The Mechanism Of Congruency Sequence Effect In The Flanker Task

Conflict control is the center of the normal human life, which plays an important role. Researches indicated that studies of conflict control were mostly focused on the congruency sequence effect (CSE). The CSE is that the interference from incongruent distractors is smaller following an incongruent trial than following a congruent one. Conflict monitoring hypothesis proposes that this effect is due to conflict adaptation, that is, the conflict detection on previous trial facilitates the solving of the upcoming conflict. Moreover, the feature integration account attributes this effect to repetition priming. However, previous researches which was used the Flanker task to prove CSE were not focused on the factors, such as the spatial distance effect or the individual difference and so on. It induces that the mechanism of CSE is indistinct until now. There were four experiments in the current study, which research CSE with the factors including the spatial distance effect or individual difference and so on in order to provide the new crucial empirical evidence on the Flanker task for CSE.The standard arrow flanker task with the spatial distance effect was used in Experiment 1. On the basic of previous studies, we added the different length of the intertrial interval (ITI) into this experiment because the arrow stimuli with two dimensions in this flanker task could not eliminate the influence of repetition completely. In Experiment 1, the influence of the following factors was examined:the presentation of a fixation and ITI were considered. Results of Experiment 1 suggested that the spatial distance effect with the standard arrow flanker task resulted in the presentation of the fixation becoming a clue during finishing the task that could directly affect the CSE.The larger set of stimulus and response with the letters flanker task was used in Experiment 2A to remove the stimulus repetition and entirely eliminate the influence of feature integration, in order to obtain the "pure" CSE and analyze the overall and individual participants respectively. All of participants in Experiment 2A was tested the Attentional Network Test (ANT) in Experiment 2B, to analyze the association between individual differences and the alerting level of the attentional network system. Results of Experiment 2A showed that the overall CSE was absent for the change trials. After analyzing further, we found that even the overall CSE was absent for the change trials, an opposite behavioral patterns for the same change trials was showed across the participants, that is, the RTCI-â…¡on almost half (30 in 65) of participants was positive named positive group, which accorded the CSE; but it was negative for the left ones named negative group, which was contrary to the CSE. Results of ANT showed that the alerting level of positive group obtaining CSE was higher than negative group that CSE was absent.Participants were tested by ANT firstly and finished the same letter flanker task with Experiment 2A in Experiment 3. We did not use RTCI-â…¡, but directly used the efficiencies of three attention networks as the criteria to divide participants to prove the conclusion of Experiment 2. Results of Experiment 3 showed that the overall CSE for the change trials was replicated with Experiment 2A. And the alerting level of positive group obtaining CSE was still higher than negative group that CSE was absent, indicating the difference of the alerting level could directly affect the presentation of CSE.Participants were still tested by ANT firstly and finished the same letter flanker task with Experiment 2A, but the probability of congruent and incongruent trials was varied between blocks, so that congruent trials occurred on 80% or 20% of the trials in Experiment 4 rather than 50% in Experiment 1 to 3. We investigated the representation of CSE under the different probabilities of congruent and incongruent trials. Results of Experiment 4 showed that when congruent trials occurred on 80%, the CSE was both present for the repetition and change trials whether the alerting level was high or not; on the contrary, when congruent trials occurred on 20%, the CSE was only present for the repetition, but disappeared for the change trials. It indicated that the CSE was not affected by the alerting level when the probability of congruent and incongruent trials was not equal.Follow-up analyses revealed that:1) Because of the spatial distance effect with the flanker task, the CSE was directly influenced by the presentation of the fixation, that is, the position of the fixation was the same with the position of upcoming target stimulus.2) The stimuli in the standard flanker task were only two dimensions and could not eliminate the influence of repetition from itself. So the factor of ITI was added in order to solve it.3) The use of a stimulus and response set lager enough to allow feature integration to be eliminated. Although the overall CSE was absent for the change trials, an opposite pattern was present across participants. Further analysis showed that the CSE was still present for the change trials for participants of the high alerting level; on the contrary, it was disappeared for ones of the low alerting level. It indicated that the alerting level of attentional network produces an effect on the ability of conflict detection and the adaptation for the upcoming trial, so influence the presentation of the CSE finally.4) When the probability of congruent and incongruent trials was equal, participants were difficult to expect the correct upcoming trial type. The ready state of participants could be affected by the different level of alerting function, that is, the high alerting participants showed a proactive control, and the low alerting ones showed a reactive control. So the CSE was directly affected by the different ready state.5) When the probability of congruent and incongruent trials was different, participants would expect the correct upcoming trial type. The alerting level of attentional network had no influence on the CSE whether the alerting level of participants was high or not, indicating the same behavioral pattern.