Neural and visual correlates of perceptual decision making in adult dyslexia

Franzen, Leon (2018) Neural and visual correlates of perceptual decision making in adult dyslexia. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2018franzenphd.pdf] PDF
Download (11MB)
Printed Thesis Information: https://eleanor.lib.gla.ac.uk/record=b3348239

Abstract

Humans have to make decisions based on visual information numerous times every day—for example, judging whether it is a friend or simply a nice stranger who is waving at us from the other side of the street, or whether the content of a contract we are about to sign is correct. In particular, perceptual decisions based on good reading comprehension might disadvantage people affected by the specific learning disorder dyslexia, characterised by impairments in reading and writing. In recent years, neuroscience has begun to uncover the neural basis of these impairments in children and adults. However, it remains unknown what neural differences might underlie impaired processing of the physical properties of written words, such as font type and style.

The current thesis sought to characterise the neural and oculomotor temporal correlates of font-modulated reading comprehension while also probing a more fundamental deficit in non-linguistic sensory perceptual decision making in adult dyslexia by using a combination of electrophysiological and eye-tracking methods. The first of our three studies (Chapter 2), investigated the impact of italics—a commonly used font style for highlighting important content—on reading comprehension in a sentence reading lexical decision task. Overall, the performance of dyslexics was worse than that of non-dyslexics. Cluster-based event-related potential (ERP) analysis revealed that brain responses within the first 300 ms following the target (decision) word differed in amplitude and spatial distribution between dyslexics and non-dyslexics when processing italicised text. The two ERP components we observed within this period showed a dissociation in peak time, spatial profile, and their ability to predict behavioural performance. These findings emphasise the importance of choosing font style carefully to optimise word processing and reading comprehension by dyslexics.

Based on these differences, our second study (Chapter 3) asked whether a specific dyslexia font can be used to alleviate difficulties with reading comprehension in adult dyslexia, and what effects such a font has on cognitive and oculomotor mechanisms. Using standardised texts coupled with validated comprehension questions, we demonstrated that reading comprehension across all participants was better on trials presented in the dyslexia font OpenDyslexic compared to those presented in traditional Times New Roman font. These benefits were larger among dyslexics. Conversely, participants’ reading speed was unaffected by OpenDyslexic. Our eye-tracking data showed increases in visual search intensity and ease of visual processing on OpenDyslexic trials in the form of decreases in median fixation duration and fixation to saccade ratio, as well as a smaller number of falsely programmed forward saccades among dyslexics. These findings provide empirical evidence for the efficacy of OpenDyslexic in longer texts and its ability to improve the visual reading strategy.

Finally, recent evidence has shown that adults with dyslexia exhibit obvious fundamental deficits spanning multiple sensory systems when performing simple perceptual decision tasks, such as integrating beeps and flashes. These deficits extend beyond the well-established linguistic difficulties. Particularly, dyslexics reading impairments are believed to be a consequence of deficient integration of congruent audio-visual information. However, it remains unclear whether dyslexic adults exhibit similar impairments when integrating audio-visual evidence in a non-linguistic perceptual decision task with noisy real-world objects. To address this question, and informed by our previous work in non-dyslexics, we used a linear multivariate discriminant analysis to investigate the extent to which audio-visual integration affects early sensory evidence encoding (‘early’) or later decision-related stages (‘late’) in dyslexia. We found increased decision accuracy and slower response times during audio-visual trials for both groups. However, overall, dyslexics showed worse performance than non-dyslexics. When comparing audio-visual to visual trials, we observed that dyslexics exhibited an increase in the magnitude of an EEG component situated between the early and late processing stages. Conversely, non-dyslexics exhibited increased component amplitudes for a later post-sensory EEG component, consistent with a post-sensory influence of audio-visual integration. Our results suggest that adult dyslexics benefit from congruent audio-visual evidence of noisy perceptual stimuli to a similar extent but rely on a different neural process to achieve these improvements.

In conclusion, our results provide novel insights into the neural dynamics, visual and cognitive mechanisms underlying adult dyslexics’ perceptual decision making. They further offer empirical evidence and practical suggestions for easily implementable applications that can improve text comprehension by everyone.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: decision making, dyslexia, font, reading, EEG, ERP, eye-tracking, perception, learning disability.
Subjects: B Philosophy. Psychology. Religion > BF Psychology
Q Science > Q Science (General)
Colleges/Schools: College of Science and Engineering > School of Psychology
Supervisor's Name: Philiastides, Dr. M.G.
Date of Award: 2018
Depositing User: Dr Leon Franzen
Unique ID: glathesis:2018-71950
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 15 May 2019 08:14
Last Modified: 24 Jun 2019 09:55
URI: https://theses.gla.ac.uk/id/eprint/71950
Related URLs:

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year