The Right Stuff: Preschoolers Generalize Reliability Across Communicative Domains When Informants Show Semantic (Not Episodic) Knowledge

ABSTRACT

Although preschoolers have strong expectations about the pedagogical nature of pointing gestures (Csibra & Gergely, 2006), more recent work has shown that preschoolers prefer to use informants’ spoken language, not their pointing gestures, to make judgments about their reliability (Palmquist & Jaswal, 2015). Here, we explored children’s inferences about pointers using a standard selective trust paradigm. Specifically, we asked whether 4- and 5-year-olds generalize reliability across communicative domains (from pointing ability to speaking ability). We found that children preferred to make generalizations about pointers’ reliability when they had conveyed semantic, but not episodic, knowledge. Individual differences in theory of mind also predicted children’s willingness to make generalizations about pointers’ reliability. Both sets of results suggest that multiple factors (i.e., the type of knowledge an informant shares and individual differences in children’s cognitive development) affect whether preschoolers generalize others’ reliability across domains.     

Inducing Knowledgeability From Niceness: Children Use Social Features for Making Epistemic Inferences

In many ways, evaluating informants based on their features is a problem of induction: Children rely on the assumption that observable informant characteristics (e.g., traits, behaviors, social categories) will predict unobservable characteristics (e.g., future behavior, knowledge states, intentions). Yet to make sensible inferences, children must recognize what informant features are relevant for what types of inferences. The current research investigated whether preschoolers use social features (e.g., niceness) for making epistemic inferences and, conversely, whether they use intellectual features (e.g., expertise) for making social inferences. In the study, 96 preschoolers (M_age = 4.96 years) were asked to attribute knowledge and behaviors to a mean informant, a nice informant, and a neutral informant. Between subjects, we varied which informant had expertise. We found that when attributing knowledge, children used both features: attributing more knowledge to nicer informants, but also attributing more knowledge to an informant when he had expertise. In contrast, when making social inferences, children relied primarily on social features.     

Preschoolers monitor the relative accuracy of informants

In 2 studies, the sensitivity of 3- and 4-year-olds to the previous accuracy of informants was assessed. Children viewed films in which 2 informants labeled familiar objects with differential accuracy (across the 2 experiments, children were exposed to the following rates of accuracy by the more and less accurate informants, respectively: 100% vs. 0%, 100% vs. 25%, 75% vs. 0%, and 75% vs. 25%). Next, children watched films in which the same 2 informants provided conflicting novel labels for unfamiliar objects. Children were asked to indicate which of the 2 labels was associated with each object. Three-year-olds trusted the more accurate informant only in conditions in which 1 of the 2 informants had been 100% accurate, whereas 4-year-olds trusted the more accurate informant in all conditions tested. These results suggest that 3-year-olds mistrust informants who make a single error, whereas 4-year-olds track the relative frequency of errors when deciding whom to trust.     

Explaining the errors away: Young children forgive understandable semantic mistakes

Errors differ in degree of seriousness. We asked whether preschoolers would use the magnitude of an informant's errors to decide if that informant would be a good source of information later. Four- and 5-year-olds observed two informants incorrectly label familiar objects, but one informant's errors were closer to the correct answer than the other's (e.g., one referred to a comb as a brush and the other referred to the same comb as a thunderstorm). When informants had an unambiguous view of the objects, children could identify which informant was closer to being correct, but they did not favor novel labels the “closer” informant later provided. When the informants had an ambiguous view of the objects (e.g., only the handle of the comb was visible), children preferred the novel labels provided later by the “closer” informant. Preschoolers are willing to overlook semantic errors that are close to being correct, but only when there is an understandable reason for the speaker's errors.     

Going With the Flow: Preschoolers Prefer Nondissenters as Informants

ABSTRACT— In two experiments, 3- and 4-year-olds were tested for their sensitivity to agreement and disagreement among informants. In pretest trials, they watched as three of four informants ( Experiment 1 ) or two of three informants ( Experiment 2 ) indicated the same referent for an unfamiliar label; the remaining informant was a lone dissenter who indicated a different referent. Asked for their own judgment, the preschoolers sided with the majority rather than the dissenter. In subsequent test trials, one member of the majority and the dissenter remained present and continued to provide conflicting information about the names of unfamiliar objects. Children remained mistrustful of the dissenter. They preferred to seek and endorse information from the informant who had belonged to the majority. The implications and scope of children's early sensitivity to group consensus are discussed.     

Preschoolers continue to trust a more accurate informant 1 week after exposure to accuracy information

To determine whether children retain a preference for a previously accurate informant only in the short term or for long‐term use, 3‐ and 4‐year‐old children were tested in two experiments. In both experiments, children were given accuracy information about two informants and were subsequently tested for their selective trust in the two informants (Experiment 1: immediately, 1 day and 1 week later; Experiment 2: immediately, 4 days and 1 week later). Both age groups preferred to trust the accurate informant not only immediately after receiving accuracy information but also at subsequent time‐points. Children who were immediately able to explicitly identify the accurate informant were significantly more likely to seek and accept information from her 1 week later. However, even when they had not been asked to explicitly identify the accurate informant both age groups still maintained their preference for her. Thus, by 3 years of age, children spontaneously choose a previously accurate informant up to 1 week after exposure to information regarding her accuracy.     

In the absence of conflicting testimony young children trust inaccurate informants

The present research investigated the nature of the inferences and decisions young children make about informants with a prior history of inaccuracies. Across three experiments, 3‐ and 4‐year‐olds (total N = 182) reacted to previously inaccurate informants who offered testimony in an object‐labeling task. Of central interest was children's willingness to accept information provided by an inaccurate informant in different contexts of being alone, paired with an accurate informant, or paired with a novel (neutral) informant. Experiments 1 and 2 showed that when a previously inaccurate informant was alone and provided testimony that was not in conflict with the testimony of another informant, children systematically accepted the testimony of that informant. Experiment 3 showed that children accepted testimony from a neutral informant over an inaccurate informant when both provided information, but accepted testimony from an inaccurate informant rather than seeking information from an available neutral informant who did not automatically offer information. These results suggest that even though young children use prior history of accuracy to determine the relative reliability of informants, they are quite willing to trust the testimony of a single informant alone, regardless of whether that informant had previously been reliable.     

Children assess informant reliability using bystanders' non-verbal cues

Recent findings show that preschool children are selective with respect to whom they ask for information and whose claims they endorse. In particular, they monitor an informant's record of past accuracy or inaccuracy and use that record to gauge future trustworthiness. We ask if preschoolers also monitor the non-verbal cues of assent or dissent that bystanders display toward an informant's claims and use that information to gauge an informant's trustworthiness. In familiarization trials, 4-year-olds watched as two adult informants made conflicting claims regarding the name of an unfamiliar object. Two adult bystanders consistently signaled assent - via nods and smiles - to the claims of one informant, and dissent - via head shakes and frowns - from the claims of the other informant. When invited to endorse one of the two claims, 4-year-olds mostly agreed with the informant who had received bystander assent. Thus, in the absence of background knowledge about an object's name, children use third-party non-verbal signals to assess the accuracy of conflicting labels. On subsequent test trials, the informants again made conflicting claims about novel object names, but in the absence of the two bystanders. Despite the lack of any informative bystander signals, children with more advanced understanding of mental states continued to display greater trust in the informant who had received bystander assent in the earlier trials.     

In beauty we trust: Children prefer information from more attractive informants

Preschool children were presented with slides on a computer screen showing a novel object, together with two informants, one with an attractive and one with a less attractive face. Children were asked which informant they would like to ask about the name of the novel object. After hearing the informants provide conflicting names, they were asked who they thought was correct. Children were more likely to endorse names provided by the person with the more attractive face, a bias that cannot be justified on epistemic grounds. The implications of this finding are discussed.     

Causal learning across domains

Five studies investigated (a) children's ability to use the dependent and independent probabilities of events to make causal inferences and (b) the interaction between such inferences and domain-specific knowledge. In Experiment 1, preschoolers used patterns of dependence and independence to make accurate causal inferences in the domains of biology and psychology. Experiment 2 replicated the results in the domain of biology with a more complex pattern of conditional dependencies. In Experiment 3, children used evidence about patterns of dependence and independence to craft novel interventions across domains. In Experiments 4 and 5, children's sensitivity to patterns of dependence was pitted against their domain-specific knowledge. Children used conditional probabilities to make accurate causal inferences even when asked to violate domain boundaries.     

When being right is not enough: four-year-olds distinguish knowledgeable informants from merely accurate informants

Recent evidence demonstrates that children are selective in their social learning, preferring to learn from a previously accurate speaker than from a previously inaccurate one. We examined whether children assessing speakers' reliability take into account how speakers achieved their prior accuracy. In Study 1, when faced with two accurate informants, 4- and 5-year-olds (but not 3-year-olds) were more likely to seek novel information from an informant who had previously given the answers unaided than from an informant who had always relied on help from a third party. Similarly, in Study 2, 4-year-olds were more likely to trust the testimony of an unaided informant over the testimony provided by an assisted informant. Our results indicate that when children reach around 4 years of age, their selective trust extends beyond simple generalizations based on informants' past accuracy to a more sophisticated selectivity that distinguishes between truly knowledgeable informants and merely accurate informants who may not be reliable in the long term.     

幼儿基于信息判断的选择性信任的发展——其持续性及与人际信任的关系

采用物品命名任务的模式, 对60名2岁、3岁、4岁幼儿的选择性信任进行研究。结果表明:(1)幼儿更信赖那些正确率高的信息传达者, 形成选择性的信任。这种选择性信任的认知能力在3岁左右出现, 并随着年龄的增长而增强, 4岁达到稳定。(2)选择性信任一旦形成, 就具有一定的稳定性和持续性, 即3岁以上的幼儿在不同时间不同地点再见到先前的信息传达者时, 依然能区分出他们。(3)基于信息判断的选择性信任可以迁移到人际信任的其他方面。     

Trust in testimony: children's use of true and false statements

The extent to which young children monitor and use the truth of assertions to gauge the reliability of subsequent testimony was examined. Three- and 4-year-old children were presented with two informants, an accurate labeler and an inaccurate labeler. They were then invited to learn names for novel objects from these informants. The children correctly monitored and identified the informants on the basis of the truth of their prior labeling. Furthermore, children who explicitly identified the unreliable or reliable informant across two tasks went on to demonstrate selective trust in the novel information provided by the previously reliable informant. Children who did not consistently identify the unreliable or reliable informant proved indiscriminate.     

幼儿能力判断发展特点研究

本研究以62名3~6岁儿童为被试,考察了幼儿在不同任务中对自身和他人能力判断的发展特点。结果表明:(1)幼儿对自身或他人能力判断的准确性随着年龄的增长而提高;(2)幼儿对他人能力判断并不比对自身能力判断更准确,随着年龄的增长,幼儿开始对自身能力做出更加积极的判断;(3)幼儿能够明确区分愿望与预期,不过当幼儿对自身能力进行判断时这种区分不容易显现出来。     

Bridging the gap: Causality-at-a-distance in children's categorization and inferences about internal properties

Previous research has shown that preschoolers extend labels and internal properties of objects based on those objects’ causal properties, even when the causal properties conflict with the objects’ perceptual appearance [Nazzi, T., & Gopnik, A. (2000). A shift in children's use of perceptual and causal cues to categorization. Developmental Science, 3, 389–396; Sobel, D. M., Yoachim, C. M., Gopnik, A., Meltzoff, A. N., & Blumenthal, E. J. (2007). The blicket within: Preschoolers’ inferences about insides and causes. Journal of Cognition and Development, 8, 159–182]. These studies, however, only presented causal relations that acted on contact. In two studies, contact causality was replaced by distance causality. In contrast to the contact causality case, 4- and 5-year-olds extended labels to objects with similar perceptual properties over objects with similar causal properties when those properties acted at a distance. When children were asked to make inferences about object's internal properties, they were more likely to make causal responses, with 5-year-olds doing so to a greater extent than 4-year-olds. In a second study, 4-year-olds registered causal properties that acted at a distance and used them to make inferences when no perceptual conflict was present. These results support a hypothesis that young children develop an understanding of the specific mechanisms that link causal relations.     

The Role of Probability and Intentionality in Preschoolers' Causal Generalizations

Three experiments examined whether preschoolers recognize that the causal properties of objects generalize to new members of the same set given either deterministic or probabilistic data. Experiment 1 found that 3- and 4-year-olds were able to make such a generalization given deterministic data but were at chance when they observed probabilistic information. Five-year-olds reliably generalized in both situations. Experiment 2 found that 4-year-olds could make some probabilistic inferences, particularly when comparing sets that had no efficacy with sets in which some members had efficacy. Children had some difficulty discriminating between completely effective sets and stochastic ones. Experiment 3 examined whether 3- and 4-year-olds could reason about probabilistic data when provided with information about the experimenter's beliefs about causal outcomes. Children who were more successful on standard false-belief measures were more likely to respond as if the data were deterministic. These data suggest that children's probabilistic inferences develop into early elementary school, but preschoolers might have some understanding of probability when reasoning about causal generalization.     

Epistemic trust: modeling children's reasoning about others' knowledge and intent

A core assumption of many theories of development is that children can learn indirectly from other people. However, indirect experience (or testimony) is not constrained to provide veridical information. As a result, if children are to capitalize on this source of knowledge, they must be able to infer who is trustworthy and who is not. How might a learner make such inferences while at the same time learning about the world? What biases, if any, might children bring to this problem? We address these questions with a computational model of epistemic trust in which learners reason about the helpfulness and knowledgeability of an informant. We show that the model captures the competencies shown by young children in four areas: (1) using informants’ accuracy to infer how much to trust them; (2) using informants’ recent accuracy to overcome effects of familiarity; (3) inferring trust based on consensus among informants; and (4) using information about mal‐intent to decide not to trust. The model also explains developmental changes in performance between 3 and 4 years of age as a result of changing default assumptions about the helpfulness of other people.     

Causal learning mechanisms in very young children: two-, three-, and four-year-olds infer causal relations from patterns of variation and covariation

Three studies investigated whether young children make accurate causal inferences on the basis of patterns of variation and covariation. Children were presented with a new causal relation by means of a machine called the "blicket detector." Some objects, but not others, made the machine light up and play music. In the first 2 experiments, children were told that "blickets make the machine go" and were then asked to identify which objects were "blickets." Two-, 3-, and 4-year-old children were shown various patterns of variation and covariation between two different objects and the activation of the machine. All 3 age groups took this information into account in their causal judgments about which objects were blickets. In a 3rd experiment, 3- and 4-year-old children used the information when they were asked to make the machine stop. These results are related to Bayes-net causal graphical models of causal learning.     

信息提供者的先前准确性对幼儿选择性信任的影响

作为社会学习的个体,幼儿从3岁起就能根据信息提供者先前准确性的差异对其可信程度进行评估。通过对以往研究效应量的整理分析发现,随年龄增长,幼儿对高先前准确性信息提供者的信任程度表现出增长趋势（3岁幼儿选择的效应量中值为d_询问=0.60,d_赞同=0.61;4岁幼儿的效应量中值为d_询问=0.80,d_赞同=0.96）。在运用先前准确性进行评估时,幼儿主要关注信息提供者是否犯错,灵活使用"整体印象"和"特质推测"两种认知策略对其未来行为进行推测。幼儿能够根据自身对材料的感知对信息提供者的可信程度进行动态调整,其心理理论发展水平和执行功能限制了他们的选择信任,而信息提供者同时具有的社会性因素对幼儿的选择也具有干扰作用（幼儿信任准确的信息提供者的效应量中值为d_询问=-0.03,d_赞同=-0.17）。未来研究可以改进实验范式,进一步探查先前准确性对幼儿选择性信任的影响,拓展信息提供者的呈现方式和其它多重特征,关注不同类别的信息提供者。