Generality of the summation effect in human causal learning

Subsampling of cues in associative learning

Theories of learning distinguish between elemental and configural stimulus processing depending on whether stimuli are processed independently or as whole configurations. Evidence for elemental processing comes from findings of summation in animals where a compound of two dissimilar stimuli is deemed to be more predictive than each stimulus alone, whereas configural processing is supported by experiments using similar stimuli in which summation is not found. However, in humans the summation effect is robust and impervious to similarity manipulations. In three experiments in human predictive learning, we show that summation can be obliterated when partially reinforced cues are added to the summands in training and tests. This lack of summation only holds when the partially reinforced cues are similar to the reinforced cues (experiment 1) and seems to depend on participants sampling only the most salient cue in each trial (experiments 2a and 2b) in a sequential visual search process. Instead of attributing our and others' instances of lack of summation to the customary idea of configural processing, we offer a formal subsampling rule that might be applied to situations in which the stimuli are hard to parse from each other.

Causal invariance as a tacit aspiration: Analytic knowledge of invariance functions

For causal knowledge to be worth learning, it must remain valid when that knowledge is applied. Because unknown background causes are potentially present, and may vary across the learning and application contexts, extricating the strength of a candidate cause requires an assumption regarding the decomposition of the observed outcome into the unobservable influences from the candidate and from background causes. Acquiring stable, useable causal knowledge is challenging when the search space of candidate causes is large, such that the reasoner's current set of candidates may fail to include a cause that generalizes well to an application context. We have hypothesized that an indispensable navigation device that shapes our causal representations toward useable knowledge involves the concept of causal invariance - the sameness of how a cause operates to produce an effect across contexts. Here, we tested our causal invariance hypothesis by making use of the distinct mathematical functions expressing causal invariance for two outcome-variable types: continuous and binary. Our hypothesis predicts that, given identical prior domain knowledge, intuitive causal judgments should vary in accord with the causal-invariance function for a reasoner's perceived outcome-variable type. The judgments are made as if the reasoner aspires to formulate causally invariant knowledge. Our experiments involved two cue-competition paradigms: blocking and overexpectation. Results show that adult humans tacitly use the appropriate causal-invariance functions for decomposition. Our analysis offers an explanation for the apparent elusiveness of the blocking effect and the adaptiveness of intuitive causal inference to the representation-dependent reality in the mind.

Beyond the "Conceptual Nervous System": Can computational cognitive neuroscience transform learning theory?

In the last century, learning theory has been dominated by an approach assuming that associations between hypothetical representational nodes can support the acquisition of knowledge about the environment. The similarities between this approach and connectionism did not go unnoticed to learning theorists, with many of them explicitly adopting a neural network approach in the modeling of learning phenomena. Skinner famously criticized such use of hypothetical neural structures for the explanation of behavior (the "Conceptual Nervous System"), and one aspect of his criticism has proven to be correct: theory underdetermination is a pervasive problem in cognitive modeling in general, and in associationist and connectionist models in particular. That is, models implementing two very different cognitive processes often make the exact same behavioral predictions, meaning that important theoretical questions posed by contrasting the two models remain unanswered. We show through several examples that theory underdetermination is common in the learning theory literature, affecting the solvability of some of the most important theoretical problems that have been posed in the last decades. Computational cognitive neuroscience (CCN) offers a solution to this problem, by including neurobiological constraints in computational models of behavior and cognition. Rather than simply being inspired by neural computation, CCN models are built to reflect as much as possible about the actual neural structures thought to underlie a particular behavior. They go beyond the "Conceptual Nervous System" and offer a true integration of behavioral and neural levels of analysis.

Exploring the Effect of Stimulus Similarity on the Summation Effect in Causal Learning

Several contemporary models anticipate that the summation effect is modulated by the similarity between the cues forming a compound. Here, we explore this hypothesis in a series of causal learning experiments. Participants were presented with two visual cues that separately predicted a common outcome and later asked for the outcome predicted by the compound of the two cues. Similarity was varied between groups through changes in shape, spatial position, color, configuration, and rotation. In variance with the predictions of these models, we observed similar and strong levels of summation in both groups across all manipulations of similarity. The effect, however, was significantly reduced by manipulations intended to impact assumptions about the causal independence of the cues forming the compound, but this reduction was independent of stimulus similarity. These results are problematic for similarity-based models and can be more readily explained by rational approaches to causal learning.

Three Ways That Non-associative Knowledge May Affect Associative Learning Processes

Associative learning theories offer one account of the way animals and humans assess the relationship between events and adapt their behavior according to resulting expectations. They assume knowledge about event relations is represented in associative networks, which consist of mental representations of cues and outcomes and the associative links that connect them. However, in human causal and contingency learning, many researchers have found that variance in standard learning effects is controlled by "non-associative" factors that are not easily captured by associative models. This has given rise to accounts of learning based on higher-order cognitive processes, some of which reject altogether the notion that humans learn in the manner described by associative networks. Despite the renewed focus on this debate in recent years, few efforts have been made to consider how the operations of associative networks and other cognitive operations could potentially interact in the course of learning. This paper thus explores possible ways in which non-associative knowledge may affect associative learning processes: (1) via changes to stimulus representations, (2) via changes to the translation of the associative expectation into behavior (3) via a shared source of expectation of the outcome that is sensitive to both the strength of associative retrieval and evaluation from non-associative influences.

Why are some dimensions integral? Testing two hypotheses through causal learning experiments

Compound generalization and dimensional generalization are traditionally studied independently by different groups of researchers, who have proposed separate theories to explain results from each area. A recent extension of Shepard's rational theory of dimensional generalization allows an explanation of data from both areas within a single framework. However, the conceptualization of dimensional integrality in this theory (the direction hypothesis) is different from that favored by Shepard in his original theory (the correlation hypothesis). Here, we report two experiments that test differential predictions of these two notions of integrality. Each experiment takes a design from compound generalization and translates it into a design for dimensional generalization by replacing discrete stimulus components with dimensional values. Experiment 1 showed that an effect analogous to summation is found in dimensional generalization with separable dimensions, but the opposite effect is found with integral dimensions. Experiment 2 showed that the analogue of a biconditional discrimination is solved faster when stimuli vary in integral dimensions than when stimuli vary in separable dimensions. These results, which are analogous to more "non-linear" processing with integral than with separable dimensions, were predicted by the direction hypothesis, but not by the correlation hypothesis. This confirms the assumptions of the unified rational theory of stimulus generalization and reveals interesting links between compound and dimensional generalization phenomena.

Mechanisms of object recognition: what we have learned from pigeons

Behavioral studies of object recognition in pigeons have been conducted for 50 years, yielding a large body of data. Recent work has been directed toward synthesizing this evidence and understanding the visual, associative, and cognitive mechanisms that are involved. The outcome is that pigeons are likely to be the non-primate species for which the computational mechanisms of object recognition are best understood. Here, we review this research and suggest that a core set of mechanisms for object recognition might be present in all vertebrates, including pigeons and people, making pigeons an excellent candidate model to study the neural mechanisms of object recognition. Behavioral and computational evidence suggests that error-driven learning participates in object category learning by pigeons and people, and recent neuroscientific research suggests that the basal ganglia, which are homologous in these species, may implement error-driven learning of stimulus-response associations. Furthermore, learning of abstract category representations can be observed in pigeons and other vertebrates. Finally, there is evidence that feedforward visual processing, a central mechanism in models of object recognition in the primate ventral stream, plays a role in object recognition by pigeons. We also highlight differences between pigeons and people in object recognition abilities, and propose candidate adaptive specializations which may explain them, such as holistic face processing and rule-based category learning in primates. From a modern comparative perspective, such specializations are to be expected regardless of the model species under study. The fact that we have a good idea of which aspects of object recognition differ in people and pigeons should be seen as an advantage over other animal models. From this perspective, we suggest that there is much to learn about human object recognition from studying the "simple" brains of pigeons.

Explaining compound generalization in associative and causal learning through rational principles of dimensional generalization

How do we apply learning from one situation to a similar, but not identical, situation? The principles governing the extent to which animals and humans generalize what they have learned about certain stimuli to novel compounds containing those stimuli vary depending on a number of factors. Perhaps the best studied among these factors is the type of stimuli used to generate compounds. One prominent hypothesis is that different generalization principles apply depending on whether the stimuli in a compound are similar or dissimilar to each other. However, the results of many experiments cannot be explained by this hypothesis. Here, we propose a rational Bayesian theory of compound generalization that uses the notion of consequential regions, first developed in the context of rational theories of multidimensional generalization, to explain the effects of stimulus factors on compound generalization. The model explains a large number of results from the compound generalization literature, including the influence of stimulus modality and spatial contiguity on the summation effect, the lack of influence of stimulus factors on summation with a recovered inhibitor, the effect of spatial position of stimuli on the blocking effect, the asymmetrical generalization decrement in overshadowing and external inhibition, and the conditions leading to a reliable external inhibition effect. By integrating rational theories of compound and dimensional generalization, our model provides the first comprehensive computational account of the effects of stimulus factors on compound generalization, including spatial and temporal contiguity between components, which have posed long-standing problems for rational theories of associative and causal learning.

Two heads are better than one, but how much? Evidence that people's use of causal integration rules does not always conform to normative standards

Many theories of causal learning and causal induction differ in their assumptions about how people combine the causal impact of several causes presented in compound. Some theories propose that when several causes are present, their joint causal impact is equal to the linear sum of the individual impact of each cause. However, some recent theories propose that the causal impact of several causes needs to be combined by means of a noisy-OR integration rule. In other words, the probability of the effect given several causes would be equal to the sum of the probability of the effect given each cause in isolation minus the overlap between those probabilities. In the present series of experiments, participants were given information about the causal impact of several causes and then they were asked what compounds of those causes they would prefer to use if they wanted to produce the effect. The results of these experiments suggest that participants actually use a variety of strategies, including not only the linear and the noisy-OR integration rules, but also averaging the impact of several causes.