Multidimensional semantic representations emerge from multi-frequency representational similarity learning
- 1University of Cambridge
Abstract
Current theories of semantic representation are agnostic about how fine-grained neurophysiological activity corresponds to representation of multidimensional semantic information. By applying an innovative multivariate technique (representational similarity learning; RSL), we adjudicated three hypotheses: (1) that multidimensional semantic structure is represented within a single frequency range (e.g., gamma/high gamma); (2) that each semantic dimension is independently represented within a different frequency range; and (3) that multidimensional semantic information is "transfrequency" (at least some information emerges only when multiple frequencies are considered together). RSL was applied to time-frequency power and phase data extracted from electrocorticography (ECoG) grid electrodes on the surface of human ventral anterior temporal lobe (vATL). We found significant decoding of graded, multidimensional semantic information from a wide range of frequencies (4 - 200 Hz), but not from individual frequency bands, providing clear evidence that multidimensional semantic information is coded in a transfrequency fashion.
Related
- Stimulus symmetries can confound representational similarity analyses
- COMPASS 2.0: psychometric representational similarity analysis distinguishes symptom structure from personal signal
- Not all solutions are created equal: An analytical dissociation of functional and representational similarity in deep linear neural networks
- Decomposing deep neural network-brain representational similarity reveals distinct sources across the human visual hierarchy
- Assessing psychological, neural and AI-based representational geometries across multiple similarity properties
BibTeX
@misc{frisby2026multidimensional,
title = {{Multidimensional semantic representations emerge from multi-frequency representational similarity learning}},
author = {S. L. Frisby and C. R. Cox and A. D. Halai and A. Shimotake and T. Kikuchi and K. Kobayashi and T. Kuneida and Y. Arakawa and A. Ikeda and R. Matsumoto and T. T. Rogers and M. A. Lambon Ralph},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.23.753859},
url = {https://www.biorxiv.org/content/10.64898/2026.09.23.753859v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Multidimensional semantic representations emerge from multi-frequency representational similarity learning
AU - S. L. Frisby
AU - C. R. Cox
AU - A. D. Halai
AU - A. Shimotake
AU - T. Kikuchi
AU - K. Kobayashi
AU - T. Kuneida
AU - Y. Arakawa
AU - A. Ikeda
AU - R. Matsumoto
AU - T. T. Rogers
AU - M. A. Lambon Ralph
PY - 2026
DA - 2026/09/23
PB - bioRxiv
DO - 10.64898/2026.09.23.753859
UR - https://www.biorxiv.org/content/10.64898/2026.09.23.753859v1
AB - Current theories of semantic representation are agnostic about how fine-grained neurophysiological activity corresponds to representation of multidimensional semantic information. By applying an innovative multivariate technique (representational similarity learning; RSL), we adjudicated three hypotheses: (1) that multidimensional semantic structure is represented within a single frequency range (e.g., gamma/high gamma); (2) that each semantic dimension is independently represented within a different frequency range; and (3) that multidimensional semantic information is "transfrequency" (at least some information emerges only when multiple frequencies are considered together). RSL was applied to time-frequency power and phase data extracted from electrocorticography (ECoG) grid electrodes on the surface of human ventral anterior temporal lobe (vATL). We found significant decoding of graded, multidimensional semantic information from a wide range of frequencies (4 - 200 Hz), but not from individual frequency bands, providing clear evidence that multidimensional semantic information is coded in a transfrequency fashion.
ER -