Hasil untuk "Consciousness. Cognition"

Tom Kortenbach, Yuan Yin, Milene Guerreiro Gonçalves et al.

Can Large Language Models (LLMs) be creative? Existing research into the creativity of LLMs frames the debate as if the construct of creativity is limited to our conception of human creativity and fails to articulate the opportunity for human and LLM creativity to co-exist and contribute to the creative act in different manners. The objective of this article is to provide a new viewpoint for reflecting on the capabilities of LLMs in the domain of creativity and to enable us to acknowledge the relative strengths and weaknesses of LLMs compared to humans. A framework is proposed that builds on existing creativity frameworks (4Ps, 5As, and 5Cs) and articulates four perspectives (Entity, Proposition, Process, and Environment) for exploring the relation between LLMs and (human) creativity. Within each perspective, empirical and theoretical research on the creativity of LLMs is positioned into a broader landscape of theoretical creativity literature from the fields of psychology, philosophy, design, and computational creativity. In the discussion section, a framework is proposed that highlights how LLMs should be regarded as complementary amplifiers of the human cognition in the domain of natural language instead of ‘on par’ in terms of nature and capacities. For researchers, the insights provide a theoretical foundation for designing empirical research. For practitioners in the creative domain, the article introduces the vocabulary to articulate how humans and LLMs relate in the creative act.

Michael Timothy Bennett

Whether machines can be conscious depends not only on what they compute, but \emph{when} they compute it. Most deployed artificial systems realise their functions via sequential or time-multiplexed updates, yet a moment of conscious experience feels unified and simultaneous. I prove that this difference matters. I augment Stack Theory with algebraic laws relating within time-window constraint satisfaction to conjunction. I introduce a temporal semantics over windowed trajectories $τ_Δ$ and prove that existential temporal realisation $\Diamond_Δ$ does not preserve conjunction. A system can realise all the ingredients of experience across time without ever instantiating the experienced conjunction itself. I then distinguish two postulates, Chord and Arpeggio. Chord is the position that conscious unity requires \textit{objective co-instantiation} of the grounded conjunction within the window, like a musical chord. Arpeggio only needs the ingredients to \textit{occur} within window, like a melody. I formalise concurrency-capacity to measure what is needed to satisfy co-instantiation. Finally, I review neurophysiological evidence suggesting that consciousness depends on phase synchrony and effective connectivity, and that loss of consciousness is associated with its breakdown. Under Chord, software consciousness on strictly sequential substrates is impossible for contents whose grounding requires two or more simultaneous contributors. The hardware matters.

Bridget Leonard, Scott O. Murray

Multimodal language models (MLMs) perform well on semantic vision-language tasks but fail at spatial reasoning that requires adopting another agent's visual perspective. These errors reflect a persistent egocentric bias and raise questions about whether current models support allocentric reasoning. Inspired by human spatial cognition, we introduce perspective tokens, specialized embeddings that encode orientation through either (1) embodied body-keypoint cues or (2) abstract representations supporting mental rotation. Integrating these tokens into LLaVA-1.5-13B yields performance on level-2 visual perspective-taking tasks. Across synthetic and naturalistic benchmarks (Isle Bricks V2, COCO, 3DSRBench), perspective tokens improve accuracy, with rotation-based tokens generalizing to non-human reference agents. Representational analyses reveal that fine-tuning enhances latent orientation sensitivity already present in the base model, suggesting that MLMs contain precursors of allocentric reasoning but lack appropriate internal structure. Overall, embedding cognitively grounded spatial structure directly into token space provides a lightweight, model-agnostic mechanism for perspective-taking and more human-like spatial reasoning.

Yueqing Hu, Xinyang Peng, Shuting Peng et al.

Recent Large Reasoning Models trained via reinforcement learning exhibit a "natural" alignment with human cognitive costs. However, we show that the prevailing paradigm of reasoning distillation -- training student models to mimic these traces via Supervised Fine-Tuning (SFT) -- fails to transmit this cognitive structure. Testing the "Hán Dān Xué Bù" (Superficial Mimicry) hypothesis across 14 models, we find that distillation induces a "Functional Alignment Collapse": while teacher models mirror human difficulty scaling ($\bar{r}=0.64$), distilled students significantly degrade this alignment ($\bar{r}=0.34$), often underperforming their own pre-distillation baselines ("Negative Transfer"). Our analysis suggests that SFT induces a "Cargo Cult" effect, where students ritualistically replicate the linguistic form of reasoning (verbosity) without internalizing the teacher's dynamic resource allocation policy. Consequently, reasoning distillation decouples computational cost from cognitive demand, revealing that human-like cognition is an emergent property of active reinforcement, not passive imitation.

Giovanni Pezzulo, Michael Levin

Achieving advanced machine intelligence remains a central challenge in AI research, often approached through scaling neural architectures and generative models. However, biological systems offer a broader repertoire of strategies for adaptive, goal-directed behavior - strategies that emerged long before nervous systems evolved. This paper advocates a genuinely life-inspired approach to machine intelligence, drawing on principles from biology that enable robustness, autonomy, and open-ended problem-solving across scales. We frame intelligence as flexible problem-solving, following William James, and develop the concept of "cognitive light cones" to characterize the continuum of intelligence in living systems and machines. We argue that biological evolution has discovered a scalable recipe for intelligence - and the progressive expansion of organisms' "cognitive light cone", predictive and control capacities. To explain how this is possible, we distill five design principles - multiscale autonomy, growth through self-assemblage of active components, continuous reconstruction of capabilities, exploitation of physical and embodied constraints, and pervasive signaling enabling self-organization and top-down control from goals - that underpin life's ability to navigate creatively diverse problem spaces. We discuss how these principles contrast with current AI paradigms and outline pathways for integrating them into future autonomous, embodied, and resilient artificial systems.

José Ricardo Vieira de Almeida, Gleydson Vieira da Silva Barros, Vitor José Monteiro Borges da Silva Valente et al.

Improvements in sleep quality and depressive symptoms are considered a cornerstone of adult health. Physical exercise is one of the interventions used to treat people with sleep disorders and improve mental health. However, there is no standardization regarding the physical exercise protocols and their effects on sleep quality and depressive symptoms in adults. The present study aims to verify, through a systematic review and meta-analysis, the effect of physical exercise on sleep quality and symptoms of depression in adults. This study adhered to the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement and was registered in the International Prospective Register of Systematic Reviews (PROSPERO). The PubMed, Cochrane Library, and Scopus databases were used to identify relevant original articles and clinical trials. Analysis was performed with Review Manager (RevMan) software (The Cochrane Collaboration, London, United Kingdom), version 5.4. The study included men and women over 18-years-old, with physical exercise as the intervention. The studies included pre- and postevaluation of sleep quality and depressive symptoms. A total of 931 articles were found, of which 15 met the eligibility criteria, encompassing 940 participants. Physical exercise significantly improved sleep quality (mean difference: -1.19; 95% confidence interval [95%CI]: −1.66 to −0.73) and depressive symptoms (mean difference: −3.51; 95%CI: −4.66 to −2.36). Aerobic exercise was the most common and effective for both outcomes. Thus, physical exercise was effective in improving sleep quality and depressive symptoms in adults. Additional studies, however, should be performed to confirm these findings.

James V. Wertsch

Anxiety about nuclear war emerged after the 1945 atomic bombings of Japan and has risen and fallen over the following decades. It is grounded in future thinking shaped by narrative form and function in policy discussions and especially in film and television. These media have repeatedly drawn on three basic narrative templates organised around three different endings: destruction, judgement, and renewal; human extinction; and permanent and irreversible societal collapse. Several film and television productions are used to illustrate the internal organisation of these narrative templates and to examine how both nuclear fear and nuclear anxiety are involved.

Wasif Khan, Lin Gu, Noah Hammarlund et al.

Physical activity is a modifiable lifestyle factor with potential to support cognitive resilience. However, the association of moderate-to-vigorous physical activity (MVPA) intensity, and timing, with cognitive function and region-specific brain structure remain poorly understood. We analyzed data from 45,892 UK Biobank participants aged 60 years and older with valid wrist-worn accelerometer data, cognitive testing, and structural brain MRI. MVPA was measured both continuously (mins per week) and categorically (thresholded using >=150 min/week based on WHO guidelines). Associations with cognitive performance and regional brain volumes were evaluated using multivariable linear models adjusted for demographic, socioeconomic, and health-related covariates. We conducted secondary analyses on MVPA timing and subgroup effects. Higher MVPA was associated with better performance across cognitive domains, including reasoning, memory, executive function, and processing speed. These associations persisted in fully adjusted models and were higher among participants meeting WHO guidelines. Greater MVPA was also associated with subcortical brain regions (caudate, putamen, pallidum, thalamus), as well as regional gray matter volumes involved in emotion, working memory, and perceptual processing. Secondary analyses showed that MVPA at any time of day was associated with cognitive functions and brain volume particularly in the midday-afternoon and evening. Sensitivity analysis shows consistent findings across subgroups, with evidence of dose-response relationships. Higher MVPA is associated with preserved brain structure and enhanced cognitive function in later life. Public health strategies to increase MVPA may support healthy cognitive aging and generate substantial economic benefits, with global gains projected to reach USD 760 billion annually by 2050.

Chi-Sheng Chen, Ying-Jung Chen, Aidan Hung-Wen Tsai

Electroencephalography provides a non-invasive window into brain activity, offering valuable insights for neurological research, brain-computer interfaces, and clinical diagnostics. However, the development of robust machine learning models for EEG analysis is hindered by the scarcity of large-scale, well-annotated datasets and the inherent variability of EEG signals across subjects and recording conditions. Inspired by the success of foundation models in natural language processing and computer vision, we propose the Large Cognition Model-a transformer-based foundation model designed to generalize across diverse EEG datasets and downstream tasks. Unlike traditional approaches, our proposed transformer-based architecture demonstrates strong generalization capabilities across datasets and tasks, even without pretraining, surpassing some existing EEG universal models on specific downstream applications. LCM leverages large-scale self-supervised learning techniques to capture universal EEG representations, enabling efficient fine-tuning for applications such as cognitive state decoding, disease classification, and neurofeedback systems. We introduce a novel architecture that integrates temporal and spectral attention mechanisms, optimizing the model's ability to extract meaningful features from raw EEG signals. Extensive evaluations demonstrate that LCM outperforms state-of-the-art approaches across multiple EEG benchmarks, exhibiting strong cross-subject and cross-task generalization. Our findings highlight the potential of pretrained EEG foundation models to accelerate advancements in neuroscience, personalized medicine, and BCI technology.

Boshi Huang, Fabio Nonato de Paula

This paper introduces a novel self-consciousness defense mechanism for Large Language Models (LLMs) to combat prompt injection attacks. Unlike traditional approaches that rely on external classifiers, our method leverages the LLM's inherent reasoning capabilities to perform self-protection. We propose a framework that incorporates Meta-Cognitive and Arbitration Modules, enabling LLMs to evaluate and regulate their own outputs autonomously. Our approach is evaluated on seven state-of-the-art LLMs using two datasets: AdvBench and Prompt-Injection-Mixed-Techniques-2024. Experiment results demonstrate significant improvements in defense success rates across models and datasets, with some achieving perfect and near-perfect defense in Enhanced Mode. We also analyze the trade-off between defense success rate improvement and computational overhead. This self-consciousness method offers a lightweight, cost-effective solution for enhancing LLM ethics, particularly beneficial for GenAI use cases across various platforms.

Bertram Opitz, Veit Kubik

Abstract Benefits of self-testing for learning have been consistently shown for simple materials such as word lists learned by rote memorization. Considerably less evidence for such benefits exists for complex, more educationally relevant materials and its application to new situations. The present study explores the mechanisms underlying this transfer. To this end, a typical retrieval-practice-effect paradigm was applied to foster the learning of an artificial language. Participants either repeatedly studied grammatically correct exemplar sentences of the artificial language or engaged in a cloze test as the interim test after learning. To assess far transfer, participants in both groups of restudy and retrieval practice engaged in a grammaticality judgment test after a delay of 5 min and 1 week. In addition, participants in both groups completed a final memory test (i.e., a cloze test identical to the initial test) 1 week after learning. In addition to a long-term memory benefit of retrieval practice, results revealed also a retrieval-practice benefit in the far-transfer test after the 1-week delay. The findings further support the view that far transfer is supported by learning the underlying grammatical rules as opposed to memorizing the material. Thus, retrieval practice is also effective for fostering learning of complex materials and, even more importantly, for promoting transfer of learning—a crucial goal in modern educational practices.

Anahita Khorrami Banaraki, Armin Toghi, Azar Mohammadzadeh

In response to shortcomings of the current classification system in translating discoveries from basic science to clinical applications, NIMH offers a new framework for studying mental health disorders called Research Domain Criteria (RDoC). This framework holds a multidimensional outlook on psychopathologies focusing on functional domains of behavior and their implementing neural circuits. In parallel, the Predictive Processing (PP) framework stands as a leading theory of human brain function, offering a unified explanation for various types of information processing in the brain. While both frameworks share an interest in studying psychopathologies based on pathophysiology, their integration still needs to be explored. Here, we argued in favor of the explanatory power of PP to be a groundwork for the RDoC matrix in validating its constructs and creating testable hypotheses about mechanistic interactions between molecular biomarkers and clinical traits. Together, predictive processing may serve as a foundation for achieving the goals of the RDoC framework.

Lorraine Subias, Noriko Katsu, Kazunori Yamada

Abstract Potential metacognitive abilities, such as monitoring and controlling cognitive processes, have been revealed in some primate species. In the tubes task, apes and macaques showed higher content-checking behavior when unaware of a reward’s location, but they also periodically inspected the tubes when aware, especially when a more appealing reward was involved. Some attribute this to the pleasure of looking at the reward. This study investigates whether the unnecessary tube-checking behavior observed in nine wild Japanese macaques, previously tested for metacognition using the tubes task, can be solely attributed to impulsivity. The macaques’ propensity to look inside a single tube containing food they cannot immediately reach was measured and compared to their behavior in the tubes task. Results indicated that looking inside the baited tube increased as reward quality improved. However, macaques displaying unnecessary tube inspections in metacognitive tests showed less impulsivity to look. This intriguing result counters the notion that excessive looking in the tubes task is solely due to impulsive looking, prompting us to advocate for further research into the relationship between inhibition and metacognitive performance.

Floriana Pintucci, Francesca Cremonini, Giulia Romagnolo et al.

Objectives To evaluate the efficacy of mandibular advancement devices (MADs) in improving apnea-hypopnea index (AHI) in positional obstructive sleep apnea (POSA), compared with a control group of nonpositional OSA (NPOSA) patients, from mild to very severe degree, in order to to find the main variables characterizing the examined group as potential predictors of treatment success.

Murat Kucukosmanoglu, Javier O. Garcia, Justin Brooks et al.

Deep neural network (DNN) models have demonstrated impressive performance in various domains, yet their application in cognitive neuroscience is limited due to their lack of interpretability. In this study we employ two structurally different and complementary DNN-based models, a one-dimensional convolutional neural network (1D-CNN) and a bidirectional long short-term memory network (BiLSTM), to classify individual cognitive states from fMRI BOLD data, with a focus on understanding the cognitive underpinnings of the classification decisions. We show that despite the architectural differences, both models consistently produce a robust relationship between prediction accuracy and individual cognitive performance, such that low performance leads to poor prediction accuracy. To achieve model explainability, we used permutation techniques to calculate feature importance, allowing us to identify the most critical brain regions influencing model predictions. Across models, we found the dominance of visual networks, suggesting that task-driven state differences are primarily encoded in visual processing. Attention and control networks also showed relatively high importance, however, default mode and temporal-parietal networks demonstrated negligible contribution in differentiating cognitive states. Additionally, we observed individual trait-based effects and subtle model-specific differences, such that 1D-CNN showed slightly better overall performance, while BiLSTM showed better sensitivity for individual behavior; these initial findings require further research and robustness testing to be fully established. Our work underscores the importance of explainable DNN models in uncovering the neural mechanisms underlying cognitive state transitions, providing a foundation for future work in this domain.

Nik Bear Brown

The Cognitive Type Project is focused on developing computational tools to enable the design of typefaces with varying cognitive properties. This initiative aims to empower typographers to craft fonts that enhance click-through rates for online ads, improve reading levels in children's books, enable dyslexics to create personalized type, or provide insights into customer reactions to textual content in media. A significant challenge in research related to mapping typography to cognition is the creation of thousands of typefaces with minor variations, a process that is both labor-intensive and requires the expertise of skilled typographers. Cognitive science research highlights that the design and form of letters, along with the text's overall layout, are crucial in determining the ease of reading and other cognitive properties of type such as perceived beauty and memorability. These factors affect not only the legibility and clarity of information presentation but also the likability of a typeface.

T. I. Petukhova, A. O. Chupakhina

Introduction. The purpose  of the research is to consider  the linguistic actualization of the main characteristics of the artist's visual perception in the interview discourse. The relevance of the research is due to its inclusion in the context  of modern anthropo-oriented study of cognition of the world by a creative person.  The novelty of the research lies in the fact that for the  first time  the  process  of perception  of the  world by an English-speaking artist  is studied as an event within the framework of a complex  linguistic and cognitive approach, which allows to analyze its specificity to the fullest extent.Methodology and sources. The methodological  basis of the research was made up of the works   of   the    linguists-cognitologists    L.V.    Laenko,   I.Yu.   Kolesov,   psychologist    V. A. Barabanshchikov who revealed distinctive characteristics of perceptual image and substantiated  the  expediency   of  perception   as  an  event.  The  research  material   was interviews  with English-speaking  artists  from  the  Internet  versions  of  the  newspapers devoted  to  the  life and work of modern  artists.  The methods  of lexico-semantic, frame, contextual  and pragmalinguistic analysis were used.Results and discussion. The article reveals the peculiarities of visual perception  of the world by contemporary English-speaking artists, which are reflected in the discourse of the interview. The author  reveals  the multidimensionality of a subject  of perception,  different  ways of its representation on the verbal level: as an observer-actor and as a “offscreen” observer.  The paper shows the specificity of visual perception of the world by the artist as a perceptual event that  possesses spatial  and  temporal  coordinates  and  such  significant  characteristics  as dynamism, motivation, focus and emotionality. The denotative situation of perception is actualized in English by personal  pronouns,  active-action  predicates,  perceptual  predicates, existential constructions, phase verbs, and state change predicates.  The important feature  of the artist's creative perception  of the world is the forming of the necessity  of representation in his  consciousness perceptive  image  by means  of  painterly  signs.  As  a  result  of  such representation the given perceptive  image receives  the further  development  and becomes the basis for creating an artistic image to be embodied in an art work.Conclusion. In the interview with the artist there  is a complex process  of cognition of the world  by  the   creative   person,   based   on  the   perceptual   event.   Visual  perception   is represented by the artist as a directed activity, which is to learn and which requires  certain efforts  and motivation. As a result of the perceptual  event that the artist experiences, an artistic image is formed that is actualized in the work of art.

Johanna M. Hoppe, Y. Walldén, M. Kanstrup et al.

Intrusive memories of trauma (memories that enter consciousness involuntarily) highjack cognitive processing, cause emotional distress, and represent a core symptom of posttraumatic stress disorder. Intrusive memories often contain the worst moment/s ('hotspots') of the trauma memory. Little is known about hotspots shortly after they are formed, i.e., in the first hours after trauma. We investigated the features of hotspots in trauma-exposed individuals (n = 21) within 72 h post-trauma, using linguistic analysis and qualitative coding. On average, participants reported three hotspots per traumatic event (M = 7.8 words/hotspot). Hotspots primarily contained words related to time, space, motion, and sensory processing. Most hotspots contained sensory features (97%) and motion (59%). Few cognitions and no emotion words were identified. Results indicate that hotspots collected shortly post-trauma are expressed as motion-rich sensory-perceptual experiences (mental imagery) with little detail about emotion/cognition. Findings are discussed in terms of the function of hotspots (e.g., preparedness for action) and clinical implications.

Andrew Chung, Michael A. Busseri, Karen M. Arnell

Abstract Several studies have investigated the effect of induced mood state on conceptual breadth (breadth and flexibility of thought). Early studies concluded that inducing a positive mood state broadened cognition, while inducing a negative mood state narrowed cognition. However, recent reports have suggested that valence and arousal can each influence conceptual breadth. Individual differences in affective dispositions may bias perceptions, thoughts, and behaviors and, in turn, may be biased by them. Here, we examine whether individual differences in valence and arousal dimensions of self-reported, naturally occurring affect relate to conceptual breadth (using the Remote Associates Test, the Object Categorization Task, and the Alternative Uses Task), with no mood manipulations or cues. The three conceptual breadth tasks loaded onto a latent conceptual breadth factor that was predicted significantly by the interaction of valence and arousal. For participants low in arousal, greater positive affect was associated with greater conceptual breadth. For participants high in arousal, greater positive affect was associated with reduced conceptual breadth. In contrast to most existing theories of conceptual breadth that highlight the importance of valence or arousal alone, the present results suggest that the interaction between arousal and valence is key to predicting individual differences in conceptual breadth. We posit that positive mood states predict greater conceptual breadth in the presence of low versus high arousal due to a relaxation of cognitive control under low arousal.

Andreas Patinkin, M. Sih Setija Utami, Erna Agustina Yudiati

Penelitian ini memiliki tujuan untuk mengetahui pengaruh pemberian Emotion-Focused Therapy (EFT) pada penurunan gejala depresi pada lansia di panti wreda. Metode yang digunakan dalam penelitian ini adalah single subject experiment atau single case experiment. Pada penelitian ini, terdapat tiga orang lansia yang menjadi partisipan (L1: 78 tahun, L2: 73 tahun, dan L4: 82 tahun). Ketiga partisipan telah mengikuti rangkaian eksperimen berupa delapan sesi EFT. Penelitian ini menggunakan sejumlah teknik pengumpulan data, seperti observasi, wawancara, dan dua buah skala. Skala yang pertama adalah Beck Depression Inventory-II (BDI-II) sebagai instrumen seleksi partisipan dan instrumen kedua adalah Geriatric Depression Scale-15 (GDS-15) yang akan diberikan kepada partisipan sepanjang proses eksperimen. Hasil perolehan data GDS-15 dianalisis menggunakan analisis visual dengan aplikasi statistik R. Hasil dari penelitian ini menunjukkan bahwa pemberian EFT dapat menurunkan depresi pada partisipan lansia. Turunnya depresi pada partisipan dapat dicermati lebih saksama pada penurunan gejala-gejala depresi setiap partisipan.