Contextual Information Shifts Young Children's Understanding of Gender Constancy
Saved in:
| Title: | Contextual Information Shifts Young Children's Understanding of Gender Constancy |
|---|---|
| Language: | English |
| Authors: | Victoria A. Varlack, Benjamin deMayo (ORCID |
| Source: | Journal of Cognition and Development. 2024 25(5):619-642. |
| Availability: | Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals |
| Peer Reviewed: | Y |
| Page Count: | 24 |
| Publication Date: | 2024 |
| Sponsoring Agency: | National Science Foundation (NSF), Office of Multidisciplinary Activities (SMA) National Science Foundation (NSF), Graduate Research Fellowship Program (GRFP) |
| Contract Number: | 1837857 2041463 |
| Document Type: | Journal Articles Reports - Research |
| Descriptors: | Young Children, Gender Issues, Sexual Identity, Context Effect, Beliefs, Child Behavior |
| DOI: | 10.1080/15248372.2024.2383563 |
| ISSN: | 1524-8372 1532-7647 |
| Abstract: | Children's understanding of the stability of gender over time has long been recognized as a hallmark of early childhood cognitive development. Prior research has argued that until roughly age 6, children do not understand that a person's gender identity remains consistent if the person undergoes a change in gendered behavior or appearance. However, this prior work has generally not given children any context as to why the behavior or appearance change is occurring. The present study showed U.S. three-to-five-year-olds a character (e.g. "Johnny") undergo such a change that occurred because of external factors (e.g. "Johnny is wearing his sister's dress because all his clothes are dirty"); we asked whether children would indicate that the character's gender was the same pre- and post-change. Children (N = 124, 52% girls, 66% non-Hispanic White) as young as 3 demonstrated an understanding of gender consistency and were significantly more likely to endorse gender consistency when they were, versus were not, given context for the change (p < 0.001, Cohen's d = 0.88). Results suggest that U.S. children take context into account when conceptualizing others' genders and believe that gender is generally consistent across transformations in behavior or appearance by age 3. We discuss the implications of these findings in light of a broader literature about children's understanding of gender constancy as well as contemporary understandings of gender that sometimes accommodate the idea that a person's gender can change. |
| Abstractor: | As Provided |
| Notes: | https://osf.io/3r5ye |
| Entry Date: | 2024 |
| Accession Number: | EJ1444843 |
| Database: | ERIC |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwH7gFXkmixF1K_Xy7E3fgysAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDEcTywUrIFY47gVXCQIBEICBmz-Wq10VeZiqUW50xFdUBYS96gTF3RLH6OD_Cu5-qUFwCXhG-4Ai5drYsHNfVKGhJ8VhGF4OfvHvr-QQlu_1dtlp0Tna4FXlsfr_fgQ6PfKFunQRQp__DqsPNSICAxNV580XrLS3LLGScA2Vdzt1rzc9VDRaDNZVMowOkT_wtHw0WD20TGgphVzIkLaETbnFaUuivqWO8hzAlKG4 Text: Availability: 1 Value: <anid>AN0180329822;7m701oct.24;2024Oct21.05:28;v2.2.500</anid> <title id="AN0180329822-1">Contextual Information Shifts Young Children's Understanding of Gender Constancy </title> <p>Children's understanding of the stability of gender over time has long been recognized as a hallmark of early childhood cognitive development. Prior research has argued that until roughly age 6, children do not understand that a person's gender identity remains consistent if the person undergoes a change in gendered behavior or appearance. However, this prior work has generally not given children any context as to why the behavior or appearance change is occurring. The present study showed U.S. three-to-five-year-olds a character (e.g. "Johnny") undergo such a change that occurred because of external factors (e.g. "Johnny is wearing his sister's dress because all his clothes are dirty"); we asked whether children would indicate that the character's gender was the same pre- and post-change. Children (N = 124, 52% girls, 66% non-Hispanic White) as young as 3 demonstrated an understanding of gender consistency and were significantly more likely to endorse gender consistency when they were, versus were not, given context for the change (p &lt; 0.001, Cohen's d = 0.88). Results suggest that U.S. children take context into account when conceptualizing others' genders and believe that gender is generally consistent across transformations in behavior or appearance by age 3. We discuss the implications of these findings in light of a broader literature about children's understanding of gender constancy as well as contemporary understandings of gender that sometimes accommodate the idea that a person's gender can change.</p> <p>Imagine that a boy named Grayson wears his sister's dress to kindergarten one day. Will his classmates think he is a boy or a girl? Classic literature in our field suggests that, upon seeing Grayson's stereotypically girl-like appearance, young children will believe Grayson is a girl, despite previously regarding him as a boy (e.g., Emmerich [<reflink idref="bib11" id="ref1">11</reflink>]; Kohlberg et al., [<reflink idref="bib24" id="ref2">24</reflink>]). This is believed to occur because young children fail to think of gender as consistent across perceptual changes, such as changes in appearance or behavior.</p> <p>But what happens if young children know <emph>why</emph> Grayson is wearing a dress and that he didn't <emph>choose</emph> to wear the dress? What if they learn that he wore the dress because the other clothes in his house were dirty? If perceptual cues (i.e., what Grayson's clothes look like in the current moment) alone are determining young children's perception of Grayson's gender, extra information should not matter in their judgments. On the other hand, if young children integrate current and prior knowledge about Grayson, such explanations may lead them to recognize that Grayson is still a boy. We test these opposing ideas in this paper.</p> <hd id="AN0180329822-2">Brief history of gender consistency research</hd> <p>Building upon Piagetian stages of cognitive development (Flavell, [<reflink idref="bib13" id="ref3">13</reflink>]), Kohlberg ([<reflink idref="bib24" id="ref4">24</reflink>]) argued that gender development reflects basic cognitive development and thus unfolds in stages (see Table 1 for definitions). His work and subsequent work on gender constancy was primarily conducted in the U.S., Canada, Australia, and Western Europe. We use the term "children" throughout to refer to these children but note that in most cases the evidence is primarily from this specific subset of countries.</p> <p>Table 1. Terms and definitions.</p> <p> <ephtml> &lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;td&gt;Concept&lt;/td&gt;&lt;td&gt;Description&lt;/td&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;Gender Constancy&lt;/td&gt;&lt;td&gt;The understanding that gender is an invariant trait that does not change over time or across situations.&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Gender Identity Stage&lt;/td&gt;&lt;td&gt;The first stage of gender constancy; recognizing the gender of oneself and others.&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Gender Stability Stage&lt;/td&gt;&lt;td&gt;The second stage of gender constancy; believing that gender is stable across time such that boys grow up to be men and girls grow up to be women.&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Gender Consistency Stage&lt;/td&gt;&lt;td&gt;The third and final stage of gender constancy; believing that gender does not change based on superficial attributes such as clothing, hairstyle, or stereotypical behavior (such as playing with particular toys).&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Preconstant&lt;/td&gt;&lt;td&gt;A term describing children who do not yet show full gender constancy understanding.&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Pseudoconstancy&lt;/td&gt;&lt;td&gt;A term describing children who provide responses indicative of gender constancy understanding without the ability to explain those beliefs.&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>The earliest Kohlbergian stage, gender identity, involves being able to label oneself and others as boys/men and girls/women, and was thought to be achieved around 3 years old. The second stage, gender stability, involves understanding that gender is invariant across time (i.e., boys grow into men and girls into women) and was thought to be achieved around 4–5 years of age.</p> <p>The final stage, according to Kohlberg, and the stage about which there has been the most debate, is gender consistency. Gender consistency is thought to reflect an understanding that gender remains stable across changes in state or appearance (e.g., a boy remains a boy if he puts on a dress). In classic work, gender consistency was explicitly connected to other kinds of concrete operational thinking (such as the ability to conserve number or liquid across changes in appearance) and was not thought to occur until 7 or 8 years of age – the same age at which children were thought to move from preoperational to concrete operational reasoning (Flavell, [<reflink idref="bib13" id="ref5">13</reflink>]).</p> <p>Since Kohlberg's theoretical contribution in 1966, many researchers in developmental psychology have attempted to empirically assess children's understanding of gender constancy, with a particularly strong focus on the consistency stage. Doing so has proven to be a methodologically complex endeavor, with different psychologists employing different tasks to probe children's gender constancy understanding and operationalizing gender constancy in varying ways (Bem, [<reflink idref="bib2" id="ref6">2</reflink>]; Eaton et al., [<reflink idref="bib9" id="ref7">9</reflink>]; Emmerich, [<reflink idref="bib10" id="ref8">10</reflink>]; Emmerich et al., [<reflink idref="bib11" id="ref9">11</reflink>]; Kuhn et al., [<reflink idref="bib25" id="ref10">25</reflink>], McConaghy, [<reflink idref="bib32" id="ref11">32</reflink>]; Gouze &amp; Nadelman, [<reflink idref="bib19" id="ref12">19</reflink>]; Marcus &amp; Overton, [<reflink idref="bib30" id="ref13">30</reflink>]; Martin &amp; Halverson, [<reflink idref="bib31" id="ref14">31</reflink>]; Ruble et al., [<reflink idref="bib36" id="ref15">36</reflink>]; Slaby &amp; Frey, [<reflink idref="bib37" id="ref16">37</reflink>]; Szkrybalo &amp; Ruble, [<reflink idref="bib38" id="ref17">38</reflink>]; Wehren &amp; De Lisi, [<reflink idref="bib41" id="ref18">41</reflink>]). Despite this complexity, a few reliable findings have emerged.</p> <p>The first is that children tend to show understanding of gender consistency at later points in development than understanding of stability (e.g., Slaby &amp; Frey, [<reflink idref="bib37" id="ref19">37</reflink>]; McConaghy, [<reflink idref="bib32" id="ref20">32</reflink>]; Szkrybalo &amp; Ruble, [<reflink idref="bib38" id="ref21">38</reflink>]). In other words, children tend to reliably endorse the notion that a boy will remain a boy and grow up to be a man at younger ages than they endorse the notion that a boy will remain a boy if he (for example) puts on a dress. The second is that children tend to report understanding that their own gender will remain constant across the lifespan (stability) and perceptual changes (consistency) developmentally prior to when they will understand these ideas as they apply to other people's gender (e.g., Gouze &amp; Nadelman, [<reflink idref="bib19" id="ref22">19</reflink>]; Eaton &amp; von Bargen, [<reflink idref="bib8" id="ref23">8</reflink>]).</p> <p>The third strand to emerge from this literature – and perhaps the most crucial for the current work – is that children's behavior on gender constancy tasks seems to be sensitive to methodological adjustments. There has been considerable variation in the types of tasks used to assess gender consistency, resulting in variability in children's demonstrated levels of gender constancy. In what has since become a classic paper, Emmerich et al. ([<reflink idref="bib11" id="ref24">11</reflink>]) showed children images of a girl and a boy, which were then "transformed" by changing the image to show the child with new clothing or a new haircut stereotypically associated with the other binary gender (i.e., the girl image would be changed to show her with a masculine haircut). Then, after observing this "perceptual transformation," participants were asked whether the child in the image was (now) a boy or a girl, and why. Only around 20% of seven-year-old children showed high levels of gender consistency understanding, indicating that most children did not consider gender to be a property that was consistent across perceptual changes (Emmerich et al., [<reflink idref="bib11" id="ref25">11</reflink>], [<reflink idref="bib10" id="ref26">10</reflink>]). Other researchers have repeatedly found that children tend to not show gender constancy on this type of task until around ages 7–9 (e.g., De Lisi &amp; Gallagher, [<reflink idref="bib6" id="ref27">6</reflink>]; Marcus &amp; Overton, [<reflink idref="bib30" id="ref28">30</reflink>]; Szkrybalo &amp; Ruble, [<reflink idref="bib38" id="ref29">38</reflink>]; Wehren &amp; De Lisi, [<reflink idref="bib41" id="ref30">41</reflink>]).</p> <p>Despite these replications, some past literature has demonstrated that methodological adjustments to the procedure used in Emmerich et al. ([<reflink idref="bib11" id="ref31">11</reflink>]) can sometimes elicit behavior from children that might indicate greater understanding of gender constancy. For example, on a measure developed by Slaby and Frey ([<reflink idref="bib37" id="ref32">37</reflink>]), children often show all forms of gender constancy (identity, stability, <emph>and</emph> consistency) by around age five, and sometimes earlier. In the consistency portion of this task, children are asked whether their gender would change if they altered their appearance (e.g., clothing or hairstyle) or activity preferences to those stereotypically associated with children of the other binary gender (e.g., asking a girl participant, "If you wore boys' clothes, would you be a boy or a girl?").</p> <p>One major difference between the consistency task used in Slaby and Frey ([<reflink idref="bib37" id="ref33">37</reflink>]) and the one in Emmerich et al. ([<reflink idref="bib11" id="ref34">11</reflink>]) is that Slaby and Frey's task did not show a picture of the target child <emph>after</emph> the perceptual transformation, while Emmerich et al. ([<reflink idref="bib11" id="ref35">11</reflink>]) did. Reflecting contemporaneous research in domains like theory of mind and the conservation of liquid (Brainerd &amp; Hooper, [<reflink idref="bib3" id="ref36">3</reflink>]), Emmerich and others designed their task to specifically require children to overcome the perceptual experience of seeing a person engage in stereotypically cross-gender behavior (e.g., a boy putting on a dress, a girl cutting her hair short, etc.) in order to be classified as truly gender-consistent. In contrast, the Slaby and Frey ([<reflink idref="bib37" id="ref37">37</reflink>]) task does not involve a mismatch between category information and the target's appearance at the time that the child answers the test question.</p> <p>The discrepancy between children's behavior on measures that do include a visual depiction of cross-gender behavior (i.e., a boy putting on a dress) versus those that do not has inspired questions about why these differences emerge. Some authors have speculated that the lack of contextualizing information when showing children images of a person engaging in stereotypically cross-gender behavior leads children to treat the situation in a way that researchers are not intending. For example, in the absence of further clarification, children may believe that a character is engaging in pretend play (Martin &amp; Halverson, [<reflink idref="bib31" id="ref38">31</reflink>]), or they might ignore facts about the biological basis of sex to which they otherwise would attend (Bem, [<reflink idref="bib2" id="ref39">2</reflink>]). In other words, some researchers have questioned whether children misinterpret what the experimenter is asking as something other than "is this kid really a boy or really a girl?"</p> <p>Even though studies have varied in their methods and results, most studies examining gender consistency have converged to suggest a relatively late onset of gender consistency understanding (at or after age 6). The idea that children might view gender as completely determined by someone's appearance is surprising, however, considering evidence from the past several decades indicating that children often view gender as an essential category, meaning that they treat a person's gender as imbued by an unobservable "essence" that is unchangeable, intrinsic, and causally powerful (Gelman, [<reflink idref="bib16" id="ref40">16</reflink>]). For example, children as young as 4 (Taylor, [<reflink idref="bib39" id="ref41">39</reflink>]) or 5 (Taylor et al., [<reflink idref="bib40" id="ref42">40</reflink>]) expect babies to develop the physical features (i.e., for males, to develop a deep voice) and grow into stereotypical gender roles (i.e., can grow up to be a daddy), even if the characters are raised in unusual environments comprised only of members of the other binary gender, suggesting that by age four, children view such stereotypical gender properties as innate, rather than determined by the environment. In addition, by ages 3–5, children treat gender as an objective and natural way to categorize people, rather than as one of many possible ways that has been culturally selected (Diesendruck et al., [<reflink idref="bib7" id="ref43">7</reflink>]; Rhodes, Gelman, &amp; Karuza, [<reflink idref="bib34" id="ref44">34</reflink>]; Taylor et al., [<reflink idref="bib40" id="ref45">40</reflink>]), suggesting that children view gender as a natural kind (in other words, a real distinction that exists in nature, versus one that is created by humans or variable across situations) beginning at least in early childhood.</p> <p>It is difficult to reconcile the ideas that children younger than age 6 view gender as a natural kind that confers stable membership and innate features, as suggested by the essentialism literature, with the idea that children think an individual's gender changes moment-to-moment based on their appearance, as suggested by the gender constancy literature. Indeed, some gender essentialism research has directly shown that children weigh category information more heavily than conflicting perceptual information when making inferences about individual features. Gelman et al. ([<reflink idref="bib17" id="ref46">17</reflink>]) presented children (age 4) with pictures of a stereotypical girl, a stereotypical boy, and a child whose appearance fell somewhere between the two. When this third person was described with a gender label (e.g., girl), children were more likely to say this child shared another property (e.g., has estro in her blood) with the same-gender child (e.g., the other girl) and not the other-gender child (e.g., the boy), indicating that children viewed gender as something that confers nonobvious properties beyond what may be visually apparent.</p> <p>Given these seemingly conflicting findings, the question of whether young children view another person's gender as completely determined by perceptual cues or have a more stable understanding of gender remains unresolved. This issue continues to have theoretical importance for our understanding of conceptual development more generally because it touches upon the broader question of whether children's social concepts go beyond what may be perceptually apparent. Therefore, we address this question in the present study by testing if children override conflicting perceptual information to reveal an understanding of gender as a stable feature of identity if they are given a plausible reason for a surprising perceptual change.</p> <hd id="AN0180329822-3">Current work</hd> <p>In the present work, we introduce a new test of whether young U.S.-based children (ages 3–5) believe that a character's gender is consistent across perceptual changes. Unlike past work, we provide participants with a reason why a character might be engaged in gender nonconforming behavior. According to classic and more recent work, children younger than age 6 tend to perceive a target's gender solely based on outward appearance such as clothing, hairstyle, and highly stereotyped behaviors (De Lisi &amp; Gallagher, [<reflink idref="bib6" id="ref47">6</reflink>]; Emmerich et al., [<reflink idref="bib11" id="ref48">11</reflink>]; Marcus &amp; Overton, [<reflink idref="bib30" id="ref49">30</reflink>]; Szkrybalo &amp; Ruble, [<reflink idref="bib38" id="ref50">38</reflink>]; Wehren &amp; De Lisi, [<reflink idref="bib41" id="ref51">41</reflink>]). In contrast, we predicted that children at these ages would believe that a target's gender has not changed, even when they engage in gender nonconforming behaviors, if a logical (non-identity-based) alternative reason for the gender nonconforming behavior is provided – such as Grayson's wearing a dress because all the other clothing in his house was dirty. This is consistent with suggestions from past work that children in this age range can override perceptual conflict (Gelman et al., [<reflink idref="bib17" id="ref52">17</reflink>]; Martin &amp; Halverson, [<reflink idref="bib31" id="ref53">31</reflink>]) in favor of endorsement of gender consistency.</p> <p>In our task, we not only ask about the perception of a target's gender category, but also a future behavioral choice – such as the outfit the target would wear the next day [inspired by property-type questions from Gelman and Markman ([<reflink idref="bib18" id="ref54">18</reflink>])]. If children believe that Grayson is now a girl, they might assume he will engage in stereotypically feminine behavior (e.g., wear a girls' swimsuit) in the future. If, instead, three-to-five-year-old children understand that gender nonconforming behaviors can be situational, they may predict that Grayson will wear masculine clothes in the future.</p> <p>In the current study, we had participants complete three versions of the consistency task: (<reflink idref="bib1" id="ref55">1</reflink>) one in which participants saw a picture of a person, and were asked what the character's gender would be if they were to engage in stereotypically gender nonconforming behavior, without explicitly illustrating this gender nonconforming behavior (<emph>Untransformed Picture Task;</emph> modeled on Slaby &amp; Frey, [<reflink idref="bib37" id="ref56">37</reflink>]); (<reflink idref="bib2" id="ref57">2</reflink>) one in which participants saw an image of a character engaging in gender-atypical behavior without explaining the transformation (<emph>Transformed Picture Task;</emph> modeled on Emmerich et al., [<reflink idref="bib11" id="ref58">11</reflink>]); and (<reflink idref="bib3" id="ref59">3</reflink>) a new consistency task in which children were shown a character engaging in gender-atypical behavior, but with an explanation provided for this behavior (<emph>Transformed Picture + Explanation Task</emph>).</p> <p>For further comparison to past work, and to generate a contemporary estimate of young children's performance on traditional measures of gender constancy, participants were also asked to complete classic measures of gender identity and gender stability (the two earlier stages of gender constancy), including questions posed from a first- and third-person perspective, as well as a first-person gender consistency task. Because these are not the central measures of interest for this paper, there is less debate about these stages, and our results were fairly consistent with past work on these tasks, we report those measures and analyses in the Supplementary Materials. Additionally, past work has often attempted to categorize children as being in a particular stage of gender constancy (i.e., understanding identity, stability, or consistency); because the stage-like progression of gender constancy is not a central question, we address this in the Supplementary Materials. In this report we focus on the third-party gender consistency tasks.</p> <p>We predicted that children would be more likely to say that a character's gender would be consistent across a change in behavior or appearance on the <emph>Untransformed Picture</emph> task and the <emph>Transformed Picture + Explanation</emph> tasks than on the <emph>Transformed Picture</emph> task; we also predicted that older children would be more likely to say that a character's gender would be consistent than younger children. All hypotheses and analysis plans were preregistered before data collection began (https://osf.io/3r5ye/). Our main preregistered analyses are confirmatory in nature, and we describe in the Results section where we included non-preregistered, exploratory analyses.</p> <hd id="AN0180329822-4">Reporting constancy results</hd> <p>Within the gender constancy literature, children's performance is reported in multiple ways, two of which are especially common and relevant to the present analyses. First, some papers report a continuous measure – the average percentage of constant responses provided at each stage of gender constancy by each participant (e.g., Marcus &amp; Overton, [<reflink idref="bib30" id="ref60">30</reflink>]). Alternatively, some papers emphasize an absolute judgment – each child is categorized as constant (gave only constant responses) or pre-constant (gave at least one non-constant response; Frey &amp; Ruble, [<reflink idref="bib14" id="ref61">14</reflink>]) at each stage (e.g., Bem, [<reflink idref="bib2" id="ref62">2</reflink>]; De Lisi &amp; Gallagher, [<reflink idref="bib6" id="ref63">6</reflink>]; Martin &amp; Halverson, [<reflink idref="bib31" id="ref64">31</reflink>]; Wehren &amp; De Lisi, [<reflink idref="bib41" id="ref65">41</reflink>]). While these two approaches can result in similar findings, there can also be major discrepancies. For example, a child who provided constant responses on five of six consistency questions would be 83% constant on the continuous assessment but "pre-constant" on the absolute assessment.</p> <p>In general, the field of psychology has moved away from all-or-nothing measures like the absolute scoring used in classic work on gender constancy, though some researchers still do dichotomize continuous variables despite evidence that this practice tends to decrease statistical power and precision in estimating effects (DeCoster et al., [<reflink idref="bib5" id="ref66">5</reflink>]). We preregistered both systems of analysis and further discuss the impact of the choice of scoring system in the Discussion section.</p> <p>Additionally, some authors (De Lisi &amp; Gallagher, [<reflink idref="bib6" id="ref67">6</reflink>]; Emmerich, [<reflink idref="bib10" id="ref68">10</reflink>]; Emmerich et al., [<reflink idref="bib11" id="ref69">11</reflink>]; Ruble et al., [<reflink idref="bib36" id="ref70">36</reflink>]; Szkrybalo &amp; Ruble, [<reflink idref="bib38" id="ref71">38</reflink>]; Wehren &amp; De Lisi, [<reflink idref="bib41" id="ref72">41</reflink>]) also require that children be able to provide explicit reasoning that supports the logic of constancy (e.g., "he's still a boy because boys cannot become girls") for an answer to be coded as indicative of gender constancy understanding. This approach to interpreting results is discussed further in the Supplementary Materials.</p> <hd id="AN0180329822-5">Method</hd> <p></p> <hd id="AN0180329822-6">Participants</hd> <p>One hundred twenty-six English-speaking participants were recruited for this study at a university preschool program or from a database of local families. All participants lived in the Northeastern United States. One was excluded due to a developmental disability. To be included in analysis for any particular task, according to our preregistration, participants needed to complete at least 50% of the multiple-choice questions for that task. One participant was excluded because they did not meet this criterion for any task in the study. We therefore include 124 participants: 48 three-year-olds (24 girls, 24 boys), 37 four-year-olds (22 girls, 15 boys), and 39 five-year-olds (18 girls, 21 boys). Of these 124, 114 completed at least 50% of all three third-person consistency measures. Nine completed at least 50% of at least one third-person consistency measure, but not all three. One did not complete at least 50% of any third-person consistency measure – their data are therefore not represented in the main manuscript but is represented in the other tasks in the Supplementary Materials.</p> <p>We pre-registered to recruit 34 participants per age group who had completed at least 50% of each measure in the study (including the gender identity, gender stability, and first-person gender consistency measures described in the Supplementary Materials). Due to scheduling errors, we recruited extra participants in each age group that count toward this preregistered N (total Ns: 36 three-year-olds, 35 four-year-olds, and 39 five-year-olds). In the main text and Supplementary Materials, our analyses include all data. However, we also conducted all analyses in the main manuscript without data from participants recruited after we had met our preregistered N for each age group. We found the same pattern of results, which can be seen in the Supplementary Materials.</p> <hd id="AN0180329822-7">Demographic description and justification</hd> <p>Demographic information of all included participants, including race, socioeconomic status, and parent education can be found in Table 2. As shown in Table 2, our sample was disproportionately composed of non-Hispanic White children from upper-income backgrounds, reflecting the demographic characteristics of the local surroundings of the research university in the Northeastern United States where we conducted the study.</p> <p>Table 2. Participant and parent demographics.</p> <p> <ephtml> &lt;table&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;&lt;bold&gt;Gender (&lt;italic&gt;N&lt;/italic&gt;)&lt;/bold&gt;&lt;/td&gt;&lt;td /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Girl&lt;/td&gt;&lt;td&gt;64 (52%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Boy&lt;/td&gt;&lt;td&gt;60 (48%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;&lt;bold&gt;Race/ethnicity (&lt;italic&gt;N&lt;/italic&gt;)&lt;/bold&gt;&lt;/td&gt;&lt;td /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;White (non-Hispanic)&lt;/td&gt;&lt;td&gt;82 (66%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Asian (non-Hispanic)&lt;/td&gt;&lt;td&gt;16 (13%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Hispanic/Latino (of any race)&lt;/td&gt;&lt;td&gt;12 (10%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Multiracial or Other race (non-Hispanic)&lt;/td&gt;&lt;td&gt;11 (9%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Did not report&lt;/td&gt;&lt;td&gt;3 (2%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;&lt;bold&gt;Parent education (&lt;italic&gt;N&lt;/italic&gt;)&lt;/bold&gt;&lt;/td&gt;&lt;td /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;High school diploma&lt;/td&gt;&lt;td&gt;1 (1%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Some college/Associate's degree&lt;/td&gt;&lt;td&gt;5 (4%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;College/Bachelor's degree&lt;/td&gt;&lt;td&gt;31 (25%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Advanced degree (MA, MD, PhD, etc.)&lt;/td&gt;&lt;td&gt;83 (67%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Did not report&lt;/td&gt;&lt;td&gt;4 (3%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;&lt;bold&gt;Parent annual household income (&lt;italic&gt;N&lt;/italic&gt;)&lt;/bold&gt;&lt;/td&gt;&lt;td /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Less than $25,000/year&lt;/td&gt;&lt;td&gt;1 (1%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;$50,001&amp;#8211;$75,000/year&lt;/td&gt;&lt;td&gt;6 (5%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;$75,001-$125,000/year&lt;/td&gt;&lt;td&gt;29 (23%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Greater than $125,000&lt;/td&gt;&lt;td&gt;81 (65%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Did not report&lt;/td&gt;&lt;td&gt;7 (6%)&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;&lt;bold&gt;Child age (months)&lt;/bold&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;M&lt;/italic&gt; = 52.40, &lt;italic&gt;SD&lt;/italic&gt; = 11.05&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;&lt;bold&gt;Parent political beliefs (1 = very liberal, 7 = very conservative)&lt;sup&gt;a&lt;/sup&gt;&lt;/bold&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;M&lt;/italic&gt; = 2.37, &lt;italic&gt;SD&lt;/italic&gt; = 1.14&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>1 <emph>Note.</emph><sups>a</sups>Thirteen parents did not report their political beliefs.</p> <hd id="AN0180329822-8">Procedure and materials</hd> <p>This study was approved under IRB Protocol #12923 at Princeton University. Starting in September 2021 and ending February 2023, participants were tested in a university testing room, a large atrium at the university, or outside (during various phases of the COVID-19 pandemic), or, for children at a preschool, in a classroom, by one of three researchers. One researcher was a White female, one was a White Hispanic male, and one was a Black female.</p> <p>When testing at the university, the researcher played with the child for a few minutes to make them comfortable with the testing environment. The experimenter obtained consent from each parent present before leading one parent and their child to the testing location. Parents were given instructions not to interfere with the task and sat in a chair a few feet away from where the participant and researcher were seated at a table. Children gave assent before beginning the task and were told they had the option, along with their parents, to skip any question or stop the task at any time. For all participants, the researcher only gave neutral reactions to participant responses (e.g., "Ok."). After the study was over families were compensated with a $10 gift card and an age-appropriate book. For data collection at the preschool, consent was obtained in advance and children were tested in a preschool classroom without a parent present; participants received a book after testing. Participation took approximately 25 minutes.</p> <p>On all three tasks of interest (<emph>Untransformed Picture</emph>, <emph>Transformed Picture</emph>, and <emph>Transformed Picture + Explanation</emph>; see Figure 1 for examples), a child's response was considered to be "constant" if they indicated that the character's gender matched their appearance on the first slide. Participants also completed tasks assessing first- and third-person identity, first- and third-person stability, and first-person consistency (see the Supplementary Materials for more information about additional tasks). Participants answered items in four blocks; all questions within a block were always shown together, but the order of the blocks was randomized. The first block included all items drawn from the Slaby and Frey ([<reflink idref="bib37" id="ref73">37</reflink>]) measure: first-person and third-person identity, first-person stability, and first-person consistency (see Table S1 for examples). For the first-person consistency task, children were always asked to justify their answer. The second block included items drawn from Fast and Olson ([<reflink idref="bib12" id="ref74">12</reflink>]): third-person stability (see Table S1 for examples), and the <emph>Untransformed Picture</emph> Task for third-person consistency (see Figure 1 for examples). The third and fourth blocks contained the two remaining third-person consistency tasks: <emph>Transformed Picture</emph> (see Table 3 for item wording) and <emph>Transformed Picture + Explanation</emph> (see Table 4 for item wording).</p> <p>Graph: Figure 1. Third-person consistency tasks.</p> <p>Table 3. Item wording on <emph>Transformed Picture</emph> task.</p> <p> <ephtml> &lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;td&gt;Item&lt;/td&gt;&lt;td&gt;Wording&lt;/td&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;1&lt;/td&gt;&lt;td&gt;If Janie played with trucks and did boy things, what would she be?&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;2&lt;/td&gt;&lt;td&gt;If Janie puts on boy clothes like this, what would she be?&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;3&lt;/td&gt;&lt;td&gt;If Janie has cut her hair short like this, what would she be?&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;4&lt;/td&gt;&lt;td&gt;If Janie has cut her hair short like this and wears boy clothes like this, what would she be?&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;5&lt;/td&gt;&lt;td&gt;If Johnny played with dolls and did girl things, what would he be?&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;6&lt;/td&gt;&lt;td&gt;If Johnny puts on girl clothes like this, what would he be?&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;7&lt;/td&gt;&lt;td&gt;If Johnny lets his hair grow long like this, what would he be?&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;8&lt;/td&gt;&lt;td&gt;If Johnny lets his hair grow long like this and wears girl clothes like this, what would he be?&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>2 <emph>Note</emph>. All "Janie" items were presented consecutively and all "Johnny" items were presented consecutively; whether "Janie" or "Johnny" came first was randomized. All items were adapted from Emmerich et al. ([<reflink idref="bib11" id="ref75">11</reflink>]).</p> <p>Table 4. Item wording on <emph>Transformed Picture</emph> + <emph>Explanation</emph> task.</p> <p> <ephtml> &lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;td&gt;Item&lt;/td&gt;&lt;td&gt;Character&lt;/td&gt;&lt;td&gt;Question setup&lt;/td&gt;&lt;td&gt;Question wording&lt;/td&gt;&lt;td&gt;Stereotypically boy-like answer choice&lt;/td&gt;&lt;td&gt;Stereotypically girl-like answer choice&lt;/td&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;1&lt;/td&gt;&lt;td&gt;"Liam" (boy)&lt;/td&gt;&lt;td&gt;One day Liam's dad didn't have time to do the laundry and when Liam was getting ready for school, he didn't have any clean clothes! But his sister had an extra clean dress, so Liam wore his sister's dress to school.&lt;/td&gt;&lt;td&gt;While he was at school, Liam's dad did all the laundry and now Liam has clean clothes. &lt;italic&gt;Which pajamas does Liam wear to bed?&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;Pajamas that have pants and an image of an airplane&lt;/td&gt;&lt;td&gt;Pajamas that resemble a nightgown&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;2&lt;/td&gt;&lt;td&gt;"Michael" (boy)&lt;/td&gt;&lt;td&gt;Michael went to his grandma and grandpa's house for the summer. His grandparents live on a farm far away from the city and there is nowhere for Michael to get his hair cut. Michael's hair got long.&lt;/td&gt;&lt;td&gt;When he got home in the fall Michael's parents took him to get his hair cut. &lt;italic&gt;Which hairstyle does Michael want?&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;Crew cut&lt;/td&gt;&lt;td&gt;Shoulder-length haircut&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;3&lt;/td&gt;&lt;td&gt;"David" (boy)&lt;/td&gt;&lt;td&gt;David was at school when he really had to go to the bathroom. When David walked to the boy's bathroom it had a sign on it saying the boy's bathroom was not working. David saw the girls' bathroom was working and empty so he went to the girls' bathroom.&lt;/td&gt;&lt;td&gt;Later, David's teacher told the class all boys should hang their art on the blue wall and all the girls should hang their art on the red wall. &lt;italic&gt;Which wall did David hang his art on?&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;Blue wall showing the boys' art&lt;/td&gt;&lt;td&gt;Red wall showing the girls' art&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;4&lt;/td&gt;&lt;td&gt;"Sarah" (girl)&lt;/td&gt;&lt;td&gt;Sarah wanted to be in her school play. Her teacher said she could be a man called Andrew in her school play. Sarah did the play.&lt;/td&gt;&lt;td&gt;After the play was over, Sarah took off her costume and picked a new outfit to wear. &lt;italic&gt;Which outfit did Sarah pick?&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;Striped polo shirt and cargo shorts&lt;/td&gt;&lt;td&gt;Spaghetti-strap dress&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;5&lt;/td&gt;&lt;td&gt;"Kaylee" (girl)&lt;/td&gt;&lt;td&gt;Kaylee loves baseball and wanted to be on a baseball team. There were no baseball teams for girls, only teams for boys, so Kaylee joined the boys' baseball team.&lt;/td&gt;&lt;td&gt;After she played baseball on the team, Kaylee went swimming with some of her friends. &lt;italic&gt;Which swimsuit did Kaylee wear?&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;Swim trunks&lt;/td&gt;&lt;td&gt;One-piece swimsuit with frilly skirt&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;6&lt;/td&gt;&lt;td&gt;"Julia" (girl)&lt;/td&gt;&lt;td&gt;Julia was chewing some bubble gum when it got stuck in her hair! Her parents couldn't get the gum out of her hair and had to cut her hair off.&lt;/td&gt;&lt;td&gt;The next day at school, Julia's teacher told the boys to line up for art and the girls to line up for music. &lt;italic&gt;Which line did Julia join?&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;Group of three boys&lt;/td&gt;&lt;td&gt;Group of three girls&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>3 <emph>Note</emph>. Items were presented in a randomized order, and participants were asked whether each character was a boy or a girl subsequently and in addition to the questions shown in this table.</p> <hd id="AN0180329822-9">Untransformed picture task</hd> <p>In this task, adapted from Ruble et al. ([<reflink idref="bib36" id="ref76">36</reflink>]), Szkrybalo and Ruble ([<reflink idref="bib38" id="ref77">38</reflink>]), and Slaby and Frey ([<reflink idref="bib37" id="ref78">37</reflink>]), participants saw a photograph of a person and were asked what gender the person in the photograph would be if they engaged in a gender nonconforming behavior, either by wearing gender nonconforming clothes (2 items; e.g., "If this kid wore boy's clothes, would this kid be a boy or a girl?") or doing work that is stereotypically associated with the other binary gender (2 items; e.g., "If this grown-up did the work that women do, would this grown-up be a man or a woman?"). Children were always then asked to justify their judgments, including if they gave a non-constant response (justifications are described in the Supplementary Materials). Critically, participants did not see the images undergo any sort of perceptual transformation in this task.</p> <hd id="AN0180329822-10">Transformed picture task</hd> <p>The second version of the third-party consistency task was based on the Boy–Girl Identity Task (Emmerich et al., [<reflink idref="bib11" id="ref79">11</reflink>]). In this task, at the time children were asked about the target's identity, the picture of the child was transformed so that the initially gender "typical" target now appeared "atypical" (i.e., counter-stereotypical). Children were always then asked to justify their judgments (justifications are described in the Supplementary Materials). In a randomized order, participants were shown two trials, one about a girl named Janie and one about a boy named Johnny. Each trial had four items; one involved change in play preferences, one involved a change in haircut, one involved a change in clothing, and one involved a change in both hair and clothing.</p> <hd id="AN0180329822-11">Transformed picture + explanation task</hd> <p>We created a new consistency task that, unlike past tasks, provided a possible reason why a child would show gender nonconforming behavior. In this task, there was again a transformed picture at the time of the child's judgment, but an explanation for the transformation was provided. Of the six trials, three involved changes in clothing, two involved changes in hair, and one involved using a gendered restroom. Importantly, the explanations provided were situational, not related to identity. Children were asked both to make categorical judgments (e.g., "Is this child a boy or girl?"), as in the <emph>Transformed Picture</emph> task, and to give behavioral predictions (e.g., while showing children pictures of stereotypically masculine swim trunks and a stereotypically feminine two-piece swimsuit, asking, "Which swimsuit will this kid wear in the future?"), which were not included in the <emph>Transformed Picture</emph> task. Children were always then asked to justify their judgments (justifications are described in the Supplementary Materials).</p> <hd id="AN0180329822-12">Results</hd> <p>We conducted analyses in accordance with our preregistration. We also include some analyses in the main text which were not preregistered and label them accordingly when they are introduced in the text. Supplementary Materials include both preregistered and non-preregistered analyses, which are also labeled accordingly. All of our analyses were conducted in R, using the tidyverse, lme4, sjPlot, emmeans, combinat, and rstatix packages (Bates et al., [<reflink idref="bib1" id="ref80">1</reflink>]; Kassambara, [<reflink idref="bib22" id="ref81">22</reflink>]; Lenth, [<reflink idref="bib26" id="ref82">26</reflink>]; Lüdecke et al., [<reflink idref="bib28" id="ref83">28</reflink>]; Wickham et al., [<reflink idref="bib42" id="ref84">42</reflink>]).</p> <hd id="AN0180329822-13">Task scoring</hd> <p>We present two approaches to assessing gender-constant performance on each task which we will refer to as <emph>continuous</emph> and <emph>absolute</emph> assessments (see Reporting Constancy Results for additional background). In the continuous assessment, we calculated the percentage of questions within each individual task (<emph>Untransformed Picture, Transformed Picture</emph>, and <emph>Transformed Picture + Explanation</emph>) on which each child provided constancy responses. For analyses with this assessment, we test whether participants (as a group) gave constancy responses significantly more (or less) than 50% of the time (chance level). In the absolute assessment, we report whether or not a child gave a constant response on all the items on a particular third-person consistency task (<emph>Untransformed Picture</emph>, <emph>Transformed Picture</emph>, and <emph>Transformed Picture + Explanation</emph>). In other words, for each measure, each child was categorized as gender constant (gave constancy responses on every item they answered) or not. When coded this way, we report the percentage of children who were "fully" constant and test whether children were more likely than chance to be gender-constant. In both assessments, we treated every question as its own response, and did not collapse questions and counter-questions together. In both assessments, we did not consider lack of an answer (e.g., a skipped question) to represent non-constancy or constancy; we excluded those items when calculating continuous and absolute scores. We also only analyze participants' data on tasks in which they answered at least 50% of the questions in the task.</p> <p>To assess the impact of age, we use linear (for the continuous assessment) and binomial (for the absolute assessment) regression, with age predicting gender consistency. We center age on the middle of our eligible age range (4.5 years old). When age does not have a significant effect, we interpret the intercept as overall performance across all participants (i.e., significantly more or less gender-consistent than chance). When age does have a significant effect, we use model estimates to determine if there are ages in our age range at which gender consistency changes with respect to chance (e.g., moves from less consistent than chance to chance or from chance to more consistent than chance).</p> <p>We also preregistered and used a multilevel modeling analysis approach, predicting gender constancy on each individual question. As the results from the multi-level models and the continuous DV were the same throughout, we report the multilevel model results in the Supplementary Materials; nonetheless, because developmental psychology is moving toward greater familiarity with multilevel modeling (Muradoglu et al., [<reflink idref="bib33" id="ref85">33</reflink>]), we are hopeful such measures will be useful (and more frequently adopted) in the future.</p> <hd id="AN0180329822-14">Age and third-person consistency: continuous assessment</hd> <p>When using continuous scoring, on the <emph>Untransformed Picture</emph> task participants said that gender would remain consistent significantly more often than expected by chance, and this did not differ by age (see Tables 5 and 6). On the <emph>Transformed Picture</emph> task, participants' responding did not differ from chance, and this not differ by age (see Tables 5 and 6). On the <emph>Transformed Picture + Explanation</emph> task participants reported that gender would remain consistent significantly more than chance, and gender consistency was higher among older children (see Tables 5 and 6). Despite a significant age-related effect, model estimates from the <emph>Transformed Picture + Explanation</emph> task show children across the entire 3–5 age range were more likely than chance to say that the character's gender would stay consistent across a change in activity, hairstyle, or clothing.</p> <p>Table 5. Continuous scores (means and standard errors).</p> <p> <ephtml> &lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;td&gt;Age group&lt;/td&gt;&lt;td&gt;Untransformed Picture&lt;/td&gt;&lt;td&gt;Transformed Picture&lt;/td&gt;&lt;td&gt;Transformed Picture + Explanation&lt;/td&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;3 years old&lt;/td&gt;&lt;td&gt;75.76% [3.53%]&lt;/td&gt;&lt;td&gt;51.28% [5.13%]&lt;/td&gt;&lt;td&gt;69.50% [2.95%]&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;4 years old&lt;/td&gt;&lt;td&gt;68.75% [6.01%]&lt;/td&gt;&lt;td&gt;36.82% [5.85%]&lt;/td&gt;&lt;td&gt;85.42% [2.33%]&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;5 years old&lt;/td&gt;&lt;td&gt;66.67% [6.52%]&lt;/td&gt;&lt;td&gt;48.40% [6.91%]&lt;/td&gt;&lt;td&gt;88.68% [2.61%]&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>4 <emph>Note</emph>. This is the mean percentage of items on which participants provided gender-constant answers (SE in brackets).</p> <p>Table 6. Linear regression – age predicting continuous scores.</p> <p> <ephtml> &lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;td /&gt;&lt;td&gt;Untransformed Picture&lt;/td&gt;&lt;td&gt;Transformed Picture&lt;/td&gt;&lt;td&gt;Transformed Picture + Explanation&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td /&gt;&lt;td&gt;&lt;italic&gt;b&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;t&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;p&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;b&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;t&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;p&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;b&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;t&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;p&lt;/italic&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;Intercept&lt;/td&gt;&lt;td&gt;0.20&lt;/td&gt;&lt;td&gt;6.55&lt;/td&gt;&lt;td&gt;&amp;#60;.001&lt;/td&gt;&lt;td&gt;&amp;#8722;0.04&lt;/td&gt;&lt;td&gt;&amp;#8722;1.20&lt;/td&gt;&lt;td&gt;0.23&lt;/td&gt;&lt;td&gt;0.32&lt;/td&gt;&lt;td&gt;19.86&lt;/td&gt;&lt;td&gt;&amp;#60;.001&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Age&lt;/td&gt;&lt;td&gt;&amp;#8722;0.00&lt;/td&gt;&lt;td&gt;&amp;#8722;1.08&lt;/td&gt;&lt;td&gt;0.29&lt;/td&gt;&lt;td&gt;0.00&lt;/td&gt;&lt;td&gt;0.07&lt;/td&gt;&lt;td&gt;0.94&lt;/td&gt;&lt;td&gt;0.01&lt;/td&gt;&lt;td&gt;5.08&lt;/td&gt;&lt;td&gt;&amp;#60;.001&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>5 <emph>Note</emph>. b is the estimated linear effect (not standardized).</p> <hd id="AN0180329822-15">Age and third-person consistency: absolute assessment</hd> <p>When using absolute scoring, on the <emph>Untransformed Picture</emph> task, participants did not differ from chance (here, meaning no more or fewer of them were classified as fully constant vs. pre-constant), and this did not differ by age (see Tables 7 &amp; 8). On the <emph>Transformed Picture</emph> task, participants were significantly less likely than chance to be absolutely gender-consistent, and this did not differ by age. For the <emph>Transformed Picture + Explanation</emph> task participants were significantly less likely than chance to be absolutely gender-consistent, and the likelihood of showing absolute gender consistency increased with age (though model estimates showed that participants were not absolutely gender-consistent more often than chance at any age in our 3–5 age range; see Tables 7 and 8).</p> <p>Table 7. Percentage of participants who gave consistent responses on every question (i.e., were judged "consistent" using absolute scoring).</p> <p> <ephtml> &lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;td&gt;Age group&lt;/td&gt;&lt;td&gt;Untransformed Picture&lt;/td&gt;&lt;td&gt;Transformed Picture&lt;/td&gt;&lt;td&gt;Transformed Picture + Explanation&lt;/td&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;3 years old&lt;/td&gt;&lt;td&gt;38.64%&lt;/td&gt;&lt;td&gt;21.43%&lt;/td&gt;&lt;td&gt;11.36%&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;4 years old&lt;/td&gt;&lt;td&gt;47.22%&lt;/td&gt;&lt;td&gt;13.51%&lt;/td&gt;&lt;td&gt;27.78%&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;5 years old&lt;/td&gt;&lt;td&gt;51.28%&lt;/td&gt;&lt;td&gt;30.77%&lt;/td&gt;&lt;td&gt;43.59%&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>Table 8. Binomial regression – age predicting absolute scores.</p> <p> <ephtml> &lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;td /&gt;&lt;td&gt;Untransformed Picture&lt;/td&gt;&lt;td&gt;Transformed Picture&lt;/td&gt;&lt;td&gt;Transformed Picture + Explanation&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td /&gt;&lt;td&gt;&lt;italic&gt;OR&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;z&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;p&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;OR&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;z&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;p&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;OR&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;z&lt;/italic&gt;&lt;/td&gt;&lt;td&gt;&lt;italic&gt;p&lt;/italic&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;Intercept&lt;/td&gt;&lt;td&gt;0.85&lt;/td&gt;&lt;td&gt;&amp;#8722;0.90&lt;/td&gt;&lt;td&gt;0.37&lt;/td&gt;&lt;td&gt;0.28&lt;/td&gt;&lt;td&gt;&amp;#8722;5.63&lt;/td&gt;&lt;td&gt;&amp;#60;.001&lt;/td&gt;&lt;td&gt;0.35&lt;/td&gt;&lt;td&gt;&amp;#8722;4.67&lt;/td&gt;&lt;td&gt;&amp;#60;.001&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Age&lt;/td&gt;&lt;td&gt;1.02&lt;/td&gt;&lt;td&gt;1.04&lt;/td&gt;&lt;td&gt;0.30&lt;/td&gt;&lt;td&gt;1.03&lt;/td&gt;&lt;td&gt;1.31&lt;/td&gt;&lt;td&gt;0.19&lt;/td&gt;&lt;td&gt;1.08&lt;/td&gt;&lt;td&gt;3.47&lt;/td&gt;&lt;td&gt;.001&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>6 <emph>Note</emph>. OR, odds ratio.</p> <p>Across all three tasks, we had expected older participants to respond in a more gender-consistent way than younger participants. Overall performance on gender consistency tasks and whether responses differed by age depended on both assessment strategy (continuous or absolute) and task (<emph>Untransformed Picture, Transformed Picture</emph>, or <emph>Transformed Picture + Explanation</emph>). If we use the continuous assessment approach, we conclude that participants show a relatively high level of gender-consistent responding. If we use the absolute assessment, we conclude that participants show a low level of gender-consistent responding. We address this issue with additional follow-up analyses below.</p> <p>When using the continuous assessment, older participants were only more likely than younger participants to endorse gender consistency on the <emph>Transformed Picture + Explanation</emph> task; there were no age effects for the <emph>Untransformed Picture</emph> or <emph>Transformed Picture</emph> tasks. When using the absolute assessment, we observed the same pattern: older participants were again more gender-consistent than younger participants for the <emph>Transformed Picture + Explanation</emph> task, with no age effects in the <emph>Untransformed Picture</emph> or <emph>Transformed Picture</emph> tasks.</p> <hd id="AN0180329822-16">Third-person consistency: task effects</hd> <p>We were interested in whether participant responses to third-person consistency questions differed across tasks (<emph>Untransformed Picture, Transformed Picture</emph>, or <emph>Transformed Picture + Explanation</emph>). For these comparisons, we include the <emph>N</emph> = 114 participants who had data from all three tasks. We use a repeated-measures ANOVA to test whether continuous performance varied by task; task had a significant effect on responding (<emph>F</emph>(1.89, 214.04) = 56.61, <emph>p</emph> &lt; 0.001; <emph>M</emph><subs>Untransformed Picture</subs> = 70.18%; <emph>SE</emph> = 3.17%; <emph>M</emph><subs>Transformed Picture</subs> = 45.65%; <emph>SE</emph> = 3.54%; <emph>M</emph><subs>Transformed Picture + Explanation</subs> = 81.72%; <emph>SE</emph> = 1.66%). As predicted, participants endorsed gender-consistency more during the <emph>Untransformed Picture</emph> task than the <emph>Transformed Picture</emph> task (<emph>t</emph>(<reflink idref="bib113" id="ref86">113</reflink>) = 7.33, <emph>p</emph> &lt; 0.001), and endorsed gender-consistency more during the <emph>Transformed Picture + Explanation</emph> task than the <emph>Transformed Picture</emph> task (<emph>t</emph>(<reflink idref="bib113" id="ref87">113</reflink>) = 9.40, <emph>p</emph> &lt; 0.001). Additionally, participants endorsed gender-consistency significantly more often on the <emph>Transformed Picture + Explanation</emph> task than the <emph>Untransformed Picture</emph> task (<emph>t</emph>(<reflink idref="bib113" id="ref88">113</reflink>) = 3.65, <emph>p</emph> &lt; 0.001).</p> <p>To test task differences under the absolute scoring system, we used a binomial multi-level model with a random intercept for participants. We predicted per-task performance with a fixed effect of task, using the <emph>Transformed Picture</emph> task as the reference task. As predicted, participants were significantly more likely to show absolute gender-consistency in the <emph>Untransformed Picture</emph> task than in the <emph>Transformed Picture</emph> task (44.74% showed absolute gender consistency on the <emph>Untransformed Picture</emph> task; 21.93% showed absolute gender consistency on the <emph>Transformed Picture</emph> task; <emph>OR</emph> = 3.55, <emph>z</emph> = 3.78, <emph>p</emph> &lt; 0.001). Contrary to our preregistered prediction, the <emph>Transformed Picture + Explanation</emph> task did not differ significantly from the <emph>Transformed Picture</emph> task when scored in this way (27.19% showed absolute gender consistency on the <emph>Transformed Picture + Explanation</emph> task; <emph>OR</emph> = 1.39, <emph>z</emph> = 0.99, <emph>p</emph> = 0.32). Using an additional non-preregistered McNemar's test on absolute performance, participants showed more endorsement of gender consistency on the <emph>Untransformed Picture</emph> task than the <emph>Transformed Picture + Explanation</emph> task (<emph>χ</emph><sups>2</sups>(df = 1) = 8.60, <emph>p</emph> = 0.003).</p> <p>Regardless of assessment approach, we see that, as expected, participants were more gender-constant when they did not see a transformation (<emph>Untransformed Picture</emph> task) relative to when they saw an unexplained perceptual transformation (<emph>Transformed Picture</emph> task). When visual transformation <emph>was</emph> present, the addition of explanations (<emph>Transformed Picture + Explanation</emph>) significantly increased children's likelihood of saying that gender would not change across the transformation when measured continuously, but not when measured with the absolute assessment (see Figure 2).</p> <p>Graph: Figure 2. Gender-constant responding on third-person gender consistency tasks (continuous scoring).</p> <hd id="AN0180329822-17">Transformed picture tasks: question effects</hd> <p>We had predicted that children would provide more gender-consistent responses on the <emph>Transformed Picture + Explanation</emph> task than on the <emph>Transformed Picture</emph> task. We found this pattern when responses were scored with the continuous assessment, but not when they were scored with absolute assessment. We were interested in this discrepancy and conducted an un-preregistered, exploratory follow-up analysis on data from both tasks, using data from the <emph>N</emph> = 115 participants who had data for both tasks.</p> <p>One possible difference between the <emph>Transformed Picture + Explanation</emph> task and <emph>Transformed Picture</emph> task was that the <emph>Transformed Picture + Explanation</emph> task includes not only questions about gender identity (as the <emph>Transformed Picture</emph> task does) but also property related questions (e.g., "Which swim suit will he wear?"). In the Supplementary Materials, we show that participants were not more likely to give gender-consistent responses on one type of question than another, suggesting that the additional question type in the <emph>Transformed Picture + Explanation</emph> task was not dragging down participants' performance on the absolute scoring and making it look indistinguishable from the <emph>Transformed Picture</emph> task. Another possible difference between the two tasks that would disproportionately impact the absolute scoring was the number of included questions. Our <emph>Transformed Picture + Explanation</emph> task includes 12 questions, while the classic <emph>Transformed Picture</emph> task includes only 8. Having a greater number of questions opens up more opportunities for a participant to make an accidental or random error, which would have minimal impact on continuous scoring but a large impact on absolute scoring. Participants need to make <emph>more</emph> gender-constant responses to count as being absolutely gender-constant on the <emph>Transformed Picture + Explanation</emph> task than on the <emph>Transformed Picture</emph> task. If the <emph>Transformed Picture + Explanation</emph> task had the same number of questions as the <emph>Transformed Picture</emph> task, would we have found our predicted difference using the absolute scoring system, as we found with the continuous scoring system?</p> <p>To test this possibility, we examined all possible subsets of questions from the <emph>Transformed Picture</emph> task (1–8 questions, 255 possible subsets) and all possible subsets of questions from the <emph>Transformed Picture + Explanation</emph> task (1–12 questions, 4095 possible subsets). For each possible subset, we determined whether each of our participants would have been classified as absolutely gender-constant or not absolutely gender-constant on each task using that subset of items. Figure 3 shows that for each task, including a smaller number of questions means a higher number of participants are classified as gender-constant (i.e., they indicated that a character's gender would remain consistent across a transformation on every included question). The number of questions included in a task has a direct relation with whether more or fewer children are judged gender-constant in the absolute assessment. This additionally means that it was harder to show absolute gender constancy on our <emph>Transformed Picture + Explanation</emph> task than the <emph>Transformed Picture</emph> task, simply because it had more questions – an essentially arbitrary feature of task design.</p> <p>Graph: Figure 3. Transformed Picture task: proportion of participants who are gender-constant according to number of questions included in absolute assessment. Lines are best-fit lines according to a LOESS approximation.</p> <p>Given this within-task relation between number of questions and absolute classification, we next compared the two tasks. If we include the same number of questions in each task, how often do we observe our expected effect (greater absolute gender-consistency on the <emph>Transformed Picture + Explanation</emph> Task than the <emph>Transformed Picture</emph> Task)? For example, there are 6336 ways we could include seven questions from each task (8 possible subsets from the <emph>Transformed Picture</emph> task × 792 possible subsets from the <emph>Transformed Picture + Explanation</emph> task). For each one, we defined the number of absolutely gender-consistent participants on each task, and tested whether consistency differed by task (using a McNemar's test). Then, we calculated how often we found our predicted effect, no effect, or the opposite effect (i.e., greater gender-constancy on the <emph>Transformed Picture</emph>). We found our expected effect in 81.49% of the 6336 possible ways to include seven questions from each task. We completed this same process for every possible number of questions, 1–8. When the same number of questions are included from each task, we find greater gender constant responding on the <emph>Transformed Picture + Explanation</emph> task compared to the <emph>Transformed Picture</emph> task more than 50% of the time, and never find the opposite effect (see Figure 4).</p> <p>Graph: Figure 4. Transformed picture task: frequency of finding task effect when we include an equal number of questions from both tasks.</p> <p>The main take-home from these additional analyses is that if both tasks had the same number of items, we would expect results on the absolute scoring system that mirrored the continuous scoring system. We return to the issue of the impact of number of items on the broader gender constancy literature in the general discussion.</p> <hd id="AN0180329822-18">Discussion</hd> <p>We assessed whether three-to-five-year-old children, who were predominantly non-Hispanic White children from upper-SES families in the Northeastern United States, believed that gender was consistent across changes in gendered appearance or behavior. We found that when this behavior was explained by a situational attribution (<emph>Transformed Picture + Explanation</emph> Task) – such as dressing up for a play or not having a clean alternative outfit – these children overcame the perceptual appearance of a child's gender nonconformity and were more likely than chance to believe the target's gender had remained consistent over time. They were more likely to endorse gender consistency in this condition than in two other conditions: one in which they saw a visual depiction of a character undergoing a change in gendered appearance or behavior without any explanation as to why this was happening (<emph>Transformed Picture</emph> Task, on which the probability of children endorsing gender consistency did not differ from chance); and another in which they did not see a visual depiction of a character's gender at all, but were just asked about it hypothetically (<emph>Untransformed Picture</emph> Task, on which children in our sample were more likely than chance to endorse gender consistency).</p> <p>Importantly, however, this pattern – that children were more likely to endorse gender consistency when they heard a verbal explanation – only emerged when their responses were scored as the percentage of time children gave gender-consistent responses. When we used absolute scoring, such that children had to give gender-consistent responses on every single item they answered for them to be considered "gender consistent," performance did not significantly differ between the <emph>Transformed Picture</emph> Task and the <emph>Transformed Picture + Explanation</emph> Task. In investigating why this might have occurred, we discovered the possible impact of an underappreciated aspect of study design in the gender constancy literature – the number of items used.</p> <p>The impact of number of items is particularly relevant to empirical research that uses absolute scoring (e.g., Bem, [<reflink idref="bib2" id="ref89">2</reflink>]; Carter &amp; Levy, [<reflink idref="bib4" id="ref90">4</reflink>]; Eaton &amp; von Bargen, [<reflink idref="bib8" id="ref91">8</reflink>]; Eaton et al., [<reflink idref="bib9" id="ref92">9</reflink>]; Frey &amp; Ruble, [<reflink idref="bib14" id="ref93">14</reflink>]; Levy &amp; Carter, [<reflink idref="bib27" id="ref94">27</reflink>]; Slaby &amp; Frey, [<reflink idref="bib37" id="ref95">37</reflink>]). In these cases, a child must reply with constancy-indicative responses on every question corresponding to a particular "stage" of constancy to "pass" that stage. However, as the number of questions increases, so too do the statistical odds of providing a nonconstant response by chance (i.e., because of simply mishearing an item or saying the wrong thing). This relates to a pattern we saw in one of our exploratory analyses: When we matched the number of questions on the <emph>Transformed Picture + Explanation</emph> Task and the <emph>Transformed Picture</emph> Task by randomly selecting subsets of questions, the results of the absolute coding mirrored the results from the continuous coding such that children tended to show more consistency on the task that involved an explanation. Therefore, the apparent lack of "absolute" gender constancy on the <emph>Transformed Picture + Explanation</emph> Task may have been an artifact of the sheer number of items participants were asked to respond to in that task. This sensitivity to the number of items is a limitation that developmental scientists should be aware of when deciding whether to assess children's cognitive competencies with variable scored in an all-or-nothing way; given our current findings, we suggest that researchers move away from this practice, as has been argued by other researchers about similar dichotomization practices (e.g., DeCoster et al., [<reflink idref="bib5" id="ref96">5</reflink>]; MacCallum et al., [<reflink idref="bib29" id="ref97">29</reflink>]). Furthermore, we encourage researchers to interpret past gender constancy research with the knowledge that much of the foundational work – which is still heavily cited today – characterized children's understanding of gender constancy using all-or-nothing scales, a design choice that may have made children's difficulties on the task look particularly dramatic.</p> <p>There are a few possible reasons why the children in our sample generally overcame a perceptual conflict and gave more gender-consistent responses when provided with a verbal explanation for a character's cross-gender behavior or appearance. One possibility is that the <emph>Transformed Picture + Explanation</emph> Task was more engaging than the less narratively detailed <emph>Transformed Picture</emph> Task, causing participants to pay more attention to the story in the former relative to the latter. Another possibility is that, as postulated in prior literature, children may be interpreting the experimenter's question (e.g., "If Johnny wears a dress like this, is he a boy or a girl?") as referring to a hypothetical or pretend situation (Martin &amp; Halverson, [<reflink idref="bib31" id="ref98">31</reflink>]) when no explanation is provided. Put differently, without an explanation as to <emph>why</emph> Johnny is wearing a dress, children may fill in the missing information by inferring that Johnny is pretending to be a girl and give an answer that is concordant with the gender they assume Johnny is pretending to be. Our results are compatible with this explanation [but see Szkrybalo and Ruble ([<reflink idref="bib38" id="ref99">38</reflink>]), for skepticism of that interpretation]. Another possible explanation is that while there is some ambiguity about whether the character in the <emph>Transformed Picture</emph> Task is changing gendered behavior or appearance because of circumstance or because of a true identity change, that ambiguity is resolved in the <emph>Transformed Picture + Explanation</emph> Task.</p> <p>Alternatively, without an explanation to anchor their understanding of the task, children may not fully grasp that the character they see before the transformation is the very same one as the character after the transformation. The explanation therefore might help them realize that the same child appears at the beginning and end of the task. This idea echoes a suggestion made by Emmerich et al. ([<reflink idref="bib11" id="ref100">11</reflink>]) about the <emph>Transformed Picture</emph> task: younger children tend to treat the task as a perceptual classification task. For example, they may see Johnny first depicted as a boy (and clearly endorse the idea that he is the boy), but when they see Johnny put on a dress, rather than keeping the fact that they are still looking at Johnny at the forefront of their attention, they may default to a simple classification task (i.e., "The person in the picture is wearing a dress, and I know girls wear dresses, therefore the person in the picture is a girl"). A similar idea was described slightly differently by Bem ([<reflink idref="bib2" id="ref101">2</reflink>]), who suggested that children are liable to interpret the question of, "If Johnny wears a dress like this, is Johnny a boy or a girl"? as actually meaning, "If Johnny were the kind of person who regularly wore dresses, probabilistically would Johnny be a boy or a girl"? Our suggestion here is that an explanation grounds children to the idea that, in watching a character undergo some kind of transformation (e.g., Johnny putting on a dress), the character is still the same person before and after this transformation in a very deep sense. While all the aforementioned reasons might be at play in explaining why children in our sample were more likely to give gender-consistent answers on the <emph>Transformed Picture + Explanation</emph> Task relative to the <emph>Transformed Picture</emph> Task, it is difficult given our current results to speculate in favor of one or more over the others.</p> <hd id="AN0180329822-19">Implications</hd> <p>Within the broader context of gender constancy research, our results indicate that past studies may have underestimated children's belief in gender consistency by not considering children's ability to sometimes overcome misleading perceptual cues when given contextual information about why these cues are present. Some studies have suggested that children do not believe that gender is consistent across change in appearance until 7 to 8 years old (De Lisi &amp; Gallagher, [<reflink idref="bib6" id="ref102">6</reflink>]; Emmerich et al., [<reflink idref="bib11" id="ref103">11</reflink>]; Marcus &amp; Overton, [<reflink idref="bib30" id="ref104">30</reflink>]; Szkrybalo &amp; Ruble, [<reflink idref="bib38" id="ref105">38</reflink>]; Wehren &amp; De Lisi, [<reflink idref="bib41" id="ref106">41</reflink>]), a finding that is consistent with some other work in conceptual development which suggests that children do not understand natural kinds as stable over perceptual transformation until ages 7–9 (Keil, [<reflink idref="bib23" id="ref107">23</reflink>]). Our results, combined with past results (e.g., Gelman et al., [<reflink idref="bib17" id="ref108">17</reflink>]), suggest that the classic account – that children rely on perceptual cues such as appearance and ignore prior knowledge in judging a person's gender category – is likely incorrect much of the time, even if children's consistency judgments become more reliable with age. In this study, we demonstrate, at the very least, the capacity to override cross-gender perceptual information even at young ages among the demographic group represented in our sample.</p> <p>Our results also accord with findings regarding children's gender essentialism which suggest that by ages 3–4, children tend to view gender as a stable, deeply held trait (Taylor, [<reflink idref="bib39" id="ref109">39</reflink>]; Diesendruck et al., [<reflink idref="bib7" id="ref110">7</reflink>]; Taylor et al., [<reflink idref="bib40" id="ref111">40</reflink>]; Rhodes et al., [<reflink idref="bib35" id="ref112">35</reflink>]; Halim et al., [<reflink idref="bib21" id="ref113">21</reflink>]). Much of the classic work on gender constancy has seemingly conflicted with research from the past three decades on children's gender essentialism, since it has typically shown that children believe that one's gender is determined by perceptual cues shown at a given moment. Our results here indicate that children's essentialist tendency to see gender as generally stable – including across perceptual transformations – is appreciable if they are given context as to why the perceptual transformation is occurring, at least in U.S.-based children.</p> <p>The idea that children take context into account when assessing someone's gender could have interesting implications in situations that differ from the one we presented to children in this study. In this study, integrating the context presented in the <emph>Transformed Picture + Explanation</emph> Task would lead one to understand that, in all likelihood, the gender of the character in question had not changed even though they had undergone some transformation in gendered behavior or appearance. Our task provided context in the form of a situational attribution for the character's change in behavior or appearance, and this situational attribution was meant to nudge children's reasoning toward endorsing the consistency of the character's gender identity. Importantly, this kind of endorsement of consistency has classically been seen in the literature as the more mature, adult-like or correct way to reason about gender as a concept (e.g., Kohlberg, [<reflink idref="bib24" id="ref114">24</reflink>]).</p> <p>However, there are contexts in which this reasoning – believing that gender is always stable, and thus there is no way that a person's gender <emph>could</emph> change – is incorrect: Some people <emph>do</emph> undergo changes in gender identity. In these cases, appropriately integrating context would mean understanding that someone's gender <emph>has</emph> actually changed, rather than <emph>a priori</emph> rejecting gender change as an impossibility in all circumstances (as has historically been considered "correct" reasoning in developmental psychology research). Our results here, which show children considering context when reasoning about someone's gender identity, raise the question of whether children are open to the idea that someone else's gender <emph>can</emph> change when shown a character undergo a change in behavior or appearance accompanied by a dispositional attribution (e.g., "Johnny is wearing a dress now because Johnny feels like a girl and wants to wear dresses"). The closest case we know of in the literature introduced 5- to 10-year-old children to a vignette about a child who socially transitioned to another gender (e.g., an infant assigned male at birth later identified as a girl) – accompanied by visual images of a perceptual transformation – and then assessed participants' perceptions of the target's gender category (Gülgöz et al., [<reflink idref="bib20" id="ref115">20</reflink>]). Participants were split such that almost half of participants, across items, believed that the target's gender had changed, while a similar number did not think change had occurred. Children who tended to like transgender children more tended to be more accepting of the change in gender, suggesting that other personal, experiential, or motivational factors may also play a role in these decisions. Additionally, parent reports of young children accepting their classmates' new gender identities after a social transition (Gallagher &amp; Olson, [<reflink idref="bib15" id="ref116">15</reflink>]) lead us to think that children at these same ages can also, under certain circumstances, acknowledge change.</p> <p>If it is ultimately the case that children are sensitive to particular cues indicating that change has not occurred (e.g., situational constraints), but are also open to other cues that indicate change has occurred (e.g., dispositional constraints), this might suggest that young children's mental models of gender would be better characterized as probabilistic than constant/nonconstant. A probabilistic model would mean that children might, after extensive exposure to people in their environments <emph>not</emph> changing genders, hold the view that (barring unusual evidence) most people's gender does not change. But unlike the classic model of gender constancy in which children hold an all-or-nothing view, it might be the case that with a probabilistic model, children remain open to updating their views. Children with exposure to change, as with a classmate's or sibling's social transition, may see change in someone else as more possible than children who never witnessed a change. Nonetheless, even with a single exposure to a change, we would expect that children might still see change as relatively rare. While entirely speculative, we believe that future work with an eye toward this type of model may not just be fruitful but also timely. Such a view could also expand beyond the cognitive models of the 1960s to better reflect contemporary views about gender.</p> <hd id="AN0180329822-20">Limitations and generalizability</hd> <p>Our study contained several noteworthy limitations. First, as is the case with most research on children's gender constancy, the visual displays we showed to children were not highly realistic, but rather were stylized cartoon images of children, so it is not certain whether and how the patterns we observed here would extend to real-life situations. Second, the gender constancy tasks used in our study imply that maintaining one gender identity across time and contexts is inherently "correct." We worked hard not to use this language because we know that continuity is not always true or correct for all people. In fact, we could imagine that children with different social histories – such as those who know gender nonconforming or transgender children – might think about gender consistency in a very different way than that which has historically been assumed to be "correct" in the gender constancy literature. An important future direction for this work is to understand the role of first-hand experience on children's understanding of gender.</p> <hd id="AN0180329822-21">Generalizability statement</hd> <p>Our sample was skewed toward white, upper-SES, highly educated, politically liberal U.S. families. The generalizability of our results to samples from different cultural, ethnic, socioeconomic, political, and educational backgrounds is unknown. Though some scholarship on gender constancy has been conducted with children in non-U.S. contexts (e.g., De Lisi &amp; Gallagher, [<reflink idref="bib6" id="ref117">6</reflink>]; Zmyj &amp; Bischoff-Kohler, [<reflink idref="bib43" id="ref118">43</reflink>]) and racially and socioeconomically diverse children within the U.S. (e.g., Emmerich et al., [<reflink idref="bib11" id="ref119">11</reflink>]), the majority of research on gender constancy has been conducted with U.S. children from largely White, upper-SES families. The scarcity of research from diverse populations of children – combined with our speculative proposal positing that children's cognition about gender constancy might be shaped by their experiences observing and understanding the genders of others in their immediate surroundings – makes it even more critical that follow-up work assess gender cognition in more diverse contexts. This will allow researchers to begin to assess whether children's understanding of gender as stable or not is robust across cultural variation – an issue that we are not able to address with the current work.</p> <hd id="AN0180329822-22">Conclusion</hd> <p>We found that three-to-five-year-old children, who were predominantly non-Hispanic white children from upper-SES families in the Northeastern United States, tend to believe that gender remains consistent even after they see that person "transform" into a person with a gender nonconforming characteristic if they have a situational explanation for that behavior. These results stand in contrast to a common perception that children are effectively unable to override perceptual cues when reasoning about a person's gender until at least age 6 and indicate that children's understanding of gender constancy might be more nuanced and sophisticated than some classic research in developmental psychology suggests.</p> <hd id="AN0180329822-23">Acknowledgments</hd> <p>We thank Sydney Peng for creating some of the visual stimuli for this study and Casey Beidel and Haley Hendrickson for their help in qualitative coding. Data collection and preparation of this article were supported by National Science Foundation (NSF) Grants SMA-1837857/SMA-2041463 (to K.R.O.) and an NSF Graduate Research Fellowship (to B.D.). The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Science Foundation.</p> <hd id="AN0180329822-24">Disclosure statement</hd> <p>No potential conflict of interest was reported by the author(s).</p> <hd id="AN0180329822-25">Data, analysis, and preregistration availability statement</hd> <p>Data, analysis, and preregistration available at https://osf.io/3r5ye/.</p> <hd id="AN0180329822-26">Ethics approval statement</hd> <p>Human subjects protocols adhered to those approved by the Institutional Review Board at Princeton University, Protocol #12923.</p> <hd id="AN0180329822-27">Materials statement</hd> <p>Materials used in this study are available from the first author.</p> <hd id="AN0180329822-28">Supplementary material</hd> <p>Supplemental data for this article can be accessed online at https://doi.org/10.1080/15248372.2024.2383563</p> <ref id="AN0180329822-29"> <title> Footnotes </title> <blist> <bibl id="bib1" idref="ref55" type="bt">1</bibl> <bibtext> Contributed equally.</bibtext> </blist> </ref> <ref id="AN0180329822-30"> <title> References </title> <blist> <bibtext> Bates, D., Mächler, M., Bolker, B., &amp; Walker, S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67 (1). https://doi.org/10.18637/jss.v067.i01</bibtext> </blist> <blist> <bibl id="bib2" idref="ref6" type="bt">2</bibl> <bibtext> Bem, S. L. (1989). Genital knowledge and gender constancy in preschool children. Child Development, 60 (3), 649 – 662. https://doi.org/10.2307/1130730</bibtext> </blist> <blist> <bibl id="bib3" idref="ref36" type="bt">3</bibl> <bibtext> Brainerd, C. J., &amp; Hooper, F. H. (1975). A methodological analysis of developmental studies of identity conservation and equivalence conservation. Psychological Bulletin, 82 (5), 725 – 737. https://doi.org/10.1037/h0077070</bibtext> </blist> <blist> <bibl id="bib4" idref="ref90" type="bt">4</bibl> <bibtext> Carter, D. B., &amp; Levy, G. D. (1988). Cognitive aspects of early sex-role development: The influence of gender schemas on preschoolers' memories and preferences for sex-typed toys and activities. Child Development, 59 (3), 782 – 792. https://doi.org/10.2307/1130576</bibtext> </blist> <blist> <bibl id="bib5" idref="ref66" type="bt">5</bibl> <bibtext> DeCoster, J., Iselin, A. M. R., &amp; Gallucci, M. (2009). A conceptual and empirical examination of justifications for dichotomization. Psychological Methods, 14 (4), 349. https://doi.org/10.1037/a0016956</bibtext> </blist> <blist> <bibl id="bib6" idref="ref27" type="bt">6</bibl> <bibtext> De Lisi, R., &amp; Gallagher, A. M. (1991). Understanding of gender stability and constancy in Argentinean children. Merrill-Palmer Quarterly, 37 (3), 483 – 502.</bibtext> </blist> <blist> <bibl id="bib7" idref="ref43" type="bt">7</bibl> <bibtext> Diesendruck, G., Goldfein-Elbaz, R., Rhodes, M., Gelman, S., &amp; Neumark, N. (2013). Cross-cultural differences in Children's beliefs about the objectivity of social categories. Child Development, 84 (6), 1906 – 1917. https://doi.org/10.1111/cdev.12108</bibtext> </blist> <blist> <bibl id="bib8" idref="ref23" type="bt">8</bibl> <bibtext> Eaton, W. O., &amp; von Bargen, D. (1981). Asynchronous development of gender understanding in preschool children. Child Development, 52 (3), 1020 – 1027. https://doi.org/10.2307/1129107</bibtext> </blist> <blist> <bibl id="bib9" idref="ref7" type="bt">9</bibl> <bibtext> Eaton, W. O., von Bargen, D., &amp; Keats, J. G. (1981). Gender understanding and dimensions of preschooler toy choice: Sex stereotype versus activity level. Canadian Journal of Behavioural Science/Revue Canadienne des Sciences du Comportement, 13 (3), 203 – 209. https://doi.org/10.1037/h0081181</bibtext> </blist> <blist> <bibtext> Emmerich, W. (1982). Nonmonotonic developmental trends in social cognition: The case of gender identity. In S. Strauss &amp; R. Stavy (Eds.), U-Shaped behavioral growth (pp. 249 – 262). Academic Press.</bibtext> </blist> <blist> <bibtext> Emmerich, W., Goldman, K. S., Kirsh, B., &amp; Sharabany, R. (1977). Evidence for a transitional phase in the development of gender constancy. Child Development, 48 (3), 930 – 936. https://doi.org/10.2307/1128343</bibtext> </blist> <blist> <bibtext> Fast, A. A., &amp; Olson, K. R. (2018). Gender development in transgender preschool children. Child Development, 89, 620 – 637. https://doi.org/10.1111/cdev.12758</bibtext> </blist> <blist> <bibtext> Flavell, J. H. (1963). The developmental psychology of Jean Piaget. D Van Nostrand. https://doi.org/10.1037/11449-000</bibtext> </blist> <blist> <bibtext> Frey, K. S., &amp; Ruble, D. N. (1992). Gender constancy and the "cost" of sex-typed behavior: A test of the conflict hypothesis. Developmental Psychology, 28 (4), 714 – 721. https://doi.org/10.1037/0012-1649.28.4.714</bibtext> </blist> <blist> <bibtext> Gallagher, N., &amp; Olson, K. R. (2023, June). The relationship between age and peer support for binary trans youth. Conference of the European Association for Social Psychology, Kraków, Poland.</bibtext> </blist> <blist> <bibtext> Gelman, S. A. (2004). Psychological essentialism in children. Trends in Cognitive Sciences, 8 (9), 404 – 409. https://doi.org/10.1016/j.tics.2004.07.001</bibtext> </blist> <blist> <bibtext> Gelman, S. A., Collman, P., &amp; Maccoby, E. E. (1986). Inferring properties from categories versus inferring categories from properties: The case of gender. Child Development, 57 (2), 396 – 404. https://doi.org/10.2307/1130595</bibtext> </blist> <blist> <bibtext> Gelman, S. A., &amp; Markman, E. M. (1986). Categories and induction in young children. Cognition, 23 (3), 183 – 209. https://doi.org/10.1016/0010-0277(86)90034-X</bibtext> </blist> <blist> <bibtext> Gouze, K. R., &amp; Nadelman, L. (1980). Constancy of gender identity for self and others in children between the Ages of three and seven. Child Development, 51 (1), 275 – 278. https://doi.org/10.2307/1129622</bibtext> </blist> <blist> <bibtext> Gülgöz, S., Gomez, E. M., DeMeules, M. R., &amp; Olson, K. R. (2018). Children's evaluation and categorization of transgender children. Journal of Cognition and Development, 19 (4), 325 – 344. https://doi.org/10.1080/15248372.2018.1498338</bibtext> </blist> <blist> <bibtext> Halim, M. L., Ruble, D. N., Tamis LeMonda, C. S., Zosuls, K. M., Lurye, L. E., &amp; Greulich, F. K. (2014). Pink frilly dresses and the avoidance of all things "girly": Children's appearance rigidity and cognitive theories of gender development. Developmental Psychology, 50 (4), 1091 – 1101. https://doi.org/10.1037/a0034906</bibtext> </blist> <blist> <bibtext> Kassambara, A. (2022). Rstatix: Pipe-friendly framework for basic statistical tests. https://CRAN.R-project.org/package=rstatix</bibtext> </blist> <blist> <bibtext> Keil, F. C. (1989). Concepts, kinds, and cognitive development. The MIT Press.</bibtext> </blist> <blist> <bibtext> Kohlberg, L. (1966). A cognitive-developmental analysis of children's sex-role concepts and attitudes. In E. Maccoby (Ed.), The development of sex differences (pp. 82 – 173). Stanford University Press.</bibtext> </blist> <blist> <bibtext> Kuhn, D., Nash, S. C., &amp; Brucken, L. (1978). Sex role concepts of two- and three-year-olds. Child Development, 49 (2), 445 – 451. https://doi.org/10.2307/1128709</bibtext> </blist> <blist> <bibtext> Lenth, R. V. (2021). Emmeans: Estimated marginal means, aka least-squares means (R package version 1.6.2-1). https://CRAN.R-project.org/package=emmeans</bibtext> </blist> <blist> <bibtext> Levy, G. D., &amp; Carter, D. B. (1989). Gender schema, gender constancy, and gender-role knowledge: The roles of cognitive factors in preschoolers' gender-role stereotype attributions. Developmental Psychology, 25 (3), 444 – 449. https://doi.org/10.1037/0012-1649.25.3.444</bibtext> </blist> <blist> <bibtext> Lüdecke, D., Bartel, A., Schwemmer, C., Powell, C., Djalovski, A., &amp; Titz, J. (2022). sjPlot: Data visualization for statistics in social science (R package version 2.8.12). https://CRAN.R-project.org/package=sjPlot</bibtext> </blist> <blist> <bibtext> MacCallum, R. C., Zhang, S., Preacher, K. J., &amp; Rucker, D. D. (2002). On the practice of dichotomization of quantitative variables. Psychological Methods, 7 (1), 19. https://doi.org/10.1037/1082-989X.7.1.19</bibtext> </blist> <blist> <bibtext> Marcus, D. E., &amp; Overton, W. F. (1978). The development of cognitive gender constancy and sex role preferences. Child Development, 49 (2), 434 – 444. https://doi.org/10.2307/1128708</bibtext> </blist> <blist> <bibtext> Martin, C. L., &amp; Halverson, C. F. (1983). Gender constancy: A methodological and theoretical analysis. Sex Roles, 9 (7), 775 – 790. https://doi.org/10.1007/BF00290030</bibtext> </blist> <blist> <bibtext> McConaghy, M. J. (1979). Gender permanence and the genital basis of gender: Stages in the development of constancy of gender identity. Child Development, 50 (4), 1223 – 1226. https://doi.org/10.2307/1129354</bibtext> </blist> <blist> <bibtext> Muradoglu, M., Cimpian, J. R., &amp; Cimpian, A. (2023). Mixed-effects models for cognitive development researchers. Journal of Cognition and Development, 24 (3), 307 – 340. https://doi.org/10.1080/15248372.2023.2176856</bibtext> </blist> <blist> <bibtext> Rhodes, M., Gelman, S. A., &amp; Karuza, J. C. (2013). Preschool ontology: The role of beliefs about category boundaries in early categorization. Journal of Cognition and Development, 15 (1), 78 – 93. https://doi.org/10.1080/15248372.2012.713875</bibtext> </blist> <blist> <bibtext> Rhodes, M., Gelman, S. A., &amp; Karuza, J. C. (2013). Preschool ontology: The role of beliefs about category boundaries in early categorization. Journal of Cognition and Development, 15 (1), 78 – 93. https://doi.org/10.1080/15248372.2012.713875</bibtext> </blist> <blist> <bibtext> Ruble, D. N., Taylor, L. J., Cyphers, L., Greulich, F. K., Lurye, L. E., &amp; Shrout, P. E. (2007). The role of gender constancy in early gender development. Child Development, 78 (4), 1121 – 1136. https://doi.org/10.1111/j.1467-8624.2007.01056.x</bibtext> </blist> <blist> <bibtext> Slaby, R. G., &amp; Frey, K. S. (1975). Development of gender constancy and selective attention to same-sex models. Child Development, 46 (4), 849 – 856. https://doi.org/10.2307/1128389</bibtext> </blist> <blist> <bibtext> Szkrybalo, J., &amp; Ruble, D. N. (1999). "God made me a girl": Sex-category constancy judgments and explanations revisited. Developmental Psychology, 35 (2), 392 – 402. https://doi.org/10.1037/0012-1649.35.2.392</bibtext> </blist> <blist> <bibtext> Taylor, M. G. (1996). The development of children's beliefs about social and biological aspects of gender differences. Child Development, 67 (4), 1555 – 1571. https://doi.org/10.2307/1131718</bibtext> </blist> <blist> <bibtext> Taylor, M. G., Rhodes, M., &amp; Gelman, S. A. (2009). Boys will be boys; cows will be cows: Children's essentialist reasoning about gender categories and animal species. Child Development, 80 (2), 461 – 481. https://doi.org/10.1111/j.1467-8624.2009.01272.x</bibtext> </blist> <blist> <bibtext> Wehren, A., &amp; De Lisi, R. (1983). The development of gender understanding: Judgments and explanations. Child Development, 54 (6), 1568 – 1578. https://doi.org/10.2307/1129820</bibtext> </blist> <blist> <bibtext> Wickham, H., Averick, M., Bryan, J., Chang, W., McGowan, L. D., François, R., Grolemund, G., Hayes, A., Henry, L., Hester, J., Kuhn, M., Pedersen, T. L., Miller, E., Bache, S. M., Müller, K., Ooms, J., Robinson, D., Seidel, D. P., Spinu, V. S., ... Yutani, H. (2019). Welcome to the tidyverse. Journal of Open Source Software, 4 (43), 1686. https://doi.org/10.21105/joss.01686</bibtext> </blist> <blist> <bibtext> Zmyj, N., &amp; Bischof-Köhler, D. (2015). The development of gender constancy in early childhood and Its Relation to Time Comprehension and false-belief understanding. Journal of Cognition and Development, 16 (3), 455 – 470. https://doi.org/10.1080/15248372.2013.824881</bibtext> </blist> </ref> <aug> <p>By Victoria A. Varlack; Benjamin deMayo; Shira Kahn-Samuelson; Natalie M. Gallagher; Marjorie Rhodes and Kristina R. Olson</p> <p>Reported by Author; Author; Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib11" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib24" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib13" firstref="ref3"></nolink> <nolink nlid="nl4" bibid="bib10" firstref="ref8"></nolink> <nolink nlid="nl5" bibid="bib25" firstref="ref10"></nolink> <nolink nlid="nl6" bibid="bib32" firstref="ref11"></nolink> <nolink nlid="nl7" bibid="bib19" firstref="ref12"></nolink> <nolink nlid="nl8" bibid="bib30" firstref="ref13"></nolink> <nolink nlid="nl9" bibid="bib31" firstref="ref14"></nolink> <nolink nlid="nl10" bibid="bib36" firstref="ref15"></nolink> <nolink nlid="nl11" bibid="bib37" firstref="ref16"></nolink> <nolink nlid="nl12" bibid="bib38" firstref="ref17"></nolink> <nolink nlid="nl13" bibid="bib41" firstref="ref18"></nolink> <nolink nlid="nl14" bibid="bib16" firstref="ref40"></nolink> <nolink nlid="nl15" bibid="bib39" firstref="ref41"></nolink> <nolink nlid="nl16" bibid="bib40" firstref="ref42"></nolink> <nolink nlid="nl17" bibid="bib34" firstref="ref44"></nolink> <nolink nlid="nl18" bibid="bib17" firstref="ref46"></nolink> <nolink nlid="nl19" bibid="bib18" firstref="ref54"></nolink> <nolink nlid="nl20" bibid="bib14" firstref="ref61"></nolink> <nolink nlid="nl21" bibid="bib12" firstref="ref74"></nolink> <nolink nlid="nl22" bibid="bib22" firstref="ref81"></nolink> <nolink nlid="nl23" bibid="bib26" firstref="ref82"></nolink> <nolink nlid="nl24" bibid="bib28" firstref="ref83"></nolink> <nolink nlid="nl25" bibid="bib42" firstref="ref84"></nolink> <nolink nlid="nl26" bibid="bib33" firstref="ref85"></nolink> <nolink nlid="nl27" bibid="bib113" firstref="ref86"></nolink> <nolink nlid="nl28" bibid="bib27" firstref="ref94"></nolink> <nolink nlid="nl29" bibid="bib29" firstref="ref97"></nolink> <nolink nlid="nl30" bibid="bib23" firstref="ref107"></nolink> <nolink nlid="nl31" bibid="bib35" firstref="ref112"></nolink> <nolink nlid="nl32" bibid="bib21" firstref="ref113"></nolink> <nolink nlid="nl33" bibid="bib20" firstref="ref115"></nolink> <nolink nlid="nl34" bibid="bib15" firstref="ref116"></nolink> <nolink nlid="nl35" bibid="bib43" firstref="ref118"></nolink> |
|---|---|
| Header | DbId: eric DbLabel: ERIC An: EJ1444843 AccessLevel: 3 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Contextual Information Shifts Young Children's Understanding of Gender Constancy – Name: Language Label: Language Group: Lang Data: English – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Victoria+A%2E+Varlack%22">Victoria A. Varlack</searchLink><br /><searchLink fieldCode="AR" term="%22Benjamin+deMayo%22">Benjamin deMayo</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0001-8723-6419">0000-0001-8723-6419</externalLink>)<br /><searchLink fieldCode="AR" term="%22Shira+Kahn-Samuelson%22">Shira Kahn-Samuelson</searchLink><br /><searchLink fieldCode="AR" term="%22Natalie+M%2E+Gallagher%22">Natalie M. Gallagher</searchLink><br /><searchLink fieldCode="AR" term="%22Marjorie+Rhodes%22">Marjorie Rhodes</searchLink><br /><searchLink fieldCode="AR" term="%22Kristina+R%2E+Olson%22">Kristina R. Olson</searchLink> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="SO" term="%22Journal+of+Cognition+and+Development%22"><i>Journal of Cognition and Development</i></searchLink>. 2024 25(5):619-642. – Name: Avail Label: Availability Group: Avail Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals – Name: PeerReviewed Label: Peer Reviewed Group: SrcInfo Data: Y – Name: Pages Label: Page Count Group: Src Data: 24 – Name: DatePubCY Label: Publication Date Group: Date Data: 2024 – Name: SourceSuprt Label: Sponsoring Agency Group: SrcSuprt Data: National Science Foundation (NSF), Office of Multidisciplinary Activities (SMA)<br />National Science Foundation (NSF), Graduate Research Fellowship Program (GRFP) – Name: NumberContract Label: Contract Number Group: NumCntrct Data: 1837857<br />2041463 – Name: TypeDocument Label: Document Type Group: TypDoc Data: Journal Articles<br />Reports - Research – Name: Subject Label: Descriptors Group: Su Data: <searchLink fieldCode="DE" term="%22Young+Children%22">Young Children</searchLink><br /><searchLink fieldCode="DE" term="%22Gender+Issues%22">Gender Issues</searchLink><br /><searchLink fieldCode="DE" term="%22Sexual+Identity%22">Sexual Identity</searchLink><br /><searchLink fieldCode="DE" term="%22Context+Effect%22">Context Effect</searchLink><br /><searchLink fieldCode="DE" term="%22Beliefs%22">Beliefs</searchLink><br /><searchLink fieldCode="DE" term="%22Child+Behavior%22">Child Behavior</searchLink> – Name: DOI Label: DOI Group: ID Data: 10.1080/15248372.2024.2383563 – Name: ISSN Label: ISSN Group: ISSN Data: 1524-8372<br />1532-7647 – Name: Abstract Label: Abstract Group: Ab Data: Children's understanding of the stability of gender over time has long been recognized as a hallmark of early childhood cognitive development. Prior research has argued that until roughly age 6, children do not understand that a person's gender identity remains consistent if the person undergoes a change in gendered behavior or appearance. However, this prior work has generally not given children any context as to why the behavior or appearance change is occurring. The present study showed U.S. three-to-five-year-olds a character (e.g. "Johnny") undergo such a change that occurred because of external factors (e.g. "Johnny is wearing his sister's dress because all his clothes are dirty"); we asked whether children would indicate that the character's gender was the same pre- and post-change. Children (N = 124, 52% girls, 66% non-Hispanic White) as young as 3 demonstrated an understanding of gender consistency and were significantly more likely to endorse gender consistency when they were, versus were not, given context for the change (p < 0.001, Cohen's d = 0.88). Results suggest that U.S. children take context into account when conceptualizing others' genders and believe that gender is generally consistent across transformations in behavior or appearance by age 3. We discuss the implications of these findings in light of a broader literature about children's understanding of gender constancy as well as contemporary understandings of gender that sometimes accommodate the idea that a person's gender can change. – Name: AbstractInfo Label: Abstractor Group: Ab Data: As Provided – Name: Note Label: Notes Group: Note Data: https://osf.io/3r5ye – Name: DateEntry Label: Entry Date Group: Date Data: 2024 – Name: AN Label: Accession Number Group: ID Data: EJ1444843 |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1444843 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1080/15248372.2024.2383563 Languages: – Text: English PhysicalDescription: Pagination: PageCount: 24 StartPage: 619 Subjects: – SubjectFull: Young Children Type: general – SubjectFull: Gender Issues Type: general – SubjectFull: Sexual Identity Type: general – SubjectFull: Context Effect Type: general – SubjectFull: Beliefs Type: general – SubjectFull: Child Behavior Type: general Titles: – TitleFull: Contextual Information Shifts Young Children's Understanding of Gender Constancy Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Victoria A. Varlack – PersonEntity: Name: NameFull: Benjamin deMayo – PersonEntity: Name: NameFull: Shira Kahn-Samuelson – PersonEntity: Name: NameFull: Natalie M. Gallagher – PersonEntity: Name: NameFull: Marjorie Rhodes – PersonEntity: Name: NameFull: Kristina R. Olson IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 01 Type: published Y: 2024 Identifiers: – Type: issn-print Value: 1524-8372 – Type: issn-electronic Value: 1532-7647 Numbering: – Type: volume Value: 25 – Type: issue Value: 5 Titles: – TitleFull: Journal of Cognition and Development Type: main |
| ResultId | 1 |
