样例学习条件下的因果力估计

在逐个呈现因果样例的条件下,考察单一因果关系因果力估计的特点,同时检验联想解释,概率对比模型,权重DP模型,效力PC理论和pCI规则。实验让65名大学生被试估计不同化学药物影响动物基因变异的能力。实验结果表明：（1）对产生原因的因果力估计符合权重DP模型;（2）对预防原因的因果力估计较多符合效力PC理论;（3）因果力估计具有复杂多样性,难以用统一的模式加以描述和概括。     

How two causes are different from one: The use of (un)conditional information in Simpson's paradox

In a causally complex world, two (or more) factors may simultaneously be potential causes of an effect. To evaluate the causal efficacy of a factor, the alternative factors must be controlled for (or conditionalized on). Subjects judged the causal strength of two potential causes of an effect that covaried with each other, thereby setting up a Simpson's paradox--a situation in which causal judgments should vary widely depending on whether or not they are conditionalized on the alternative potential cause. In Experiments 1 (table format) and 2 (trial-by-trial format), the subjects did conditionalize their judgments for one causal factor on a known alternative cause. The subjects also demonstrated that they knew what information was needed to properly make causal judgments when two potential causes are available. In Experiment 3 (trial-by-trial), those subjects who were not told about the causal mechanism by which the alternative cause operated were less likely to conditionalize on it. However, the more a subject recognized the covariation between the alternative cause and the effect, the more the subject conditionalized on it. Such behavior may arise from the interaction between bottom-up and top-down processing.     

Non-bayesian inference: causal structure trumps correlation

The study tests the hypothesis that conditional probability judgments can be influenced by causal links between the target event and the evidence even when the statistical relations among variables are held constant. Three experiments varied the causal structure relating three variables and found that (a) the target event was perceived as more probable when it was linked to evidence by a causal chain than when both variables shared a common cause; (b) predictive chains in which evidence is a cause of the hypothesis gave rise to higher judgments than diagnostic chains in which evidence is an effect of the hypothesis; and (c) direct chains gave rise to higher judgments than indirect chains. A Bayesian learning model was applied to our data but failed to explain them. An explanation-based hypothesis stating that statistical information will affect judgments only to the extent that it changes beliefs about causal structure is consistent with the results.     

The role of learning data in causal reasoning about observations and interventions

Recent studies have shown that people have the capacity to derive interventional predictions for previously unseen actions from observational knowledge, a finding that challenges associative theories of causal learning and reasoning (e.g., Meder, Hagmayer, & Waldmann, 2008). Although some researchers have claimed that such inferences are based mainly on qualitative reasoning about the structure of a causal system (e.g., Sloman, 2005), we propose that people use both the causal structure and its parameters for their inferences. We here employ an observational trial-by-trial learning paradigm to test this prediction. In Experiment 1, the causal strength of the links within a given causal model was varied, whereas in Experiment 2, base rate information was manipulated while keeping the structure of the model constant. The results show that learners’ causal judgments were strongly affected by the observed learning data despite being presented with identical hypotheses about causal structure. The findings show furthermore that participants correctly distinguished between observations and hypothetical interventions. However, they did not adequately differentiate between hypothetical and counterfactual interventions.     

Recency and primacy in causal judgments: Effects of probe question and context switch on latent inhibition and extinction

Traditional associative models assume that associative weights are updated on a trial-by-trial basis. As a result, it is usually expected that responses based on these weights will tend to reflect the most recently presented contingencies. However, a number of studies of human causal judgments have shown primacy effects, wherein judgments obtained at the end of a series of trials are more strongly influenced by a contingency that was in force early in the sequence than by a contingency that was in force later in the sequence. The experiments described in this article replicated other work showing that requesting causal judgments during a sequence can reverse primacy and produce strong recency effects. Evidence was also obtained to suggest that primacy effects are produced by an interaction between latent inhibition and extinction processes and that requesting a judgment affects both of these processes.     

When learning order affects sensitivity to base rates: challenges for theories of causal learning

In three experiments we investigated whether two procedures of acquiring knowledge about the same causal structure, predictive learning (from causes to effects) versus diagnostic learning (from effects to causes), would lead to different base-rate use in diagnostic judgments. Results showed that learners are capable of incorporating base-rate information in their judgments regardless of the direction in which the causal structure is learned. However, this only holds true for relatively simple scenarios. When complexity was increased, base rates were only used after diagnostic learning, but were largely neglected after predictive learning. It could be shown that this asymmetry is not due to a failure of encoding base rates in predictive learning because participants in all conditions were fairly good at reporting them. The findings present challenges for all theories of causal learning.     

Transfer of control between causal predictive judgments and instrumental responding

Four experiments were conducted to explore outcome-specific transfer from causal predictive judgments to instrumental responding. A video game was designed in which participants had to defend Andalusia from navy and air force attacks. First, they learned the relationship between two instrumental responses (two key on a standard keyboard) and two different outcomes (destruction of the ships or destruction of the planes). Then they learned to predict which of two different stimuli predicted which outcome. Finally, they had the opportunity of making either of the two instrumental responses in the presence of either stimulus. Transfer was shown as a preference for the response that shared an outcome with the current stimulus. The presentation of the stimulus during the test produced a decrease in the overall rate of response. Responding to a neutral stimulus in Experiments 2 and 3 suggested that this overall decrease in responding was due to a combination of the time needed to process the meaning of the stimulus and the activation of the representation of the outcome in the presence of the stimulus during the test. Transfer between predictive judgments and instrumental responding mirrors the outcome-specific Pavlovian instrumental transfer observed in conditioning studies with rats.     

Contrasting cue-density effects in causal and prediction judgments

Many theories of contingency learning assume (either explicitly or implicitly) that predicting whether an outcome will occur should be easier than making a causal judgment. Previous research suggests that outcome predictions would depart from normative standards less often than causal judgments, which is consistent with the idea that the latter are based on more numerous and complex processes. However, only indirect evidence exists for this view. The experiment presented here specifically addresses this issue by allowing for a fair comparison of causal judgments and outcome predictions, both collected at the same stage with identical rating scales. Cue density, a parameter known to affect judgments, is manipulated in a contingency learning paradigm. The results show that, if anything, the cue-density bias is stronger in outcome predictions than in causal judgments. These results contradict key assumptions of many influential theories of contingency learning.     

Asymmetries in predictive and diagnostic reasoning

In this article, we address the apparent discrepancy between causal Bayes net theories of cognition, which posit that judgments of uncertainty are generated from causal beliefs in a way that respects the norms of probability, and evidence that probability judgments based on causal beliefs are systematically in error. One purported source of bias is the ease of reasoning forward from cause to effect (predictive reasoning) versus backward from effect to cause (diagnostic reasoning). Using causal Bayes nets, we developed a normative formulation of how predictive and diagnostic probability judgments should vary with the strength of alternative causes, causal power, and prior probability. This model was tested through two experiments that elicited predictive and diagnostic judgments as well as judgments of the causal parameters for a variety of scenarios that were designed to differ in strength of alternatives. Model predictions fit the diagnostic judgments closely, but predictive judgments displayed systematic neglect of alternative causes, yielding a relatively poor fit. Three additional experiments provided more evidence of the neglect of alternative causes in predictive reasoning and ruled out pragmatic explanations. We conclude that people use causal structure to generate probability judgments in a sophisticated but not entirely veridical way.     

Effect of local context of responding on human judgment of causality

Two experiments examined the effect of various relationships between a response (pressing the space bar of a computer) and an outcome (a triangle flashing on a screen) on judgments of the causal effectiveness of the response. In Experiment 1, when responses were required to be temporarily isolated from each other prior to an outcome, ratings of the causal effectiveness of the responses were higher than in a condition in which the probability of an outcome following a response was the same but in which no temporal isolation was required. In Experiment 2, when a number of responses were required to be emitted temporally close to the outcome, ratings of the causal effectiveness of the responses were lower than in a condition in which the probability of an outcome following a response was the same but in which no temporal proximity was required. These results suggest that, in addition to the overall probability that an outcome will follow a response, the local context of responding at the time an outcome is presented is critical in influencing ratings of causal effectiveness.     

Schedules of reinforcement as determinants of human causality judgments and response rates

Experiments examined the effect of relationships between a response and an outcome on human judgments of causal effectiveness. In Experiment 1, the time between outcomes obtained on a variable ratio (VR) schedule became the intervals for a yoked variable interval (VI) schedule. Response rates were higher on the VR than on the VI schedule. In Experiment 2, the number of responses required per outcome on a VR schedule were matched to that on a master VI 20-s schedule. Both ratings of causal effectiveness and response rates were higher in the VR schedule. In Experiment 3, tandem VI fixed-ratio (FR) schedules produced higher rates and judgments than equivalent conjunctive VI FR schedule. In Experiment 4, a VI schedule with a reinforcement requirement for a short interresponse time (IRT) produced higher rates and judgments than a simple VI schedule. These results corroborate the view that schedules are a determinant of both response rates and causal judgments. Few current theories of causal judgment predict this pattern of results.     

A dissociation between causal judgment and outcome recall

It has been suggested that causal learning in humans is similar to Pavlovian conditioning in animals. According to this view, judgments of cause reflect the degree to which an association exists between the cause and the effect. Inferential accounts, by contrast, suggest that causal judgments are reasoning based rather than associative in nature. We used a direct measure of associative strength, identification of the outcome with which a cause was paired (cued recall), to see whether associative strength translated directly into causal ratings. Causal compounds AB+ and CD+ were intermixed with A+ and C- training. Cued-recall performance was better for cue B than for cue D; thus, associative strength was inherited by cue B from the strongly associated cue A (augmentation). However, the reverse was observed on the causal judgment measure: Cue B was judged to be less causal than D (cue competition). These results support an inferential over an associative account of causal judgments.     

Cue interaction and judgments of causality: Contributions of causal and associative processes

In four experiments, the predictions made by causal model theory and the Rescorla-Wagner model were tested by using a cue interaction paradigm that measures the relative response to a given event based on the influence or salience of an alternative event. Experiments 1 and 2 uncorrelated two variables that have typically been confounded in the literature (causal order and the number of cues and outcomes) and demonstrated that overall contingency judgments are influenced by the causal structure of the events. Experiment 3 showed that trial-by-trial prediction responses, a second measure of causal assessment, were not influenced by the causal structure of the described events. Experiment 4 revealed that participants became less sensitive to the influence of the causal structure in both their ratings and their predictions as trials progressed. Thus, two experiments provided evidence for high-level (causal reasoning) processes, and two experiments provided evidence for low-level (associative) processes. We argue that both factors influence causal assessment, depending on what is being asked about the events and participants' experience with those events.     

Causal Learning: Association Versus Computation

Causal learning enables humans and other animals not only to predict important events or outcomes, but also to control their occurrence in the service of needs and desires. Computational theories assume that causal judgments are based on an estimate of the contingency between a causal cue and an outcome. However, human causal learning exhibits many of the characteristics of the associative learning processes thought to underlie animal conditioning. One problem for associative theory arises from the finding that judgments of the causal power of a cue can be revalued retrospectively after learning episodes when that cue is not present. However, if retrieved representations of cues can support learning, retrospective revaluation is anticipated by modified versions of standard associative theories.     

The effects of assessment type on verbal ratings of conditional stimulus valence and contingency judgments: implications for the extinction of evaluative learning

Recent research on causal learning found (a) that causal judgments reflect either the current predictive value of a conditional stimulus (CS) or an integration across the experimental contingencies used in the entire experiment and (b) that postexperimental judgments, rather than the CS's current predictive value, are likely to reflect this integration. In the current study, the authors examined whether verbal valence ratings were subject to similar integration. Assessments of stimulus valence and contingencies responded similarly to variations of reporting requirements, contingency reversal, and extinction, reflecting either current or integrated values. However, affective learning required more trials to reflect a contingency change than did contingency judgments. The integration of valence assessments across training and the fact that affective learning is slow to reflect contingency changes can provide an alternative interpretation for researchers' previous failures to find an effect of extinction training on verbal reports of CS valence.     

The impact of domain-specific beliefs on decisions and causal judgments

Extensive evidence suggests that people often rely on their causal beliefs in their decisions and causal judgments. To date, however, there is a dearth of research comparing the impact of causal beliefs in different domains. We conducted two experiments to map the influence of domain-specific causal beliefs on the evaluation of empirical evidence when making decisions and subsequent causal judgments. Participants made 120 decisions in a two-alternative forced-choice task, framed in either a medical or a financial domain. Before each decision, participants could actively search for information about the outcome (“occurrence of a disease” or “decrease in a company's share price”) on the basis of four cues. To analyze the strength of causal beliefs, we set two cues to have a generative relation to the outcome and two to have a preventive relation to the outcome. To examine the influence of empirical evidence, we manipulated the predictive power (i.e., cue validities) of the cues. Both experiments included a validity switch, where the four selectable cues switched from high to low validity or vice versa. Participants had to make a causal judgment about each cue before and after the validity switch. In the medical domain, participants stuck to the causal information in causal judgments, even when evidence was contradictory, while decisions showed an effect of both empirical and causal information. In contrast, in the financial domain, participants mainly adapted their decisions and judgments to the cue validities. We conclude that the strength of causal beliefs (1) is shaped by the domain, and (2) has a differential influence on the degree to which empirical evidence is taken into account in causal judgments and decision making.     

The power of causal beliefs and conflicting evidence on causal judgments and decision making

In two experiments, we investigated the relative impact of causal beliefs and empirical evidence on both decision making and causal judgments, and whether this relative impact could be altered by previous experience. Participants had to decide which of two alternatives would attain a higher outcome on the basis of four cues. After completing the decision task, they were asked to estimate to what extent each cue was a reliable cause of the outcome. Participants were provided with instructions that causally related two of the cues to the outcome, whereas they received neutral information about the other two cues. Two of the four cues—a causal and a neutral cue—had high validity and were both generative. The remaining two cues had low validity, and were generative in Experiment 1, but almost not related to the outcome in Experiment 2. Selected groups of participants in both experiments received pre-training with either causal or neutral cues, or no pre-training was provided. Results revealed that the impact of causal beliefs and empirical evidence depends on both the experienced pre-training and cue validity. When all cues were generative and participants received pre-training with causal cues, they mostly relied on their causal beliefs, whereas they relied on empirical evidence when they received pre-training with neutral cues. In contrast, when some of the cues were almost not related to the outcome, participants’ responses were primarily influenced by validity and—to a lesser extent—by causal beliefs. In either case, however, the influence of causal beliefs was higher in causal judgments than in decision making. While current theoretical approaches in causal learning focus either on the effect of causal beliefs or empirical evidence, the present research shows that both factors are required to explain the flexibility involved in human inferences.     

Judgement frequency, belief revision, and serial processing of causal information

The main aim of this research was to study the cognitive architecture underlying causal/covariation learning by investigating the frequency of judgement effect. Previous research has shown that decreasing the number of trials between opportunities to make a judgement in a covariation learning task led to a higher score after an a or d type of trial (positive cases) than after b and c trials (negative cases). Experiment 1 replicated this effect using a trial-by-trial procedure and examined the conditions under which it occurs. Experiment 2 demonstrated a similar frequency of judgement effect when the information was presented in the form of contingency tables. Associative or statistical single-mechanism accounts of causal and covariation learning do not provide a satisfactory explanation for these findings. An alternative belief revision model is presented.     

Influence of the cost of responding

In three experiments, the effect of costs associated with responding on judgments of the causal effectiveness of the response was examined. In Experiment 1, the temporal interval between outcomes was matched on a variable interval (VI) and a variable ratio (VR) schedule. When each response was made at some “cost,” and outcomes represented some “gain” for the subject, the rating of causal effectiveness for responses was higher on the VR than on the VI schedule. This relationship was absent when the outcome was a triangle Hash. In Experiment 2, the number of responses required per outcome on a VI and a VR schedule were matched, and responses on the VR schedule were rated as more causally effective. In Experiment 3, a VI-to-VR yoking procedure was used. With minimal response costs, judgments were similar on the VI and VR schedules, but with greater response costs, responses performed on the VR schedule were rated as more causally effective than those emitted on the VI schedule.     

Inferring interventional predictions from observational learning data

Previous research has shown that people are capable of deriving correct predictions for previously unseen actions from passive observations of causal systems (Waldmann & Hagmayer, 2005). However, these studies were limited, since learning data were presented as tabulated data only, which may have turned the task more into a reasoning rather than a learning task. In two experiments, we therefore presented learners with trial-by-trial observational learning input referring to a complex causal model consisting of four events. To test the robustness of the capacity to derive correct observational and interventional inferences, we pitted causal order against the temporal order of learning events. The results show that people are, in principle, capable of deriving correct predictions after purely observational trial-by-trial learning, even with relatively complex causal models. However, conflicting temporal information can impair performance, particularly when the inferences require taking alternative causal pathways into account.