View in EDS HTML Full Text PDF Full Text

Counterfactual Choices and Moral Judgments in Children

Saved in:
Bibliographic Details
Title: Counterfactual Choices and Moral Judgments in Children
Language: English
Authors: Gautam, Shalini (ORCID 0000-0001-6565-7458), Owen Hall, Ruby, Suddendorf, Thomas, Redshaw, Jonathan
Source: Child Development. Sep-Oct 2023 94(5):e296-e307.
Availability: Wiley. Available from: John Wiley & Sons, Inc. 111 River Street, Hoboken, NJ 07030. Tel: 800-835-6770; e-mail: cs-journals@wiley.com; Web site: https://www.wiley.com/en-us
Peer Reviewed: Y
Page Count: 12
Publication Date: 2023
Document Type: Journal Articles
Reports - Research
Descriptors: Decision Making, Moral Values, Developmental Stages, Child Development, Cognitive Processes, Story Telling, Foreign Countries
Geographic Terms: Australia
DOI: 10.1111/cdev.13943
ISSN: 0009-3920
1467-8624
Abstract: When making moral judgments of past actions, adults often think counterfactually about what could have been done differently. Considerable evidence suggests that counterfactual thinking emerges around age 6, but it remains unknown how this development influences children's moral judgments. Across two studies, Australian children aged 4-9 (N = 236, 142 Females) were told stories about two characters who had a choice that led to a good or bad outcome, and two characters who had no choice over a good or bad outcome. Results showed that 4- and 5-year-olds' moral judgments were influenced only by the actual outcome. From age 6, children's moral judgments were also influenced by the counterfactual choices that had been available to the characters.
Abstractor: As Provided
Notes: https://osf.io/aw5ks/?view_only=45577dcf234142eb95bd37e29815c505
Entry Date: 2023
Accession Number: EJ1394005
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwF_K_QrYWVF11UWVxmiUCu-AAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDAFSewbpoEZsIskJUwIBEICBmxOYM6lBA5yHpDN5ogejJnsGWy0TBnSbMbR59vb6kbkv53JVtJW-LCJIwzHdgLg4ot-Fzf9GOuouIcnxFXOiC3uar8Ch4t5iraAUepZNVh_dA8tR2RtYFHxogIUvedSXH5BSfV_7Svqlpn-FOIxkHSz9XpV7OpRNtj2uQ8AYkr-kyTTWhcVzWyJB59Y0SrzACI1Gcy4Q7xVGSsOo
Text:
  Availability: 1
  Value: <anid>AN0171385518;cdv01sep.23;2023Sep07.05:46;v2.2.500</anid> <title id="AN0171385518-1">Counterfactual choices and moral judgments in children </title> <p>When making moral judgments of past actions, adults often think counterfactually about what could have been done differently. Considerable evidence suggests that counterfactual thinking emerges around age 6, but it remains unknown how this development influences children's moral judgments. Across two studies, Australian children aged 4–9 (N = 236, 142 Females) were told stories about two characters who had a choice that led to a good or bad outcome, and two characters who had no choice over a good or bad outcome. Results showed that 4‐ and 5‐year‐olds' moral judgments were influenced only by the actual outcome. From age 6, children's moral judgments were also influenced by the counterfactual choices that had been available to the characters.</p> <p>When making moral judgments about past actions, adults typically account for the degree of choice an actor had. When evaluating behavior that led to a negative outcome, for example, someone who had more of a choice is generally judged more harshly than someone who had no choice at all (Cushman, [<reflink idref="bib7" id="ref1">7</reflink>]; Darley et al., [<reflink idref="bib9" id="ref2">9</reflink>]; Martin & Cushman, [<reflink idref="bib23" id="ref3">23</reflink>]; Monroe & Malle, [<reflink idref="bib26" id="ref4">26</reflink>]). "Choice" here entails that the person could have pursued a different course of action than the one actually taken. Incorporating choice into one's moral judgments, therefore, presupposes a capacity for <emph>counterfactual thinking</emph>, which refers to the ability to think about how the past may have turned out differently (Branscombe et al., [<reflink idref="bib2" id="ref5">2</reflink>]; Byrne, [<reflink idref="bib3" id="ref6">3</reflink>], [<reflink idref="bib4" id="ref7">4</reflink>]; Migliore et al., [<reflink idref="bib25" id="ref8">25</reflink>]; Williams et al., [<reflink idref="bib39" id="ref9">39</reflink>]).</p> <p>Children can imagine certain kinds of alternatives to reality early in life, as evidenced in pretend play from around age 2 (e.g., Lillard et al., [<reflink idref="bib21" id="ref10">21</reflink>]) and an ability to answer questions about alternative outcomes from around age 3 (e.g., Harris et al., [<reflink idref="bib14" id="ref11">14</reflink>]). However, when we refer to counterfactual thinking in this paper, we strictly mean the ability to imagine an alternative version of the past and a consequent alternative version of the present. This form of counterfactual thinking involves mentally adjusting one aspect of a past scenario while keeping other key aspects constant (McCormack et al., [<reflink idref="bib24" id="ref12">24</reflink>]; Rafetseder et al., [<reflink idref="bib32" id="ref13">32</reflink>]). For example, in a situation where a child has missed the bus to school and is late, they should consider not only the alternative scenario where they did not miss the bus to school but also keep in mind other pertinent facts (such that the bus ended up breaking down and so they would have been late regardless). Although one recent paper suggests that children may begin to think in this manner by age 4 (Nyhout & Ganea, [<reflink idref="bib29" id="ref14">29</reflink>]), many studies have found that it is not until 6 years or older that children show competence (e.g., Beck et al., [<reflink idref="bib1" id="ref15">1</reflink>]; Gautam et al., [<reflink idref="bib12" id="ref16">12</reflink>]; Kominsky et al., [<reflink idref="bib17" id="ref17">17</reflink>]; Leahy et al., [<reflink idref="bib20" id="ref18">20</reflink>]; McCormack et al., [<reflink idref="bib24" id="ref19">24</reflink>]; Rafetseder et al., [<reflink idref="bib32" id="ref20">32</reflink>]; Rafetseder & Perner, [<reflink idref="bib33" id="ref21">33</reflink>]). Similarly, only around age 6 do children show consistent evidence of experiencing counterfactual emotions such as regret or relief over alternative choices they could have made (e.g., O'Connor et al., [<reflink idref="bib30" id="ref22">30</reflink>]; Weisberg & Beck, [<reflink idref="bib38" id="ref23">38</reflink>]), including when their actual choices have negatively impacted others (Corbett et al., [<reflink idref="bib6" id="ref24">6</reflink>]; Uprichard & McCormack, [<reflink idref="bib36" id="ref25">36</reflink>]). Without the capacity for counterfactual thinking, younger children may not be capable of reflecting on their own or others' past options, and so may not be able to judge the morality of actions in the same manner that adults do.</p> <p>Two recent studies directly tested children's capacity to consider choices when judging a character's intentions. Zhao et al. ([<reflink idref="bib41" id="ref26">41</reflink>]) told children aged 4–6, as well as a group of adults, a story about someone who had a choice between one of three snacks, and picked in such a way that they either left an option of snacks for another person or not. That is, if there was an initial choice between two bananas and an orange, the character could pick a banana and leave an orange and a banana (i.e., an option of snacks), or they could pick the orange and leave two bananas (i.e., leaving no alternative option). The scenario varied according to whether the characters could leave an option of snacks, as in the previous example, or whether there was no choice to leave an option of snacks (e.g., there were only three bananas available to begin with). Six‐year‐old children and adults judged that it was nicer to leave options for the next person than to not leave options. However, they only judged it to be nicer to leave options when the character could have chosen to not leave options. Similarly, if the character had no choice but to leave no option of snacks for the next person, children judged that they were nicer than a character who did not leave options but could have chosen to. The 4‐ and 5‐year‐olds' judgments, by contrast, did not account for whether the characters could have chosen to leave an option of snacks. However, these young children also did not judge that it was nicer to leave an option of snacks at all. Therefore, they did not appear to understand the basic premise that leaving an option is the nicer thing to do.</p> <p>In another study by Pesowski et al. ([<reflink idref="bib31" id="ref27">31</reflink>]), 5‐ and 6‐year‐old children demonstrated that they could infer the preferences of a character by reflecting on that character's choices. The children were told that a character had made a choice between two toy castles. Children subsequently judged that the character preferred the chosen toy castle over the unchosen toy castle. However, if one of the toy castles was out of the reach of the character, 5‐ and 6‐year‐old children did <emph>not</emph> judge that the character preferred the chosen castle, whereas 3‐ and 4‐year‐olds did. This is presumably because the 5‐ and 6‐year‐old children understood that the character may have retrieved that castle merely because it was the only one available to them. Similar to the finding of (Zhao et al., [<reflink idref="bib41" id="ref28">41</reflink>]), the older children appeared to rely on information about counterfactual choices to infer a character's intentions.</p> <p>Understanding someone's intentions is a key factor in moral judgments, at least in Western cultures (Cushman, [<reflink idref="bib7" id="ref29">7</reflink>]). Research has found that between the ages of 4 and 8, children increasingly judge accidental harms based on the intentions of others, rather than based on the negative outcome per se (e.g., Chernyak & Sobel, [<reflink idref="bib5" id="ref30">5</reflink>]; Cushman et al., [<reflink idref="bib8" id="ref31">8</reflink>]; Margoni & Surian, [<reflink idref="bib22" id="ref32">22</reflink>]; Nobes et al., [<reflink idref="bib28" id="ref33">28</reflink>]; Yuill & Perner, [<reflink idref="bib40" id="ref34">40</reflink>]). For example, they begin to understand that an accidental harm is 'less wrong' than a failed attempt to harm (Cushman et al., [<reflink idref="bib8" id="ref35">8</reflink>]). However, children may not need to think counterfactually about what a character could have done when making such judgments. Rather, they could simply reflect on what the character actually meant to do. To clarify the difference, consider one of the stories told to children by Cushman et al. ([<reflink idref="bib8" id="ref36">8</reflink>]) about accidental and attempted wrongdoings:</p> <p>One boy accidentally steals an apple after it rolls into his shopping basket when he isn't looking. Another boy attempts to steal an apple, but it rolls out of his shopping basket when he isn't looking.</p> <p>Following this story, children were asked if the characters were being "bad or naughty." Indeed, children might answer such questions by thinking counterfactually about what could have happened (i.e., the character who accidentally took the apple did not choose to do so, whereas the character who did steal could have chosen not to). However, they could also simply understand the premise—grounded in actual reality—that an intention to steal is naughty, but accidents are not. Both lines of reasoning would lead to the same response: that accidental harms are less naughty than attempted harms. Therefore, the extent to which children rely on counterfactual thinking when making such moral judgments remains unclear.</p> <p>There is some evidence that children make judgments of fairness based partly on the choices someone had available to them (Jaroslawska et al., [<reflink idref="bib16" id="ref37">16</reflink>]; Sutter, [<reflink idref="bib35" id="ref38">35</reflink>]). For example, Jaroslawska et al. ([<reflink idref="bib16" id="ref39">16</reflink>]) asked children aged 6–8 years to judge the fairness of an offer from a proposer in the context of a mini‐ultimatum game. Results showed that the alternatives available to the proposer influenced children's judgments, suggesting that children were considering what the proposer could have chosen to do when judging what they actually did. However, given that multiple lines of evidence point to critical transformations in counterfactual thinking around age 6 (e.g., Beck et al., [<reflink idref="bib1" id="ref40">1</reflink>]; Kominsky et al., [<reflink idref="bib17" id="ref41">17</reflink>]; Leahy et al., [<reflink idref="bib20" id="ref42">20</reflink>]; McCormack et al., [<reflink idref="bib24" id="ref43">24</reflink>]; Rafetseder et al., [<reflink idref="bib32" id="ref44">32</reflink>]; Rafetseder & Perner, [<reflink idref="bib33" id="ref45">33</reflink>]), it stands to reason that younger children might not be similarly influenced by the presence of counterfactual alternatives when making moral judgments. It also remains unknown whether, and if so when, children consider counterfactuals in the context of moral judgments about third‐party interactions.</p> <p>Across two studies, we sought to confirm the role of counterfactual thinking in children's moral judgments. In Study 1, children aged 4–9 years were told a story about four characters who brought a snack to share with a friend, on that friend's birthday. Two of the characters had a choice over which snack they brought—either one their friend likes or one their friend dislikes—and two of the characters had no choice. If children can account for counterfactual choices in their moral judgments, then we would expect them to judge the characters who had a choice as nicer if they chose the snack their friend likes, or meaner if they chose the snack their friend does not like than the corresponding characters who had no choice. Conversely, children who do not account for counterfactual choices should rate characters who bring the bad snack (i.e., the snack their friend dislikes) as being similarly "mean," and those who bring the good snack (i.e., the snack their friend likes) as being similarly "nice," regardless of whether they had a choice.</p> <p>Finally, it is plausible that choosing to be mean may be more salient to children than choosing to be nice since the latter is the typical response when giving birthday gifts. There is also evidence that children more readily consider "upward" counterfactuals (comparing what happened to a better alternative), than "downward" counterfactuals (comparing what happened to a worse alternative), as they seem to experience the negative counterfactual emotion of regret before they experience its positive counterpart, relief (Gautam et al., [<reflink idref="bib11" id="ref46">11</reflink>]; Weisberg & Beck, [<reflink idref="bib37" id="ref47">37</reflink>]). We therefore suspect that across ages, children may find it easier to judge the difference between the two characters who brought the bad snack than the two characters who brought the good snack.</p> <hd id="AN0171385518-2">STUDY 1</hd> <p></p> <hd id="AN0171385518-3">Method</hd> <p></p> <hd id="AN0171385518-4">Participants</hd> <p>The final sample consisted of 122 children ranging from 4 to 9 years (<emph>M</emph> = 83.83 months, SD = 19.87 months, range = 48–118 months), comparable to similar studies (Pesowski et al., [<reflink idref="bib31" id="ref48">31</reflink>]; Zhao et al., [<reflink idref="bib41" id="ref49">41</reflink>]). We chose this age range with the goal of tracking children's moral judgments both before and after the age at which they typically begin to show compelling evidence of counterfactual thinking. As per our preregistration (https://osf.io/aw5ks/?view%5fonly=45577dcf234142eb95bd37e29815c505), we collected a minimum of 40 children per age group: forty‐one 4‐ to 5‐year‐olds (18 females and 23 males, <emph>M</emph> = 61.24 months, SD = 6.98 months, range = 48–71 months), forty 6‐ to 7‐year‐olds (20 females and 20 males, <emph>M</emph> = 83.55 months, SD = 7.75 months, range = 72–95 months), and forty‐one 8‐ to 9‐year‐olds (26 females and 15 males, <emph>M</emph> = 106.68 months, SD = 5.36 months, range = 96–118 months). An additional five children were excluded due to inattention (<emph>n</emph> = 3) or parental interference (<emph>n</emph> = 2). The sample was mostly White and middle‐class, from an Australian city. All children were tested at a local museum and received a small armband for participating.</p> <hd id="AN0171385518-5">Materials and procedure</hd> <p>Children were presented with a story on an iPad. Before the main story was told, children were shown a training procedure in which images of items in a cupboard dropped down into a character's possession. This was to make sure children understood the relationship between the items and characters they would see in the main story. Children were also introduced to a 5‐point "nice‐to‐mean" scale (see Figure 2). In past research on moral judgments (e.g., Nobes et al., [<reflink idref="bib27" id="ref50">27</reflink>], [<reflink idref="bib28" id="ref51">28</reflink>]), children have been asked to respond to questions about how good or bad a character is acting without any visible scale. Therefore, we designed the current scale of faces to make responses to such questions more intuitive to children (e.g., Ingram & Romero, [<reflink idref="bib15" id="ref52">15</reflink>]).</p> <p>Following this introduction, children were told a story about four friends who were going to bring a snack to another friend's birthday to share. Characters were gender matched with the participating child, such that they were either shown stories with all boys or all girls. Children were told that the birthday girl/boy loved liquorice but did not like chocolate, whereas the four friends of the birthday girl/boy all loved chocolate but did not like liquorice. Following this, each friend proceeded to select a snack from their cupboard at home (see Figure 1). Critically, two of the friends had a choice between chocolate and liquorice, and the other two friends only had one option (i.e., either chocolate <emph>or</emph> liquorice). Henceforth, we will refer to liquorice as the "good snack" and chocolate as the "bad snack." Children were asked several comprehension questions that checked their understanding of the important aspects of the story, such as what each character liked best (see OSF for script). If children answered any of these comprehension questions incorrectly, they were reminded of the correct answer.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/CDV/01sep23/cdev13943-fig-0001.jpg?ephost1=dGJyMNHX8kSepq84v%2bvlOLCmsE6epq5Srqa4SK6WxWXS" alt="cdev13943-fig-0001.jpg" title="1 Depiction of four characters with the various contents of their cupboards. Two characters had a choice, and two characters did not. Arrows indicate each character's selection. Note that in the actual iPad story, each friend and cupboard combination was shown on a separate page to minimize confusion. Further, in the iPad story, selected snacks were animated to drop down from the cupboard, rather than being represented with arrows as in this diagram." /> </p> <p></p> <hd id="AN0171385518-7">Nice‐to‐mean scale</hd> <p>Of the two characters who had a choice, one selected the bad snack, and the other selected the good snack. The two characters without a choice simply selected whatever snack was in their cupboard. Children were then asked to rate how nice or mean each character was being on the 5‐point nice‐to‐mean scale (e.g., "How is Sophia acting towards Clara? Very nice, a little nice, just ok, a little mean or very mean? Why?"). To reduce memory demands when children made these ratings, all the information regarding the choice (or lack thereof) that each character had was again depicted on the screen, and the key elements of the story were repeated (see Figure 2).</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/CDV/01sep23/cdev13943-fig-0002.jpg?ephost1=dGJyMNHX8kSepq84v%2bvlOLCmsE6epq5Srqa4SK6WxWXS" alt="cdev13943-fig-0002.jpg" title="2 An example of a slide where children made the rating on the nice‐to‐mean scale. When children rated each character, they saw the entirety of the situation, and key elements such as the chosen snack, what the birthday child liked best, and what was left in the cupboard were repeated. In this slide, the character had a choice and selected the "good snack." The scale of faces was depicted and explained to children before testing began. Children were asked which faces they thought looked nice and which faces they thought looked mean. Next, the scale was described as follows (from left to right): this face is very nice, this face is a little nice, this face is not nice or mean—it is just okay, this face is a little mean, and this face is very mean." /> </p> <p></p> <hd id="AN0171385518-9">Comparison questions</hd> <p>Children were shown the two characters who were bringing the bad snack (i.e., the one with a choice and the one without a choice) side by side and asked to rate who was the <emph>meanest</emph> of the two (i.e., "Who was the meanest?"). They were also shown the two characters who were bringing the good snack (i.e., the one with a choice and the one without a choice) side by side and asked to rate who was the <emph>nicest</emph> of the two (i.e., "Who was the nicest?"). (see Figure 3).</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/CDV/01sep23/cdev13943-fig-0003.jpg?ephost1=dGJyMNHX8kSepq84v%2bvlOLCmsE6epq5Srqa4SK6WxWXS" alt="cdev13943-fig-0003.jpg" title="3 Two separate slides depicting the comparison between the characters who selected the bad snack (top) and the good snack (bottom). The chosen snack is located next to the cartoon character." /> </p> <p></p> <hd id="AN0171385518-11">Counterbalancing</hd> <p>We counterbalanced the order of snack selections shown to children, the order of the test questions, and the cartoon characters depicted as making each selection (see Supporting Information). We did not counterbalance the target snack as we wanted to keep the two choices consistent with typical preferences. That is, it is more common for children to like chocolate and less common for them to not.</p> <hd id="AN0171385518-12">Results</hd> <p></p> <hd id="AN0171385518-13">Nice‐to‐mean scale</hd> <p>As preregistered, a within‐subjects 2 (choice/no choice) × 2 (good snack/bad snack) GLM was conducted on children's judgments on the nice‐to‐mean scale, with age in months (mean‐centered) included as a between‐subjects factor. Results revealed a main effect of snack type, such that characters who brought the good snack (<emph>M</emph> = 1.17, SD = 0.39) were rated as nicer than those who brought the bad snack (<emph>M</emph> = 3.81, SD = 1.13), irrespective of choice, <emph>F</emph>(<reflink idref="bib1" id="ref53">1</reflink>, 121) = 525.03, <emph>p</emph> < .001, <ephtml> <math altimg="urn:x-wiley:00093920:media:cdev13943:cdev13943-math-0001" display="inline" overflow="scroll" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msubsup><mi>η</mi><mi mathvariant="normal">p</mi><mn>2</mn></msubsup></mrow></semantics></math> </ephtml>  = .814. There was no main effect of choice <emph>F</emph>(<reflink idref="bib1" id="ref54">1</reflink>, 121) = 0.109, <emph>p</emph> = .742, <ephtml> <math altimg="urn:x-wiley:00093920:media:cdev13943:cdev13943-math-0002" display="inline" overflow="scroll" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msubsup><mi>η</mi><mi mathvariant="normal">p</mi><mn>2</mn></msubsup></mrow></semantics></math> </ephtml>  = .001. There was an interaction between choice and snack, such that children's judgments of the snack characters selected depended on the choice characters had available to them, <emph>F</emph>(<reflink idref="bib1" id="ref55">1</reflink>, 121) = 56.44, <emph>p</emph> < .001, <ephtml> <math altimg="urn:x-wiley:00093920:media:cdev13943:cdev13943-math-0003" display="inline" overflow="scroll" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msubsup><mi>η</mi><mi mathvariant="normal">p</mi><mn>2</mn></msubsup></mrow></semantics></math> </ephtml>  = .320. Critically, this relationship between choice and snack significantly interacted with age, <emph>F</emph>(<reflink idref="bib1" id="ref56">1</reflink>, 121) = 13.10, <emph>p</emph> < .001, <ephtml> <math altimg="urn:x-wiley:00093920:media:cdev13943:cdev13943-math-0004" display="inline" overflow="scroll" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msubsup><mi>η</mi><mi mathvariant="normal">p</mi><mn>2</mn></msubsup></mrow></semantics></math> </ephtml>  = .098. As preregistered, this three‐way interaction was followed up with within‐subjects <emph>t</emph> tests that investigated children's responses across conditions separately for the three age groups (4‐ to 5‐year‐olds, 6‐ to 7‐year‐olds, and 8‐ to 9‐year‐olds). We used a Bonferroni correction for multiple analyses, resulting in an adjusted alpha of.017. For each age group, we compared children's ratings on the nice‐to‐mean scale for characters who chose to bring the good snack to the character who brought the good snack but did not have a choice. Likewise, we compared children's ratings for characters who chose to bring the bad snack to the character who brought the bad snack but did not have a choice.</p> <p>As seen in Figure 4, these follow‐up tests showed that 4‐ to 5‐year‐olds did not rate characters as significantly different based on choice, either for the good snack, <emph>t</emph>(<reflink idref="bib40" id="ref57">40</reflink>) = −1.09, <emph>p</emph> = .281, or for the bad snack <emph>t</emph>(<reflink idref="bib40" id="ref58">40</reflink>) = 1.83, <emph>p</emph> = .075. Six‐ to 7‐year‐olds, however, rated the character who chose to bring the bad snack as significantly meaner (<emph>M</emph> = 4.10, SD = 1.00) than the character who brought the bad snack but had no choice (<emph>M</emph> = 3.50, SD = 1.28), <emph>t</emph>(<reflink idref="bib39" id="ref59">39</reflink>) = 3.59, <emph>p</emph> = .001, although children of this age did not rate the characters who brought the good snack significantly differently, <emph>t</emph>(<reflink idref="bib39" id="ref60">39</reflink>) = −0.374, <emph>p</emph> = .711. Finally, the 8‐ to 9‐year‐olds rated the character who chose the bad snack as significantly meaner (<emph>M</emph> = 4.07, SD = .96) than the one who had no choice but to bring the bad snack (<emph>M</emph> = 3.37, SD = .92), <emph>t</emph>(<reflink idref="bib39" id="ref61">39</reflink>) = 5.57, <emph>p</emph> < .001, and also rated the character who had a choice and brought the good snack as significantly nicer (<emph>M</emph> = 1.02, SD = 0.16) than the character who brought the good snack but had no choice (<emph>M</emph> = 1.61, SD = 0.83), <emph>t</emph>(<reflink idref="bib39" id="ref62">39</reflink>) = −4.48, <emph>p</emph> < .001.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/CDV/01sep23/cdev13943-fig-0004.jpg?ephost1=dGJyMNHX8kSepq84v%2bvlOLCmsE6epq5Srqa4SK6WxWXS" alt="cdev13943-fig-0004.jpg" title="4 Children's rating of characters on the nice‐to‐mean scale. **p < .001. Error bars represent standard error. Very nice = 1 to Very mean = 5 (see Figure 2)." /> </p> <p></p> <hd id="AN0171385518-15">Comparison questions</hd> <p>To analyze the answers to the direct comparison questions (see Figure 3), we conducted preregistered binomial tests against a 50% chance level. During testing, we did not correct children who made statements to the effect that both characters were equally mean or nice. There were 27 children who reported this sentiment at least once: 22 in the 4‐ and 5‐year‐old age group (<emph>n</emph> = 21 for the good snack, <emph>n</emph> = 11 for the bad snack), 5 in the 6‐ and 7‐year‐old age group (<emph>n</emph> = 5 for the good snack, <emph>n</emph> = 3 for the bad snack), and none in the 8‐ to 9‐year old age group. These children were excluded from the binomial tests against chance, and we discuss the consequences of this response method when introducing Study 2 below.</p> <p>Within the remaining sample, children were significantly more likely than chance to report that the character who chose to bring the good snack was nicer than the character who had no choice and brought the good snack (78/96 = 81.25%, <emph>p</emph> < .001), and that the character who chose to bring the bad snack was meaner than the character who brought the bad snack but had no choice (97/108 = 89.81%, <emph>p</emph> < .001). Point‐biserial correlations revealed that with increasing age, children were significantly more likely to account for choice when judging which character was meaner or nicer, both for the characters who brought the good snack, <emph>r</emph><subs>pb</subs> = .405, <emph>p</emph> < .001, and for the characters who brought the bad snack, <emph>r</emph><subs>pb</subs> = .427, <emph>p</emph> < .001. We, therefore, conducted follow‐up binomial tests for each age group (see Table 1). Once again, we used an adjusted alpha of.017 to correct for multiple analyses.</p> <p>1 TABLE Binomial test results by age group.</p> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left">Age group</th><th align="left">Good snack, (number correct/included <italic>n</italic>)</th><th align="left"><italic>p</italic></th><th align="left">Bad snack, (number correct/included <italic>n</italic>)</th><th align="left"><italic>p</italic></th></tr></thead><tbody valign="top"><tr><td>4–5 year‐olds (n = 41)</td><td>11/20 (50%)</td><td>.584</td><td>20/30 (66.67%)</td><td>.049</td></tr><tr><td>6–7 year‐olds (n = 40)</td><td>29/35 (80.56%)</td><td><.001</td><td>36/37 (97.30%)</td><td><.001</td></tr><tr><td>8–9 year‐olds (n = 41)</td><td>38/41 (95%)</td><td><.001</td><td>41/41 (100%)</td><td><.001</td></tr></tbody></table> </ephtml> </p> <p>1 <emph>Note</emph>: 'Number correct' refers to children who identified that the character with a choice was nicer for choosing the good snack, and meaner for choosing the bad snack, compared with the characters who had no choice but brought the good or bad snack respectively. "Included <emph>n</emph>" indicates the number of children who responded by selecting a character as either being nicer or meaner for the comparison questions (children who stated that both characters were equal have been excluded).</p> <p>When asked who was meaner, 6‐ to 7‐ and 8‐ to 9‐year‐olds selected the character with a choice significantly more often than chance, whereas 4‐ to 5‐year‐olds selected at levels no different from chance. When asked who was nicer, 6‐ to 7‐ and 8‐ to 9‐year‐olds again selected the character who had a choice significantly more often than chance, whereas 4‐ to 5‐year‐olds again selected at levels no different from chance. A McNemar test revealed that across all ages, children were more likely to select the correct option (i.e., the character with a choice) for bad snacks than for good snacks, <emph>p</emph> = .013.</p> <hd id="AN0171385518-16">Discussion</hd> <p>In study 1, we found evidence that from the age of 6 children's judgments were influenced by what that character could have chosen to do on both the nice‐to‐mean scale and the comparison questions. By contrast, there was no evidence that 4‐ and 5‐year‐olds accounted for what a character could have done when making such judgments. These findings provide initial evidence that counterfactual thinking has important consequences for children's moral judgments. In study 2, we aimed to replicate this pattern in a second story with an alternative narrative context. We also made a number of changes to the methodology to address some potential limitations with study 1.</p> <hd id="AN0171385518-17">STUDY 2</hd> <p>It is possible that some children in study 1 rated characters differently not because of the counterfactual items available to the characters with a choice, but simply because the characters with a choice always selected between two items, whereas the characters without a choice always selected the single item available. For example, some children might have rated characters differently depending on whether they selected the last item in the cupboard, or if there was another item left over. In study 2, we, therefore, increased the structural similarity between the choice and no choice trials by including two options across both trials, although for the choice trials, the two items were different, whereas for the no choice trials, the two items were identical. It is also possible that some children in study 1 believed that the character would get to keep whatever they did not bring as it was left behind in the cupboard. Indeed, children may have believed that a character who had a choice and who brought liquorice (a good snack) to the party was primarily motivated by self‐interest in keeping the chocolate for themselves. In study 2, the characters therefore made their selections at a shop, so that any item left behind would not remain in possession of these characters afterward.</p> <p>In study 1, we did not compel children to select one of the two characters for the comparison questions, such that responses equating the morality of the characters were not corrected. And consistent with Guttentag and Ferrell's ([<reflink idref="bib13" id="ref63">13</reflink>]) research on counterfactual emotions, it was the younger children who were more likely to equate characters with and without a counterfactual alternative. In study 2, we aimed to increase statistical sensitivity by encouraging children to select one character only. Further, because the intentions of characters without a choice are ambiguous, this time we asked the children "who was <emph>definitely</emph> being mean/nice?" rather than "who was the nicest/meanest?" Finally, we removed the color from the scale of faces in study 2 (in case this factor influenced their judgments), and we also made all the characters have neutral expressions to avoid children simply matching each character's facial expressions to the face scale.</p> <hd id="AN0171385518-18">Method</hd> <p></p> <hd id="AN0171385518-19">Participants</hd> <p>Study 2 consisted of 114 children ranging from 4 to 9 years (<emph>M</emph> = 82.63 months, SD = 19.22 months, range = 48 to 117 months). As per our preregistration (https://osf.io/aw5ks/?view%5fonly=45577dcf234142eb95bd37e29815c505), we collected a minimum of 30 children per age group: thirty‐nine 4‐ to 5‐year‐olds (24 females and 15 males, <emph>M</emph> = 60.89 months, SD = 7.36 months, range = 48 to 71 months), forty‐one 6‐ to 7‐year‐olds (33 females and 8 males, <emph>M</emph> = 84.00 months, SD = 6.52 months, range = 72 to 95 months), and thirty‐four 8‐ to 9‐year‐olds (21 females and 13 males, <emph>M</emph> = 106.03 months, SD = 5.81 months, range = 96 to 117 months). One child was excluded due to inattention. The sample consisted of mainly White, middle‐class children from an Australian city. Children were tested at a local museum (<emph>n</emph> = 88) and within a developmental psychology laboratory (<emph>n</emph> = 26) and received a small prize for participating.</p> <hd id="AN0171385518-20">Procedure</hd> <p>Children were told a story that was depicted on an iPad. As in study 1, we provided initial training to make sure they understood key animations, for example that an item dropping down from a shelf indicated that it had been selected. In study 2, the story centered around a main character who was hosting a "Lego Party." Although the main character loved playing with Lego, friends that were invited to the party all preferred to play with playdough. The character hosting the Lego party asked the friends to bring something to the party, and each child visited a shop to buy one item to bring. Following this introduction to the story, we asked children a number of comprehensions check questions to make sure they understood the key components of the story, such as what toy each character liked best (see Supporting Information for the full script). As in study 1, two characters ended up bringing Lego, and two characters brought playdough. Two characters had a choice between different items and two characters had no choice, although this time all characters selected between two items in the shop (see Figure 5). Children were then asked to make ratings on how nice or mean each character was acting on a black‐and‐white scale of faces. When children made their ratings they were reminded of the key aspects of the story, such as what the character had selected, and what options had been available (see Figure 6).</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/CDV/01sep23/cdev13943-fig-0005.jpg?ephost1=dGJyMNHX8kSepq84v%2bvlOLCmsE6epq5Srqa4SK6WxWXS" alt="cdev13943-fig-0005.jpg" title="5 A depiction of the four trial types. Characters in the story either had a choice and chose the "good toy" (i.e., the Lego) or the "bad toy" (i.e., the playdough) or they had no choice but to select the good toy or bad toy." /> </p> <p></p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/CDV/01sep23/cdev13943-fig-0006.jpg?ephost1=dGJyMNHX8kSepq84v%2bvlOLCmsE6epq5Srqa4SK6WxWXS" alt="cdev13943-fig-0006.jpg" title="6 An example of the images children saw when rating how nice or mean characters were acting for characters with a choice (top panel) and without a choice (bottom panel). These examples both show a character selecting a good toy." /> </p> <p></p> <hd id="AN0171385518-23">Comparison questions</hd> <p>We asked children to select who was definitely being nice (see Figure 7) between the two characters who selected the "good toy" (i.e., Lego for the Lego party), and who was definitely being mean between the two characters who selected the "bad toy" (i.e., playdough for the Lego party).</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/CDV/01sep23/cdev13943-fig-0007.jpg?ephost1=dGJyMNHX8kSepq84v%2bvlOLCmsE6epq5Srqa4SK6WxWXS" alt="cdev13943-fig-0007.jpg" title="7 An example of the images children saw for the comparison questions (for this example children were asked who was definitely being nice)." /> </p> <p></p> <hd id="AN0171385518-25">Counterbalancing</hd> <p>We followed the same counterbalancing procedures as in study 1 (see Supporting Information).</p> <hd id="AN0171385518-26">Results and discussion</hd> <p></p> <hd id="AN0171385518-27">Nice‐to‐mean scale</hd> <p>As preregistered, a within‐subjects 2 (choice/no choice) × 2 (good toy/bad toy) GLM was conducted on children's judgments on the nice‐to‐mean scale, with age in months (mean‐centered) included as a between‐subjects factor. Results revealed a main effect of toy type, such that characters who brought the good toy (<emph>M</emph> = 1.37, SD = 0.75) were rated as nicer than those who brought the bad toy (<emph>M</emph> = 3.47, SD = 1.07), irrespective of choice, <emph>F</emph>(<reflink idref="bib1" id="ref64">1</reflink>, 113) = 334.44, <emph>p</emph> < .001, <ephtml> <math altimg="urn:x-wiley:00093920:media:cdev13943:cdev13943-math-0005" display="inline" overflow="scroll" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msubsup><mi>η</mi><mi mathvariant="normal">p</mi><mn>2</mn></msubsup></mrow></semantics></math> </ephtml>  = .751. There was no main effect of choice, <emph>F</emph>(<reflink idref="bib1" id="ref65">1</reflink>, 113) = 0.145 <emph>p</emph> = .704, <ephtml> <math altimg="urn:x-wiley:00093920:media:cdev13943:cdev13943-math-0006" display="inline" overflow="scroll" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msubsup><mi>η</mi><mi mathvariant="normal">p</mi><mn>2</mn></msubsup></mrow></semantics></math> </ephtml>  = .001, but there was a significant interaction between toy and choice, <emph>F</emph>(<reflink idref="bib1" id="ref66">1</reflink>, 113) = 69.25, <emph>p</emph> < .001, <ephtml> <math altimg="urn:x-wiley:00093920:media:cdev13943:cdev13943-math-0007" display="inline" overflow="scroll" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msubsup><mi>η</mi><mi mathvariant="normal">p</mi><mn>2</mn></msubsup></mrow></semantics></math> </ephtml>  = .384. Critically, and as in study 1, this relationship between choice and toy significantly interacted with age, <emph>F</emph>(<reflink idref="bib1" id="ref67">1</reflink>, 113) = 28.68, <emph>p</emph> = <.001, <ephtml> <math altimg="urn:x-wiley:00093920:media:cdev13943:cdev13943-math-0008" display="inline" overflow="scroll" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msubsup><mi>η</mi><mi mathvariant="normal">p</mi><mn>2</mn></msubsup></mrow></semantics></math> </ephtml>  = .205. As preregistered, this three‐way interaction was followed up with within‐subjects <emph>t</emph> tests that investigated children's responses across conditions separately for the three age groups (4‐ to 5‐year‐olds, 6‐ to 7‐year‐olds, and 8‐ to 9‐year‐olds). We used a Bonferroni correction for multiple analyses, resulting in an adjusted alpha of.017. For each age group, we compared children's ratings on the nice‐to‐mean scale for characters who chose to bring the good toy to the character who brought the good toy but did not have a choice. Likewise, we compared children's ratings for characters who chose to bring the bad toy to the character who brought the bad toy but did not have a choice.</p> <p>As seen in Figure 7, these follow‐up tests showed that, as in study 1, 4‐ to 5‐year‐olds did not rate characters as significantly different based on choice, either for the good snack, <emph>t</emph>(<reflink idref="bib38" id="ref68">38</reflink>) = 0.00, <emph>p</emph> > .999, or the bad snack <emph>t</emph>(<reflink idref="bib38" id="ref69">38</reflink>) = 2.32, <emph>p</emph> = .026. However, somewhat in contrast to study 1, 6‐ to −7‐ year‐olds rated characters who chose to bring the good toy as significantly nicer (<emph>M</emph> = 1.22, SD = 0.57) than characters who brought the good toy but had no choice (<emph>M</emph> = 1.61, SD = 1.0), <emph>t</emph>(<reflink idref="bib40" id="ref70">40</reflink>) = −2.90, <emph>p</emph> = .006, but did not rate the characters who brought the bad toy as significantly differently, <emph>t</emph>(<reflink idref="bib40" id="ref71">40</reflink>) = 1.64, <emph>p</emph> = .110 (i.e., the opposite pattern of significance compared to study 1). Finally, and as in study 1, the 8‐ to 9‐ year‐olds rated the character who chose the bad toy as significantly meaner (<emph>M</emph> = 3.82, SD = .94) than the one who had no choice but to bring the bad toy (<emph>M</emph> = 3.03, SD = .94), <emph>t</emph>(<reflink idref="bib33" id="ref72">33</reflink>) = 4.13, <emph>p</emph> < .001, and also rated the character who had a choice and brought the good snack as significantly nicer (<emph>M</emph> = 1.15, SD = 0.44) than the character who brought the good snack but had no choice (<emph>M</emph> = 1.97, SD = 0.94), <emph>t</emph>(<reflink idref="bib33" id="ref73">33</reflink>) = −5.13, <emph>p</emph> < .001 (Figure 8).</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/CDV/01sep23/cdev13943-fig-0008.jpg?ephost1=dGJyMNHX8kSepq84v%2bvlOLCmsE6epq5Srqa4SK6WxWXS" alt="cdev13943-fig-0008.jpg" title="8 Children's rating of characters on the nice‐to‐mean scale. *p < .01, **p < .001. Error bars represent standard error. Very nice = 1 to very mean = 5 (see Figure 6)." /> </p> <p></p> <hd id="AN0171385518-29">Comparison questions</hd> <p>As in study 2, we conducted preregistered binomial tests against a 50% chance level to analyze the comparison questions. In study 2, we also compelled children to select only one character, and all children did so for both comparison questions. Overall, children were significantly more likely to report that the character who chose to bring the good toy was definitely nice than the character who had no choice and brought the good toy (84/114 = 73.68%, <emph>p</emph> < .001), and that the character who chose to bring the bad toy was definitely mean than the character who brought the bad toy but had no choice (83/114 = 72.81%, <emph>p</emph> < .001). As in study 1, point‐biserial correlations revealed that with increasing age, children were significantly more likely to account for choice when judging which character was meaner or nicer, both for the characters who brought the good toy, <emph>r</emph><subs>pb</subs> = .35, <emph>p</emph> < .001, and for the characters who brought the bad toy, <emph>r</emph><subs>pb</subs> = .56, <emph>p</emph> < .001. We, therefore, conducted follow‐up binomial tests for each age group with an alpha level of.017 (see Table 2).</p> <p>2 TABLE Binomial test results by age group study 2.</p> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left">Age group</th><th align="left">Good toy, number correct</th><th align="left"><italic>p</italic></th><th align="left">Bad toy, number correct</th><th align="left"><italic>p</italic></th></tr></thead><tbody valign="top"><tr><td>4‐ to 5‐year‐olds (n = 39)</td><td>21 (53.85%)</td><td>.375</td><td>14 (35.90%)</td><td>.973</td></tr><tr><td>6‐ to 7‐year‐olds (n = 41)</td><td>31 (75.61%)</td><td>.001</td><td>37 (90.24%)</td><td><.001</td></tr><tr><td>8‐ to 9‐year‐olds (n = 34)</td><td>32 (94.12%)</td><td><.001</td><td>32 (94.12%)</td><td><.001</td></tr></tbody></table> </ephtml> </p> <p>2 <emph>Note</emph>: "Number correct"' refers to children who identified that the character with a choice was nicer for choosing the good toy, and meaner for choosing the bad toy, compared with the characters who had no choice but brought the good or bad toy respectively.</p> <p>Overall, the pattern of results across the age groups was similar to study 1. When asked who was definitely mean, 6‐ to 7‐ and 8‐ to 9‐year‐olds selected the character with a choice significantly more often than chance, whereas 4‐ to 5‐year‐olds were selecting at chance. Similarly, when asked who was definitely nice, 6‐ to 7‐year‐olds and 8‐ to 9‐year‐olds selected the character who had a choice significantly more often than chance, whereas 4‐ to 5‐year‐olds were at chance. A McNemar test revealed that, unlike in study 1, children across ages were no more likely to select the correct option (i.e., the character with a choice) for bad snacks than for good snacks, <emph>p</emph> > .999.</p> <hd id="AN0171385518-30">GENERAL DISCUSSION</hd> <p>Our two studies examined whether children consider counterfactual alternatives when making moral judgments of past actions. We asked children to make individual character judgments on a nice‐to‐mean scale and to directly compare characters side‐by‐side. Across both studies and measures, we found consistent and compelling evidence that 8‐ and 9‐year‐olds were considering what characters could have chosen to do when judging their actions. That is, 8‐ and 9‐year‐olds rated characters as nicer or meaner if they made a choice that led to a good or bad outcome than if they did not have a choice over the respective outcome. Both studies also yielded good evidence that 6‐ and 7‐year‐olds were considering what characters could have chosen to do, although this tendency was not always consistent. By contrast, there was no evidence that 4‐ and 5‐year‐olds were considering what characters could have chosen to do when making moral judgments or comparisons. Overall, this pattern across the three age groups is in line with the broader literature on the emergence of counterfactual thinking (e.g., Kominsky et al., [<reflink idref="bib17" id="ref74">17</reflink>]; McCormack et al., [<reflink idref="bib24" id="ref75">24</reflink>]; Rafetseder & Perner, [<reflink idref="bib33" id="ref76">33</reflink>]) and emotions (e.g., O'Connor et al., [<reflink idref="bib30" id="ref77">30</reflink>]; Weisberg & Beck, [<reflink idref="bib38" id="ref78">38</reflink>]). Next, we will discuss the findings for the 4‐ and 5‐year‐olds and 6‐ and 7‐year‐olds in more detail.</p> <p>For the nice‐to‐mean scale ratings, one potential explanation for the lack of effects in 4‐ and 5‐year‐olds is that these young children tended to fixate on the end points of the scales. However, we note that across both studies, even the 4‐ and 5‐year‐olds were not always making ceiling ratings, especially for their judgments of how mean characters were being (see Figures 4 and 7 and see Supporting Information for frequency information). There is also evidence that children of this age can use such scales to report emotions of varying intensity and do not simply fixate on the end points (Gautam et al., [<reflink idref="bib10" id="ref79">10</reflink>]). Further, unlike in previous work (Zhao et al., [<reflink idref="bib41" id="ref80">41</reflink>]), it is clear from our results that children across all ages understood the premise that one should bring something to a host's house that the host likes (either in the form of food or a toy), and that it is mean to not do so. Therefore, it appears that across both studies, the 4‐ and 5‐year‐old children were simply rating characters based on the selected item itself, without considering counterfactuals. Indeed, in both study 1 and study 2 there was also no evidence that the 4‐ and 5‐year‐old children were considering what characters could have chosen to do on the more sensitive direct comparison questions.</p> <p>On study 1's face scale measure, 6‐and 7‐year‐olds rated characters who had a choice and selected an item that led to a bad outcome as significantly meaner than characters who had no choice but to select the item that led to a bad outcome, whereas they did not show this pattern for the characters who selected the item leading to the good outcome. In study 2, however, children of this age showed the opposite pattern of significance. This cross‐study variation in the face scale effects for 6‐ to‐7‐year‐olds may reflect the relative insensitivity of the scale, including the aforementioned tendency for some children to respond using the end points only. Indeed, when asked to directly compare two options, children of this age—but not younger children—consistently identified the character with the counterfactual option as being nicer/meaner (in study 1) or definitely nice/mean (in study 2) than the one without a choice. It is also plausible that children's conceptual knowledge of the specific story elements in each study may have influenced their consideration of the relevant counterfactuals (e.g., Kominsky et al., [<reflink idref="bib17" id="ref81">17</reflink>]; Kushnir, [<reflink idref="bib19" id="ref82">19</reflink>]), and this could be an interesting direction for future research to explore. Overall, however, 6‐ to 7‐year‐olds clearly showed that they are able to consider what characters could have chosen to do.</p> <p>It seems likely that 6‐ to 9‐year‐old children were inferring the intentions of the characters based on the choices that were available to them (as in Jaroslawska et al., [<reflink idref="bib16" id="ref83">16</reflink>]; Pesowski et al., [<reflink idref="bib31" id="ref84">31</reflink>]; Zhao et al., [<reflink idref="bib41" id="ref85">41</reflink>]). That is, when a character had a choice, children could be more certain that they intended to either be mean or nice, whereas when characters did not have a choice, children could not be sure of their intentions. Interestingly, even these older children—who evidently can consider counterfactual choices in their moral judgments—still rated characters without any choice as meaner if they brought the bad snack or toy than if they brought the good snack or toy, across both studies. This may suggest that a script‐like knowledge of "gift‐giving" behavior influenced children's responses, in that across ages children presumed that it is nice to give a good gift and mean to give a bad one. Alternatively, although children were told that each character <emph>had</emph> to bring a snack or toy, perhaps some children simply overlooked this fact. That is, some children may have judged the character who brought a bad snack as mean simply because it would have been nicer to bring no snack at all.</p> <p>We expected that children would be more likely to consider a character's past choices when evaluating a negative outcome compared with a positive one, because the negative outcome of not bringing the appropriate gift might be more salient to young children, and because some earlier research suggested that children experience the negative counterfactual emotion of regret before the positive one of relief (e.g., Weisberg & Beck, [<reflink idref="bib37" id="ref86">37</reflink>]). However, our results were mixed with the findings of study 2 running counter to this expectation. Future research could explore what factors influence which counterfactuals children are more likely to bring to mind. Indeed, adults tend to imagine upward versus downward counterfactuals (i.e., either considering what could have been better or what could have been worse) based on a variety of factors, including whether an action that led to a particular outcome conforms to a norm, and how controllable the action was (see Byrne, [<reflink idref="bib3" id="ref87">3</reflink>] for a review).</p> <p>Counterfactual thinking about what could have happened is implicated in many of the judgments adults make (Branscombe et al., [<reflink idref="bib2" id="ref88">2</reflink>]; Kühberger et al., [<reflink idref="bib18" id="ref89">18</reflink>]; Martin & Cushman, [<reflink idref="bib23" id="ref90">23</reflink>]; Redshaw & Ganea, [<reflink idref="bib34" id="ref91">34</reflink>]), and it stands to reason that children may likewise incorporate counterfactuals into their judgments once they become capable of such thinking. Across two studies, we find good evidence that 6‐ to 9‐year‐olds, unlike 4‐ and 5‐year‐olds, considered what a character could have chosen to do when making moral judgments, and this pattern was elicited for both positive and negative outcomes. These findings suggest an integral role of counterfactual thinking in children's emerging moral judgments, and they align with an expanding body of literature suggesting that children begin to maturely reason counterfactually about past alternatives from around age 6 onwards. The findings also raise questions about how children's counterfactual thinking may influence more complex moral judgments, for example when people intend to do the right thing but find themselves lacking the choice to do so. But here, we provide the first direct evidence that children account for counterfactuals in their moral judgments at all.</p> <hd id="AN0171385518-31">ACKNOWLEDGMENTS</hd> <p>This research was supported by the Australian Research Council Discovery Project (grant no. DP210101572) awarded to TS and JR. We would like to thank the Queensland Museum for their support of this project throughout the data collection period and the parents and children who participated. Open access publishing facilitated by The University of Queensland, as part of the Wiley ‐ The University of Queensland agreement via the Council of Australian University Librarians.</p> <hd id="AN0171385518-32">CONFLICT OF INTEREST STATEMENT</hd> <p>The authors declare that they have no conflict of interest.</p> <hd id="AN0171385518-33">DATA AVAILABILITY STATEMENT</hd> <p>The data necessary to reproduce the analyses presented here are publicly accessible at: https://osf.io/aw5ks/?view_only=45577dcf234142eb95bd37e29815c505.</p> <p>GRAPH: Appendix S1.</p> <ref id="AN0171385518-34"> <title> REFERENCES </title> <blist> <bibl id="bib1" idref="ref15" type="bt">1</bibl> <bibtext> Beck, S. R., Robinson, E. J., Carroll, D. J., & Apperly, I. A. (2006). Children's thinking about counterfactuals and future hypotheticals as possibilities. Child Development, 77 (2), 413 – 426.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref5" type="bt">2</bibl> <bibtext> Branscombe, N. R., Owen, S., Garstka, T. A., & Coleman, J. J. J. (1996). Rape and accident counterfactuals: Who might have done otherwise and would it have changed the outcome? Journal of Applied Social Psychology, 26 (12), 1042 – 1067.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref6" type="bt">3</bibl> <bibtext> Byrne, R. M. (2016). Counterfactual thought. Annual Review of Psychology, 67 (1), 135 – 157. https://doi.org/10.1146/annurev‐psych‐122414‐033249</bibtext> </blist> <blist> <bibl id="bib4" idref="ref7" type="bt">4</bibl> <bibtext> Byrne, R. M. (2017). Counterfactual thinking: From logic to morality. Current Directions in Psychological Science, 26 (4), 314 – 322.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref30" type="bt">5</bibl> <bibtext> Chernyak, N., & Sobel, D. M. (2016). "But he didn't mean to do it": Preschoolers correct punishments imposed on accidental transgressors. Cognitive Development, 39, 13 – 20. https://doi.org/10.1016/j.cogdev.2016.03.002</bibtext> </blist> <blist> <bibl id="bib6" idref="ref24" type="bt">6</bibl> <bibtext> Corbett, B., Feeney, A., & McCormack, T. (2021). Interpersonal regret and prosocial risk taking in children. Cognitive Development, 58, 101036. https://doi.org/10.1016/j.cogdev.2021.101036</bibtext> </blist> <blist> <bibl id="bib7" idref="ref1" type="bt">7</bibl> <bibtext> Cushman, F. (2015). Deconstructing intent to reconstruct morality. Current Opinion in Psychology, 6, 97 – 103. https://doi.org/10.1016/j.copsyc.2015.06.003</bibtext> </blist> <blist> <bibl id="bib8" idref="ref31" type="bt">8</bibl> <bibtext> Cushman, F., Sheketoff, R., Wharton, S., & Carey, S. (2013). The development of intent‐based moral judgment. Cognition, 127 (1), 6 – 21.</bibtext> </blist> <blist> <bibl id="bib9" idref="ref2" type="bt">9</bibl> <bibtext> Darley, J. M., Carlsmith, K. M., & Robinson, P. H. (2000). Incapacitation and just deserts as motives for punishment. Law and Human Behavior, 24 (6), 659 – 683. https://doi.org/10.1023/A:1005552203727</bibtext> </blist> <blist> <bibtext> Gautam, S., Bulley, A., von Hippel, W., & Suddendorf, T. (2017). Affective forecasting bias in preschool children. Journal of Experimental Child Psychology, 159, 175 – 184.</bibtext> </blist> <blist> <bibtext> Gautam, S., Suddendorf, T., & Redshaw, J. (2022). Counterfactual thinking elicits emotional change in young children. Philosophical Transactions of the Royal Society B: Biological Sciences, 377 (1866), 20210346. https://doi.org/10.1098/rstb.2021.0346</bibtext> </blist> <blist> <bibtext> Gautam, S., Suddendorf, T., Henry, J. D., & Redshaw, J. (2019). A taxonomy of mental time travel and counterfactual thought: Insights from cognitive development. Behavioural Brain Research, 374, 112108. https://doi.org/10.1016/j.bbr.2019.112108</bibtext> </blist> <blist> <bibtext> Guttentag, R., & Ferrell, J. (2004). Reality compared with its alternatives: Age differences in judgments of regret and relief. Developmental Psychology, 40 (5), 764 – 775.</bibtext> </blist> <blist> <bibtext> Harris, P. L., German, T., & Mills, P. (1996). Children's use of counterfactual thinking in causal reasoning. Cognition, 61 (3), 233 – 259.</bibtext> </blist> <blist> <bibtext> Ingram, G., & Romero, C. O. M. (2019). Children consider both outcome and intent in moral judgements, punishment recommendations, and reward allocations. https://doi.org/10.31234/osf.io/zpv6h</bibtext> </blist> <blist> <bibtext> Jaroslawska, A. J., McCormack, T., Burns, P., & Caruso, E. M. (2020). Outcomes versus intentions in fairness‐related decision making: School‐aged children's decisions are just like those of adults. Journal of Experimental Child Psychology, 189, 104704.</bibtext> </blist> <blist> <bibtext> Kominsky, J. F., Gerstenberg, T., Pelz, M., Sheskin, M., Singmann, H., Schulz, L., & Keil, F. C. (2021). The trajectory of counterfactual simulation in development. Developmental Psychology, 57 (2), 253.</bibtext> </blist> <blist> <bibtext> Kühberger, A., Großbichler, C., & Wimmer, A. (2011). Counterfactual closeness and predicted affect. Thinking & Reasoning, 17 (2), 137 – 155.</bibtext> </blist> <blist> <bibtext> Kushnir, T. (2022). Imagination and social cognition in childhood. Wiley Interdisciplinary Reviews: Cognitive Science, 13, e1603.</bibtext> </blist> <blist> <bibtext> Leahy, B., Rafetseder, E., & Perner, J. (2014). Basic conditional reasoning: How children mimic counterfactual reasoning. Studia Logica, 102 (4), 793 – 810. https://doi.org/10.1007/s11225‐013‐9510‐7</bibtext> </blist> <blist> <bibtext> Lillard, A. S., Pinkham, A. M., & Smith, E. (2011). Pretend play and cognitive development. The Wiley‐Blackwell Handbook of Childhood Cognitive Development, 32, 285.</bibtext> </blist> <blist> <bibtext> Margoni, F., & Surian, L. (2017). Children's intention‐based moral judgments of helping agents. Cognitive Development, 41, 46 – 64. https://doi.org/10.1016/j.cogdev.2016.12.001</bibtext> </blist> <blist> <bibtext> Martin, J. W., & Cushman, F. (2016). Why we forgive what can't be controlled. Cognition, 147, 133 – 143.</bibtext> </blist> <blist> <bibtext> McCormack, T., Ho, M., Gribben, C., O'Connor, E., & Hoerl, C. (2018). The development of counterfactual reasoning about doubly‐determined events. Cognitive Development, 45, 1 – 9.</bibtext> </blist> <blist> <bibtext> Migliore, S., Curcio, G., Mancini, F., & Cappa, S. F. (2014). Counterfactual thinking in moral judgment: An experimental study. Frontiers in Psychology, 5, 451. https://doi.org/10.3389/fpsyg.2014.00451</bibtext> </blist> <blist> <bibtext> Monroe, A. E., & Malle, B. F. (2017). Two paths to blame: Intentionality directs moral information processing along two distinct tracks. Journal of Experimental Psychology: General, 146 (1), 123 – 133. https://doi.org/10.1037/xge0000234</bibtext> </blist> <blist> <bibtext> Nobes, G., Panagiotaki, G., & Bartholomew, K. J. (2016). The influence of intention, outcome and question‐wording on children's and adults' moral judgments. Cognition, 157, 190 – 204.</bibtext> </blist> <blist> <bibtext> Nobes, G., Panagiotaki, G., & Engelhardt, P. E. (2017). The development of intention‐based morality: The influence of intention salience and recency, negligence, and outcome on children's and adults' judgments. Developmental Psychology, 53 (10), 1895 – 1911.</bibtext> </blist> <blist> <bibtext> Nyhout, A., & Ganea, P. A. (2019). Mature counterfactual reasoning in 4‐and 5‐year‐olds. Cognition, 183, 57 – 66.</bibtext> </blist> <blist> <bibtext> O'Connor, E., McCormack, T., & Feeney, A. (2012). The development of regret. Journal of Experimental Child Psychology, 111 (1), 120 – 127.</bibtext> </blist> <blist> <bibtext> Pesowski, M. L., Denison, S., & Friedman, O. (2016). Young children infer preferences from a single action, but not if it is constrained. Cognition, 155, 168 – 175.</bibtext> </blist> <blist> <bibtext> Rafetseder, E., Cristi‐Vargas, R., & Perner, J. (2010). Counterfactual reasoning: Developing a sense of "nearest possible world". Child Development, 81 (1), 376 – 389.</bibtext> </blist> <blist> <bibtext> Rafetseder, E., & Perner, J. (2014). Counterfactual reasoning: Sharpening conceptual distinctions in developmental studies. Child Development Perspectives, 8 (1), 54 – 58.</bibtext> </blist> <blist> <bibtext> Redshaw, J., & Ganea, P. A. (2022). Thinking about possibilities: Mechanisms, ontogeny, functions and phylogeny. Philosophical Transactions of the Royal Society B: Biological Sciences, 377 (1866), 20210333. https://doi.org/10.1098/rstb.2021.0333.</bibtext> </blist> <blist> <bibtext> Sutter, M. (2007). Outcomes versus intentions: On the nature of fair behavior and its development with age. Journal of Economic Psychology, 28 (1), 69 – 78.</bibtext> </blist> <blist> <bibtext> Uprichard, B., & McCormack, T. (2019). Becoming kinder: Prosocial choice and the development of interpersonal regret. Child Development, 90 (4), e486 – e504.</bibtext> </blist> <blist> <bibtext> Weisberg, D. P., & Beck, S. R. (2010). Children's thinking about their own and others' regret and relief. Journal of Experimental Child Psychology, 106 (2), 184 – 191.</bibtext> </blist> <blist> <bibtext> Weisberg, D. P., & Beck, S. R. (2012). The development of children's regret and relief. Cognition & Emotion, 26 (5), 820 – 835.</bibtext> </blist> <blist> <bibtext> Williams, C. W., Lees‐Haley, P. R., & Price, J. R. (1996). The role of counterfactual thinking and causal attribution in accident‐related judgments. Journal of Applied Social Psychology, 26 (23), 2100 – 2112.</bibtext> </blist> <blist> <bibtext> Yuill, N., & Perner, J. (1988). Intentionality and knowledge in children's judgments of actor's responsibility and recipient's emotional reaction. Developmental Psychology, 24 (3), 358 – 365. https://doi.org/10.1037/0012‐1649.24.3.358</bibtext> </blist> <blist> <bibtext> Zhao, X., Zhao, X., Gweon, H., & Kushnir, T. J. C. (2021). Leaving a choice for others: Children's evaluations of considerate, socially‐mindful actions. Child Development, 92 (4), 1238 – 1253.</bibtext> </blist> </ref> <aug> <p>By Shalini Gautam; Ruby Owen Hall; Thomas Suddendorf and Jonathan Redshaw</p> <p>Reported by Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib23" firstref="ref3"></nolink> <nolink nlid="nl2" bibid="bib26" firstref="ref4"></nolink> <nolink nlid="nl3" bibid="bib25" firstref="ref8"></nolink> <nolink nlid="nl4" bibid="bib39" firstref="ref9"></nolink> <nolink nlid="nl5" bibid="bib21" firstref="ref10"></nolink> <nolink nlid="nl6" bibid="bib14" firstref="ref11"></nolink> <nolink nlid="nl7" bibid="bib24" firstref="ref12"></nolink> <nolink nlid="nl8" bibid="bib32" firstref="ref13"></nolink> <nolink nlid="nl9" bibid="bib29" firstref="ref14"></nolink> <nolink nlid="nl10" bibid="bib12" firstref="ref16"></nolink> <nolink nlid="nl11" bibid="bib17" firstref="ref17"></nolink> <nolink nlid="nl12" bibid="bib20" firstref="ref18"></nolink> <nolink nlid="nl13" bibid="bib33" firstref="ref21"></nolink> <nolink nlid="nl14" bibid="bib30" firstref="ref22"></nolink> <nolink nlid="nl15" bibid="bib38" firstref="ref23"></nolink> <nolink nlid="nl16" bibid="bib36" firstref="ref25"></nolink> <nolink nlid="nl17" bibid="bib41" firstref="ref26"></nolink> <nolink nlid="nl18" bibid="bib31" firstref="ref27"></nolink> <nolink nlid="nl19" bibid="bib22" firstref="ref32"></nolink> <nolink nlid="nl20" bibid="bib28" firstref="ref33"></nolink> <nolink nlid="nl21" bibid="bib40" firstref="ref34"></nolink> <nolink nlid="nl22" bibid="bib16" firstref="ref37"></nolink> <nolink nlid="nl23" bibid="bib35" firstref="ref38"></nolink> <nolink nlid="nl24" bibid="bib11" firstref="ref46"></nolink> <nolink nlid="nl25" bibid="bib37" firstref="ref47"></nolink> <nolink nlid="nl26" bibid="bib27" firstref="ref50"></nolink> <nolink nlid="nl27" bibid="bib15" firstref="ref52"></nolink> <nolink nlid="nl28" bibid="bib13" firstref="ref63"></nolink> <nolink nlid="nl29" bibid="bib10" firstref="ref79"></nolink> <nolink nlid="nl30" bibid="bib19" firstref="ref82"></nolink> <nolink nlid="nl31" bibid="bib18" firstref="ref89"></nolink> <nolink nlid="nl32" bibid="bib34" firstref="ref91"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1394005
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Counterfactual Choices and Moral Judgments in Children
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Gautam%2C+Shalini%22">Gautam, Shalini</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0001-6565-7458">0000-0001-6565-7458</externalLink>)<br /><searchLink fieldCode="AR" term="%22Owen+Hall%2C+Ruby%22">Owen Hall, Ruby</searchLink><br /><searchLink fieldCode="AR" term="%22Suddendorf%2C+Thomas%22">Suddendorf, Thomas</searchLink><br /><searchLink fieldCode="AR" term="%22Redshaw%2C+Jonathan%22">Redshaw, Jonathan</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Child+Development%22"><i>Child Development</i></searchLink>. Sep-Oct 2023 94(5):e296-e307.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Wiley. Available from: John Wiley & Sons, Inc. 111 River Street, Hoboken, NJ 07030. Tel: 800-835-6770; e-mail: cs-journals@wiley.com; Web site: https://www.wiley.com/en-us
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 12
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2023
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Decision+Making%22">Decision Making</searchLink><br /><searchLink fieldCode="DE" term="%22Moral+Values%22">Moral Values</searchLink><br /><searchLink fieldCode="DE" term="%22Developmental+Stages%22">Developmental Stages</searchLink><br /><searchLink fieldCode="DE" term="%22Child+Development%22">Child Development</searchLink><br /><searchLink fieldCode="DE" term="%22Cognitive+Processes%22">Cognitive Processes</searchLink><br /><searchLink fieldCode="DE" term="%22Story+Telling%22">Story Telling</searchLink><br /><searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink>
– Name: Subject
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Australia%22">Australia</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1111/cdev.13943
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0009-3920<br />1467-8624
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: When making moral judgments of past actions, adults often think counterfactually about what could have been done differently. Considerable evidence suggests that counterfactual thinking emerges around age 6, but it remains unknown how this development influences children's moral judgments. Across two studies, Australian children aged 4-9 (N = 236, 142 Females) were told stories about two characters who had a choice that led to a good or bad outcome, and two characters who had no choice over a good or bad outcome. Results showed that 4- and 5-year-olds' moral judgments were influenced only by the actual outcome. From age 6, children's moral judgments were also influenced by the counterfactual choices that had been available to the characters.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: Note
  Label: Notes
  Group: Note
  Data: https://osf.io/aw5ks/?view_only=45577dcf234142eb95bd37e29815c505
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2023
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1394005
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1394005
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1111/cdev.13943
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 12
        StartPage: e296
    Subjects:
      – SubjectFull: Decision Making
        Type: general
      – SubjectFull: Moral Values
        Type: general
      – SubjectFull: Developmental Stages
        Type: general
      – SubjectFull: Child Development
        Type: general
      – SubjectFull: Cognitive Processes
        Type: general
      – SubjectFull: Story Telling
        Type: general
      – SubjectFull: Foreign Countries
        Type: general
      – SubjectFull: Australia
        Type: general
    Titles:
      – TitleFull: Counterfactual Choices and Moral Judgments in Children
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Gautam, Shalini
      – PersonEntity:
          Name:
            NameFull: Owen Hall, Ruby
      – PersonEntity:
          Name:
            NameFull: Suddendorf, Thomas
      – PersonEntity:
          Name:
            NameFull: Redshaw, Jonathan
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2023
          Identifiers:
            – Type: issn-print
              Value: 0009-3920
            – Type: issn-electronic
              Value: 1467-8624
          Numbering:
            – Type: volume
              Value: 94
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: Child Development
              Type: main
ResultId 1