Send to

Choose Destination
Neuropsychologia. 2015 Sep;76:254-63. doi: 10.1016/j.neuropsychologia.2015.03.023. Epub 2015 Mar 21.

For richer or poorer? Imageability effects in semantic dementia patients' reading aloud.

Author information

Neuroscience and Aphasia Research Unit, School of Psychological Sciences, Zochonis Building, University of Manchester, Brunswick Street, Manchester M13 9PL, England. Electronic address:


The degree to which a word's meaning evokes a mental image exerts an influence on performance across a variety of conceptual and linguistic tasks. In normal healthy participants, this effect takes the form of an advantage for high over low imageability words. Consideration of the influence of imageability on performance of patients with semantic dementia can provide information concerning its cognitive and neural bases. Semantic dementia patients show deficits in conceptual processing tasks, and an associated enhancement of the advantage for high over low imageability words. Semantic dementia patients also show deficits in linguistic processing tasks, including reading aloud words with inconsistent spelling-sound correspondences. This study provides the first systematic exploration of the influence of imageability on semantic dementia patients' reading aloud performance. Over 10 cases, the imageability effect seen for inconsistent words was actually reversed in reaction times, with faster performance for low than high imageability items. The same reversal was observed for inconsistent words when the frequency of legitimate alternative reading of components errors was considered, and this reversed effect grew larger with increasing semantic impairment. This result is interpreted in terms of the development of stronger connections along the direct pathway between spelling and sound for low than high imageability items that are then revealed under diminished semantic activation. This interpretation emphasises the interaction between form and meaning that occurs throughout learning in connectionist models.


Imageability; Reading; Semantic dementia; Spelling–sound consistency; Task differences

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center