Is there evidence for a noisy computation deficit in developmental dyslexia?

Area of Science:

• Neuroscience
• Cognitive Science
• Developmental Psychology

Background:

• The noisy computation hypothesis suggests developmental dyslexia (DD) arises from neural processing variability.
• This hypothesis predicts less stable neural representations of words in individuals with DD.

Purpose of the Study:

• To test the noisy computation hypothesis of DD.
• To investigate whether neural representations are noisier in dyslexic readers compared to typical readers.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used with dyslexic and typical readers.
• Participants repeatedly viewed words twelve times.
• Behavioral (reaction time variability) and neural (univariate, multivariate pattern analyses) measures were analyzed.

Main Results:

• No significant differences in neural representation variability were found between dyslexic and typical readers.
• Robust findings within each group indicated high representational similarity and good classification of words.

Conclusions:

• The study found no evidence to support the noisy computation hypothesis of developmental dyslexia.
• Neural representations do not appear to be significantly noisier in dyslexic readers.