A biased competition model accounts for negative priming in a free choice situation as residual effects of inhibitory conflict resolution

The processing of a stimulus can be facilitated (positive priming) or impeded (negative priming), depending on whether a repeated stimulus has recently been attended or ignored. The current experiment presented consecutive dichotic syllable pairs with instructions to report any syllable from each pair. The results showed that trials which repeated a syllable from the previous trial had increased response times, and the repeated syllable had a decreased likelihood of being selected. This indicates that inhibition is involved when selecting between stimuli, and that the inhibition associated with stimuli has a residual effect on subsequent selection. Closer examination showed that a repeated syllable was more likely to be ignored if the syllable had been ignored rather than attended on the previous presentation, and trials showing this response pattern had faster response times than trials that did not, thus representing an effect similar to negative priming. We suggest that a biased competition network model, in which pathways are made stronger or weaker as a residual effect of selection, can be applied to account for the observed effect. The model has support from other experimental tasks, and in contrast to prominent accounts of negative priming, it focuses on the processing performed by the network rather than on pre-onset selection criteria for processing.