Hearing What We Read

Psychological scientists have discovered new evidence of what goes on in the brain when people read printed words. The scientists, led by Maria Dimitropoulou of the Basque Center on Cognition, Brain, and Language, in Donostia, Spain, used Greek and Spanish, two languages with common phonemes and partially overlapping graphemes, to investigate how knowledge about the relationship between written language and sound influences our ability to recognize words.

The study was published in the Journal of Cognitive Psychology. In two experiments, the scientists used the masked priming paradigm, a method used to study visual word recognition. The paradigm works like this: First, volunteers are briefly shown one string of letters, called the prime. The prime’s appearance and disappearance happen so fast that the volunteers aren’t even aware of seeing it. Immediately following, the volunteers are shown another string of letters, called the target. Even though primes are undetected by the volunteers, experiments show that certain primes can influence how quickly volunteers react to or recognize targets.

Dimitropoulou and her colleagues designed a version of the masked priming paradigm to test how the phonological (speech and sound–related) codes that we associate with written language influence our ability to visually identify words. In the past, scientists who conducted similar experiments had found it difficult to separate the cognitive influence of phonology from the influence of orthography (visual elements like spelling and script) because many words that sound the same look similar, too — even in the case of words that belong to different languages. Some have tried to overcome this limitation by incorporating made-up pseudowords into masked priming experiments; however, Dimitropoulou and colleagues object that the brain may process real words and made-up words differently.

To avoid orthographic interference in their experiment on phonology, the team chose to mix Greek words with Spanish words, noting that the two languages sound similar but look very different when written. They designed an experiment in which native Greeks who had studied Spanish for several years were primed with Greek words that were phonologically similar to but orthographically distinct from Spanish target words. Because Greek and Spanish have some alphabetic characters in common, the scientists were also able to test the effect of Greek primes that were both phonologically similar and orthographically similar to Spanish targets. In the control condition, volunteers were primed with words that were not related to the target. The same procedure was followed in a second experiment, but this time the primes were in Spanish and the targets in Greek.

In both experiments, purely phonological primes led volunteers to identify their targets more quickly than in the control condition. Curiously, when primes had both phonological and orthographical properties in common with the target words, no significant difference in identification time emerged between the experimental and control condition in either language. The results suggest that phonological cues by themselves — but not phonological cues in conjunction with orthographic cues — play a special role in visual word recognition.

Dimitropoulou, M., Dunabeitia, J. A., & Carreiras, M. (2011). Phonology by itself: Masked phonological priming effects with and without orthographic overlap Journal of Cognitive Psychology, 23 (2), 185-203 DOI: 10.1080/20445911.2011.477811