| The purpose of this study is to validate the usefulness of eye tracking as a tool in investigating priming effects. The aim is to investigate which spontaneous eye-movement dependent measures best capture semantic priming effects for isolated words in a cross format priming context, in which written word primes are highly associated or not associated with a target.; The picture stimuli were selected from 260 grey-scale pictures developed by Rossion and Pourtois (2004). The first phase consisted of establishing prime words for each picture target, assigning two pictures as low association nontargets for each prime word, and a traditional priming experiment with a naming task to ensure that the above pairs of related stimuli showed semantic priming effects. Thirty-four pairs of related prime-picture stimuli were selected for the eye-movement experiment. A total of 40 language-normal native English speakers (M = 20.17, SD = 1.68) participated in the eye-movement experiment. For each trial, the word duration was 400 milliseconds and the picture array was displayed for a total of four seconds. Eye movements of the participants were monitored by the ISCAN RK426 remote pupil center/corneal reflection system.; Multifactorial analysis of variance indicates that all the fixation duration measures to the target item, the proportion of fixation time, the average fixation duration, and the first pass fixation duration measures are significantly greater in the related than in the unrelated condition. Fixation duration measures on targets are greater than on nontargets.; In case of the first pass fixation duration measure, fixation latency measure and a fixation location measure, previewing the display prior to presenting the prime plays a significant role in increasing the priming effect size between the related/unrelated, target/nontarget conditions. Additionally, there is a significant relationship between eye-movement dependent measures and the traditional priming reaction time measure.; Spontaneous eye movement measures hold promise as valid indicators of semantic priming effects in a multiple-choice priming format. Similar measures may be useful for investigating priming effects that are less well established, such as form, morphological, and syntactic priming in normal populations and in individuals with neurogenic communication disorders. |
