Times are displayed in (UTC-05:00) Central Time (US & Canada) Change
About this paper symposium
| Panel information |
|---|
| Panel 20. Social Cognition |
| Paper #1 | |
|---|---|
| The impact of conventional language on imitation of an inefficient tool in a pressure-based situation | |
| Author information | Role |
| Cara DiYanni, Rider University | Presenting author |
| Kathleen Corriveau, Boston University, United States | Non-presenting author |
| Jennifer Clegg, Texas State University, United States | Non-presenting author |
| Abstract | |
|
From a young age, there is a strong motivation for children to follow conventions (Haun & Tomasello, 2011; Haun et al., 2012). Evidence from multiple cultural backgrounds suggests children are attuned to social cues that indicate when tasks are conventional and therefore need to be imitated with higher fidelity (Clegg & Legare, 2016). This motivation is enforced by parents and caregivers (Clegg & Legare, 2017). Children even copy a model’s selection of an inefficient tool more often when social cues indicate the task is conventional (DiYanni et al., 2022). The current study explored how children would respond to a model’s selection of an inefficient tool in a timed, pressure-based situation. Ronfard et al. (2016) suggested that if children are asked the fastest way to perform a task, their subsequent actions would clarify how they interpret a model’s behavior. If children who initially copy the use of an inefficient tool stick with it even in a pressured situation, it might indicate they have encoded it as being for the task. If, however, they switch to a more efficient tool, it might suggest that an initial intention of adhering to convention is overridden when efficiency becomes paramount. Four- and 5-year-olds watched two videos of a model demonstrating a task (cookie-crushing; moving rice between two bowls). The model either used conventional language (“Everyone always does it this way,” N = 48) or instrumental language ( “I am going to…” (data collection in progress)). After pointedly considering both tools, the model intentionally selected the less efficient option. Children were asked to choose a tool to complete each task in an initial trial and in a speeded trial where they had to complete the goal as quickly as possible. In the conventional condition, imitation rates in the initial trial were high: 68.8% of children imitated the model’s selection of the inefficient rice tool, 70.8% copied the model’s choice of the inefficient cookie tool, 58.3% imitated the model’s selection both times, and 87.5% copied at least once. In the speeded trial, however, imitation rates dropped sharply. Only 25% of children used the inefficient rice mover, and only 29.2% used the inefficient cookie crusher. Just 12.5% of children imitated in both trials, and 41.7% imitated once. These results to date suggest that while conventional language has an initial impact on children’s decisions, in a pressured, speeded situation, children are likely to switch to a more efficient tool. Data collection for the instrumental condition is in progress. It is predicted that rates of imitation will be lower in the initial trial than in the conventional condition. It is uncertain whether imitation rates in the speeded trial will look similar to those in the conventional condition, or may drop even further. Between 1/4 and 1/3 of children did choose the inefficient tool in the speeded task after hearing a model use conventional language. If these numbers drop to near-zero after hearing instrumental language, it would suggest that although the effect of conventional language is diminished in a pressured situation, it does not disappear entirely. |
|
| Paper #2 | |
|---|---|
| Chinese children favour model expertise over copying a majority in a social learning task | |
| Author information | Role |
| Emily Burdett, University of Nottingham, United Kingdom | Presenting author |
| Liqi Zhu, Chinese Academy of the Sciences, China | Non-presenting author |
| Nicola McGuigan, University of the West of Scotland, United Kingdom | Non-presenting author |
| Andrew Whiten, University of St. Andrews, United Kingdom | Non-presenting author |
| Abstract | |
|
Children are selective social learners who often copy those who are particularly competent, kind, or in the majority. However, cultural values may influence the way in which young children learn and copy from others. Research suggests that cultural groups in Eastern contexts are more likely to copy others than cultural groups in Western contexts (Chang et al, 2011; Mesoudi et al., 2014; Toelch et al., 2014). This difference has been suggested to be a consequence of Asian cultures being more collectivistic in their ideologies as compared to Western cultures that are often more individualistic (Triandis, 1995). Other work suggests that Asian cultures have tight social norms (Gelfand et al., 2017; Gelfand et al., 2011) where children may be more likely to employ a majority social learning strategy (Corriveau et al., 2017; Corriveau & Harris, 2010; Corriveau et al., 2013; DiYanni, Corriveau et al., 2015). These findings suggest that Asian children may be more likely to imitate a consensus over an individual, compared to Western children who are likely to choose an individual demonstrating an optimal solution. Previous work has shown that when given the choice, Western children will choose the optimal solution over imitation of irrelevant actions (Evans et al., 2018). However, when children are given the option to imitate either a consensus or a competent expert (Burdett at al., 2016), British children demonstrate a consistent individual preference to copy either a majority, or an expert when these two alternative social learning strategies were pitted directly against each other (overall British children did not prefer one strategy over the other). In the current study, we explored whether Eastern cultural values would influence such selective social learning preferences. Specifically we tested children in China, a nation typically biased towards the collective, allowing us to test the hypothesis that Chinese children may prefer to copy a majority over an expert. In Experiments 1 and 2 children aged 6 and 7 years observed a neutral majority and a competent expert model each perform a different action on two puzzle boxes presented in either a normative (actions called ‘fai’/’bu’) or an instrumental (extracting a prize) context. In Experiment 1, Chinese children copied the expert in both instrumental and normative conditions. In Experiment 2, Chinese children were more likely to copy the expert than the majority in the normative condition, but not the instrumental, a preference that was evident when the models were members of the cultural outgroup (British models), but not when they viewed their cultural ingroup (Chinese models, Experiment 3). Experiment 4 revealed that the bias towards the expert displayed in Experiments 2 and 3 was not a consequence of an inherent instrumental emphasis associated with puzzle boxes, with children continuing to copy an expert over a majority in a task with no consequent functional outcome. Chinese children may also hold particular cultural values to copy an authority figure (Wang & Meltzoff, 2020). We conclude that different transmission biases occur in different cultures, thereby affecting the information transmitted at a population level (cultural transmission). |
|
| Paper #3 | |
|---|---|
| The influence of contextual variability on children’s imitative tendencies across different cultures | |
| Author information | Role |
| Frankie Fong, Victoria University of Wellington, New Zealand | Presenting author |
| Daniel B. M. Haun, Max Planck Institute for Evolutionary Anthropology, Germany | Non-presenting author |
| Abstract | |
|
One prevailing view of social learning is that children in urban populations predominantly learn via first-party pedagogical learning, instructed by adults, whereas those from small-scale communities primarily learn via third-party observational learning with high autonomy. Cross-cultural studies suggest that children from urbanized Western countries and small-scale communities are equivalently likely to overimitate, in first-party contexts. It remains unknown if children from small-scale communities will copy with high-fidelity (at the expense of efficiency) in third-party contexts, given that their daily observational learning is typically highly autonomous. Some evidence suggests that Western children are just as likely to overimitate in a third-party context (by observing others) as in a first-party context (when the demonstration is directed towards them). However, causal opacity in overimitation tasks may serve as a strong social cue that drives imitation regardless of learning contexts. To address these gaps, this study examined 5- to 10-year-old children’s imitative tendencies in two imitation phases across Leipzig (Germany), Kuala Lumpur (KL, Malaysia), a Batek hunter-gatherer community (Malaysia Indigenous), and a remote community in Uganda. Counterbalanced across all participants, one phase (OI) involved an overimitation task (with a series of causally redundant actions) and another phase (SI) involved two simple tool-use imitation tasks (with an inefficient method vs an efficient alternative). 40 children in each population watched the task demonstrations in a first-party context, where the model faced the child when demonstrating the tasks, and another 40 children in a third-party context, with the model facing away from the child. All children were offered 3 attempts to complete the OI task, and 2 attempts to complete each of the SI tasks. We have just completed data collection in Malaysia, data collection is in progress in Leipzig, and we are about to start data collection in Uganda. For OI, we predict a condition x culture interaction, where Batek and Uganda children will be less likely to overimitate in the 3rd party condition than in the 1st party condition. We predict overimitation rates to decrease over the repeated trials for all children in the 3rd party condition only, but not in the1st party condition. For SI, we predict a main effect of culture, where Batek and Uganda children will be less likely than Leipzig and KL children to copy the suboptimal methods regardless of conditions. We expect imitative tendency to decrease across repeated trials for Leipzig children in the 3rd party condition only. For Leipzig and KL children, we predict a strong correlation of imitative tendencies between the two phases, for both 1st and 3rd party conditions. For Batek and Uganda children, we predict a strong correlation of imitative tendencies between the two phases for the 3rd party condition only. |
|
| Paper #4 | |
|---|---|
| Tool innovation across diverse cultures: The role of academic achievement | |
| Author information | Role |
| Bruce Rawlings, Durham University, United Kingdom | Presenting author |
| Oskar Burger, Omni International, United States | Non-presenting author |
| Lydia Chen, University of Texas at Austin, United States | Non-presenting author |
| Emily Messer, Heriot-Watt University, United Kingdom | Non-presenting author |
| Cristine Legare, University of Texas at Austin, United States | Non-presenting author |
| Abstract | |
|
Humans’ ability to create and use tools is one of our species’ defining characteristics and has irreversibly altered our social and ecological environments. All human societies create and use tools, and no other species does so with the complexity and diversity of humans. Yet, paradoxically, it is well-established that children are remarkably poor at simple tool innovation challenges, such as reshaping pipe cleaners to retrieve-out-of reach rewards (widely known as the hook task). Much work has shown that children from geographically and culturally diverse populations struggle on tasks such as the hook task, and it is not until mid-late childhood that children are consistently successful. Further, some studies have shown that children in non-western populations may perform worse than those in western ones with some small-scale societies seemingly performing at floor level on these kinds of tasks. Why children struggle so markedly on simple tool innovation challenges, and the potential factors shaping population differences in performance, remain unclear. In the first part of my talk, I will present data from a large-scale cross-cultural project examining whether academic achievement (measured by numeracy and literacy) is associated with children’s tool innovation success (as measured by the hook task), in 4–12-year-old children (N = ~1400) from 11 countries (Brazil, Ecuador, Colombia, Mexico, USA, Vanuatu, R. Congo, Ghana, Australia, India, Malaysia), with varying exposure to, and engagement with, formal education. We find marked variation in performance across populations, and that numeracy and literacy skills are not associated with tool innovation success, within and across populations. This suggests that core skills associated with attending school do not impact innovative skills early in life but that other cultural factors may do. The second part of my talk will explore some methodological challenges when measuring constructs such as innovation across culturally diverse populations. The global replication of low success rates on the hook task is indicative of its robustness, and using the same measure across populations allows direct cross-cultural performance comparisons. However, the practice of taking western designed and validated cognitive measures and transporting them to non-western populations needs to be carefully considered to ensure they are measuring what they are intended to across communities. I will discuss my ongoing ESRC funded project, aiming to for the first time, develop culturally fair measures of tool innovation for 4-12-year-old children in four geographically and culturally diverse populations; urban Western children (UK), two small-scale subsistence societies (R. Congo) and a rural Indigenous population (Malaysia). The project is taking a bottom-up approach to developing culturally grounded tool innovation measures, using ethnographic data alongside cycles of feedback from community members and systematic protocol refinement to ensure maximal validity. The ultimate aim is to develop culturally grounded measures of tool innovation for each of the four communities and compare performance on them to that of the hook task. At the time of the SRCD meeting, the project will be approaching the end of its first year. |
|
⇦ Back to session
The Influence of Time Constraints, Language, Model, and Culture on Children’s Tool Imitation and Innovation
Submission Type
Paper Symposium
Description
| Session Title | The Influence of Time Constraints, Language, Model, and Culture on Children’s Tool Imitation and Innovation |