Satoshi Nishida
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Satoshi Nishida has a total of 93 co-authors. Here are the top co-authors by number of shared publications:
- Shinji Nishimoto
Center for Information and Neural Networks, National Institute of Information and Communications Technology, Suita, Japan
18 shared publications - Naoya Maeda
NTT DATA Japan Corporation, Tokyo, 135-6033, Japan
5 shared publications Tomohiro Tanaka
Department of Integrative Brain Science, Graduate School of Medicine, Kyoto University, Kyoto, Japan; and
5 shared publicationsAntoine Blanc
Center for Information and Neural Networks (CiNet), Advanced ICT Research Institute, National Institute of Information and Communications Technology (NICT)
4 shared publications- Hideki Asoh
Artificial Intelligence Research Center, National Institute of Advanced Industrial Science and Technology, Tsukuba, Japan
4 shared publications - Katsunori Miyahara
Center for Human Nature, Artificial Intelligence, and Neuroscience (CHAIN), Hokkaido University, Kita, Sapporo, Japan
4 shared publications
Latest Publications
- Open Access
- 23 October 2026
- Published by Elsevier in NeuroImage
- Vol. 342, 122312
Decoding individual differences in mental information from human brain responses predicted by deep neural networks
- 23 October 2026
- Published by Elsevier in NeuroImage
- Vol. 342, 122312
Open AccessDecoding individual differences in mental information from human brain responses predicted by deep neural networks
- Open Access
- 15 September 2026
- Published by SAGE Publications in I-Perception
- Vol. 17 (5)
Illusions induce contradictory perception of motion
Can we perceive motion and stillness simultaneously? Philosophical studies suggest that the perception of illusory motion can involve both of these contradictory contents. We examined this possibility by asking participants to rate both perceived motion and perceived stillness using stimuli with illusory or physical motion. The results showed that the negative correlations between the two ratings were much weaker in illusory motion, suggesting that motion and stillness can be disentangled in illusory motion perception. This is the first empirical examination of the hypothesis that perception can involve contradictory contents.
- 15 September 2026
- Published by SAGE Publications in I-Perception
- Vol. 17 (5)
Open AccessIllusions induce contradictory perception of motion
Can we perceive motion and stillness simultaneously? Philosophical studies suggest that the perception of illusory motion can involve both of these contradictory contents. We examined this possibility by asking participants to rate both perceived motion and perceived stillness using stimuli with illusory or physical motion. The results showed that the negative correlations between the two ratings were much weaker in illusory motion, suggesting that motion and stillness can be disentangled in illusory motion perception. This is the first empirical examination of the hypothesis that perception can involve contradictory contents.
- Open Access
Bridging Perception and Language: A Systematic Benchmark for LVLMs' Understanding of Amodal Completion Reports
One of the main objectives in developing large vision-language models (LVLMs) is to engineer systems that can assist humans with multimodal tasks, including interpreting descriptions of perceptual experiences. A central phenomenon in this context is amodal completion, in which people perceive objects even when parts of those objects are hidden. Although numerous studies have assessed whether computer-vision algorithms can detect or reconstruct occluded regions, the inferential abilities of LVLMs on texts related to amodal completion remain unexplored. To address this gap, we constructed a benchmark grounded in Basic Formal Ontology to achieve a systematic classification of amodal completion. Our results indicate that while many LVLMs achieve human-comparable performance overall, their accuracy diverges for certain types of objects being completed. Notably, in certain categories, some LLaVA-NeXT variants and Claude 3.5 Sonnet exhibit lower accuracy on original images compared to blank stimuli lacking visual content. Intriguingly, this disparity emerges only under Japanese prompting, suggesting a deficiency in Japanese-specific linguistic competence among these models.
Open AccessBridging Perception and Language: A Systematic Benchmark for LVLMs' Understanding of Amodal Completion Reports
One of the main objectives in developing large vision-language models (LVLMs) is to engineer systems that can assist humans with multimodal tasks, including interpreting descriptions of perceptual experiences. A central phenomenon in this context is amodal completion, in which people perceive objects even when parts of those objects are hidden. Although numerous studies have assessed whether computer-vision algorithms can detect or reconstruct occluded regions, the inferential abilities of LVLMs on texts related to amodal completion remain unexplored. To address this gap, we constructed a benchmark grounded in Basic Formal Ontology to achieve a systematic classification of amodal completion. Our results indicate that while many LVLMs achieve human-comparable performance overall, their accuracy diverges for certain types of objects being completed. Notably, in certain categories, some LLaVA-NeXT variants and Claude 3.5 Sonnet exhibit lower accuracy on original images compared to blank stimuli lacking visual content. Intriguingly, this disparity emerges only under Japanese prompting, suggesting a deficiency in Japanese-specific linguistic competence among these models.
- 2 April 2025
- Published by Springer Nature
- p. 167-178
Sex Differences in Brain Information Processing: Insights from Specific Brain Regions and Hemispheric Connectivity
- 2 April 2025
- Published by Springer Nature
- p. 167-178
Sex Differences in Brain Information Processing: Insights from Specific Brain Regions and Hemispheric Connectivity
- 28 March 2025
- Published by Springer Nature
- p. 412-417
Enhancing Emotion Prediction via Self-supervised Learning Using Neural Data
- 28 March 2025
- Published by Springer Nature
- p. 412-417
Enhancing Emotion Prediction via Self-supervised Learning Using Neural Data
- Open Access
- 9 February 2025
- Published by Elsevier in Neuroimage: Reports
- Vol. 5 (1) , 100243
Asymmetric representation of symmetric semantic information in the human brain
- 9 February 2025
- Published by Elsevier in Neuroimage: Reports
- Vol. 5 (1) , 100243
Open AccessAsymmetric representation of symmetric semantic information in the human brain
- Open Access
- 8 November 2024
- Published by Elsevier in Computers in Human Behavior: Artificial Humans
- Vol. 2 (2)
Behavioral and neural evidence for the underestimated attractiveness of faces synthesized using an artificial neural network
- 8 November 2024
- Published by Elsevier in Computers in Human Behavior: Artificial Humans
- Vol. 2 (2)
Open AccessBehavioral and neural evidence for the underestimated attractiveness of faces synthesized using an artificial neural network
- Open Access
- 28 August 2024
- Published by Wiley in Psychiatry and Clinical Neurosciences
- Vol. 78 (11) , 687-695
Homogenization of word relationships in schizophrenia: Topological analysis of cortical semantic representations
Aim: Patients with schizophrenia typically exhibit symptoms of disorganized thought and display concreteness and over‐inclusion in verbal reports, depending on the level of abstraction. While concreteness and over‐inclusion may appear contradictory, the underlying psychopathology that explains these symptoms remains unclear. In the current study, we used functional magnetic resonance imaging with an encoding modeling approach to examine how concepts of various words, represented as brain activity, are anomalously connected at different levels of abstraction in patients with schizophrenia.Methods: Fourteen individuals diagnosed with schizophrenia and 17 healthy controls underwent functional magnetic resonance imaging to measure brain activity representing concepts of various words. We used a persistent homology (PH) method to analyze the topological structures of word representations in schizophrenia patients, healthy controls, and random data, across different levels of abstraction by varying dissimilarity scales in the representation space.Results: The results revealed that patients with schizophrenia exhibited more homogeneous word relationships across different levels of abstraction compared with healthy controls. Additionally, topological structures exhibited a shift toward a random network structure in patients with schizophrenia compared with controls. The PH method successfully distinguished semantic representations of patients with schizophrenia from those of controls.Conclusions: The current results provide an explanation for the mechanisms underlying the deficits in abstraction ability observed in schizophrenia. The isotopic connection of individual concepts reflects both the reduction of contextual connections at a semantically fine‐grained scale and the absence of clear boundaries between related concepts at a coarse scale, which lead to concreteness and over‐inclusion, respectively.
- 28 August 2024
- Published by Wiley in Psychiatry and Clinical Neurosciences
- Vol. 78 (11) , 687-695
Open AccessHomogenization of word relationships in schizophrenia: Topological analysis of cortical semantic representations
Aim: Patients with schizophrenia typically exhibit symptoms of disorganized thought and display concreteness and over‐inclusion in verbal reports, depending on the level of abstraction. While concreteness and over‐inclusion may appear contradictory, the underlying psychopathology that explains these symptoms remains unclear. In the current study, we used functional magnetic resonance imaging with an encoding modeling approach to examine how concepts of various words, represented as brain activity, are anomalously connected at different levels of abstraction in patients with schizophrenia.Methods: Fourteen individuals diagnosed with schizophrenia and 17 healthy controls underwent functional magnetic resonance imaging to measure brain activity representing concepts of various words. We used a persistent homology (PH) method to analyze the topological structures of word representations in schizophrenia patients, healthy controls, and random data, across different levels of abstraction by varying dissimilarity scales in the representation space.Results: The results revealed that patients with schizophrenia exhibited more homogeneous word relationships across different levels of abstraction compared with healthy controls. Additionally, topological structures exhibited a shift toward a random network structure in patients with schizophrenia compared with controls. The PH method successfully distinguished semantic representations of patients with schizophrenia from those of controls.Conclusions: The current results provide an explanation for the mechanisms underlying the deficits in abstraction ability observed in schizophrenia. The isotopic connection of individual concepts reflects both the reduction of contextual connections at a semantically fine‐grained scale and the absence of clear boundaries between related concepts at a coarse scale, which lead to concreteness and over‐inclusion, respectively.
- Open Access
Neural correlates of phenomenological attitude toward perceptual experience
Phenomenology is one of the most promising approaches to study conscious experience. It holds that a rigorous study of conscious experience requires a transition in the subject from the “natural attitude” (NA) to the “phenomenological attitude” (PA). NA describes our ordinary stance, in which our attention is directed at external objects and events. PA is a distinctive, reflective stance in which our attention is directed at our conscious experience itself. Despite its theoretical importance in philosophy and science of consciousness, the neural mechanisms underlying PA remain unknown. To clarify this point, we developed a novel behavioral task in which participants alternate between NA and PA in relation to their stimulus-evoked subjective experiences. Participants are presented with two sentences and requested to identify the one that best captures their experience. These sentences are designed to induce either NA or PA. We found that participants had lower error rates but slower reaction times in the PA condition compared to the NA condition, suggesting a difference beyond task difficulty. Using fMRI, we also found that multivoxel activation patterns in the premotor cortex, posterior parietal cortex, supplementary motor area, and cerebellum successfully classified the task conditions. Furthermore, the activation strength in these regions was lower in the PA condition, indicating that PA depends on neural processes that suppress action-related information. These findings provide the first evidence for the neural signature of PA, contributing to a better understanding of phenomenological method and its underlying neural mechanisms. Significance statement Phenomenology is one of the most promising approaches to study conscious experience. A key step is a transition from the natural attitude (NA)—where our attention is directed at external objects and events—to the phenomenological attitude (PA)—where our attention is directed toward our conscious experience itself. However, the neural processes underlying PA remain unclear. This study aimed to clarify this point by analyzing fMRI signals measured during a cognitive task that forced participants to repeatedly alternate between NA and PA. We observed that multiple action-related brain regions exhibited different activation patterns between NA and PA. Our findings provide the first neuroscientific evidence that illuminates the core process of the phenomenological method.
Open AccessNeural correlates of phenomenological attitude toward perceptual experience
Phenomenology is one of the most promising approaches to study conscious experience. It holds that a rigorous study of conscious experience requires a transition in the subject from the “natural attitude” (NA) to the “phenomenological attitude” (PA). NA describes our ordinary stance, in which our attention is directed at external objects and events. PA is a distinctive, reflective stance in which our attention is directed at our conscious experience itself. Despite its theoretical importance in philosophy and science of consciousness, the neural mechanisms underlying PA remain unknown. To clarify this point, we developed a novel behavioral task in which participants alternate between NA and PA in relation to their stimulus-evoked subjective experiences. Participants are presented with two sentences and requested to identify the one that best captures their experience. These sentences are designed to induce either NA or PA. We found that participants had lower error rates but slower reaction times in the PA condition compared to the NA condition, suggesting a difference beyond task difficulty. Using fMRI, we also found that multivoxel activation patterns in the premotor cortex, posterior parietal cortex, supplementary motor area, and cerebellum successfully classified the task conditions. Furthermore, the activation strength in these regions was lower in the PA condition, indicating that PA depends on neural processes that suppress action-related information. These findings provide the first evidence for the neural signature of PA, contributing to a better understanding of phenomenological method and its underlying neural mechanisms. Significance statement Phenomenology is one of the most promising approaches to study conscious experience. A key step is a transition from the natural attitude (NA)—where our attention is directed at external objects and events—to the phenomenological attitude (PA)—where our attention is directed toward our conscious experience itself. However, the neural processes underlying PA remain unclear. This study aimed to clarify this point by analyzing fMRI signals measured during a cognitive task that forced participants to repeatedly alternate between NA and PA. We observed that multiple action-related brain regions exhibited different activation patterns between NA and PA. Our findings provide the first neuroscientific evidence that illuminates the core process of the phenomenological method.
- Open Access
Asymmetric representation of symmetric semantic information in the human brain
Specific pairs of semantic entities have symmetric relationships, such as word pairs with opposite meanings (e.g., “intelligent” and “stupid”; “human” and “mechanical”). Such semantic symmetry is a key feature of semantic information. However, the representation of symmetric semantic information in the brain is not yet understood. Additionally, it is unclear whether symmetric pairs of semantic information do have symmetric representations in the brain? We addressed this question in a data-driven manner by using the voxelwise modeling of movie-evoked cortical response measured by functional magnetic resonance imaging. In this modeling, response in each voxel was predicted from semantic labels designated for each movie scene. The semantic labels consisted of 30 different items, including 15 pairs of semantically symmetric items. Each item was manually evaluated using a 5-point scale. By localizing the semantic representation associated with each item based on the voxelwise accuracy of brain-response predictions, we found that semantic representations of symmetric item pairs are broadly distributed but with little overlap in the cortex. Additionally, the weight of voxelwise models revealed highly complex, various patterns of cortical representations for each item pair. These results suggest that symmetric semantic information is rather asymmetric and heterogeneous representations in the human brain. Significance statement This study aimed to investigate if symmetric pairs of semantic information have symmetric representations in the human brain using statistical modeling of functional magnetic resonance imaging signals evoked by naturalistic movies. We built a model based on movie labeling of symmetric semantic items to quantify the cortical representations of symmetric semantic entities. These findings showed that symmetric pairs of semantic entities are represented in widespread cortical regions; however, still exhibit little overlap in localization and heterogeneous representations. These results offer significant insights into the cortical representations of semantic symmetry and advance the understanding of representational structure of semantic information in the human brain.
Open AccessAsymmetric representation of symmetric semantic information in the human brain
Specific pairs of semantic entities have symmetric relationships, such as word pairs with opposite meanings (e.g., “intelligent” and “stupid”; “human” and “mechanical”). Such semantic symmetry is a key feature of semantic information. However, the representation of symmetric semantic information in the brain is not yet understood. Additionally, it is unclear whether symmetric pairs of semantic information do have symmetric representations in the brain? We addressed this question in a data-driven manner by using the voxelwise modeling of movie-evoked cortical response measured by functional magnetic resonance imaging. In this modeling, response in each voxel was predicted from semantic labels designated for each movie scene. The semantic labels consisted of 30 different items, including 15 pairs of semantically symmetric items. Each item was manually evaluated using a 5-point scale. By localizing the semantic representation associated with each item based on the voxelwise accuracy of brain-response predictions, we found that semantic representations of symmetric item pairs are broadly distributed but with little overlap in the cortex. Additionally, the weight of voxelwise models revealed highly complex, various patterns of cortical representations for each item pair. These results suggest that symmetric semantic information is rather asymmetric and heterogeneous representations in the human brain. Significance statement This study aimed to investigate if symmetric pairs of semantic information have symmetric representations in the human brain using statistical modeling of functional magnetic resonance imaging signals evoked by naturalistic movies. We built a model based on movie labeling of symmetric semantic items to quantify the cortical representations of symmetric semantic entities. These findings showed that symmetric pairs of semantic entities are represented in widespread cortical regions; however, still exhibit little overlap in localization and heterogeneous representations. These results offer significant insights into the cortical representations of semantic symmetry and advance the understanding of representational structure of semantic information in the human brain.