Purchase this article with an account.
David Rothlein, Brenda Rapp; The similarity structure of distributed neural responses reveals abstract and modality-specific representations of letters. Journal of Vision 2013;13(9):786. doi: https://doi.org/10.1167/13.9.786.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Most cognitive theories of visual letter processing posit modality-specific representations of letter shapes, spoken letter names, and motor plans, and also abstract, amodal letter representations that unify the various modality-specific formats. Importantly, abstract letter representations encode identity in a case-invariant manner such that markedly different letter-forms can share the same abstract letter representation (e.g. r – R). However, fundamental questions remain regarding the very existence of abstract letter representations, the neuro-topography of the different types of letter representations, and the degree of cortical selectivity for orthographic information. We use Multivariate Pattern Analysis-Representational Similarity Analysis (MVPA-RSA) specifically applying searchlight methods to directly test quantitative models of the similarity/dissimilarity structure of distributed neural representations of letters. These analyses identify substrates selectively tuned to both modality-specific (visual, phonological and motoric) representations of letters, as well as a left hemisphere occipital-temporal region within the ventral visual stream selectively tuned to abstract letter representations. We also find that these different formats of letter representation are closely integrated with neural networks used for word reading. The approaches applied address various shortcomings of previous studies that have investigated these questions and the findings serve to advance our understanding of the format of the letter representations found within sub-regions of the large-scale networks used in reading and spelling. Furthermore, the evidence of abstract letter representation provides a clear example of ventral stream encoding of object identity information that is invariant to fundamental differences in visual form.
Meeting abstract presented at VSS 2013
This PDF is available to Subscribers Only