ISPC effect is not observed when the word comes too late: a time course analysis.

The item-specific proportion congruency (ISPC) effect is demonstrated by a smaller Stroop effect observed for mostly incongruent items compared to mostly congruent items. Currently, there is a continuing debate on whether conflict driven item-specific control processes or stimulus-response contingency learning account for the ISPC effect. In the present study, we conducted two experiments to investigate the time course of the ISPC effect with a stimulus onset asynchrony (SOA) manipulation. Both negative and positive SOAs were used in order to manipulate the contingency learning between the word and the color dimensions. We also combined this SOA manipulation with a set size manipulation (Bugg and Hutchison, 2013) to moderate the contribution of contingency learning and item-specific processes to the observed ISPC effect. We expected that the change in the magnitude of the ISPC effect as a result of SOA would follow different patterns for the 2-item and 4-item set conditions. Results showed that the SOA manipulation influenced the ISPC effect. Specifically, when the word followed the color with a 200 ms delay, the observed ISPC effect was smaller, if at all present, than the ISPC effects in other negative and positive SOA conditions, regardless of set size. In conclusion, our results showed that the ISPC effect was not observed if the word arrived too late. We also conducted additional awareness and RT distribution analyses (delta plots) to further investigate the ISPC effect. These analyses showed that a higher percentage of participants were aware of the ISPC manipulation in the 2-item set condition compared to the 4-item set condition. Delta plots revealed that the ISPC effect was smaller for fastest responses and increased as the responses got slower.