It is largely acknowledged that natural languages emerge not just from human brains but also from rich communities of interacting human brains (Senghas, ). Yet the precise role of such communities and such interaction in the emergence of core properties of language has largely gone uninvestigated in naturally emerging systems, leaving the few existing computational investigations of this issue at an artificial setting. Here, we take a step toward investigating the precise role of community structure in the emergence of linguistic (...) conventions with both naturalistic empirical data and computational modeling. We first show conventionalization of lexicons in two different classes of naturally emerging signed systems: (a) protolinguistic “homesigns” invented by linguistically isolated Deaf individuals, and (b) a natural sign language emerging in a recently formed rich Deaf community. We find that the latter conventionalized faster than the former. Second, we model conventionalization as a population of interacting individuals who adjust their probability of sign use in response to other individuals' actual sign use, following an independently motivated model of language learning (Yang, , ). Simulations suggest that a richer social network, like that of natural (signed) languages, conventionalizes faster than a sparser social network, like that of homesign systems. We discuss our behavioral and computational results in light of other work on language emergence, and other work of behavior on complex networks. (shrink)

Cheap talk has often been thought incapable of supporting the emergence of cooperation because costless signals, easily faked, are unlikely to be reliable (Zahavi and Zahavi, 1997). I show how, in a social network model of cheap talk with reinforcement learning, cheap talk does enable the emergence of cooperation, provided that individuals also temporally discount the past. This establishes one mechanism that suffices for moving a population of initially uncooperative individuals to a state of mutually beneficial cooperation even in the (...) absence of formal institutions. (shrink)

Yang returns to the problem of referential ambiguity, addressed in the opening paper by Gleitman and Trueswell. Using a computational approach, he argues that “big data” approaches to resolving referential ambiguity are destined to fail, because of the inevitable computational explosion needed to keep track of contextual associations present when a word is uttered. Yang tests several computational models, two of which depend on one‐trial learning, as described in Gleitman and Trueswell’s paper. He concludes that such models outperform cross‐situational learning (...) models, thereby formalizing, reinforcing, and advancing the conclusions offered by Gleitman and Trueswell. (shrink)

1. Introduction 2. Reward-Guided Decision Making 3. Content in the Model 4. How to Deflate a Metarepresentational Reading Proust and Carruthers on metacognitive feelings 5. A Deflationary Treatment of RPEs? 5.1 Dispensing with prediction errors 5.2 What is use of the RPE focused on? 5.3 Alternative explanations—worldly correlates 5.4 Contrast cases 6. Conclusion Appendix: Temporal Difference Learning Algorithms.

The last couple of years have seen a rapid growth of interest in the study of crossmodal correspondences – the tendency for our brains to preferentially associate certain features or dimensions of stimuli across the senses. By now, robust empirical evidence supports the existence of numerous crossmodal correspondences, affecting people’s performance across a wide range of psychological tasks – in everything from the redundant target effect paradigm through to studies of the Implicit Association Test, and from speeded discrimination/classification tasks through (...) to unspeeded spatial localisation and temporal order judgment tasks. However, one question that has yet to receive a satisfactory answer is whether crossmodal correspondences automatically affect people’s performance , as opposed to reflecting more of a strategic, or top-down, phenomenon. Here, we review the latest research on the topic of crossmodal correspondences to have addressed this issue. We argue that answering the question will require researchers to be more precise in terms of defining what exactly automaticity entails. Furthermore, one’s answer to the automaticity question may also hinge on the answer to a second question: Namely, whether crossmodal correspondences are all ‘of a kind’, or whether instead there may be several different kinds of crossmodal mapping . Different answers to the automaticity question may then be revealed depending on the type of correspondence under consideration. We make a number of suggestions for future research that might help to determine just how automatic crossmodal correspondences really are. (shrink)

Learning in natural environments is often characterized by a degree of inconsistency from an input. These inconsistencies occur, for example, when learning from more than one source, or when the presence of environmental noise distorts incoming information; as a result, the task faced by the learner becomes ambiguous. In this study, we investigate how learners handle such situations. We focus on the setting where a learner receives and processes a sequence of utterances to master associations between objects and their labels, (...) where the source is inconsistent by design: It uses both “correct” and “incorrect” object‐label pairings. We hypothesize that depending on the order of presentation, the result of the learning may be different. To this end, we consider two types of symbolic learning procedures: the Object‐Label (OL) and the Label‐Object (LO) process. In the OL process, the learner is first exposed to the object, and then the label. In the LO process, this order is reversed. We perform experiments with human subjects, and also construct a computational model that is based on a nonlinear stochastic reinforcement learning algorithm. It is observed experimentally that OL learners are generally better at processing inconsistent input compared to LO learners. We show that the patterns observed in the learning experiments can be reproduced in the simulations if the model includes (a) an ability to regularize the input (and also to do the opposite, i.e., undermatch) and (b) an ability to take account of implicit negative evidence (i.e., interactions among different objects/labels). The model suggests that while both types of learners utilize implicit negative evidence in a similar way, there is a difference in regularization patterns: OL learners regularize the input, whereas LO learners undermatch. As a result, OL learners are able to form a more consistent system of image‐utterance associations, despite the ambiguous learning task. (shrink)

Time is a fundamental dimension of consciousness. Many studies of the “sense of agency” have investigated whether we attribute actions to ourselves based on a conscious experience of intention occurring prior to action, or based on a reconstruction after the action itself has occurred. Here, we ask the same question about a lower level aspect of action experience, namely awareness of the detailed spatial form of a simple movement. Subjects reached for a target, which unpredictably jumped to the side on (...) some trials. Participants expressed their expectancy of a target shift during the upcoming movement, pointed at the target as quickly and accurately as possible before returning to the start posiment to the target shift if required and reproduced the spatial path of the movement they had just made, as accurately as possible, to give an indication of their awareness of the pointing movement. We analysed the spatial disparity between the initial and the reproduced movements on those with a target shift. A negative disparity value, or undershoot, suggests that motor awareness merely reflects a sluggish record of coordinated motor performance, while a positive value, or overshoot, suggests that participants’ intention to point to the shifting target contributes more to their awareness of action than their actual pointing movement. Undershoot and overshoot thus measure the reconstructive and the preconstuctive aspects of action awareness, respectively. We found that trials on which subjects strongly expected a target shift showed greater overshoot and less undershoot than trials with lower expectancy. Conscious expectancy therefore strongly influences the experience of the detailed motor parameters of our actions. Further, a delay inserted either between the expectancy judgement and the pointing movement, or between the pointing movement and the reproduction of the movement, had no effect on visuomotor adjustment but strongly influenced action awareness. Delays during either interval boosted undershoots, suggesting increased reliance on a time-limited sensory memory for action. The experience of action is thus strongly influenced by prior thoughts and expectations, but only over a short time period. Thus, awareness of our actions is a dynamic and relatively flexible mixture of what we intend to do, and what our motor system actually does. (shrink)

It is argued that current attempts to model human learning behavior commonly fail on one of two counts: either the model assumptions are artificially restricted so as to permit the application of mathematical techniques in deriving their consequences, or else the required complex assumptions are imbedded in computer programs whose technical details obscure the theoretical content of the model. The first failing is characteristic of so-called mathematical models of learning, while the second is characteristic of computer simulation models. An approach (...) to model building which avoids both these failings is presented under the title of a black-box theory of learning. This method permits the statement of assumptions of any desired complexity in a language which clearly exhibits their theoretical content.Section II of the paper is devoted to the problem of testing and comparing alternative learning theories. The policy advocated is to abandon attempts at hypothesis testing. It is argued that, in general, we not only lack sufficient data and sufficiently powerful techniques to test hypotheses, but that the truth of a model is not really the issue of basic interest. A given model may be true in the sense that on the basis of available evidence we cannot statistically reject it, but not interesting in the sense that it provides little information about the processes underlying behavior. Rather, we should accept or reject models on the basis of how much information they provide about the way in which subjects respond to environmental structure. This attitude toward model testing is made precise by introducing a formal measure of the information content of a model. Finally, it is argued that the statistical concept of degrees-of-freedom is misleading when used in the context of model testing and should be replaced by a measure of the information absorbed from the data in estimating parameters. (shrink)

According to the reward-prediction error hypothesis of dopamine, the phasic activity of dopaminergic neurons in the midbrain signals a discrepancy between the predicted and currently experienced reward of a particular event. It can be claimed that this hypothesis is deep, elegant and beautiful, representing one of the largest successes of computational neuroscience. This paper examines this claim, making two contributions to existing literature. First, it draws a comprehensive historical account of the main steps that led to the formulation and subsequent (...) success of the RPEH. Second, in light of this historical account, it explains in which sense the RPEH is explanatory and under which conditions it can be justifiably deemed deeper than the incentive salience hypothesis of dopamine, which is arguably the most prominent contemporary alternative to the RPEH. (shrink)