Competition between orthographically and phonologically similar words during sentence reading: Evidence from eye movements

• 2 Eye-tracking experiments examined the effects of neighbor words during reading.
• Inhibition effect only when prime and target both look and sound alike.
• No effects when prime and target only look alike or only sound alike.
• Inhibition effect disappears with increased distance/time between prime and target.
• Affected by syntactic structure and readers’ comprehension skill.

Two eye movement experiments tested the effect of orthographic and/or phonological overlap between prime and target words embedded in a sentence. In Experiment 1, four types of overlap were tested: phonological and orthographic overlap (O+P+) occurring word initially (strain–strait) or word finally (wings–kings), orthographic overlap alone (O+P−, bear–gear) and phonological overlap alone (O−P+, smile–aisle). Only O+P+ overlap resulted in inhibition, with the rhyming condition showing an immediate inhibition effect on the target word and the non-rhyming condition on the spillover region. No priming effects were found on any eye movement measure for the O+P− or the O−P+ conditions. Experiment 2 demonstrated that the size of this inhibition effect is affected by both the distance between the prime and target words and by syntactic structure. Inhibition was again observed when primes and targets appeared close together (approximately 3 words). In contrast, no inhibition was observed when the separation was nine words on average, with the prime and target either appearing in the same sentence or separated by a sentence break. However, when the target was delayed but still in the same sentence, the size of the inhibitory effect was affected by the participants’ level of reading comprehension. Skilled comprehenders were more negatively impacted by related primes than less skilled comprehenders. This suggests that good readers keep lexical representations active across larger chunks of text, and that they discard this activation at the end of the sentence. This pattern of results is difficult to accommodate in existing competition or episodic memory models of priming.