Unobserved Causes Research Articles

Using unobserved causes to explain unexpected outcomes: The effect of existing causal knowledge on protection from extinction by a hidden cause.

People often rely on the covariation between events to infer causality. However, covariation between cues and outcomes may change over time. In the associative learning literature, extinction provides a model to study updating of causal beliefs when a previously established relationship no longer holds. Prediction error theories can explain both extinction and protection from extinction when an inhibitory (preventive) cue is present during extinction. In three experiments using the allergist causal learning task, we found that protection could also be achieved by a hidden cause that was inferred but not physically present, so long as that cause was a plausible preventer of the outcome. We additionally showed complete protection by a physically presented cue that was neutral rather than inhibitory at the outset of extinction. Both findings are difficult to reconcile with dominant prediction error theories. However, they are compatible with the idea of theory protection, where the learner attributes the absence of the outcome to the added cue (when present) or to a hidden cause, and therefore does not need to revise causal beliefs about A. Our results suggest that prediction error encourages changes in causal beliefs, but the nature of the change is determined by reasoning processes that incorporate existing knowledge of causal mechanisms and may be biased toward preservation of existing beliefs. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Neural Causal Information Extractor for Unobserved Causes.

Causal inference aims to faithfully depict the causal relationships between given variables. However, in many practical systems, variables are often partially observed, and some unobserved variables could carry significant information and induce causal effects on a target. Identifying these unobserved causes remains a challenge, and existing works have not considered extracting the unobserved causes while retaining the causes that have already been observed and included. In this work, we aim to construct the implicit variables with a generator-discriminator framework named the Neural Causal Information Extractor (NCIE), which can complement the information of unobserved causes and thus provide a complete set of causes with both observed causes and the representations of unobserved causes. By maximizing the mutual information between the targets and the union of observed causes and implicit variables, the implicit variables we generate could complement the information that the unobserved causes should have provided. The synthetic experiments show that the implicit variables preserve the information and dynamics of the unobserved causes. In addition, extensive real-world time series prediction tasks show improved precision after introducing implicit variables, thus indicating their causality to the targets.

Female "Paradox" in Atrial Fibrillation-Role of Left Truncation Due to Competing Risks.

Female sex in patients with atrial fibrillation (AF) is a controversial and paradoxical risk factor for stroke-controversial because it increases the risk of stroke only among older women of some ethnicities and paradoxical because it appears to contradict male predominance in cardiovascular diseases. However, the underlying mechanism remains unclear. We conducted simulations to examine the hypothesis that this sex difference is generated non-causally through left truncation due to competing risks (CR) such as coronary artery diseases, which occur more frequently among men than among women and share common unobserved causes with stroke. We modeled the hazards of stroke and CR with correlated heterogeneous risk. We assumed that some people died of CR before AF diagnosis and calculated the hazard ratio of female sex in the left-truncated AF population. In this situation, female sex became a risk factor for stroke in the absence of causal roles. The hazard ratio was attenuated in young populations without left truncation and in populations with low CR and high stroke incidence, which is consistent with real-world observations. This study demonstrated that spurious risk factors can be identified through left truncation due to correlated CR. Female sex in patients with AF may be a paradoxical risk factor for stroke.

Effects of Carbon Emission Trading on Companies’ Market Value: Evidence from Listed Companies in China

Emissions trading schemes (ETSs) are effective measures that facilitate economic growth and carbon mitigation, especially for developing countries such as China. These schemes can further affect the cash flow, production, and investment decisions of regulated companies. However, few empirical studies have explored how ETSs promote companies’ market value. We systematically evaluate the influence of the carbon emission trading (CET) policy on companies’ market value and explore the influential mechanism. We use the data of listed companies from the Chinese stock “A” markets and employ the difference-in-difference method to account for the unobserved cause of the CET policy regarding companies’ market value. Robust benchmark regression results reveal that the CET policy promotes companies’ market value significantly. The mechanism analysis reveals that the CET policy can improve the market value of listed companies by influencing the carbon price, innovative activities, and carbon disclosure. The results of the heterogeneity analysis show that the CET policy’s impact on companies’ market value is heterogeneous in terms of marketization degree, industry, firm ownership, and different regions. We suggest that the carbon pricing mechanism, degree of market perfection, carbon disclosure policy, and carbon finance should be optimized to improve the efficiency of ETSs.

Control variables approach to estimate semiparametric models of mismeasured endogenous regressors with an application to U.K. twin data

We study the identification and estimation of semiparametric models with mismeasured endogenous regressors using control variables that ensure the conditional covariance restriction on endogenous regressors and unobserved causes. We provide a set of sufficient conditions for identification, which control for both endogeneity and measurement error. We propose a sieve-based estimator and derive its asymptotic properties. Given the sieve approximation, our proposed estimator is easy to implement as weighted least squares. Monte Carlo simulations illustrate that our proposed estimator performs well in the finite samples. In an empirical application, we estimate the return to education on earnings using U.K. twin data, in which self-reported education is potentially measured with error and is also correlated with unobserved factors. Our approach utilizes the twin’s reported education as a control variable to obtain consistent estimates. We find that a one-year increase in education leads to an 11% increase in hourly wage. The estimate is significantly higher than those from OLS and IV approaches which are potentially biased. The application underscores that our proposed estimator is useful to correct for both endogeneity and measurement error in estimating returns to education.

Robust point-process Granger causality analysis in presence of exogenous temporal modulations and trial-by-trial variability in spike trains.

Assessing directional influences between neurons is instrumental to understand how brain circuits process information. To this end, Granger causality, a technique originally developed for time-continuous signals, has been extended to discrete spike trains. A fundamental assumption of this technique is that the temporal evolution of neuronal responses must be due only to endogenous interactions between recorded units, including self-interactions. This assumption is however rarely met in neurophysiological studies, where the response of each neuron is modulated by other exogenous causes such as, for example, other unobserved units or slow adaptation processes. Here, we propose a novel point-process Granger causality technique that is robust with respect to the two most common exogenous modulations observed in real neuronal responses: within-trial temporal variations in spiking rate and between-trial variability in their magnitudes. This novel method works by explicitly including both types of modulations into the generalized linear model of the neuronal conditional intensity function (CIF). We then assess the causal influence of neuron i onto neuron j by measuring the relative reduction of neuron j's point process likelihood obtained considering or removing neuron i. CIF's hyper-parameters are set on a per-neuron basis by minimizing Akaike's information criterion. In synthetic data sets, generated by means of random processes or networks of integrate-and-fire units, the proposed method recovered with high accuracy, sensitivity and robustness the underlying ground-truth connectivity pattern. Application of presently available point-process Granger causality techniques produced instead a significant number of false positive connections. In real spiking responses recorded from neurons in the monkey pre-motor cortex (area F5), our method revealed many causal relationships between neurons as well as the temporal structure of their interactions. Given its robustness our method can be effectively applied to real neuronal data. Furthermore, its explicit estimate of the effects of unobserved causes on the recorded neuronal firing patterns can help decomposing their temporal variations into endogenous and exogenous components.

Medicine and Paradoxography in the Ancient World by George Kazantzidis

BOOK REVIEWS/COMPTES RENDUS 423 the Romanization of Greek culture during the empire obviated the role of the novel as a form of resistance; historical relevance was supplanted by literary conceit (19). Times changed again by the fourth century a.d.: in Heliodorus’ Aethiopica only people who had once been slaves can be enslaved (216), which echoes contemporary sentiment in the Constantinian legislation restricting slavery to vernae, i.e., home-born slaves. This is not the first book one should read on the Greek novel, but it is a necessary one. Greek is often cited, usually in parentheses after a summation, but, in spite of this, the book remains accessible for student term essays or oral reports. Owens avoids jargon and his conviction in his thesis never leads him to belabour the point. The discussion of scholarly debates is largely restricted to the notes, and Owens is generous in assessing the contributions of his colleagues to the field. I admit an interest in reading this book for review since a consistent feature of Silenos and the satyr chorus in satyr drama is to declare themselves “slaves of Dionysos.” Epigraphical and material evidence for the continued performance of satyr drama well into the Roman empire is now becoming known,3 and one wonders if the Greek novel and imperial performance of satyr drama could have influenced one another. Satyrus in Achilles Tatius (155–160) raises the possibility of cross-pollination between the two genres in imperial salons and performance. Altogether, the book well illustrates the centrality of slavery to the Greek novel, and shows how the ancient Greek novel became a cultural entrepot through which new comedy, Roman comedy, and other genres came to be appreciated in the Greek-speaking world under Roman rule. Carleton University George W. M. Harrison Medicine and Paradoxography in the Ancient World. Edited by George Kazantzidis. Berlin: De Gruyter. 2019. Pp. viii, 225. Since at least P. M. Fraser’s Ptolemaic Alexandria (Oxford 1972), Greek paradoxography (the term is a coinage of the twelfth-century poet-grammarian Tzetzes) has been reproached as a derivative genre that pilfers and mutilates the fruits of true scientific, geographic, and historical research. The growing literature on paradoxography, however, has begun to challenge this prevailing conception by situating the genre, and ancient technical literature generally, in its Sitz im Leben to be judged according to its own development and relative merits. This volume, which contributes to these scholarly advances, collects ten wide-ranging papers given at a conference in Patras in 2016, and presents a concerted effort to uncover the sympathies and antipathies between paradoxography and the respectable field of ancient medicine. The editor, George Kazantzidis, identifies in his substantial and genre-traversing introduction (1–40) the synergies between ancient Greek medicine and discourses of wonder . He highlights, for instance, how the stupefying effects and lingering traces of disease could often, like thaumata, provoke intense aesthetic and intellectual reactions. Many of the essays that follow tacitly hinge on Kazantzidis’s unsurprising claim that ancient medicine strived, mostly, to demystify wonders by probing their unobserved causes, in contrast to paradoxography, which eschewed rational explanations to safeguard a sense of 3 For example, M. Skotheim and G. W. M. Harrison in Reconstructing Satyr Drama from Its Remnants (Berlin forthcoming). 424 PHOENIX the marvelous. Katerina Oikonomopoulou (53–71), for example, compares the rhetoric and ethos of two Aristotelian texts (the gargantuan Problemata and the paradoxographical On Marvellous Things Heard) to argue that although paradoxes and anomalies stand at the heart of both, the natural-scientific Problemata endeavors to subordinate their affective qualities to the intellectual and cognitive opportunities that they afford. On Marvellous Things Heard, by contrast, drops causes by the wayside, as if to heighten the pleasurable dimensions of wonders. The problemata genre reappears in Michiel Meeusen’s study (199–214) on the epistemological and moral limits that ps.-Alexander sets for his medical and scientific program in Medical Puzzles and Natural Problems: the causes of preternatural phenomena possessing “unspeakable properties (€di—thtew Ärrhtoi, Med. Puzz. 1, Praef. 59)” (208) are ascertainable by divine reason alone and thus excluded from inquiry. In demarcating a class of marvels to be excluded from investigation, ps...

Statistical resolutions for large variabilities in hair mineral analysis.

Measuring biomaterials is usually subject to error. Measurement errors are classified into either random errors or biases. Random errors can be well controlled using appropriate statistical methods. But, biases due to unknown, unobserved, or temporary causes, may lead to biased conclusions. This study describes a verification method to examine whether measurement errors are random or not and to determine efficient statistical methods.A number of studies have dealt with associations between hair minerals and exposures such as health, dietary or environmental conditions. Most review papers, however, emphasize the necessity for validation of hair mineral measurements, since large variations can cause highly variable results. To address these issues, we answer the following questions: How can we ascertain the reliability of measurements?How can we assess and control the variability of measurements?How do we efficiently determine associations between hair minerals and exposures?How can we concisely present the reference values?Since hair minerals all have distinctive natures, it would be unproductive to examine each mineral individually to find significant and consistent answers that apply to all minerals. To surmount this difficulty, we used one simple model for all minerals to explore quantitative answers. Hair mineral measurements of six-year-old children were analyzed based on the statistical model. The analysis verified that most of the measurements were reliable, and their inter-individual variations followed two-parameter distributions. These results allow for sophisticated study designs and efficient statistical methods to examine the effects of various kinds of exposures on hair minerals.

Analysing causal structures with entropy

A central question for causal inference is to decide whether a set of correlations fits a given causal structure. In general, this decision problem is computationally infeasible and hence several approaches have emerged that look for certificates of compatibility. Here, we review several such approaches based on entropy. We bring together the key aspects of these entropic techniques with unified terminology, filling several gaps and establishing new connections, all illustrated with examples. We consider cases where unobserved causes are classical, quantum and post-quantum, and discuss what entropic analyses tell us about the difference. This difference has applications to quantum cryptography, where it can be crucial to eliminate the possibility of classical causes. We discuss the achievements and limitations of the entropic approach in comparison to other techniques and point out the main open problems.

Endogeneity bias modeling using observables

This paper proposes an alternative solution to the endogeneity problem by explicitly modeling the joint interaction of the endogenous variables and the unobserved causes of the dependent variable as a function of additional observables. We derive identification of the parameters, develop an estimator, and establish its consistency and asymptotic normality.

Using Bayesian models to assess the capability of a manufacturing process in the presence of unobserved assignable causes

The capability of a manufacturing process is a measure of how well the process is able to manufacture items within required engineering specifications. A process capability analysis is traditionally considered only after all assignable causes of variation have been removed. Under these conditions, a manufacturing process is considered to be in control. This is often an idealized situation since assignable causes such as tool wear, environmental conditions and batch variation in raw materials are inherent in many processes. In this article, the capability of a process is modelled in such a way that the process distribution is allowed to change between observed subgroups due to assignable causes. Standard and hierarchical Bayesian models are used. The Bayesian framework allows for prior information that is known about the manufacturing processes to be formally incorporated into the model. The hierarchical Bayesian framework is useful when the within group variation dominates the between group variation, and when an overall measure of process capability is desired for the entire process. The usefulness of the proposed methods is demonstrated through the application of several examples and issues such as practical implementation and computation are addressed.

The Invisible Hand: Toddlers Connect Probabilistic Events With Agentive Causes.

Children posit unobserved causes when events appear to occur spontaneously (e.g., Gelman & Gottfried, 1996). What about when events appear to occur probabilistically? Here toddlers (M=20.1months) saw arbitrary causal relationships (Cause A generated Effect A; Cause B generated Effect B) in a fixed, alternating order. The relationships were then changed in one of two ways. In the Deterministic condition, the event order changed (Event B preceded Event A); in the Probabilistic condition, the causal relationships changed (Cause A generated Effect B; Cause B generated Effect A). As intended, toddlers looked equally long at both changes (Experiment 1). We then introduced a previously unseen candidate cause. Toddlers looked longer at the appearance of a hand (Experiment 2) and novel agent (Experiment 3) in the Deterministic than the Probabilistic conditions, but looked equally long at novel non-agents (Experiment 4), suggesting that by 2 years of age, toddlers connect probabilistic events with unobserved agents.

What Is Going on Inside the Arrows? Discovering the Hidden Springs in Causal Models.

Using Gebharter's (2014) representation, we consider aspects of the problem of discovering the structure of unmeasured sub-mechanisms when the variables in those sub-mechanisms have not been measured. Exploiting an early insight of Sober's (1998), we provide a correct algorithm for identifying latent, endogenous structure-sub-mechanisms-for a restricted class of structures. The algorithm can be merged with other methods for discovering causal relations among unmeasured variables, and feedback relations between measured variables and unobserved causes can sometimes be learned.

Time Delayed Causal Gene Regulatory Network Inference with Hidden Common Causes.

Inferring the gene regulatory network (GRN) is crucial to understanding the working of the cell. Many computational methods attempt to infer the GRN from time series expression data, instead of through expensive and time-consuming experiments. However, existing methods make the convenient but unrealistic assumption of causal sufficiency, i.e. all the relevant factors in the causal network have been observed and there are no unobserved common cause. In principle, in the real world, it is impossible to be certain that all relevant factors or common causes have been observed, because some factors may not have been conceived of, and therefore are impossible to measure. In view of this, we have developed a novel algorithm named HCC-CLINDE to infer an GRN from time series data allowing the presence of hidden common cause(s). We assume there is a sparse causal graph (possibly with cycles) of interest, where the variables are continuous and each causal link has a delay (possibly more than one time step). A small but unknown number of variables are not observed. Each unobserved variable has only observed variables as children and parents, with at least two children, and the children are not linked to each other. Since it is difficult to obtain very long time series, our algorithm is also capable of utilizing multiple short time series, which is more realistic. To our knowledge, our algorithm is far less restrictive than previous works. We have performed extensive experiments using synthetic data on GRNs of size up to 100, with up to 10 hidden nodes. The results show that our algorithm can adequately recover the true causal GRN and is robust to slight deviation from Gaussian distribution in the error terms. We have also demonstrated the potential of our algorithm on small YEASTRACT subnetworks using limited real data.

Evaluating the inverse reasoning account of object discovery

People routinely make inferences about unobserved objects. A hotel guest with welts on his arms, for example, will often worry about bed bugs. The discovery of unobserved objects almost always involves a backward inference from some observed effects (e.g., welts) to unobserved causes (e.g., bed bugs). The inverse reasoning account, which is typically formalized as Bayesian inference, posits that the strength of a backward inference is closely connected to the strength of the corresponding forward inference from the unobserved causes to the observed effects. We evaluated the inverse reasoning account of object discovery in three experiments where participants were asked to discover the unobserved “attractors” and “repellers” that controlled a “particle” moving within an arena. Experiments 1 and 2 showed that participants often failed to provide the best explanations for various particle motions, even when the best explanations were simple and when participants enthusiastically endorsed these explanations when presented with them. This failure demonstrates that object discovery is critically dependent on the processes that support hypothesis generation—processes that the inverse reasoning account does not explain. Experiment 3 demonstrated that people sometimes generate explanations that are invalid even according to their own forward inferences, suggesting that the psychological processes that support forward and backward inference are less intertwined than the inverse reasoning account suggests. The experimental findings support an alternative account of object discovery in which people rely on heuristics to generate possible explanations.

Cutaneous leishmaniasis: an unusual cause of nasal obstruction

Nasal and sinus-related illnesses are among the most common reasons for appointments to otolaryngologists and allergists. The usual causes include septal deviation, allergic rhinitis, polyps and so on. Apart from the common cold, allergic rhinitis is probably the most common cause of nasal obstruction. Allergic rhinitis is most common during adolescence and improves significantly during middle age and wanes during old age. The anatomical causes including septal deviation and adenoids constitute other common causes of nasal obstruction. The unusual causes include middle turbinate osteoma, concha bullosa of the inferior turbinate, congenital nasal pyriform aperture stenosis, and congenital inferior turbinate hypertrophy. We describe the case of a 44-year-old man with localized cutaneous leishmaniasis, a hitherto unobserved cause of nasal obstruction, cured with intravenous sodium stibogluconate.

Toddlers infer unobserved causes for spontaneous events.

Previous research suggests that children infer the presence of unobserved causes when objects appear to move spontaneously. Are such inferences limited to motion events or do children assume that unexplained physical events have causes more generally? Here we introduce an apparently spontaneous event and ask whether, even in the absence of spatiotemporal and co-variation cues linking the events, toddlers treat a plausible variable as a cause of the event. Toddlers (24 months) saw a toy that appeared to light up either spontaneously or after an experimenter’s action. Toddlers were also introduced to a button but were not shown any predictive relation between the button and the light. Across three different dependent measures of exploration, predictive looking (Study 1), prompted intervention (Study 2), and spontaneous exploration (Study 3), toddlers were more likely to represent the button as a cause of the light when the event appeared to occur spontaneously. In Study 4, we found that even in the absence of a plausible candidate cause, toddlers engaged in selective exploration when the light appeared to activate spontaneously. These results suggest that toddlers’ exploration is guided by the causal explanatory power of events.

Understanding of and reasoning about object–object relationships in long-tailed macaques?

Diagnostic reasoning, defined as the ability to infer unobserved causes based on the observation of their effects, is a central cognitive competency of humans. Yet, little is known about diagnostic reasoning in non-human primates, and what we know is largely restricted to the Great Apes. To track the evolutionary history of these skills within primates, we investigated long-tailed macaques’ understanding of the significance of inclinations of covers of hidden food as diagnostic indicators for the presence of an object located underneath. Subjects were confronted with choices between different objects that might cover food items. Based on their physical characteristics, the shape and orientation of the covers did or did not reveal the location of a hidden reward. For instance, hiding the reward under a solid board led to its inclination, whereas a hollow cup remained unaltered. Thus, the type of cover and the occurrence or absence of a change in their appearance could potentially be used to reason diagnostically about the location of the reward. In several experiments, the macaques were confronted with a varying number of covers and their performance was dependent on the level of complexity and on the type of change of the covers’ orientation. The macaques could use a board’s inclination to detect the reward, but failed to do so if the lack of inclination was indicative of an alternative hiding place. We suggest that the monkeys’ performance is based on a rudimentary understanding of causality, but find no good evidence for sophisticated diagnostic reasoning in this particular domain.

Common Factors and Causal Networks

The target article touches upon some of the most difficult and essential questions in personality psychology. Questioning the notion of a common factor as an as–yet–unobserved common cause of a behaviour domain's exemplars, the authors propose using graphical representations to inspire hypotheses of more complex causal structures. I do not find the case for the de–emphasis of the common factor model to be compelling for those behaviour domains (cognitive abilities) with which I am most familiar. It behoves all personality psychologists, however, to question the foundations of their favoured tools. Copyright © 2012 John Wiley & Sons, Ltd.

Effect of grouping of evidence types on learning about interactions between observed and unobserved causes.

When a cause interacts with unobserved factors to produce an effect, the contingency between the observed cause and effect cannot be taken at face value to infer causality. Yet it would be computationally intractable to consider all possible unobserved, interacting factors. Nonetheless, 6 experiments found that people can learn about an unobserved cause participating in an interaction with an observed cause when the unobserved cause is stable over time. Participants observed periods in which a cause and effect were associated followed by periods of the opposite association ("grouped condition"). Rather than concluding a complete lack of causality, participants inferred that the observed cause does influence the effect (Experiment 1), and they gave higher causal strength estimates when there were longer periods during which the observed cause appeared to influence the effect (Experiment 2). Consistent with these results, when the trials were grouped, participants inferred that the observed cause interacted with an unobserved cause (Experiments 3 and 4). Indeed, participants could even make precise predictions about the pattern of interaction (Experiments 5 and 6). Implications for theories of causal reasoning are discussed.