The ability to appropriately sequence a list of discrete items is an important facet in performing routine cognitive tasks and may play a significant role in the acquisition of early communication skills. Though the serial learning abilities of some species, such as chimpanzees and rhesus macaques are well documented, there is virtually no information on the extent of these skills with gorillas. In this study, a young female western lowland gorilla has demonstrated the ability to learn a list of (...) seven Arabic numerals by selecting them on a monitor behind an infrared touchframe. As list length increased, Rollie required fewer trials to reach the criterion, suggesting an acquisition of skills associated with serial learning. Rollie was most likely to correctly select the first item in the sequence and least likely to select the penultimate item in the sequence. Her performance suggests serial learning abilities in the range of other non-verbal species and provides preliminary evidence that supports the assertion that gorillas can succeed at this form of serial learning task. (shrink)

In popular discussions, the nuances of AI are often abridged as “the algorithm”, as the specific arrangements of machine learning, deep learning and automated decision-making on social media platforms are typically shrouded in proprietary secrecy punctuated by press releases and transparency initiatives. What is clear, however, is that AI embedded on social media functions to recommend content, personalize ads, aggregate news stories, and moderate problematic material. It is also increasingly apparent that individuals are concerned with the uses, implications, (...) and fairness of algorithmic systems. Perhaps in response to concerns about “the algorithm” by individuals and governments, social media platforms utilize transparency initiatives and official statements, in part, to deflect official regulation. In the following paper, I draw from transparency initiatives and statements from representatives of Facebook and TikTok as case studies of how AI is embedded in these platforms, with attention to the promotion of AI content moderation as a solution to the circulation of problematic material and misinformation. This examination considers the complexity of embedded AI as a material-discursive apparatus, predicated on discursive techniques—what is seeable, sayable, knowable in a given time period—as well as the material arrangements—algorithms, datasets, users, platforms, infrastructures, moderators, etc. As such, the use of AI as part of the immensely popular platforms Facebook and TikTok demonstrates that AI does not exist in isolation, instead functioning as human–machine ensemble reliant on strategies of acceptance via discursive techniques and the changing material arrangements of everyday embeddedness. (shrink)

Within the planning discourse two poles have materialised over the last decades: a participatory ideal guided by substantive rationality, opposed to an algorithmic governmentality subordinated to instrumental reason. This rift within socialist thought is also observable when it comes to the discovery of needs. The paper understands this discovery procedure primarily as a forecasting problem and demonstrates how many authors dedicated to a participatory planning process call for consumers to write down their desires in the form of wish lists. As (...) a response to this epistemically questionable discovery procedure, the state of the art in capitalist demand-forecasting at enterprises like Amazon is presented, where machine-learning algorithms excel at modelling interrelated time series on a global level by extrapolating demand patterns in real-time. The paper closes with a proposal to reconfigure this predictive apparatus for socialist ends and raises questions concerned with the political implications of centralising decision-making in black-box algorithms. (shrink)

The position advanced in this paper is that the bedrock of emotional feelings is contained within the evolved emotional action apparatus of mammalian brains. This dual-aspect monism approach to brain–mind functions, which asserts that emotional feelings may reflect the neurodynamics of brain systems that generate instinctual emotional behaviors, saves us from various conceptual conundrums. In coarse form, primary process affective consciousness seems to be fundamentally an unconditional “gift of nature” rather than an acquired skill, even though those systems facilitate (...) skill acquisition via various felt reinforcements. Affective consciousness, being a comparatively intrinsic function of the brain, shared homologously by all mammalian species, should be the easiest variant of consciousness to study in animals. This is not to deny that some secondary processes cannot be evaluated in animals with sufficiently clever behavioral learning procedures, as with place-preference procedures and the analysis of changes in learned behaviors after one has induced re-valuation of incentives. Rather, the claim is that a direct neuroscientific study of primary process emotional/affective states is best achieved through the study of the intrinsic , albeit experientially refined, emotional action tendencies of other animals. In this view, core emotional feelings may reflect the neurodynamic attractor landscapes of a variety of extended trans-diencephalic, limbic emotional action systems—including SEEKING, FEAR, RAGE, LUST, CARE, PANIC, and PLAY. Through a study of these brain systems, the neural infrastructure of human and animal affective consciousness may be revealed. Emotional feelings are instantiated in large-scale neurodynamics that can be most effectively monitored via the ethological analysis of emotional action tendencies and the accompanying brain neurochemical/electrical changes. The intrinsic coherence of such emotional responses is demonstrated by the fact that they can be provoked by electrical and chemical stimulation of specific brain zones—effects that are affectively laden. For substantive progress in this emerging research arena, animal brain researchers need to discuss affective brain functions more openly. Secondary awareness processes, because of their more conditional, contextually situated nature, are more difficult to understand in any neuroscientific detail. In other words, the information-processing brain functions, critical for cognitive consciousness, are harder to study in other animals than the more homologous emotional/motivational affective state functions of the brain. (shrink)

This work reflects on hypochondria as well as on the global functioning of the human mind and on the place of the patient/physician relationship in the wider organisation of society. First published in 1711, revised and enlarged in 1730, and now edited and published with a critical apparatus for the first time, this is a major work in the history of medical literature as well as a complex literary creation. Composed of three dialogues between a physician and two of (...) his patients, Mandeville’s Treatise mirrors the digressive structure of a talking cure. Thanks to the soothing and enlightening effects of this casual conversation, the physician Mandeville demonstrates the healing power of words for a class of patients that he presents as men of learning who need above all to be addressed in their own language. Mandeville’s aim was to delineate his own cure for hypochondria and hysteria, which consisted of a talking cure followed by diet and exercise, but also to discuss the practice of medicine in England and continental Europe at a time when physicians were beginning to lose ground to apothecaries. Opposing a purely theoretical approach to medicine, Mandeville takes up the principles presented by Francis Bacon, Thomas Sydenham, and Giorgio Baglivi, and advocates a medical practice based on experience and backed up by time-tested theories. (shrink)

Gesture-first theories of language origins often raise two unsubstantiated arguments against vocal origins. First, they argue that great ape vocal behavior is highly constrained, limited to a fixed, species-typical repertoire of reflexive calls. Second, they argue that vocalizations lack any significant potential to ground meaning through iconicity, or resemblance between form and meaning. This paper reviews the considerable evidence that debunks these two “myths”. Accumulating evidence shows that the great apes exercise voluntary control over their vocal behavior, including their breathing (...) apparatus, larynx, and supralaryngeal articulators. They are also able to learn new vocal behaviors, and even show some rudimentary ability for vocal imitation. In addition, an abundance of research demonstrates that the vocal modality affords rich potential for iconicity. People can understand iconicity in sound symbolism, and they can produce iconic vocalizations to communicate a diverse range of meanings. Thus, two of the primary arguments against vocal origins theories are not tenable. As an alternative, the paper concludes that the origins of language – going as far back as our last common ancestor with great apes – are rooted in iconicity in both gesture and vocalization. (shrink)

Two variables are usually recognised as determinants of human causal learning: the contingency between a candidate cause and effect, and the temporal and/or spatial contiguity between them. A common finding is that reductions in temporal contiguity produce concomitant decrements in causal judgement. This finding had previously (Shanks & Dickinson, 1987) been interpreted as evidence that causal induction is based on associative learning processes. Buehner and May (2002, 2003, 2004) have challenged this notion by demonstrating that the impact of (...) temporal delay depends on expectations about the timeframe between cause and effect. A corollary of this knowledge-mediation account is that in certain situations longer delays could facilitate, while shorter delays should impair, causal learning. Here we present two experiments involving a physical apparatus that demonstrate a detrimental effect of contiguity under certain conditions. In contrast to all previous studies, delays universally promoted causal learning. This evidence is clearly at variance with the notion of a bottom-up contiguity bias in causal induction. A new, more general timeframe bias is discussed. (shrink)

This volume makes available for the first time critical editions of John Locke's A Vindication and A Second Vindication of the Reasonableness of Christianity, in which Locke defends his interpretation of the New Testament and of the Christian Religion against charges of heterodoxy. These works contribute greatly to our understanding of Locke's Christian commitments, which it is now recognized played an important role in shaping his philosophical opinions; they also demonstrate his sophistication as a biblical scholar, and the breadth of (...) his theological learning. The texts are accompanied by a historical introduction explaining the origin of the works and setting them in context. In addition to a textual introduction and critical apparatus, editorial notes help to clarify the text. The volume also includes a French translation and abridgment by Pierre Coste, a Huguenot scholar, who was patronized by Locke and worked on his translations while residing in Locke's household. This definitive edition is an important contribution to an understanding of the development of modern enlightened Christian attitudes. (shrink)

Sociability is essential for modern robots to increase their acceptability in human environments. Traditional techniques use manually engineered utility functions inspired by observing pedestrian behaviors to achieve social navigation. However, social aspects of navigation are diverse, changing across different types of environments, societies, and population densities, making it unrealistic to use hand-crafted techniques in each domain. This paper presents a data-driven navigation architecture that uses state-of-the-art neural architectures, namely Conditional Neural Processes, to learn global and local controllers of the mobile (...) robot from observations. Additionally, we leverage a state-of-the-art, deep prediction mechanism to detect situations not similar to the trained ones, where reactive controllers step in to ensure safe navigation. Our results demonstrate that the proposed framework can successfully carry out navigation tasks regarding social norms in the data. Further, we showed that our system produces fewer personal-zone violations, causing less discomfort. (shrink)

This book present interdisciplinary research in the social sciences and humanities by connecting seemingly disparate sources through a sensitivity to endangered human values. It links reflections on the contemporary relationship between art and technology in a post-modern context, seeing art in terms of crossing boundaries and exploring virtuality. It deals with the consequences of economics colonising other disciplines, in terms of the processes by which the social becomes the economic. Using Jantsch''s evolutionary paradigm, the concept of self-transcendence is seen as (...) crucial for the understanding of human beings and their social systems. Incorporating recent thinking from the natural sciences, the learning process can be conceived as the life and activity of all complex systems, including those once conceived as organisms, machines, cultures or economies. Without the societal embrace of scientific and technological development no collective or individual meaning can be assigned to the production of new knowledge. The book seeks to recognise the point where a collective learning process becomes the heart of productivity, and where the shift from the hegemony of material labour to immaterial labour becomes fundamental. The author brings new understandings of art, the social, and technology together, based on the idea that history is not a story told in separate physical, social and spiritual spaces and that the most fundamental problem of today is how to find some shared meaning in a fractured world. The author analyses, at a global level, the process of the co-production of scientific and social order, of culture and technology, of life sciences and economic and political regimes. It rightly identifies the rise of the role of knowledge and the move of capital into life sciences as a new stage in the history of capitalism: what we can qualify as cognitive capitalism or biocapitalism. In this new era of capitalism, what is being manufactured and sold are not just tangible and non-tangible goods, whose increasing importance, as the author shows, poses unsurmountable theoretical problems to the theoretical apparatus of economic science. The increasing mercantilisation of the world appears at the same time as a bio-power, i.e. a set of instruments creating and controlling different forms of life, forms of communication, standards of socialisation, education, the individual and collective imaginary, etc. More fundamentally, the encounter of life sciences and the information technology integrates and subjects the most essential mechanisms of biological and social reproduction to the logic of capital valorization. To understand the complexity of these changes and the ethical and philosophical questions that the development of technology and sciences poses to the future of mankind one must break through the disciplinary barriers delineating different disciplines in social sciences and those separating social sciences from natural sciences. Professor Matko Mestrovic manages to tackle this challenge not only because of his impressive and masterful knowledge of different disciplines in the social and natural sciences, but he does it also owing to his capacity for theoretical elaboration that allows him to lay the foundations of a new transdisciplinary paradigm. This is why his work can raise the awareness of the general public on two issues: on a global and profound vision of the challenges posed by the new millennium; and on the need for a radical theoretical innovation bringing into question the disciplinary certitudes in develop a social science able to better understand the movement and the ambiguity of history. - Carlo Vercellone, Universite de Paris 1, Pantheon-Sorbonne Mestrovic provides a unique insight into the often forgotten relation between economics on the one hand and arts and culture on the other, demonstrating that these adomains function as a total social facticity and not as separate, entirely independent elements thereof. In doing so he is dispelling the illusions about the disciplinary self-containdness of individual forms of knowledge and is relying on those paradigms of contemporary scientific thought whose aepistemiological programs are based on close cooperation and aopening up and not on the persuasion about one''s own positions and dispositions. - Prof. dr. Rade Kalanj, Redovni profesor na Odjelu za sociologiju. (shrink)

Imitation in robotics is seen as a powerful means to reduce the complexity of robot programming. It allows users to instruct robots by simply showing them how to execute a given task. Through imitation robots can learn from their environment and adapt to it just as human newborns do. Despite different facets of imitative behaviours observed in humans and higher primates, imitation in robotics has usually been implemented as a process of copying demonstrated actions onto the movement apparatus of (...) the robot. While the results being reached are impressive, we believe that a shift towards a higher expression of imitation, namely the comprehension of human actions and inference of its intentions, is needed. In order to be useful as human companions, robots must act for a purpose by achieving goals and fulfilling human expectations. In this paper we present ConSCIS, an architecture for goal-level imitation in robotics where the focus is put on final effects of actions on objects. The architecture tightly links low-level data with high-level knowledge, and integrates, in a unified framework, several aspects of imitation, such as perception, learning, knowledge representation, action generation and robot control. Some preliminary experimental results with an anthropomorphic arm/hand robotic system are shown. (shrink)

To what degree can language be acquired from linguistic input alone? This question has vexed scholars for millennia and is still a major focus of debate in the cognitive science of language. The complexity of human language has hampered progress because studies of language–especially those involving computational modeling–have only been able to deal with small fragments of our linguistic skills. We suggest that the most recent generation of Large Language Models (LLMs) might finally provide the computational tools to determine empirically (...) how much of the human language ability can be acquired from linguistic experience. LLMs are sophisticated deep learning architectures trained on vast amounts of natural language data, enabling them to perform an impressive range of linguistic tasks. We argue that, despite their clear semantic and pragmatic limitations, LLMs have already demonstrated that human-like grammatical language can be acquired without the need for a built-in grammar. Thus, while there is still much to learn about how humans acquire and use language, LLMs provide full-fledged computational models for cognitive scientists to empirically evaluate just how far statistical learning might take us in explaining the full complexity of human language. (shrink)

Imitation in robotics is seen as a powerful means to reduce the complexity of robot programming. It allows users to instruct robots by simply showing them how to execute a given task. Through imitation robots can learn from their environment and adapt to it just as human newborns do. Despite different facets of imitative behaviours observed in humans and higher primates, imitation in robotics has usually been implemented as a process of copying demonstrated actions onto the movement apparatus of (...) the robot. While the results being reached are impressive, we believe that a shift towards a higher expression of imitation, namely the comprehension of human actions and inference of its intentions, is needed. In order to be useful as human companions, robots must act for a purpose by achieving goals and fulfilling human expectations. In this paper we present ConSCIS, an architecture for goal-level imitation in robotics where the focus is put on final effects of actions on objects. The architecture tightly links low-level data with high-level knowledge, and integrates, in a unified framework, several aspects of imitation, such as perception, learning, knowledge representation, action generation and robot control. Some preliminary experimental results with an anthropomorphic arm/hand robotic system are shown. (shrink)

In three studies, we explored the retention and transfer of tool-making knowledge, learnt from an adult demonstration, to other temporal and task contexts. All studies used a variation of a task in which children had to make a hook tool to retrieve a bucket from a tall transparent tube. Children who failed to innovate the hook tool independently saw a demonstration. In Study 1, we tested children aged 4 to 6 years (N = 53) who had seen the (...) original demonstration 3 months earlier. Performance was excellent at the second time, indicating that children’s knowledge was retained over the 3 month period. In Studies 2 and 3 we explored transfer of the new knowledge to other tasks. In Study 2, children were given two variants of the apparatus that differed in surface characteristics (e.g. shape and colour). Participants generalised their knowledge to these new apparatuses even though the new pipecleaner also differed in size and colour. Five- to 6-year-olds (n = 22) almost always transferred their knowledge to problems where the same tool had to be made. Younger, 3- to 5-year-olds’ (n = 46), performance was more variable. In Study 3, 4- to 7-year-olds (N = 146) saw a demonstration of hook making with a pipecleaner, but then had to make a tool by combining pieces of wooden dowel (or vice versa: original training on dowel, transfer to pipecleaner). Children did not transfer their tool-making knowledge to the new material. Children retained tool-making knowledge over time and transferred their knowledge to new situations in which they needed to make a similar tool from similar materials, but not different materials. We concluded that children’s ability to use tool-making knowledge in novel situations is likely to depend on memory and analogical reasoning, with the latter continuing to develop during middle childhood. (shrink)

A critical challenge in robot learning from demonstration is the ability to map the behavior of the trainer onto a robot’s existing repertoire of basic/primitive capabilities. In part, this problem is due to the fact that the observed behavior of the teacher may consist of a combination of the robot’s individual primitives. The problem becomes more complex when the task involves temporal sequences of goals. We introduce an autonomous control architecture that allows for learning of hierarchical task (...) representations, in which: every goal is achieved through a linear superposition of robot primitives and sequencing across goals is achieved through arbitration. We treat learning of the appropriate superposition as a state estimation problem over the space of possible linear fusion weights, inferred through a particle filter. We validate our approach in both simulated and real world environments with a Pioneer 3DX mobile robot. (shrink)

Husserl’s theory of passive experience first came to systematic and detailed expression in the lectures on passive synthesis from the early 1920s, where he discusses pure passivity under the rubric of affection and association. In this paper I suggest that this familiar theory of passive experience is a first approximation leaving important questions unanswered. Focusing primarily on affection, I will show that Husserl did not simply leave his theory untouched. In later manuscripts he significantly reworks the theory of affection in (...) terms of instinctive intentionality and a passive experience of desire aimed at satisfaction and enjoyment. This paper will show that the theory of affection and the theory of instincts in Husserl are really one and the same, differing only in the superior theoretical apparatus with which Husserl treats the phenomenon in his more considered theory of the instincts. I demonstrate the connection between the two theories by showing how what he generically calls “affection” in earlier texts is the same phenomenon he calls “curiosity” in later texts. The connection is further supported by the way curiosity does the same work as affection in its function within Husserl’s theory of association, serving as the basic connective tissue linking diverse experiences. In closing, I deal with the problem of how to integrate the experience of the body into the theory of instincts, displaying in another way how Husserl improves his theory of affection by making it more concrete when he recasts it as a theory of instincts. (shrink)

Exploiting the skills of others enables individuals to reduce the risks and costs of resource innovation. Social corvids are known to possess sophisticated social and physical cognitive abilities. However, their capacity for imitative learning and its inter-individual transmission pattern remains mostly unexamined. Here we demonstrate the large-billed crows' ability to learn problem-solving techniques by observation and the dominance-dependent pattern in which this technique is transmitted. Crows were allowed to observe one of two box-opening behaviours performed by a dominant or (...) subordinate demonstrator and then tested regarding action and technique. The observers successfully opened the box on their first attempts by using non-matching actions but matching techniques to those observed, suggesting emulation. In the subsequent test sessions, dominant observers (i.e. those dominant to the bird acting as demonstrator) consistently used the learned technique, whereas subordinates (i.e. those subordinate to the bird acting as demonstrator) learned alternative techniques by explorative trial and error. Our findings demonstrate crows' capacity to learn by observing behaviours and the effect of dominance on transmission patterns of behavioural skills. Keywords: social learning; imitation; emulation; affordance; culture; innovation. (shrink)

Previous research suggests that implicit sequence learning is superior for believers in the paranormal and individuals with increased cerebral dopamine. Thirty-five healthy participants performed feedback-guided anticipations of four arrow directions. A 100-trial random sequence preceded two 100-trial biased sequences in which visual targets on trial t tended to be displaced 90° clockwise or counter-clockwise from those on t − 1. ISL was defined as a positive change during the course of the biased run in the difference between pro-bias and (...) counter-bias responses. It was hypothesized that this difference would be greater for believers in the paranormal than for skeptics, for those who received dopamine than for those who received placebo, and for believers who received dopamine than for the other groups. None of the hypotheses were supported by the data. It is suggested that a simple binary guessing task with a focus on prediction accuracy during early trials should be considered for future explorations. (shrink)

Philosophers have recently focused on critical, epistemological challenges that arise from the opacity of deep neural networks. One might conclude from this literature that doing good science with opaque models is exceptionally challenging, if not impossible. Yet, this is hard to square with the recent boom in optimism for AI in science alongside a flood of recent scientific breakthroughs driven by AI methods. In this paper, I argue that the disconnect between philosophical pessimism and scientific optimism is driven by a (...) failure to examine how AI is actually used in science. I show that, in order to understand the epistemic justification for AI-powered breakthroughs, philosophers must examine the role played by deep learning as part of a wider process of discovery. The philosophical distinction between the 'context of discovery' and the 'context of justification' is helpful in this regard. I demonstrate the importance of attending to this distinction with two cases drawn from the scientific literature, and show that epistemic opacity need not diminish AI's capacity to lead scientists to significant and justifiable breakthroughs. (shrink)

(I) MATHEMATICAL LECTURES. LECTURE I. Of the Name and general Division of the Mathematical Sciences. BEING about to treat upon the Mathematical Sciences, ...

Jerry Fodor’s arguments against the possibility of concept learning, and the responses that have been offered in defense of the coherence of concept learning, have both by and large assumed that concept learning is a descriptive process. I offer an alternative, ostensive approach to concept learning and explain how descriptive concept learning can be explained as a version of ostensive concept learning. I argue that an ostensive view of concept learning offers an empirically (...) plausible and philosophically adequate account of concept learning that is explanatorily superior to traditional descriptive views. (shrink)

In recent years, the family of algorithms collected under the term ``deep learning'' has revolutionized artificial intelligence, enabling machines to reach human-like performances in many complex cognitive tasks. Although deep learning models are grounded in the connectionist paradigm, their recent advances were basically developed with engineering goals in mind. Despite of their applied focus, deep learning models eventually seem fruitful for cognitive purposes. This can be thought as a kind of biological exaptation, where a physiological structure becomes (...) applicable for a function different from that for which it was selected. In this paper, it will be argued that it is time for cognitive science to seriously come to terms with deep learning, and we try to spell out the reasons why this is the case. First, the path of the evolution of deep learning from the connectionist project is traced, demonstrating the remarkable continuity, and the differences as well. Then, it will be considered how deep learning models can be useful for many cognitive topics, especially those where it has achieved performance comparable to humans, from perception to language. It will be maintained that deep learning poses questions that cognitive sciences should try to answer. One of such questions is the reasons why deep convolutional models that are disembodied, inactive, unaware of context, and static, are by far the closest to the patterns of activation in the brain visual system. (shrink)

Just as humans can draw conclusions responsibly or irresponsibly, so too can computers. Machine learning systems that have been trained on data sets that include irresponsible judgments are likely to yield irresponsible predictions as outputs. In this paper I focus on a particular kind of inference a computer system might make: identification of the intentions with which a person acted on the basis of photographic evidence. Such inferences are liable to be morally objectionable, because of a way in which (...) they are presumptuous. After elaborating this moral concern, I explore the possibility that carefully procuring the training data for image recognition systems could ensure that the systems avoid the problem. The lesson of this paper extends beyond just the particular case of image recognition systems and the challenge of responsibly identifying a person’s intentions. Reflection on this particular case demonstrates the importance (as well as the difficulty) of evaluating machine learning systems and their training data from the standpoint of moral considerations that are not encompassed by ordinary assessments of predictive accuracy. (shrink)

This paper develops an account of how the implementation of ML models into healthcare settings requires revising the methodological apparatus of philosophical bioethics. On this account, ML models are cognitive interventions that provide decision-support to physicians and patients. Due to reliability issues, opaque reasoning processes, and information asymmetries, ML models pose inferential problems for them. These inferential problems lay the grounds for many ethical problems that currently claim centre-stage in the bioethical debate. Accordingly, this paper argues that the best (...) way to make progress in remedying these ethical problems is to distil their epistemic core and to identify appropriate epistemic norms as guardrails. The viability of this approach will be highlighted based on four key issues: trust, responsibility, paternalism, and fairness. In that respect, the paper also contributes to a more pronounced view of the specific methodological challenges for ethical theorizing, when the point of reference are complex statistical models that make predictions, rather than individual agents, acting for reasons. (shrink)

This Special Issue addresses a range of educational issues linked to main themes from our 2019 book The Emergence of Complexity: Rethinking Education as a Social Science. This book elaborated two major theses that raise fundamental questions for philosophy of education. First, that learning by groups is typically a distinctive kind of learning that is not reducible to learning by individuals. Second, that a degree of holism, as against a focus on individuals, is essential for achieving a (...) convincing understanding of this distinctive type of group learning. These two theses are of direct interest to philosophy of education since they challenges the resources of received theories of learning. Yet this kind of group learning characterises the vast bulk of human learning situations that occur outside of formal education systems. In this Special Issue introductory article we, firstly, introduce some basic principles of complexity theory, and demonstrate that, together with the concept of a ‘co-present group’ (between 2 and about 12 individuals), these ideas offer develop novel understandings of the distinctive learning that occurs within such groups. Secondly, we outline and illustrate with further examples the main features of co-present groups. Thirdly, we consolidate our conceptual work by showing five ways in which our account of group learning serves to refurbish the concept of ‘learning’. (shrink)

From a theoretical perspective, most discussions of statistical learning have focused on the possible “statistical” properties that are the object of learning. Much less attention has been given to defining what “learning” is in the context of “statistical learning.” One major difficulty is that SL research has been monitoring participants’ performance in laboratory settings with a strikingly narrow set of tasks, where learning is typically assessed offline, through a set of two-alternative-forced-choice questions, which follow a (...) brief visual or auditory familiarization stream. Is that all there is to characterizing SL abilities? Here we adopt a novel perspective for investigating the processing of regularities in the visual modality. By tracking online performance in a self-paced SL paradigm, we focus on the trajectory of learning. In a set of three experiments we show that this paradigm provides a reliable and valid signature of SL performance, and it offers important insights for understanding how statistical regularities are perceived and assimilated in the visual modality. This demonstrates the promise of integrating different operational measures to our theory of SL. (shrink)

This paper examines the learning and performance of the air traffic control (ATC) work domain. This domain was chosen because it embodies features that represent future work for many other industries (e.g., information service provision mediated by information technologies; a high reliance on communication skills and collaborative work; increasing complexity and intensity of the work activity), within an organisational context undergoing considerable change. In ATC work learning occurs formally as part of accredited training and informally, as part of (...) everyday practice. In this way learning and experience of ATC work shapes – and is shaped by – the way work is organised, which includes the divisions of labour and technological artefacts. ATC work is experienced temporally, complexly, through affect and socially. Controllers learn to use a variety of socio-material resources which include embodiment and artefacts to undertake the work. In so doing the article demonstrates the ways in which learning and performance is not just a cognitive undertaking, but one that draws on embodiment to achieve what here is called “interdependent sentience”. This involves using all their senses to gain an awareness of problems as they emerge to collectively accomplish a seamless air traffic flow in a complex interdependent socio-technical system. (shrink)

Self-relevance has been demonstrated to impair instrumental learning. Compared to unfamiliar symbols associated with a friend, analogous stimuli linked with the self are learned more slowly. What is not yet understood, however, is whether this effect extends beyond arbitrary stimuli to material with intrinsically meaningful properties. Take, for example, stimulus valence an established moderator of self-bias. Does the desirability of to-be-learned material influence self-learning? Here, in conjunction with computational modelling (i.e. Reinforcement Learning Drift Diffusion Model analysis), a (...) probabilistic selection task was used to establish if and how stimulus valence (i.e. desirable/undesirable posters) impacts the acquisition of knowledge relating to object-ownership (i.e. owned-by-self vs. owned-by-friend). Several interesting results were observed. First, undesirable posters were learned more rapidly for self compared to friend, an effect that was reversed for desirable posters. Second, learning rates were accompanied by associated differences in reward sensitivity toward desirable and undesirable choice selections as a function of ownership. Third, decisional caution was greater for self-relevant (vs. friend relevant) responses. Collectively, these findings inform understanding of self-function and how valence and stimulus relevance mutually influence probabilistic learning. (shrink)

As scientists start to adopt machine learning (ML) as one research tool, the security of ML and the knowledge generated become a concern. In this paper, I explain how supervised ML can be improved with better data ontology, or the way we make categories and turn information into data. More specifically, we should design data ontology in such a way that is consistent with the knowledge that we have about the target phenomenon so that such ontology can help us (...) make the inductive leap. I do so by thinking through a thought experiment, Goodman’s New Riddle of Induction (Fact, fiction, and forecast, Harvard University Press, 1955). Goodman’s riddle helps flesh out three problems of induction: (1) the problem of equal goodies, that there are often too many equally good inductive results given the same data; (2) the problem of diverging performance, that these equally good results can give opposite predictions in the future; and (3) the problem of mediocrity, that when averaged across all equally possible datasets and tasks, no inductive algorithm outperforms any other. I show that all these three problems are manifested as real obstacles in ML practice, namely, the Rashomon effect (Breiman in Stat Sci 16(3):199–231, 2001), the problem of underspecification (D’Amour et al. in J Mach Learn Res, 2020, https://doi.org/10.48550/arXiv.2011.03395 ), and the No Free Lunch theorem (Wolpert in Neural Comput 8(7):1341–90, 1996, https://doi.org/10.1162/neco.1996.8.7.1341 ). Lastly, I argue that proper data ontology can help mitigate these problems and I demonstrate how using concrete examples from climate science. This research highlights the links between philosophers’ discussions of induction and implications in ML practice. (shrink)