties and aging, although some confusion remained. We reasoned that children who had more experience with missed or multiple birthday parties might be less subject to the misconception that parties play a role in aging. However, the results from our parent questionnaire did not support this hypothesis. Because children with higher levels of general biological knowledge better understood the inevitability of aging, one possible scenario is as follows. As children acquire stronger beliefs about the autonomous natures of basic biological processes, they find it easier to refrain from positing a causal link between parties and aging, even if they lack an understanding of the true cause of their seemingly sudden change in age. One concern that arises from these data is whether the apparent misattribution of a causal role to birthday parties might result from a lack of understanding of the questions or of the tasks we used. One possibility is that children were confused about how many fingers represented each age. This is unlikely because children of all ages performed close to ceiling on the control stories asking them to indicate how old each child was on his or her first, second, and third birthday. An additional concern is the fact that the 3-year-olds performed poorly on the control tasks in both studies. Thus, it is possible that 3-year-olds possess a level of understanding that our tasks did not capture. However, the misconception that birthday parties cause aging was also shown in both 4- and 5-year-olds, and these age-groups performed significantly above chance on the control task. Thus, the misconception about causality is likely to be real. Regardless, future studies should attempt to devise even simpler procedures to be used with these young children. Finally, we also observed significant development in children’s knowledge of their own age and birthday—a potentially important, but understudied, component of self-identity. Anecdotally, a child’s name and age seem prominent candidates for the earliest components of socially communicated autobiographical knowledge in young children. Although there is little research on what aspects of self-knowledge children consider most important, name and age are certainly those that are uniformly solicited from them by adults. Research shows that by 4½ months, infants recognize their own names (Mandel, Jusczyk, & Pisoni, 1995), and most preschoolers know and can produce their names (Zill, Resnick, & McKey, 1999). Our research indicates that most 3-year-olds also know their age. Perhaps underscoring the importance placed on birthday celebrations in this country, the majority of preschool-age children also know in what month they were born. A considerable number of preschool-age children are also able to report the exact date of their birthday. Clearly, one’s date of birth is considered a central component of self-knowledge by adults in our culture, and its importance has not gone unnoticed by the preschool child. In future research, it will be important to probe the role of adults in these conceptions, perhaps exploring cultural variation in the emphasis placed on age in conversations. It is interesting to consider how knowledge of one’s age and birthday are acquired in the broader context of other early acquired self-knowledge, such as sex and race. Of these three, sex has the most clear physical manifestations that children can utilize to self-identify, even if these are not necessarily those used by adults (e.g., hair length and clothing). Although physical characteristics do offer cues to age, the relationship is more complex in two primary ways: (a) age is a continuous (vs. discrete) variable and (b) age changes, whereas sex and race remain constant. However, there may also be similarities between development of self-knowledge of sex and age because both are often verbally marked when adults say things like, “How old are you now? Three! What a big boy you are!” whereas race is not explicitly marked as such (Bigler & Liben, 2006). Supporting this, research indicates that by the age of 2 or 3, children identify themselves and others as being either male or female (Fagot, Leinbach, & Hagan, 1986; Martin & Little, 1990; O’Brien & Huston, 1985; Weinraub et al., 1984), whereas self-identification of race exhibits a more protracted development. In conclusion, age plays a large role in Western society, from inspiring annual celebrations of birthdays to anchoring driving and drinking laws. Yet, developmental psychologists know almost nothing about what young children know and how they think about age, both their own and that of others. From our research, the following picture emerges: Three-year-olds have begun to acquire a small but basic set of age-related concepts; most know their age, about half know the month in which they were born, and many show a beginning understanding that aging occurs independent of nonbiological processes like desire and magic. At the same time, however, many exhibit an intriguing misconception that birthday parties play a causal role in aging. By age 4, significantly more children know their birthday month and date, and they have a stronger understanding of the independence of aging from both desires and birthday parties, although many 4-year-olds still misattribute a causal role to birthday parties in aging. Significant development continues to occur between 4 and 5, with the majority of 5-year-olds knowing their age, month, and even exact birth date, all critical components of self-knowledge. Many 5-year-olds also have a much more mature understanding of the inevitability of aging, including the fact that desire and birthday parties do not cause aging, but some confusion remains. Overall, the findings from this research extend to work on children’s naïve biology and begin to address how children integrate their biological knowledge with their cultural experiences. Our findings also add to our understanding of very young children’s early self-knowledge and pave the way for future research into children’s age-based reasoning about other people.

Appendix A. Study 1 Tasks Birthday Party Task Warm up (picture of baby) This is Kennon. Kennon just had her first birthday. She had a really fun party. How old is she—can you show me how many fingers? Actually/That’s right. She’s one year old. (picture of toddler) Then, when Kennon was 2 years old, she had her second birthday. She had a really fun party that year too. How old is she—can you show me how many fingers? Actually/That’s right. She’s two years old. (picture of preschooler) Then, the next year Kennon had her next birthday. She had a really fun party that year too. Can you show me how old she is now? How many fingers? No party story (picture of baby) This is Avery. When Avery was one year old, she had her first birthday. She had a really fun party. Can you show me how many fingers? Actually/That’s right. She’s one year old. (picture of toddler) Then, when Avery was 2 years old, she had her second birthday. She had a really fun party that year too. Can you show me how many fingers? Actually/That’s right. She’s two years old. (picture of preschooler) Then, the next year Avery had her next birthday. But do you know what happened? That year, her parents did not have enough money to have a birthday party. So she did not have a party. Can you show me how old she is now? Two parties story (picture of baby) This is Elena. When Elena was one year old she had her first birthday. She had a really fun party. Can you show me how many fingers? (picture of toddler) Then, when Elena was 2 years old, she had her second birthday. She had a really fun party that year too. Can you show me how many fingers? (picture of preschooler) Then, the next year Elena had her next birthday. But do you know what happened? Her mom had a really fun birthday party for her at her house on one day. And her grandma had another really fun birthday party for her at the park, the next day. Can you show me how old she is now? Control story (picture of baby) This is Laura. When was one year old, she had her first birthday. She had a really fun party. Can you show me how many fingers? (picture of toddler) Then, when Laura was 2 years old, she had her second birthday. (picture of preschooler) Then, the next year Laura had her next birthday. Can you show me how old she is now? Aging Process Task 1. This is Nicole. [show picture of girl] Nicole is 3 years old. She loves being 3 years old and never wants to get any older. She does not want to turn 4 years old; she wants to stay 3 years old forever. Can Nicole stay three years old and not get older? Does Nicole have to turn 4 years old even if she does not want to? 2. This is Mrs. Jamison. She is a teacher. How old do you think she is? Tomorrow is her birthday. But she does not want to get older. She wants to stay the same age for the rest of her life. Can Mrs. Jamison stay the same age and not get older? Does Mrs. Jamison have to get older even if she does not want to? 3. This is Mrs. Johnson. How old do you think she is? Well, Mrs. Johnson decided that she wants to get younger. Can she do that? How can she do that?/Why can’t she do that? Some other kids I talked to had some ideas about how she could do that. I’ll tell you what they said and you tell me if you think they are good ideas or not good ideas. One child said that she could get younger by drinking magic potion. What do you think—would that work or would that not work? Another child said that if she stopped having birthday parties she could get younger. Would that work? Another child said that she could not get younger. But he said that if she stopped having birthday parties she could stay the same age. Would that work? Naïve Biology Task 1. If you want to, can you stop breathing for a couple of days?

2. If you want to, can you stop your heart from beating?” Appendix B. Parent Questionnaire 1. Do you celebrate birthdays in your family?

2. Please list the children in your family, along with their age, sex, and birthdate (you do not need to list names—if you prefer you may simply write “Child1, Child2,” etc.) The following questions refer to the child that is participating in the birthday study: 3. When do you normally have your child’s birthday party? (e.g., the exact date, the weekend after …)

4. How many times have you had a birthday party for your child? Every year Most years One or two years Never had one 5. Has your child ever had multiple birthday parties (not just dinner celebrations, but actual parties, with cake, presents, decorations, etc.) for the same birthday?

5b. (If yes) How often has this occurred? Once 2–3 times every year 6. Have you ever not held a birthday party for your child?

6a. What was the reason you did not have a party that year?

6b. Was your child aware of the reason?

7. Has your child ever attended a birthday party, besides his or her own? YES, A LOT JUST A FEW TIMES NO 8. (If your child has siblings) Does your child attend his or her sibling’s birthday parties? ALWAYS SOMETIMES NO 9. Does your family hold birthday celebrations for the adults in your family? YES, ALWAYS SOMETIMES NO 10. (If so) Does your child attend these? YES, ALWAYS SOMETIMES NO

Declaration of Conflicting Interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The authors received no financial support for the research, authorship, and/or publication of this article.