Search

During class, taking notes helps the encoding process, especially when students reformulate what is presented. After class, notes can be used for revision. Previous studies, mostly conducted in laboratory contexts, showed that note-taking on paper seems to be more efficient than on computer, probably because students are more likely to reformulate on paper. The originality of this study is to investigate students’ perceptions and strategies depending on their note-taking method during but also after class. Data of 108 students showed that students on paper (30%) declare making more reformulations and less multitasking that students on computer (70%). After class, review sheets are written for a large majority of the students on paper, whatever the method they used during class. These results suggest that even if computer is the most used method during class, paper is still used after class, and with paper are associated better strategies.

How did monolingual bias emerge and why does it persist? What are the epistemologies concerning human language? How do language and languages develop? What do the terms monolingual, bilingual, and multilingual mean? What does “typical” language development entail? What constitutes “atypical” language development? To answer these questions, this paper foregrounds that language is a human cognitive skill for communication, and that Anglo-American and European notions of language diversity, monolingualism, bilingualism, and multilingualism disseminate monolingual bias. Monolingual bias extends to educational and clinical contexts where children from bi/multilingual households and/or communities face inappropriate scrutiny when language practitioners assess and support them based on normative monolingual expectations or standards. Monolingual bias is then linked to conceptual issues on “typical” and “atypical” language development. Ultimately, this paper attempts to explain why Anglo-American and European notions of bi/multilingualism are less relevant for most, if not all, of the world’s multilingual contexts, especially in the Philippines.

Since the landmark paper of Spivey et al. (2005), there has been an explosion in the use of continuous trajectories for studying decision-making. In these studies, the participant makes their decision by reaching to a target. By extracting trajectory-based measures, it is possible to observe how the decision unfolds over time. However, studies have shown that these trajectories may not continuously track the decision-making process, but rather result from several discrete submovements executed intermittently. Due to this, the conclusions of papers using purely kinematic measures may be affected by these motor control properties. This study presents simulations of trajectories using human motor control models and demonstrates the relationship between non-decision-related parameters and outcome measures typically used. We also comment on the implications of these findings for using kinematic measures in decision-making and make recommendations for how the decision variable can be more appropriately sampled.

This paper introduces Waymaking, a philosophy that defines cognition by the trajectories of an agent’s movement through its ongoing encounter, whereby those trajectories are themselves dynamic patterns of nested (i.e., whole-body, neural) spatiotemporal movement. Here, an agent’s encounter incudes all realms traditionally named mental, physical, and virtual, distinguishing these by their affordances and sensory landscapes relative to the agent-base. Recent research on the hippocampal formation and entorhinal cortex has opened a way for cognition to be understood as trajectories within nested landscapes: We can now posit knowledge-acquisition, remembering, and spatiotemporal navigation as a common process through statistically diverse clustered regularities. This paper formulates these findings into a general framework that can be used heuristically to provide a practical definition of cognition applicable across disciplines and species, thus alleviating stubborn dichotomies such as those at the heart of the so-called mind-body problem and arguments around potential plant or animal cognition.

The aim of this paper is to import basic concepts from information theory into the epistemological debate about probabilistic measures of coherence. Rather than putting forward and defending a new measure, this paper will sketch an account of `external' coherence, which will be defined as a relation between a target variable of interest, sources containing (more or less) information about the target, and a rule or theoretical hypothesis that postulates the relevant connections between source and target. Relations with and potential insights for standard notions of coherence in formal epistemology are explored. More generaly, the paper explores the potential benefits of applying information theory to the epistemological debate about coherence.

People tend to avoid wasting resources, but little is known about when this emerges in development. Though young children are often wasteful with food and other items, previous work suggests that children consider waste in other judgments. Here, we examined if children anticipate that others should minimize waste. In two experiments (total N = 195), children chose which of two foods someone should eat (Experiment 1; 3-7-year-olds) or two papers someone should make a snowflake with (Experiment 2; 5-year-olds). One of the options would result in minimal waste (i.e., a small food/paper) while the other would result in greater waste (i.e., a large food/paper). Children did not anticipate that others would choose smaller foods, however, at around five years they predicted that others would choose smaller paper. These findings contribute to our knowledge of the development of waste aversion and may extend our understanding of waste aversion as a form of efficiency.

The lexica of natural languages are ambiguous, but the de- gree of ambiguity is unequal between words. Some words have more meanings than others. However, the exact prop- erties that favor some words over others when acquiring a new meaning are not very well understood. In recent years, several studies suggested that some words gain more meanings than others based on selection for efficient communication, which could explain the correlation between meaning and frequency discovered by Zipf (Piantadosi, Tily, & Gibson, 2012; Gib- son et al., 2019). The object of this study is to assess the role of selection in the meaning-frequency correlation using a neu- tral model that yields a meaning-frequency correlation without selection pressures. We provide a model where words gain additional meanings through reuse. In the neutral model pre- sented in this paper, words are chosen to be reused at random, independently of their frequency, hence there is no selection mechanism favoring efficient communication. Unlike previous attempts to introduce null models of the meaning-frequency correlation (Caplan, Kodner, & Yang, 2020; Trott & Bergen, 2020), it truly does not rely on selection for frequency. We show that statistical regularities related to ambiguity, such as Zipf’s meaning-frequency correlation, can arise in conditions when words are not undergoing any selective pressures. This model has the additional property of matching word frequency distributions of natural languages. It can provide the baseline against which the presence of selection for efficient communi- cation in natural languages can be assessed.

Embodied simulation hypothesis supposes that language processing involves the activation of perceptual-motor systems to recreate the described scene (Bergen, 2012, 2019). The paper investigates whether and how adolescent second language (L2) learners’ online processing of prepositions engages mental simulation. Specifically, the study examines whether any observed mental simulation effect was modulated by prepositions, abstractness of senses, and Stimulus Onset Asynchrony (SOA). 40 Chinese adolescents completed a diagram-picture matching task followed by a semantic priming task in English, where participants saw a diagrammatic prime and made phrasal acceptability judgement. Results showed a compatibility effect of schematic diagrams on adolescent L2 English learners’ accuracy rates (ARs) of processing prepositional phrases (PPs), while response times (RTs) results did not reveal mental simulation effects. The findings suggest that schematic diagrams could serve as effective perceptual cues to prime adolescent L2 learners’ processing of schema-compatible English PPs by improving judgement accuracy but not processing speed.

In this paper, we study human sequential behavior by integrating cognitive, evolutionary, and computational approaches. Our work centers around the emergence of shared vocabularies in the Embodied Communication Game (ECG). Here, participant pairs solve a shared task without access to conventional means of communication, enforcing the emergence of a new communication system. This problem is solved typically by negotiating a shared set of sequential signals that acquire meaning through interactions. Individual differences in Personal Need for Structure (PNS) have been found to influence how this process develops. We trained deep neural networks to mimic the emergence of new communicative systems and used hyperparameter optimization to approximate latent human cognitive variables to explain human behavior. We demonstrate that models based on bidirectional LSTM networks are better at capturing human behavior than unidirectional LSTM networks. This suggests that human sequence processing in the ECG is influenced by expected future states. The approximated variables cannot explain the differences in PNS, but we do provide evidence suggesting that random and uncertainty-directed exploration strategies are combined to develop optimal behavior.

Causal perception is a cornerstone of early cognitive development. A large database of research attests to the fact that infant causal perception emerges between 6 and 10 months of age. However, it remains unknown how infants learn about the causal properties of more realistic categories such as people and objects. For example, how do infants learn that people can cause other people to act and behave at a distance, whereas inanimate objects require contact to move and act? One answer to this question is that this knowledge is present from birth or shortly thereafter and is underpinned by core knowledge systems. An alternative perspective maintains that infants acquire this knowledge via domain-general associative learning. The goal of the present paper is to demonstrate that this alternative perspective—implemented in a connectionist computational model—is sufficient to explain infants’ developing knowledge about people and object causal action.

This paper describes the principles of a research programme for cognitive science that exploits recent developments in machine learning (ML) to generate novel hypotheses about the structure of human cognition. Current debate over the interpretability and explainability algorithms usually focuses on the properties of the algorithms themselves in virtue of which they are either interpretable or explainable. However, we argue that there is value in conceptualizing these categories as inherently psychological constructs. Given certain mathematical features of machine learning algorithms – specifically, that many useful ML algorithms are members of Rashomon sets – it is possible to exploit their utility to reason using a principle of parsimony about the inferential structure of certain human cognitive tasks. Algorithms that do something the human mind can do, and are both interpretable and explicable could be, we shall argue, inferential homologues of certain core cognitive processes. We illustrate this proposal with an example drawn from clustering models used in exploratory data analysis, and then conclude with a discussion of some of the philosophical limitations of our proposal.

Humans' capacity for cumulative culture is remarkable: we can build up vast bodies of knowledge over generations. Communication, particularly via language, is a key component of this process. Previous work has described language as enabling posterior passing, where one Bayesian agent transmits a posterior distribution to the next. In practice, we cannot exactly copy our beliefs into the minds of others--we must communicate over the limited channel language provides. In this paper, we analyze cumulative culture as Bayesian reinforcement learning with communication over a rate-limited channel. We implement an agent that solves a crafting task and communicates to the next agent by approximating the optimal rate-distortion trade-off. Our model produces documented effects, such as the benefits of abstraction and selective social learning. It also suggests a new hypothesis: selective social learning can be harmful in tasks where initial exploration is required.

This paper delves into the possible correlations between cognitive impairment due to Alzheimer’s Disease and differences in the ways spatial relations are decoded by groups of individuals with and without AD. We tested participants’ judgments of spatial scenes coded according to intrinsic, relative or undissociated Frames of Reference, ground rotation and use of a locative construction. We found significant differences in the costs of processing spatial information which were considerably higher in elderly and AD groups. On the other hand, judgment ratings reveal that those groups tend to maintain the same Frame of Reference applied by health young individuals to resolve the spatial ambiguity. Furthermore, a general preference for non-relative Frames of Reference was found among participants, all native speakers of Brazilian Portuguese.

A growing body of literature has revealed ordinary causal judgement to be sensitive to normative factors, such that a norm-violating agent is regarded more causal than their non-norm-violating counterpart. In this paper, we explore two competing explanations for this phenomenon: the Responsibility View and the Bias View. The Bias View, but not the Responsibility View, predicts features peripheral to the agent’s responsibility to impact causal attributions. In a series of three preregistered experiments (N = 1162), we present new evidence that the Norm Effect arises from such peripheral features, namely from nonpertinent or entirely silly norm violations. Furthermore, we show that this effect cannot be explained by recourse to the agent’s foreknowledge or desire of the outcome, nor by its foreseeability: the Norm Effect arises even when participants judge the norm-violating agent’s doing as equally foreseeable. This, we argue, provides evidence in favour of the Bias View.

Previous research under the cue-based retrieval framework has assumed that general and discrete retrieval cues, such as [+subject] and [+singular] are employed in selecting a retrieval target during dependency building. However, to explain the effects of semantic compatibility between the target and the head of a dependency, as demonstrated in Cunnings and Sturt (2018), an infinite amount of lexically specific features and retrieval cues are needed. Smith and Vasishth (2020) offered a principled way for feature selection using word embeddings. However, even with a giant corpus, some dependencies are missing. To solve the coverage problem, we leverage pre-trained neural language models (GPT-2) to select features. The metric used in this paper is highly correlated with those in Smith and Vasishth (2020) and can predict the results reported in Cunnings and Sturt (2018). We argue that it the method offers a broader-coverage yet convenient way for feature selection.

Technology to predict responses is a key element in human-to-human messaging that has increasingly been utilized to enable AI-mediated communication. When response suggestions from AI are incorporated into human messaging, it will have an impact on the flow and content of the dialogue. In this paper, we investigated the effect of AI response suggestion sentences used for chat messaging on dialogue flow from two aspects: dialogue act and initiative. Usage rates of response suggestions for different dialogue acts were measured with BLEU scores, and we found that the response suggestions contributed to the establishment of a proper dialogue flow, such as an answer in response to a question. The results of our case study indicated that users who take the initiative in the dialogue tend to utilize response suggestions less frequently. We also found that some written responses were based on the suggested sentence structure but conveyed different messages.

In assessing the sense of agency (SoA) in the field, researchers are often confronted with the limitations of conventional methods (e.g., retrospectivity, task-disruption). Generally, facial analysis offers a promising possibility to overcome these obstacles by directly collecting objective data in real-time and without any need for contact. This paper examines the applicability of face tracking for SoA assessment by investigating whether facial expressions can be used to derive information about the extent to which users feel in control in interactions with technical systems. In a laboratory experiment (N = 27) based on the dot control task, we manipulated SoA via task difficulty and explored the effects on facial expression patterns. Three action units in the region of the mouth were observed to be characteristic facial indicators of SoA. These findings support the suitability of facial expressions for SoA evaluation as well as the potential of face tracking for monitoring SoA in various practical fields of application (e.g., automated driving).

In this paper, we simulated true and false recognition in the Deese/Roediger/McDermott (DRM; Deese, 1959; Roediger & McDermott, 1995) paradigm by incorporating word embeddings derived from distributed semantic models (word2vec) into an instance memory model (MINERVA2). Previously, Arndt and Hirshman (1998) demonstrated that MINERVA2 (Hintzman, 1984) could capture multiple classic false recognition findings with randomly generated word representations. However, as randomized representations deviate systematically from semantic representations learned from the natural language environment, there remains uncertainty about whether MINERVA2 can capture the false memory illusion when scaling up to real-life complexity in word representations. To address this uncertainty, we used word2vec embeddings that are derived from large corpora of natural language instead of randomized representations in MINERVA2. Our results showed that MINERVA2 can still capture the standard true and false recognition, and it can also accommodate the true and false recognition effects of various classic manipulations (e.g., associative strength, number of associates, divided attention, retention interval).

Predictability is a prerequisite for effective human control of artificial intelligence (AI). For example, the inability to predict the malfunctioning of AI impedes timely human intervention. In this paper, we employ a computerized navigation task, a lunar lander game, to investigate empirically how AI's predictability compares to humans' predictability. To this end, we ask participants to guess whether the landings of a spaceship in this game performed by AI and humans will succeed. We show that humans are worse at predicting AI performance than at predicting human performance in this environment. Significantly, participants underestimate the differences in the relative predictability of AI and, at times, overestimate their prediction skills. We link the difference in predictability to differences in the approaches, i.e. different landing patterns, employed by AI and humans to succeed in the task. These results highlight important differences in perception of AI and human with implications for human-computer interaction.

Knowledge Graph Reasoning (KGR) is an effective method to solve the incompleteness and sparsity problems of Knowledge Graph (KG), which infers new knowledge based on existing knowledge. Especially, the Graph Convolution Network (GCN)-based approaches can obtain state-of-the-art effectiveness, but there are still some problems such as weak reasoning ability, incomplete local information acquisition, insufficient attention score, and high learning cost, which lead to limited prediction accuracy. This paper proposes a multi-head self-attention mechanism architecture based on cognitive transfer learning, named CogTrans, to make effective improvements in the above problems. Shaped like a cross, CogTrans horizontally includes intuition and reasoning stages, which can achieve a faster convergence rate and obtain prediction results that are more in line with human intuition. Furthermore, CogTrans longitudinally includes source and target domains, and benefit from transfer learning, it can not only obtain the advantages of the horizontal architecture but also can “draw inferences from one instance”, which is more conducive to realizing the human brain-like reasoning effect of the architecture. Extensive experimental results show that our CogTrans architecture can obtain the most advanced accuracy of current GCN-based methods.