Generalization is a fundamental problem solved by every cognitive system in essentially every domain. Although it is known that how people generalize varies in complex ways depending on the context or domain, it is an open question how people learn the appropriate way to generalize for a new context. To understand this capability, we cast the problem of learning how to generalize as a problem of learning the appropriate hypothesis space for generalization. We propose a normative mathematical framework for learning (...) how to generalize by learning inductive biases for which properties are relevant for generalization in a domain from the statistical structure of features and concepts observed in that domain. More formally, the framework predicts that an ideal learner should learn to generalize by either taking the weighted average of the results of generalizing according to each hypothesis space, with weights given by how well each hypothesis space fits the previously observed concepts, or by using the most likely hypothesis space. We compare the predictions of this framework to human generalization behavior with three experiments in one perceptual (rectangles) and two conceptual (animals and numbers) domains. Across all three studies we find support for the framework's predictions, including individual‐level support for averaging in the third study. (shrink)

The widely cited Nuffield Council on Bioethics ‘Intervention Ladder’ structurally embodies the assumption that personal autonomy is maximized by non-intervention. Consequently, the Intervention Ladder encourages an extreme ‘negative liberty’ view of autonomy. Yet there are several alternative accounts of autonomy that are both arguably superior as accounts of autonomy and better suited to the issues facing public health ethics. We propose to replace the one-sided ladder, which has any intervention coming at a cost to autonomy, with a two-sided ‘Balanced Intervention (...) Ladder,’ where intervention can either enhance or diminish autonomy. We show that not only the alternative, richer accounts of autonomy but even Mill’s classic version of negative liberty puts some interventions on the positive side of the ladder. (shrink)

Encyclopedia entry on the concepts of adaptation and adaptationism.

The selected effects or ‘etiological’ theory of Proper function is a naturalistic and realist account of biological teleology. It is used to analyse normativity in philosophy of language, philosophy of mind, philosophy of medicine, and elsewhere. The theory has been developed with a simple and intuitive view of natural selection. Traits are selected because of their positive effects on the fitness of the organisms that have them. These ‘selected effects’ are the Proper functions of the traits. Proponents argue that this (...) analysis of biological teleology has the unique advantage that the selected effect function of a trait is also a causal explanation of the trait: the trait exists because it performs this function. We show, however, that selected effect functions as currently defined explain the existence of traits only under highly restrictive assumptions about evolutionary dynamics. In many common scenarios in which traits evolve by natural selection, selected effect functions do not explain those traits. This is because definitions of selected effect function extract from any evolutionary scenario only the information that would be explanatorily relevant in the simple evolutionary scenario implicit in those definitions. When applied to more complex scenarios, selected effect functions omit the key information that is explanatorily relevant in those scenarios. The assumptions required for selected effect functions to be explanatory are particularly unlikely to hold in the domain that its proponents care most about—the evolution of representation. A more adequate selected effects theory of Proper functions may be possible, but will require much greater attention to the structure of actual evolutionary explanations. (shrink)

The rapid rise of generative AI, including LLMs, has prompted a great deal of concern, both within and beyond academia. One of these concerns is that generative models embed, reproduce, and therein potentially perpetuate all manner of bias. The present study offers an alternative perspective: exploring the potential of LLMs to detect bias in human generated text. Our target is genetic essentialism in obesity discourse in Australian print media. We develop and deploy an LLM-based classification model to evaluate a large (...) sample of relevant articles (n=26,163). We show that our model detects genetic essentialist biases as reliably as human experts; and find that, while genes figure less prominently in popular discussions of obesity than previous work might suggest, when genetic information is invoked, it is often presented in a biased way. Implications for future work are discussed. (shrink)

Machine learning algorithms may radically improve our ability to diagnose and treat disease. For moral, legal, and scientific reasons, it is essential that doctors and patients be able to understand and explain the predictions of these models. Scalable, customisable, and ethical solutions can be achieved by working together with relevant stakeholders, including patients, data scientists, and policy makers.

This anthology of articles provides contemporary international feminist perspectives on issues of identity, agency, and difference as they pertain to both feminist politics in particular, and contemporary western politics more generally.

Recent work in defence of direct realism has concentrated on the representationalist and disjunctivist responses to the arguments from illusion and hallucination, whilst relatively little attention has been given to the argument from causation which has been dismissed lightly as irrelevant or confused. However such charges arise from an ambiguity in the thesis which is being defended and the failure to distinguish between metaphysical and epistemological issues and between factual and conceptual claims. The argument from causation, as an argument against (...) the conjunction of metaphysical direct realism and an explanation of the perceptual process in terms of a naturalistically understood causal chain of events, has not been answered in the philosophical literature. Moreover when the process of perception is fleshed out in terms of contemporary cognitive science, the difficulties are compounded. Neither representation-friendly mainstream cognitive science, nor representation-averse radical embodied cognitive science, is compatible with a theory of perception which is at the same time both direct and robustly realist. (shrink)

Evidence suggests that psychedelics bring about their therapeutic outcomes in part through the subjective or qualitative effects they engender and how the individual interprets the resulting experiences. However, psychedelics are contraindicated for individuals who have been diagnosed with certain mental illnesses, on the grounds that these subjective effects may be disturbing or otherwise counter-therapeutic. Substantial resources are therefore currently being devoted to creating psychedelic substances that produce many of the same biological changes as psychedelics, but without their characteristic subjective effects. (...) In this article, we consider ethical issues arising from the prospect of such potential “nonsubjective” psychedelics. We are broadly supportive of efforts to produce such substances for both scientific and clinical reasons. However, we argue that such nonsubjective psychedelics should be reserved for those special cases in which the subjective effects of psychedelics are specifically contraindicated, whereas classic psychedelics that affect subjective experience should be considered the default and standard of care. After reviewing evidence regarding the subjective effects of psychedelics, we raise a number of ethical concerns around the prospect of withholding such typically positive, meaningful, and therapeutic experiences from most patients. (shrink)

Sustained attention, a key cognitive skill that improves during childhood and adolescence, tends to be worse in some emotional and behavioural disorders. Sustained attention is typically studied in non-affective task contexts; here, we used a novel task to index performance in affective versus neutral contexts across adolescence (N = 465; ages 11–18). We asked whether: (i) performance would be worse in negative versus neutral task contexts; (ii) performance would improve with age; (iii) affective interference would be greater in younger adolescents; (...) (iv) adolescents at risk for depression and higher in anxiety would show overall worse performance; and (v) would show differential performance in negative contexts. Results indicated that participants performed more poorly in negative contexts and showed age-related performance improvements. Those at risk of depression performed more poorly than those at lower risk. However, there was no difference between groups as a result of affective context. For anxiety there was no difference in performance as a function of severity. However, those with higher anxiety showed less variance in their reaction times to negative stimuli than those with lower anxiety. One interpretation is that moderate levels of emotional arousal associated with anxiety make individuals less susceptible to the distracting effects of negative stimuli. (shrink)

Mental disorders such as depression and anxiety are increasingly common. Yet there are too few specialists to offer help to everyone, and negative attitudes to psychological problems and their treatment discourage people from seeking it. As a result, many people never receive help for these problems. The Oxford Guide to Low Intensity CBT Interventions marks a turning point in the delivery of psychological treatments for people with depression and anxiety. Until recently, the only form of psychological intervention available for patients (...) with depression and anxiety was traditional one-to-one 60 minute session therapy - usually with private practitioners for those patients who could afford it. Now Low Intensity CBT Interventions are starting to revolutionize mental health care by providing cost effective psychological therapies which can reach the vast numbers of people with depression and anxiety who did not previously have access to effective psychological treatment. The Oxford Guide to Low Intensity CBT Interventions is the first book to provide a comprehensive guide to Low Intensity CBT interventions. It brings together researchers and clinicians from around the world who have led the way in developing evidence-based low intensity CBT treatments. It charts the plethora of new ways that evidence-based low intensity CBT can be delivered: for instance, guided self-help, groups, advice clinics, brief GP interventions, internet-based or book-based treatment and prevention programs, with supported provided by phone, email, internet, sms or face-to-face. These new treatments require new forms of service delivery, new ways of communicating, new forms of training and supervision, and the development of new workforces. They involve changing systems and routine practice, and adapting interventions to particular community contexts. The Oxford Guide to Low Intensity CBT Interventions is a state-of-the-art handbook, providing low intensity practitioners, supervisors, managers commissioners of services and politicians with a practical, easy-to-read guide - indispensible reading for those who wish to understand and anticipate future directions in health service provision and to broaden access to cost-effective evidence-based psychological therapies. (shrink)

Humans can learn complex functional relationships between variables from small amounts of data. In doing so, they draw on prior expectations about the form of these relationships. In three experiments, we show that people learn to adjust these expectations through experience, learning about the likely forms of the functions they will encounter. Previous work has used Gaussian processes—a statistical framework that extends Bayesian nonparametric approaches to regression—to model human function learning. We build on this work, modeling the process of learning (...) to learn functions as a form of hierarchical Bayesian inference about the Gaussian process hyperparameters. (shrink)

Paul E. Griffiths argues that most research on the emotions has been as misguided as Aristotelian efforts to study "superlunary objects" - objects...

Since the cognitive revolution, psychologists have developed formal theories of cognition by thinking about the mind as a computer. However, this metaphor is typically applied to individual minds. Humans rarely think alone; compared to other animals, humans are curiously dependent on stores of culturally transmitted skills and knowledge, and we are particularly good at collaborating with others. Rather than picturing the human mind as an isolated computer, we can imagine each mind as a node in a vast distributed system. Viewing (...) human cognition through the lens of distributed systems motivates new questions about how humans share computation, when it makes sense to do so, and how we can build institutions to facilitate collaboration. (shrink)

Convolutional neural networks (CNNs) are increasingly widely used in psychology and neuroscience to predict how human minds and brains respond to visual images. Typically, CNNs represent these images using thousands of features that are learned through extensive training on image datasets. This raises a question: How many of these features are really needed to model human behavior? Here, we attempt to estimate the number of dimensions in CNN representations that are required to capture human psychological representations in two ways: (1) (...) directly, using human similarity judgments and (2) indirectly, in the context of categorization. In both cases, we find that low-dimensional projections of CNN representations are sufficient to predict human behavior. We show that these low-dimensional representations can be easily interpreted, providing further insight into how people represent visual information. A series of control studies indicate that these findings are not due to the size of the dataset we used and may be due to a high level of redundancy in the features appearing in CNN representations. (shrink)

It’s an honour to have so many major contributors to the literature respond to our article and we thank them for their thoughtful responses. There are clear shared themes across these commentaries,...

Integrating the study of human diversity into the human evolutionary sciences requires substantial revision of traditional conceptions of a shared human nature. This process may be made more difficult by entrenched, 'folkbiological' modes of thought. Earlier work by the authors suggests that biologically naive subjects hold an implicit theory according to which some traits are expressions of an animal's inner nature while others are imposed by its environment. In this paper, we report further studies that extend and refine our account (...) of this aspect of folkbiology. We examine biologically naive subjects' judgments about whether traits of an animal are 'innate', 'in its DNA' or 'part of its nature'. Subjects do not understand these three descriptions to be equivalent. Both innate and in its DNA have the connotation that the trait is species-typical. This poses an obstacle to the assimilation of the biology of polymorphic and plastic traits by biologically naive audiences. Researchers themselves may not be immune to the continuing pull of folkbiological modes of thought. (shrink)

Elliott Sober''s selection for/selection of distinction has been widely used to clarify the idea that some properties of organisms are side-effects of selection processes. It has also been used, however, to choose between different descriptions of an evolutionary product when assigning biological functions to that product. We suggest that there is a characteristic error in these uses of the distinction. Complementary descriptions of function are misrepresented as mutually excluding one another. This error arises from a failure to appreciate that selection (...) processes can be described at several different theoretical levels. (shrink)

Inferring an individual's preferences from their observable behavior is a key step in the development of assistive decision-making technology. Although machine learning models such as neural networks could in principle be deployed toward this inference, a large amount of data is required to train such models. Here, we present an approach in which a cognitive model generates simulated data to augment limited human data. Using these data, we train a neural network to invert the model, making it possible to infer (...) preferences from behavior. We show how this approach can be used to infer the value that people assign to food items from their eye movements when choosing between those items. We demonstrate first that neural networks can infer the latent preferences used by the model to generate simulated fixations, and second that simulated data can be beneficial in pretraining a network for predicting human-reported preferences from real fixations. Compared to inferring preferences from choice alone, this approach confers a slight improvement in predicting preferences and also allows prediction to take place prior to the choice being made. Overall, our results suggest that using a combination of neural networks and model-simulated training data is a promising approach for developing technology that infers human preferences. (shrink)

The selected effects or ‘etiological’ theory of Proper function is a naturalistic and realist account of biological teleology. It is used to analyse normativity in philosophy of language, philosophy of mind, philosophy of medicine and elsewhere. The theory has been developed with a simple and intuitive view of natural selection. Traits are selected because of their positive effects on the fitness of the organisms that have them. These ‘selected effects’ are the Proper functions of the traits. Proponents argue that this (...) analysis of biological teleology has the unique advantage that the selected effect function of a trait is also a causal explanation of the trait: the trait exists because it performs this function. We show, however, that selected effect functions as currently defined explain the existence of traits only under highly restrictive assumptions about evolutionary dynamics. In many common scenarios in which traits evolve by natural selection, selected effect functions do not explain those traits. This is because definitions of selected effect function extract from any evolutionary scenario only the information that would be explanatorily relevant in the simple evolutionary scenario implicit in those definitions. When applied to more complex scenarios selected effect functions omit the key information that is explanatorily relevant in those scenarios. The assumptions required for selected effect functions to be explanatory are particularly unlikely to hold in the domain that its proponents care most about - the evolution of representation. A more adequate selected effects theory of Proper functions may be possible, but will require much greater attention to the structure of actual evolutionary explanations. (shrink)

The Hare Psychopathy Checklist (PCL; Hare, Neumann, & Mokros 2018) scales are among the most widely used forensic assessment tools. Their perceived utility rests partly on their ability to assess stable personality traits indicative of a lack of conscience, which then facilitates behavioral predictions useful in forensic decisions. In this systematic review, we evaluate the empirical evidence behind 3 fundamental justifications for using the PCL scales in forensics, namely, that they are empirically predictive of (1) criminal behavior, (2) treatment outcomes, (...) and (3) a lack of conscience. We found the PCL scales can predict criminal behavior to a statistically significant degree, though with important limitations. We found no evidence of PCL psychopathy being predictive of treatment and rehabilitation outcomes. We found no evidence of PCL psychopathy being predictive of a lack of conscience. These findings disprove widespread beliefs about PCL psychopathy among forensic practitioners and questions the current and future role of the PCL scales in forensic settings. (shrink)

Identifying patterns in the world requires noticing not only unusual occurrences, but also unusual absences. We examined how people learn from absences, manipulating the extent to which an absence is expected. People can make two types of inferences from the absence of an event: either the event is possible but has not yet occurred, or the event never occurs. A rational analysis using Bayesian inference predicts that inferences from absent data should depend on how much the absence is expected to (...) occur, with less probable absences being more salient. We tested this prediction in two experiments in which we elicited people's judgments about patterns in the data as a function of absence salience. We found that people were able to decide that absences either were mere coincidences or were indicative of a significant pattern in the data in a manner that was consistent with predictions of a simple Bayesian model. (shrink)