Prior Beliefs Research Articles

Whose Pants Are on Fire? Journalists Correcting False Claims are Distrusted More Than Journalists Confirming Claims

Do people trust journalists who provide fact-checks? Building upon research on negativity bias, two studies support the hypothesis that people generally trust journalists when they confirm claims as true, but are relatively distrusting of journalists when they correct false claims. In Study 1, participants read a real fact-check that corrected or confirmed a claim about politics or economics. In Study 2, participants read a real report that corrected or confirmed a marketing claim for one of several products. Participants in both studies had higher levels of distrust for journalists providing corrections, perceiving them as more likely to be lying and possessing ulterior motives. This effect held even among corrections consistent with respondents’ prior beliefs (i.e., for claims that participants thought might be false). The results represent a novel reason why people distrust journalists and resist belief correction. We discuss implications for transparency in journalism, and for how journalists frame fact-checks.

Trust in a national anti-corruption agency: A survey experiment among citizens and experts

Are anti-corruption agencies able to secure public trust, promote public integrity and fight political corruption in the eyes of the general public? The paper investigates this question by focusing on France, which became a leader in the fight against corruption after the launch in 2013 of the High Authority for the Transparency in Public Life (HATVP). We run a survey experiment among 3000 citizens and 33 experts to collect their prior beliefs about political corruption, and then evaluate the impact of granting information about the track record of the national agency on citizens' perceptions of its effectiveness and legitimacy. The paper provides four main results. First, as expected, information provision has meaningful and positive impacts on citizens’ perceptions of the HATVP, political transparency, and representative democracy. Second, while most beneficial impacts are broad-based, treatment effects are as large or even larger among the most poorly informed and distrustful citizens. Third, the experiment points toward the existence of a modest “integrity paradox”, i.e., an increase in the salience or perceived severity of corruption when citizens are better informed about the anti-corruption agency. Fourth, information provision reduces the divergence of opinions between the average citizen and experts about the effectiveness of the HATVP and the dynamics of political integrity.

Learning From Aggregated Opinion

The capacity to leverage information from others’ opinions is a hallmark of human cognition. Consequently, past research has investigated how we learn from others’ testimony. Yet a distinct form of social information—aggregated opinion—increasingly guides our judgments and decisions. We investigated how people learn from such information by conducting three experiments with participants recruited online within the United States (N = 886) comparing the predictions of three computational models: a Bayesian solution to this problem that can be implemented by a simple strategy for combining proportions with prior beliefs, and two alternatives from epistemology and economics. Across all studies, we found the strongest concordance between participants’ judgments and the predictions of the Bayesian model, though some participants’ judgments were better captured by alternative strategies. These findings lay the groundwork for future research and show that people draw systematic inferences from aggregated opinion, often in line with a Bayesian solution.

Explainable Gaussian processes: a loss landscape perspective

Prior beliefs about the latent function to shape inductive biases can be incorporated into a Gaussian process (GP) via the kernel. However, beyond kernel choices, the decision-making process of GP models remains poorly understood. In this work, we contribute an analysis of the loss landscape for GP models using methods from chemical physics. We demonstrate ν-continuity for Matérn kernels and outline aspects of catastrophe theory at critical points in the loss landscape. By directly including ν in the hyperparameter optimisation for Matérn kernels, we find that typical values of ν can be far from optimal in terms of performance. We also provide an a priori method for evaluating the effect of GP ensembles and discuss various voting approaches based on physical properties of the loss landscape. The utility of these approaches is demonstrated for various synthetic and real datasets. Our findings provide insight into hyperparameter optimisation for GPs and offer practical guidance for improving their performance and interpretability in a range of applications.

How the manner in which data is visualized affects and corrects (mis)perceptions of political polarization.

While the mechanisms underlying polarization are complex, scholars have consistently found a pervasive overestimation of perceptions of polarization to be a contributing factor. We argue that one mitigation strategy that can work at scale to address such misperceptions might be relatively straightforward: better data visualizations of cross-party attitudes on key issues. In a large-scale (N = 6603), international replication, we find that mode of presentation-or the manner in which data are visually presented-plays a significant role in moderating perceptions of polarization, even for longstanding, divisive issues for which partisans would likely hold strong prior beliefs. Additionally, we find the effects that different modes of presentation have on issue-specific polarization also extend to participant beliefs about overall interparty polarization, with certain modes proving capable of not only promoting less polarized views but also enabling more accurate estimates of the extent to which political groups agree. Finally, our findings also suggest that the manner in which intergroup data are visualized may also exert influence over the degree to which political groups are essentialized-a finding with implications for not only political perception but also for apolitical social psychological phenomena such as dehumanization.

Bayesian parameter estimation using truncated normal distributions as priors for parameters in fundamental models of chemical processes

AbstractModellers of chemical processes with knowledge about plausible parameter values use Bayesian parameter estimation methods to account for their prior beliefs. Some modellers specify prior distributions with finite parameter ranges, such as uniform distributions and truncated normal distributions, because they better account for knowledge about realistic parameter ranges than normal prior distributions with parameter values ranging between and . We derive closed‐form objective functions for Bayesian parameter estimation with truncated normal priors and uniform priors, for the first time, so that parameter estimation can be performed by solving simple optimization problems rather than using complex sampling‐based techniques. A parametric bootstrapping method that considers truncated normal priors and model nonlinearity is proposed to determine 95% confidence intervals and joint confidence regions. A pharmaceutical case study is used to show the effectiveness of the proposed objective functions and bootstrapping methodology. Confidence regions from bootstrapping are similar to linearization‐based confidence regions that do not account for truncation when truncated areas in normal prior distributions are relatively small. More truncation, which corresponds to more‐precise prior knowledge about the parameters, results in smaller joint confidence regions. The proposed methods will be attractive for parameter estimation in complex process models because they can be less computationally intensive than Markov chain Monte Carlo methods that provide similar results.

The disconnect between metamemory and memory for emotional images

ABSTRACT Emotional information is reliably predicted to be remembered better than neutral information, and this has been found for words, images, and facial expressions. However, many studies find that these judgments of learning (JOLs) are not predictive of memory performance (e.g. [Hourihan, K. L. (2020). Misleading emotions: Judgments of learning overestimate recognition of negative and positive emotional images. Cognition and Emotion, 34(4), 771–782. https://doi.org/10.1080/02699931.2019.1682972]). The present study investigates and rules out numerous potential causes for this discrepancy between memory predictions and performance, including (1) reactivity to making JOLs, (2) idiosyncrasies of specific images used, (3), type of memory test, and (4) effects of fluency. Three additional experiments investigate whether JOLs can become more predictive of memory performance, either by experience with the task or by manipulating prior beliefs about memory for emotional images. In all experiments, we found the same effect: Emotional images are inaccurately predicted to be remembered better than neutral images. The results suggest that emotion is used as a heuristic for learning, resulting in low metamnemonic accuracy for emotional stimuli.

Erroneous predictive coding across brain hierarchies in a non-human primate model of autism spectrum disorder

In autism spectrum disorder (ASD), atypical sensory experiences are often associated with irregularities in predictive coding, which proposes that the brain creates hierarchical sensory models via a bidirectional process of predictions and prediction errors. However, it remains unclear how these irregularities manifest across different functional hierarchies in the brain. To address this, we study a marmoset model of ASD induced by valproic acid (VPA) treatment. We record high-density electrocorticography (ECoG) during an auditory task with two layers of temporal control, and applied a quantitative model to quantify the integrity of predictive coding across two distinct hierarchies. Our results demonstrate a persistent pattern of sensory hypersensitivity and unstable predictions across two brain hierarchies in VPA-treated animals, and reveal the associated spatio-spectro-temporal neural signatures. Despite the regular occurrence of imprecise predictions in VPA-treated animals, we observe diverse configurations of underestimation or overestimation of sensory regularities within the hierarchies. Our results demonstrate the coexistence of the two primary Bayesian accounts of ASD: overly-precise sensory observations and weak prior beliefs, and offer a potential multi-layered biomarker for ASD, which could enhance our understanding of its diverse symptoms.

Global food expenditure patterns diverge between low-income and high-income countries.

Globalization, income growth and changing cultural trends are believed to prompt consumers in low-income countries to adopt the more affluent diet of high-income countries. This study investigates the convergence of food expenditure patterns worldwide, focusing on total food expenditure, raw food categories and ultra-processed foods and beverages across more than 90 countries over the past decades. Contrary to prior belief, we find that food expenditure patterns of lower-income countries do not universally align with those of higher-income nations. This trend is evident across most raw food categories and ultra-processed foods and beverages, as the income level of a country continues to play a crucial role in determining its food expenditure patterns. Importantly, expenditure patterns offer estimates rather than a precise idea of dietary intake, reflecting consumer choices shaped by economic constraints rather than exact dietary consumption.

Predictive Processing During Cue-Outcome Associative Learning in Autistic Children.

Predictive coding theories posit that autism is characterized by an over-adjustment to prediction errors, resulting in frequent updates of prior beliefs. Atypical weighting of prediction errors is generally considered to negatively impact the construction of stable models of the world, but may also yield beneficial effects. In a novel associative learning paradigm, we investigated whether unexpected events trigger faster learning updates in favour of subtle but fully predictive cues in autistic children compared to their non-autistic counterparts. We also explored the relationship between children's language proficiency and their predictive performances. Anticipatory fixations and explicit predictions were recorded during three associative learning tasks with deterministic or probabilistic contingencies. One of the probabilistic tasks was designed so that a fully predictive but subtle cue was overshadowed by a less predictive salient one. Both autistic and non-autistic children based their learning on the salient cue, and, contrary to our predictions, showed no signs of updating in favour of the subtle cue. While both groups demonstrated associative learning, autistic children made less accurate explicit predictions than their non-autistic peers in all tasks. Explicit prediction performances were positively correlated with language proficiency in non-autistic children, but no such correlation was observed in autistic children. These results suggest no over-adjustment to prediction errors in autistic children and highlight the need to control for general performance in cue-outcome associative learning in predictive processing studies. Further research is needed to explore the nature of the relationship between predictive processing and language development in autism.

Updating Prospective Self-Efficacy Beliefs About Cardiac Interoception in Anorexia Nervosa: An Experimental and Computational Study.

Patients with anorexia nervosa (AN) typically hold altered beliefs about their body that they struggle to update, including global, prospective beliefs about their ability to know and regulate their body and particularly their interoceptive states. While clinical questionnaire studies have provided ample evidence on the role of such beliefs in the onset, maintenance, and treatment of AN, psychophysical studies have typically focused on perceptual and 'local' beliefs. Across two experiments, we examined how women at the acute AN (N = 86) and post-acute AN state (N = 87), compared to matched healthy controls (N = 180) formed and updated their self-efficacy beliefs retrospectively (Experiment 1) and prospectively (Experiment 2) about their heartbeat counting abilities in an adapted heartbeat counting task. As preregistered, while AN patients did not differ from controls in interoceptive accuracy per se, they hold and maintain 'pessimistic' interoceptive, metacognitive self-efficacy beliefs after performance. Modelling using a simplified computational Bayesian learning framework showed that neither local evidence from performance, nor retrospective beliefs following that performance (that themselves were suboptimally updated) seem to be sufficient to counter and update pessimistic, self-efficacy beliefs in AN. AN patients showed lower learning rates than controls, revealing a tendency to base their posterior beliefs more on prior beliefs rather than prediction errors in both retrospective and prospective belief updating. Further explorations showed that while these differences in both explicit beliefs, and the latent mechanisms of belief updating, were not explained by general cognitive flexibility differences, they were explained by negative mood comorbidity, even after the acute stage of illness.

Financial information and diverging beliefs

Standard Bayesians’ beliefs converge when they receive the same piece of new information. However, when agents initially disagree and have uncertainty about the precision of a signal, their disagreement might instead increase, despite receiving the same information. We demonstrate that this divergence of beliefs leads to a unimodal effect of the absolute surprise in the signal on trading volume. We show that this prediction is consistent with the empirical evidence using trading volume around earnings announcements of U.S. firms. We find evidence of elevated volume following moderate surprises and depressed volume following more extreme surprises, a pattern that is more pronounced when investors hold more distant prior beliefs and are more uncertain about earnings’ precision. The evidence is consistent with the model where investors disagree about stocks’ expected returns and do not know the precision of earnings as a signal about the firm’s value.

Impacts of supply chain leadership on dynamic pricing under consumer patience information asymmetry

AbstractAs the online market rapidly develops, firms will face strategic consumers who differ in buying times. Given that the leadership structure in supply chains affects the chain members' decision‐making and performance, we study this impact in a two‐period dynamic pricing setting with an upstream manufacturer and a downstream retailer, where the patience degree of consumers is the retailer's private information. We examine two leadership scenarios: retailer dominance and manufacturer dominance. The two members play a signaling game under retailer dominance, whereas the manufacturer can only set its wholesale price based on its prior belief under manufacturer dominance. Our analysis indicates that when consumers are more patient, the retailer tends to share its private information with the manufacturer under retailer dominance to avoid the reduction of its retail margin and enhance its profitability. However, the retailer may keep the consumers' patience information private when the manufacturer dominates the supply chain because the manufacturer will set a compromised wholesale price and passively cede profits to the retailer. Finally, we examine consumer surplus and social welfare. We demonstrate that retailer dominance is the Pareto optimal strategy.

Why don’t students draw when learning from science texts?

This study explored why students rarely create drawings when learning from science texts despite potential learning benefits. Undergraduates (n = 114) studied a 10-part text on the human respiratory system and took notes by choosing their own strategies (free choice group) or by choosing to create a drawing or write a verbal summary (forced choice group). Other students were instructed to create drawings (draw group) or write summaries (summarize group). All students then completed a series of post-tests. The forced choice group chose to draw significantly more frequently than the free choice group; however, both groups still overwhelmingly chose summarizing. Participants across all groups reported lower prior experience, lower expectancies for success, lower perceived value, and higher perceived cost of drawing compared to summarizing. Students’ prior experiences and beliefs about drawing were also associated with how frequently they chose to draw, providing implications for future instructional interventions.

Search with learning in the retail gasoline market

Abstract This article estimates a model of optimal search where consumers learn the distribution of gasoline prices during their driving trips. Our model incorporates traffic information and leverages this ordered search environment to recover parameters of the search and learning process using only station‐level price and market share data. We find that learning is a crucial component of search in this market. Consumers' prior beliefs regularly deviate from the true price distribution but are updated quickly following each new price observation. Counterfactuals reveal how these learning dynamics generate asymmetric search patterns commonly associated with asymmetric cost pass‐through.

Transdiagnostic failure to adapt interoceptive precision estimates across affective, substance use, and eating disorders: A replication and extension of previous results

Recent Bayesian theories of interoception suggest that perception of bodily states rests upon a precision-weighted integration of afferent signals and prior beliefs. In a previous study, we fit a computational model of perception to behavior on a heartbeat tapping task to test whether aberrant precision-weighting could explain misestimation of cardiac states in psychopathology. We found that, during an interoceptive perturbation designed to amplify afferent signal precision (inspiratory breath-holding), healthy individuals increased the precision-weighting assigned to ascending cardiac signals (relative to resting conditions), while individuals with anxiety, depression, substance use disorders, and/or eating disorders did not. In this pre-registered study, we aimed to replicate and extend our prior findings in a new transdiagnostic patient sample (N = 285) similar to the one in the original study. As expected, patients in this new sample were also unable to adjust beliefs about the precision of cardiac signals – preventing the ability to accurately perceive changes in their cardiac state. Follow-up analyses combining samples from the previous and current study (N = 719) also afforded power to identify group differences between narrower diagnostic categories, and to examine predictive accuracy when logistic regression models were trained on one sample and tested on the other. With this confirmatory evidence in place, future studies should examine the utility of interoceptive precision measures in predicting treatment outcomes and test whether these computational mechanisms might represent novel therapeutic targets.

Keeping up with ourselves: Multimodal processes underlying body ownership across the lifespan

The sense of a bodily self is thought to depend on adaptive weighting and integration of bodily afferents and prior beliefs. While the physical body changes in shape, size, and functionality across the lifespan, the sense of body ownership remains relatively stable. Yet, little is known about how multimodal integration underlying such sense of ownership is altered in ontogenetic periods of substantial physical changes. We aimed to study this link for the motor and the tactile domain in a mixed-realty paradigm where participants ranging from 7 to 80 years old saw their own body with temporally mismatching multimodal signals. Participants were either stroked on their hand or moved it, while they saw it in multiple trials with different visual delays. For each trial, they judged the visuo-motor/tactile synchrony and rated the sense of ownership for the seen hand. Visual dependence and proprioceptive acuity were additionally assessed. The results show that across the lifespan body ownership decreases with increasing temporal multisensory mismatch, both in the tactile and the motor domain. We found an increased sense of ownership with increasing age independent of delay and modality. Delay sensitivity during multisensory conflicts was not consistently related to age. No effects of age were found on visual dependence or proprioceptive accuracy. The results are at least partly in line with an enhanced weighting of top-down and a reduced weighting of bottom-up signals for the momentary sense of bodily self with increasing age.

Public service users’ responses to performance information: Bayesian learning or motivated reasoning?

Abstract Although performance information is widely promoted to improve the accountability of public service provision, behavioral research has revealed that motivated reasoning leads recipients to update their beliefs inaccurately. However, the reasoning processes of service users have been largely neglected. We develop a theory of public service users’ motivated reasoning about performance information stemming from their identification with the organization providing their services. We address a significant challenge to studying motivated reasoning—that widely used existing research designs cannot rule out alternative cognitive explanations, especially Bayesian learning, such that existing findings could be driven by strong prior beliefs rather than biased processing of new information. We use a research design incorporating Bayesian learning as a benchmark to identify departures from accuracy motivated reasoning process. We assess the empirical implications of the theory using a preregistered information provision experiment among parents with children using public schools. To assess their identity-based motivated reasoning, we provide them with noisy, but true, performance information about their school. Overall, we find no evidence of directionally motivated reasoning. Instead, parents change their beliefs in response to performance feedback in a way that largely reflects conservative Bayesian learning. Performance reporting to service users is less vulnerable to motivational biases in this context than suggested by the general literature on motivated reasoning. Furthermore, exploratory findings show that performance information can correct erroneous beliefs among misinformed service users, suggesting that investment in reporting performance to service users is worthwhile to inform their beliefs and improve accountability.

Promoting teaching improvement in elementary mathematics: A case with inverse relations

In this study, we examine two U.S. elementary teachers’ use of worked examples, representations, and deep questions in elementary mathematics lessons both before and after a project intervention designed to promote teaching improvement. This intervention was guided by a cognitive construct containing these three aspects, which were demonstrated via cross-cultural video lessons of both U.S. and Chinese teachers. We specifically analyze two teachers (Ann and Bea) who taught first-grade lessons on the topic of inverse relations. Although their lessons were rated similarly in year 1, the rating of their lessons in year 4, post-intervention, demonstrated different patterns of change. While Ann's use of worked examples and representations underwent no significant improvement, Bea spent more time unpacking a single worked example with a representational sequence fading from concrete to abstract (also called “concreteness fading”). Neither teacher showed improvement in asking deep questions. An analysis of the teachers’ responses to the project intervention revealed that their video noticing skills and personal takeaways were related to the areas of teaching where they improved. Other factors such as teachers’ prior knowledge, beliefs, and textbook resources, also played a role in teaching changes. These findings shed light on a recent endeavor called the “science of improvement” in mathematics teaching, which seeks to document teaching changes and the underlying mechanisms that might induce these changes. Implications for both research and practice are discussed.

Promoting multiple-text comprehension through thinking dispositions: a randomized controlled trial

ABSTRACT The present study tested the efficacy of an intervention on thinking dispositions on multiple-text comprehension performance. The participants in the study were 75 undergraduate students. First, participants completed pretest assessment tests (prior beliefs and knowledge, thinkinPromoting multiple-text comprehension through thinking dispositions: a randomized controlled trialg dispositions). Second, the participants were randomly assigned to the experimental (n = 36) or control condition (n = 35). In each condition they were asked to read instructional material about critical thinking when reading from multiple texts. The experimental group students were specifically instructed on the definition and functioning of thinking dispositions. Third, participants read two texts about the controversial topic of red meat consumption, answered to a comprehension test, and wrote an argumentative essay (coded by length and argumentative quality). Overall, the thinking disposition intervention was effective in promoting a higher engagement with the task and a higher level of argumentative quality.