AUTHOR=Winsler Kurt , Holcomb Phillip J. , Midgley Katherine J. , Grainger Jonathan 

TITLE=Evidence for Separate Contributions of High and Low Spatial Frequencies during Visual Word Recognition

JOURNAL=Frontiers in Human Neuroscience

VOLUME=Volume 11 - 2017

YEAR=2017

URL=https://www.frontiersin.org/journals/human-neuroscience/articles/10.3389/fnhum.2017.00324

DOI=10.3389/fnhum.2017.00324

ISSN=1662-5161

ABSTRACT=Previous studies have shown that different spatial frequency information processing streams interact during the recognition of visual stimuli. However it is a matter of debate as to the contributions of high and low spatial frequency information for visual word recognition. This study examined the role of different spatial frequencies in visual word recognition using ERP masked priming. EEG was recorded from 32 scalp sites in 30 English-speaking adults in a go/no-go semantic categorization task. Stimuli were white characters on a neutral grey background. Targets were uppercase 5 letter words preceded by a forward-mask (#######) and a 50ms lowercase prime. Primes were either the same word (repeated) or a different word (un-repeated) than the subsequent target and either contained only high, only low, or full spatial frequency information. Additionally within each condition, half of the prime-target pairs were high lexical frequency, and half were low. In the full spatial frequency condition, typical ERP masked priming effects were found with an attenuated N250 (sub-lexical) and N400 (lexical-semantic) for repeated compared to un-repeated primes. For high spatial frequency primes there was a weaker N250 effect which interacted with lexical frequency, a significant reversal of the effect around 300ms, and an N400-like effect for only high lexical frequency word pairs. Low spatial frequency primes did not produce any of the classic ERP repetition priming effects, however they did elicit a distinct early effect around 200ms in the opposite direction of typical repetition effects. High spatial frequency information accounted for many of the masked repetition priming ERP effects and therefore suggests that high spatial frequencies are more crucial for word recognition. However low spatial frequencies did produce their own pattern of priming effects indicating that larger scale information may still play a role in word recognition.