Dataset for 'The neural correlates of logical-mathematical symbol systems processing resemble that of spatial cognition more than natural language processing'The file zips for each figure contain the raw data, analysis code, and a 'readme' file for explanation. figure1_domain-level meta-maps and overlap figure2_sub-domain of LMS meta-maps and overlap figure3_sub-domain of language meta-maps and overlap figure4_subject-level fMRI maps and overlap figure5_6_ROI multivariate pattern similarityThe research abstract:The ability to manipulate logical-mathematical symbols (LMS), encompassing tasks such as calculation, reasoning, and programming, is a cognitive skill arguably unique to humans. Considering the relatively recent emergence of this ability in human evolutionary history, it has been suggested that LMS processing may build upon more fundamental cognitive systems, possibly through neuronal recycling. Previous studies have pinpointed two primary candidates, natural language processing and spatial cognition. Existing comparisons between these domains largely relied on task-level comparison, which may be confounded by task idiosyncrasy. The present study instead compared the neural correlates at the domain level with both automated meta-analysis and synthesized maps based on three representative LMS tasks, reasoning, calculation, and mental programming. Our results revealed a more substantial cortical overlap between LMS processing and spatial cognition, in contrast to language processing. Furthermore, in regions activated by both spatial and language processing, the multivariate activation pattern for LMS processing exhibited greater multivariate similarity to spatial cognition than to language processing. A hierarchical clustering analysis further indicated that typical LMS tasks were indistinguishable from spatial cognition tasks at the neural level, suggesting an inherent connection between these two cognitive processes. Taken together, our findings support the hypothesis that spatial cognition is likely the basis of LMS processing, which may shed light on the limitations of large language models in logical reasoning, particularly those trained exclusively on textual data without explicit emphasis on spatial content.

Dataset for 'The neural correlates of logical-mathematical symbol systems processing resemble that of spatial cognition more than natural language processing'The file zips for each figure contain the raw data, analysis code, and a 'readme' file for explanation. figure1_domain-level meta-maps and overlap figure2_sub-domain of LMS meta-maps and overlap figure3_sub-domain of language meta-maps and overlap figure4_subject-level fMRI maps and overlap figure5_6_ROI multivariate pattern similarityThe research abstract:The ability to manipulate logical-mathematical symbols (LMS), encompassing tasks such as calculation, reasoning, and programming, is a cognitive skill arguably unique to humans. Considering the relatively recent emergence of this ability in human evolutionary history, it has been suggested that LMS processing may build upon more fundamental cognitive systems, possibly through neuronal recycling. Previous studies have pinpointed two primary candidates, natural language processing and spatial cognition. Existing comparisons between these domains largely relied on task-level comparison, which may be confounded by task idiosyncrasy. The present study instead compared the neural correlates at the domain level with both automated meta-analysis and synthesized maps based on three representative LMS tasks, reasoning, calculation, and mental programming. Our results revealed a more substantial cortical overlap between LMS processing and spatial cognition, in contrast to language processing. Furthermore, in regions activated by both spatial and language processing, the multivariate activation pattern for LMS processing exhibited greater multivariate similarity to spatial cognition than to language processing. A hierarchical clustering analysis further indicated that typical LMS tasks were indistinguishable from spatial cognition tasks at the neural level, suggesting an inherent connection between these two cognitive processes. Taken together, our findings support the hypothesis that spatial cognition is likely the basis of LMS processing, which may shed light on the limitations of large language models in logical reasoning, particularly those trained exclusively on textual data without explicit emphasis on spatial content.

This is the dataset for BodySize_Affordance manuscript.
--- The dataset contains all behaviour data, fMRI data and stimuli we used in the experiment.
## Description of the data and file structure The folder "27objectimages" contains the stimuli we used in behaviour experiment. In the "behaviour data" folder, "AllBehviourData.xlsx" contains all action possibilities related to each object for human data (both male and female), the imagined body data (cat and elephant), and all model outputs (GPT-4, GPT-3.5, GPT-2, BERT).
"singleobject_random100.csv" contains the similarity and averagedsize between object pairs generated by randomly chosen object within each rank. In the "fMRI data" folder, the third-level (group-level) z-maps of WITHIN versus baseline, BEYOND versus baseline, and the conjunction map were provided.
The second-level (all participants) maps for four conditions (2(congurency) by 2(object type)) were also provided. "BCN"for BEYOND-CONGRUENT, "BINC"for BEYOND-INCONGRUENT, "WCN"for WITHIN-CONGRUENT, "WINC"for WITHIN-INCONGRUENT. *Note: the detailed calculation is provided in the Methods part of the manuscript.

This is the dataset for BodySize_Affordance manuscript.
--- The dataset contains all behaviour data, fMRI data and stimuli we used in the experiment.
## Description of the data and file structure The folder "27objectimages" contains the stimuli we used in behaviour experiment. In the "behaviour data" folder, "AllBehviourData.xlsx" contains all action possibilities related to each object for human data (both male and female), the imagined body data (cat and elephant), and all model outputs (GPT-4, GPT-3.5, GPT-2, BERT).
"singleobject_random100.csv" contains the similarity and averagedsize between object pairs generated by randomly chosen object within each rank. In the "fMRI data" folder, the third-level (group-level) z-maps of WITHIN versus baseline, BEYOND versus baseline, and the conjunction map were provided.
The second-level (all participants) maps for four conditions (2(congurency) by 2(object type)) were also provided. "BCN"for BEYOND-CONGRUENT, "BINC"for BEYOND-INCONGRUENT, "WCN"for WITHIN-CONGRUENT, "WINC"for WITHIN-INCONGRUENT. *Note: the detailed calculation is provided in the Methods part of the manuscript.