Neural correlates of morphological decomposition during visual word recognition

  • Authors:
  • Brian T. Gold;Kathleen Rastle

  • Affiliations:
  • -;-

  • Venue:
  • Journal of Cognitive Neuroscience
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Considerable behavioral research has demonstrated that the visual word recognition system is sensitive to morphological structure. It has typically been assumed that analysis of morphologically complex words occurs only when the meaning of these words can be derived from the meanings of their constituents (e.g., hunter = hunt + er). However, results from recent behavioral research using the masked priming technique have demonstrated that morphological analysis can occur at an earlier orthographic level, in cases in which the meanings of complex words cannot be derived from their constituents (e.g., corner = corn + er). Here, we combine the logic of behavioral masked priming with the neurophysiological phenomenon of functional magnetic resonance imaging priming suppression to look for evidence of nonsemantic morphological priming at the neural level. Both behavioral and functional magnetic resonance imaging results indicated priming effects associated with the mere appearance of morphological structure (corner---CORN). In addition, these effects were distinguishable from lexical-semantic effects (bucket---PAIL) and orthographic effects (brothel---BROTH). Three left-lateralized occipito-temporal regions showed sensitivity to early morphological components of visual word recognition. Two of these regions also showed orthographic priming (∼BA 37, peak: -48 -60 -17; ∼BA 19, peak: -40 -77 -1), whereas one was sensitive only to morphological similarity between primes and targets (∼BA 19, peak: ∼37 ∼67 ∼7). These findings provide a neurobiological basis for a purely structural morphemic segmentation mechanism operating at early stages of visual word recognition.