Emergence of a compositional neural code for written words: Recycling of a convolutional neural network for reading
-
- T. Hannagan
- Cognitive Neuroimaging Unit, Commissariat à l'Énergie Atomique et aux Énergies Alternatives, INSERM, Université Paris-Saclay, NeuroSpin, Gif-Sur-Yvette 91191, France;
-
- A. Agrawal
- Cognitive Neuroimaging Unit, Commissariat à l'Énergie Atomique et aux Énergies Alternatives, INSERM, Université Paris-Saclay, NeuroSpin, Gif-Sur-Yvette 91191, France;
-
- L. Cohen
- Sorbonne Université, INSERM U1127, CNRS UMR 7225, Institut du Cerveau et de la Moelle épinièr, Hôpital de la Pitié-Salpêtrière, Paris 75013, France;
-
- S. Dehaene
- Cognitive Neuroimaging Unit, Commissariat à l'Énergie Atomique et aux Énergies Alternatives, INSERM, Université Paris-Saclay, NeuroSpin, Gif-Sur-Yvette 91191, France;
説明
<jats:title>Significance</jats:title> <jats:p>Learning to read results in the formation of a specialized region in the human ventral visual cortex. This region, the visual word form area (VWFA), responds selectively to written words more than to other visual stimuli. However, how neural circuits at this site implement an invariant recognition of written words remains unknown. Here, we show how an artificial neural network initially designed for object recognition can be retrained to recognize words. Once literate, the network develops a sparse neuronal representation of words that replicates several known aspects of the cognitive neuroscience of reading and leads to precise predictions concerning how a small set of neurons implement the orthographic stage of reading acquisition using a compositional neural code.</jats:p>
収録刊行物
-
- Proceedings of the National Academy of Sciences
-
Proceedings of the National Academy of Sciences 118 (46), 2021-11-08
Proceedings of the National Academy of Sciences