True Belief Tasks Research Articles

Perceptual Access Reasoning: What are the alternatives?

Karen Bartsch (2021), Charlie Lewis (2021), and Beate Sodian (2021) provided thoughtful commentaries on our Society for Research in Child Development Monograph, “Perceptual Access Reasoning (PAR) in Developing a Representational Theory of Mind” (Fabricius et al., 2021). The commentators suggested alternative accounts of our empirical findings, an alternative approach to studying ToM development more generally, and alternative ways to test PAR theory specifically. Here we provide an in-depth response to the various alternatives, as well as a brief overview of the theory and a discussion of two recent alternative accounts that challenge the theory’s counter-intuitive prediction that PAR-users will fail true belief tasks.

How do children overcome their pragmatic performance problems in the true belief task? The role of advanced pragmatics and higher-order theory of mind.

The true belief (TB) control condition of the classical location-change task asks children to ascribe a veridical belief to an agent to predict her action (analog to the false belief (FB) condition to test Theory of Mind (ToM) abilities). Studies that administered TB tasks to a broad age range of children yielded surprising findings of a U-shaped performance curve in this seemingly trivial task. Children before age four perform competently in the TB condition. Children who begin to solve the FB condition at age four, however, fail the TB condition and only from around age 10, children succeed again. New evidence suggests that the decline in performance around age four reflects pragmatic confusions caused by the triviality of the task rather than real competence deficits in ToM. Based on these results, it can be hypothesized that the recovery of performance at the end of the U-shaped curve reflects underlying developments in children's growing pragmatic awareness. The aim of the current set of studies, therefore, was to test whether the developmental change at the end of the U-shaped performance curve can be explained by changes in children's pragmatic understanding and by more general underlying developmental changes in recursive ToM or recursive thinking in general. Results from Study 1 (N = 81, 6-10 years) suggest that children's recursive ToM, but not their advanced pragmatic understanding or general recursive thinking abilities predict their TB performance. However, this relationship could not be replicated in Study 2 (N = 87, 6-10 years) and Study 3 (N = 64, 6-10 years) in which neither recursive ToM nor advanced pragmatic understanding or recursive thinking explained children's performance in the TB task. The studies therefore remain inconclusive regarding explanations for the end of the U-shaped performance curve. Future research needs to investigate potential pragmatic and general cognitive foundations of this developmental change more thoroughly.

The Path to Fully Representational Theory of Mind: Conceptual, Executive, and Pragmatic Challenges.

Although an explicit Theory of Mind (ToM) has been found to develop around 4 years of age in Western societies, recent work showing that 4- and 5-year-olds fail modified versions of False Belief tasks as well as seemingly easier True Belief tasks calls into question the robustness of preschoolers’ belief understanding. Some have argued these findings illustrate children’s conceptual limitations in their understanding of belief that are masked by standard False Belief tasks. However, others claim these examples of children’s failure can be explained by pragmatics of the testing situation, rather than conceptual limitations. Given the documented relation between ToM and executive function, an unexamined possibility is that children’s failure can be explained by certain executive demands. In the current study, we examined the relation between typically developing 4- (n = 43) and 5-year-olds’ (n = 42) performance on traditional and modified False Belief tasks, True Belief tasks, and one component of executive functioning - working memory. We found that children performed worse on modified False Belief tasks and True Belief tasks compared to standard 2-option False Belief tasks, and that working memory was related to modified 3-option contents False Belief performance. These results suggest that a fully representational ToM, one that is stable in the context of increased conceptual, executive, and pragmatic demands, may develop later than traditional accounts have assumed.

Correction to: In defense of pluralist theory

Unfortunately there is a typo in section 2.3.1, paragraph 7 “Referring to recent findings from verbal versions of the true belief task that suggest that 4- to 5-year-old children pass such tasks not via belief reasoning but simpler heuristics that draw on perceptual access (Fabricius et al. 2010), Fiebich argues that 4- to 5-year-olds are still engaged in cognitively demanding belief reasoning when passing verbal versions of the false belief task.”

Mistaken Max befriends Duplo girl: No difference between a standard and an acted-out false belief task

With their Duplo task, Rubio-Fernández and Geurts (2013) challenged the assumption that children under 4 years of age cannot pass the standard false belief test. In an attempt to replicate this task on a sample of 73 children aged 32–51 months, we added a standard change of location false belief task as well as a Duplo true belief task. Performance on the latter is crucial for interpreting answers in the Duplo false belief task as to whether they reflect evidence for understanding or merely exhibit a difference in guessing rate. We found (a) a greater variability of response types in both Duplo tasks, (b) no evidence that responses in the Duplo tasks reveal earlier competence than those in the standard false belief test, and (c) a reassuring correlation between false belief tasks, suggesting that the Duplo task does pick up understanding of belief in light of the standard test.

Infants’ false belief understanding: A non-replication of the helping task

False belief understanding in infancy can be measured through helping tasks. However, there has been controversy surrounding the interpretation of such data. In order to better understand the nature of infants’ false belief understanding, replication of the original findings is critical. The objective of the present study was to conduct a conceptual replication of the helping task designed by Buttelmann, Carpenter, and Tomasello (2009). In Experiment 1, 41 18-month-olds were tested using the false belief task. A larger sample (n = 97) was tested in Experiment 2 with this task in order to increase statistical power. Additionally, 33 infants were tested on a true belief task. Results from both experiments failed to replicate those from the original study. The discussion addresses the potential reasons for this lack of replication, as well as the implication of these findings regarding the debate on the interpretation of infants’ behavior in this prompted-action task.

Children’s difficulty with true belief tasks: Competence deficit or performance problem?

According to the standard picture of explicit theory of mind (ToM) development, children begin to (explicitly) ascribe beliefs to others and themselves from around age 4. The empirical basis of this picture comes from numerous studies consistently showing that children master verbal false belief (FB) tasks from around age 4 while children much younger have no difficulty in mastering structurally analogous true belief (TB) tasks. The standard picture, though, has come under serious attack from recent studies using TB tasks with wider age ranges. These studies have found that, paradoxically, children begin to fail TB tasks once they master FB tasks. Such findings cast doubt on the standard picture and suggest, instead, that FB tasks may be solved by much simpler strategies than proper belief reasoning. In the present study, we tested for the development of FB and TB performance in comprehensive and systematic ways. In particular, we tested the competing predictions of competence accounts (according to which TB failure reflects lack of conceptual competence) versus performance limitation accounts (according to which the standard picture is true yet children from around age 4 fail TB tasks due to performance factors). Studies 1 and 2 showed that performance in a variety of novel TB tasks showed a clear U-shaped curve, with children until age 3 and from age 10 performing competently and children in between failing, with strong negative correlations between TB and FB. Crucially, these patterns were found for various kinds of TB tasks, including those for which existing competence limitation accounts would not even predict any difficulty. Study 3, therefore, directly tested performance limitation accounts in terms of pragmatic and related factors and found that these patterns (failure in TB and negative TB-FB correlations) disappear once the relevant performance factors have been removed from the TB tasks. Taken together, these findings suggest that previous TB findings constitute false negatives, clearly speak for performance limitation accounts and thus corroborate the standard picture of the development of explicit theory of mind.

Belief Reasoning and Emotion Understanding in Balanced Bilingual and Language-Dominant Mexican American Young Children

ABSTRACTBelief reasoning and emotion understanding were measured among 102 Mexican American bilingual children ranging from 4 to 7 years old. All children were tested in English and Spanish after ensuring minimum comprehension in each language. Belief reasoning was assessed using 2 false and 1 true belief tasks. Emotion understanding was measured using subtests from the Test for Emotion Comprehension. The influence of family background variables of yearly income, parental education level, and number of siblings on combined Spanish and English vocabulary, belief reasoning, and emotion understanding was assessed by regression analyses. Age and emotion understanding predicted belief reasoning. Vocabulary and belief reasoning predicted emotion understanding. When the sample was divided into language-dominant and balanced bilingual groups on the basis of language proficiency difference scores, there were no significant differences on belief reasoning or emotion understanding. Language groups were demographically similar with regard to child age, parental educational level, and family income. Results suggest Mexican American language-dominant and balanced bilinguals develop belief reasoning and emotion understanding similarly.

Are Infants’ Theory-of-Mind Abilities Well Integrated? Implicit Understanding of Intentions, Desires, and Beliefs

The development of theory of mind (ToM) in infancy has been mainly documented through studies conducted on a single age group with a single task. Very few studies have examined ToM abilities other than false belief, and very few studies have used a within-subjects design. During 2 testing sessions, infants aged 14 and 18 months old were administered ToM tasks based on the violation-of-expectation paradigm which measured intention, true belief, desire, and false-belief understanding. Infants’ looking times at the congruent and incongruent test trials of each task were compared, and results revealed that both groups of infants looked significantly longer at the incongruent trial on the intention and true-belief tasks. In contrast, only 18-month-olds looked significantly longer at the incongruent trial of the desire task and neither age group looked significantly longer at the incongruent trial on the false-belief task. Additionally, intertask comparisons revealed only a significant relation between performance on the false-belief and intention task. These findings suggest that implicit intention and true-belief understanding emerge earlier than desire and false-belief understanding and that ToM constructs do not appear to be integrated, as is the case for explicit ToM.

Perceptual access reasoning: developmental stage or system 1 heuristic?

In contrast with the two dominant views in Theory of Mind development, the Perceptual Access Reasoning hypothesis of Fabricius and colleagues is that children don’t understand the mental state of belief until around 6 years of age. Evidence for this includes data that many children ages 4 and 5, who pass the standard 2-location false belief task, nonetheless fail the true belief task, and often fail a 3-location false belief task by choosing the irrelevant option. These findings can be explained by the PAR hypothesis but pose challenges for the two dominant views. I argue against an alternate hypothesis which is proposed by Anika Fiebich in a recent paper. According to Fiebich, PAR is not a distinct transitional stage in children’s theory of mind development, but is a fast and frugal System 1 heuristic which fades once children become fluent in social reasoning. However, I point out a number of problems with Fiebich’s proposal and argue for the superiority of the PAR hypothesis. I also present five reasons to be skeptical about the findings of Perner and Horn which purportedly show that 4- and 5-year-olds can pass the 3-location false belief task when suitably modified. This is a further difficulty for Fiebich’s proposal, since she relies on these findings in her fluency theory. Finally, I sketch a dual systems theory of mind account based upon the PAR hypothesis which is different from Fiebich’s.

The Weird World of Cross-Cultural False-Belief Research: A True- and False-Belief Study Among Samoan Children Based on Commands

Previous cross-cultural research using false-belief tasks has explored whether children's theory of mind develops synchronously across cultures. Success on false-belief tasks is usually interpreted as an important indicator of children's mental state understanding, but inconsistent findings have led to questions regarding the interpretation of children's success and failure. Based on the assumptions of perceptual access reasoning (Hedger & Fabricius, 2011) and reflecting on inconsistencies in cross-cultural false-belief research, we argue for the advantages of the additional use of true-belief tasks, which can help to differentiate between different levels of children's reasoning. Consequently, a false-belief task and a true-belief task were derived from typical Samoan adult–child interactions. The performance of 40 Samoan children aged 5 to 7 years old was compared to the performance of 40 age-matched German children. While German children passed both tasks, Samoan children failed the false-belief task and did not reply above chance level in the true-belief task. According to our knowledge, this is the first study using both a false-belief task and true-belief task in a cross-cultural setting. Our results reveal additional patterns of reasoning that are neither in line with perceptual access reasoning nor with a representational understanding of false beliefs. The study is discussed in terms of a more general problem of experimental research in non-Western settings.

Mindreading with ease? Fluency and belief reasoning in 4- to 5-year-olds

For decades, philosophers and psychologists have assumed that children understand other people’s behavior on the basis of Belief Reasoning (BR) at latest by age 5 when they pass the false belief task. Furthermore, children’s use of BR in the true belief task has been regarded as being ontogenetically prior. Recent findings from developmental studies challenge this view and indicate that 4- to 5-year-old children make use of a reasoning strategy, which is cognitively less demanding than BR and called perceptual access reasoning (PAR), in true belief tasks. I appeal to research on fluency to explain these findings. On my account, 4- to 5- year-old children understand other people’s behavior by means of BR if they experience cognitive strain (such as in false belief tasks) but they revert to simpler heuristics PAR when such an experience is missing (such as in true belief tasks).

Assessing true and false belief in young children with cerebral palsy through anticipatory gaze behaviours: A pilot study

Children with a clinical description of cerebral palsy (CP) commonly experience cognitive and sensory difficulties that co-occur with motor impairment, and for some children this can include impairments in social communication. While research has begun to examine theory of mind abilities in children with CP, relatively little is known about social communication difficulties in this population. Assessing theory of mind abilities in children with CP using traditional procedures such as the classic Sally-Anne task can be problematic if performance is affected by physical difficulties in signalling responses and/or by cognitive and language demands inherent to the task itself. The central aim of this study therefore was to examine the potential of using a new action anticipation task and eye-tracking technique to assess implicit true and false belief understanding in four developmentally young children with quadriplegic cerebral palsy who had little or no functional speech, and one language age matched child with Down syndrome who did not have severe motor impairment. All children in this study consistently demonstrated anticipatory gaze behaviours in the context of the true belief task. One child with CP and the child with Down syndrome demonstrated anticipatory gaze behaviours indicative of an ability to attribute false belief. The findings are discussed in relation to the application of action anticipation and eye-tracking paradigms in research and clinical practice.

True Belief Belies False Belief: Recent Findings of Competence in Infants and Limitations in 5-Year-Olds, and Implications for Theory of Mind Development

False belief tasks have enjoyed a monopoly in the research on children’s development of a theory of mind. They have been granted this status because they promise to deliver an unambiguous assessment of children’s understanding of the representational nature of mental states. Their poor cousins, true belief tasks, have been relegated to occasional service as control tasks. That this is their only role has been due to the universal assumption that correct answers on true belief tasks are inherently ambiguous regarding the level of the child’s understanding of mental states. It has also been due to the universal assumption that nothing in the child’s developing theory of mind would lead to systematically incorrect answers on true belief tasks. We review new findings that 4- and 5-year-olds do err, systematically and profoundly, on the true belief versions of all the extant belief tasks. This reveals an intermediate level of understanding in the development of children’s theory of mind. Researchers have been unaware of this intermediate level because it produces correct answers in false belief tasks. A simple two-task battery—one true belief task and one false belief task—is sufficient to remove the ambiguity from each task. The new findings show that children do not acquire an understanding of beliefs, and hence a representational theory of mind, until after 6 years of age, or 2 years later than most developmental psychologists have concluded. This raises the question of how to interpret other new findings that infants are able to pass false belief tasks. We review these new infant studies, as well as recent studies on chimpanzees, in light of older children’s failure on true belief tasks, and end with some speculation about how all of these new findings might be reconciled.

Where will the triangle look for it? Attributing false beliefs to a geometric shape at 17 months

Prior research on implicit mind-reading skills has focussed on how infants anticipate other persons' actions. This study investigated whether 11- and 17-month-olds spontaneously attribute false beliefs (FB) even to a simple animated geometric shape. Infants were shown a triangle chasing a disk through a tunnel. Using an eye-tracker, we found that 17-month-olds in a change-of-location true belief (TB) task anticipated that the triangle would search for the disk in the correct place while in a FB test they anticipated that it would search for it in the wrong, belief congruent place. These results suggest that 17-month-olds' psychological-reasoning system is applied to the actions of unfamiliar agents and it is employed to anticipate agents' actions even in the absence of any morphological features that are typical of natural agents. These findings provide support for theoretical accounts that emphasize continuity in the development of theory of mind core concepts and belief reasoning skills.

Lifespan aging and belief reasoning: Influences of executive function and social cue decoding

Older adults often perform poorly on Theory of Mind (ToM) tests that require understanding of others’ beliefs and intentions. The course and specificity of age changes in belief reasoning across the adult lifespan is unclear, as is the cause of the age effects. Cognitive and neuropsychological models predict that two types of processing might influence age differences in belief reasoning: executive functioning and social cue detection. In the current study we assessed 129 adults aged between 18 and 86 on novel measures of ToM (video clips and verbal vignettes), which manipulated whether true or false belief reasoning was required. On both video and verbal tasks, older adults (aged 65–88) had specific impairments in false belief reasoning, but showed no such problem in performing true belief tasks. Middle-aged adults (aged 40–64) generally performed as well as the younger adults (aged 18–39). Difficulties in updating information in working memory (but not inhibitory problems) partially mediated the age differences in false belief reasoning. Also, the ability to decode biological motion, indexing social cue detection, partially mediated age-related variance in the ability to interpret false beliefs. These results indicate that age differences in decoding social cues and updating information in memory may be important influences on the specific problems encountered when reasoning about false beliefs in old age.

True or false: Do 5-year-olds understand belief?

In 3 studies (N = 188) we tested the hypothesis that children use a perceptual access approach to reason about mental states before they understand beliefs. The perceptual access hypothesis predicts a U-shaped developmental pattern of performance in true belief tasks, in which 3-year-olds who reason about reality should succeed, 4- to 5-year-olds who use perceptual access reasoning should fail, and older children who use belief reasoning should succeed. The results of Study 1 revealed the predicted pattern in 2 different true belief tasks. The results of Study 2 disconfirmed several alternate explanations based on possible pragmatic and inhibitory demands of the true belief tasks. In Study 3, we compared 2 methods of classifying individuals according to which 1 of the 3 reasoning strategies (reality reasoning, perceptual access reasoning, belief reasoning) they used. The 2 methods gave converging results. Both methods indicated that the majority of children used the same approach across tasks and that it was not until after 6 years of age that most children reasoned about beliefs. We conclude that because most prior studies have failed to detect young children's use of perceptual access reasoning, they have overestimated their understanding of false beliefs. We outline several theoretical implications that follow from the perceptual access hypothesis.

The effect of a looker’s past reliability on infants’ reasoning about beliefs.

We investigated whether 16-month-old infants' past experience with a person's gaze reliability influences their expectation about the person's ability to form beliefs. Infants were first administered a search task in which they observed an experimenter show excitement while looking inside a box that either contained a toy (reliable looker condition) or was empty (unreliable looker condition). The infants were then administered a true belief task in which they watched as the same experimenter hid a toy in 1 of 2 locations. In the test trial, the infants witnessed the experimenter search for the toy in a location that was consistent or inconsistent with her belief about the toy's location. Results for the true belief task indicated that only the infants in the reliable looker condition looked longer at the incongruent than at the congruent search behavior. These findings are consistent with evidence suggesting that infants encode the identity of agents based on past reliability and implicitly attribute beliefs to others during the 2nd year of life.

Problems with the Seeing = Knowing Rule

Abstract The view that children understand the mind via a coherent theory is supported by evidence that children rigidly follow a Seeing = Knowing Rule: seeing, and only seeing, leads to knowing. This paper presents two kinds of evidence that children do not follow this rule. First, we critically review previous findings that children neglect the role of inference and argue that these studies do not in fact support the view that children follow a Seeing = Knowing Rule. We then present two studies in which children who correctly attributed ignorance and false belief to an observer in a false belief task also attributed ignorance (Study 1) and false belief (Study 2) in true belief tasks. These findings demonstrate that children sometimes attribute ignorance and false belief to an observer who is granted visual access, an outcome that should not occur if children rigidly follow the Seeing = Knowing Rule. We end by discussing some problems associated with modifying the Seeing = Knowing Rule to account for children's failure on the true belief task.