Activation Followed by Suppression in the Simon Task: Influence of an Accessory Peripheral Signal

Introduction
In a Simon task subjects react to the identity (e.g. colour) of a stimulus, while ignoring the stimulus location. Even though stimulus location is irrelevant, responses are faster and more accurate when stimulus (S) and response (R) location correspond.
The Simon effect (SE) can be interpreted in terms of a dual-route model, where relevant information is processed by a controlled route and irrelevant information by an automatic route. Hommel (1993) suggested that the SE is the consequence of the overlap in time between these two processes (TOM = Temporal Overlap Model). Evidence for this idea was found by presenting the stimulus gradually so that the stimulus location is processed before the relevant information. The SE was reduced in comparison to an abrupt presentation. According to Hommel (1993) the decrease of the effect reflects spontaneous decay of the irrelevant spatial code.
Previous experiments with a central coloured stimulus and a peripheral go/nogo stimulus revealed a Simon effect, using a fixed stimulus-onset asynchrony of 0, 100 and 600 ms. To investigate the temporal characteristics of the Simon effect, we dissociated the processing of the relevant and irrelevant stimulus information by presenting both features at different points in time. An accessory signal was used so we could present the peripheral information before and after the response initiation was already started. In order to prevent anticipations, we used go and no-go trials.
Method
Subjects were instructed to respond to the colour of a central stimulus, presented during 100 ms. Instead of responding immediately, they had to postpone their response until a go stimulus appeared 600 ms later. At different intervals, an irrelevant peripheral stimulus could appear before, after or simultaneously with the go or no-go stimulus.
Results
We found an interaction between congruency and the time when the irrelevant information was presented. A SE was found when the irrelevant location information was shown 150 ms before (p=0.01), or together with, the go-signal (p=0.02). When the irrelevant signal was presented 150 ms after the go-signal started, we found a significant reversal of the SE (p=0.03). Presenting the irrelevant information more than 150 ms before or after the go-signal had no influence on RT (no SE).
Discussion
In support of the TOM and the decay hypothesis, the SE only shows up when the accessory peripheral signal is presented shortly before or simultaneously with the go signal. However, when the peripheral signal appears shortly after the go signal, the SE reverses, probably showing the presence of an active suppression mechanism that inhibits the activation caused by the irrelevant information, as suggested in the activation-suppression model of Ridderinkhof (2002).
Alternatively these results can be explained by S-R binding. At the start of the trial, colour and response code bind. When the go-signal appears, the response is initiated leading to an occupation of the response code. The slower responses on congruent trials, when the accessory signal appears shortly after the go signal, indicates that the code occupation interferes with the attempt to integrate the same code (of the accessory signal) during response initiation. This is supported by a study where relevant information is not given in advance and no reversed SE was found.
Acknowledgments
This research was funded by the Fund for Scientific Research of Flanders, Belgium.
References
Hommel, B. (1993). The relation between stimulus processing and response selection in the Simon task: Evidence for a temporal overlap. Psychological Research, 55, 280-290.
Ridderinkhof, R. (2002). Activation and suppression in conflict tasks: Empirical clarification through distributional analyses. In W. Prinz & B. Hommel (Eds.), Attention and performance XIX: Common mechanisms in perception and action, Oxford: Oxford University Press.