Foveal crowding in posterior cortical atrophy: A specific early-visual-processing deficit affecting word reading

Abstract

Visual crowding is a form of masking in which single-letter identification is compromised by the presence of additional letters or other simple visual forms in close proximity. This behavioural phenomenon has been studied most frequently in the context of amblyopic and normal peripheral vision. In the current study, we investigate this phenomenon in the context of two patients with peripheral dyslexia and a third with visual disorientation consequent to bilateral posterior cortical atrophy. In one case, reading showed the effects of word length typical of letter-by-letter reading, whereas the second case was unable to read any whole words. In a series of letter identification tasks, recognition accuracy was shown to decrease significantly in the presence of a range of flanking stimuli (e.g., letters, digits, letter fragments). Compatible with previous reports of the crowding phenomenon, the flanking effect was strengthened by increasing flanker proximity but was unaffected by target or flank size, flank contrast, target–flank lexicality, or flank category. One patient also showed amelioration of the flanking effect when the target and flankers were of opposite contrast polarity. To the best of our knowledge, this is the first demonstration of visual crowding in individuals with posterior cortical atrophy. We consider the relevance of these empirical findings to accounts of the letter-by-letter reading form of peripheral dyslexia. In particular, we suggest that crowding constitutes one specific form of early-visual-processing deficit, which impairs the reading process.

Keywords:

• Foveal crowding
• Posterior cortical atrophy
• Peripheral dyslexia
• Early visual processing
• Alzheimer's disease

Acknowledgments

We would like to thank Martin Rossor for allowing us to study patients under his care. We are grateful to Basil Ridha for his help with the clinical assessments and to Basil Ridha, Jason Warren, and John Stevens for their opinions on the neuroimaging data. We would also like to thank Marina Danilova for directing our interest towards work on crowding and to Gordon Plant for very helpful discussions and advice. This work was undertaken at UCLH/UCL (University College London Hospital/University College London) who received a proportion of funding from the Department of Health's NIHR Biomedical Research Centres funding scheme. The Dementia Research Centre is an Alzheimer's Research Trust Coordinating Centre. This work was supported by a Research Fellowship to SC and an equipment grant from the Alzheimer's Research Trust.

Notes

¹ We note that not all patients held to have an attentional dyslexia show a difference in target identification with letter and number (different category) flankers (e.g., Humphreys & Mayall, 2001). Thus the performance of our patients on the current tasks distinguishes their letter- and word-processing abilities from the original description of attentional dyslexia (Shallice & Warrington, 1977), but additional tasks would be required to formally rule out any overlap with possible other subtypes of attentional dyslexia (see Davis & Coltheart, 2002).