Vol.:(0123456789) 1 3
Psychological Research
DOI 10.1007/s00426-017-0860-z
ORIGINAL ARTICLE
On the time-course of automatic response activation in the Simon
task
Ruben Ellinghaus
1
· Matthias Karlbauer
1
· Karin M. Bausenhart
1
· Rolf Ulrich
1
Received: 19 September 2016 / Accepted: 21 March 2017
© Springer-Verlag Berlin Heidelberg 2017
Proctor, 1995). In this paradigm, participants respond to
a certain attribute (e.g., color) of stimuli which randomly
appear on the left or right side of a fxation point. Usu-
ally, a congruency efect is found: Participants react faster
when the stimulus is presented on the same side as the cor-
rect response key (congruent trials) compared to trials in
which the stimulus is presented on the side opposite to the
response key (incongruent trials). Hence, although irrel-
evant to the task, the location information of the stimulus
nevertheless infuences the response selection process.
Experimental psychologists interested in the cognitive
mechanisms underlying performance in the Simon task
have moved beyond the analysis of mean reaction time
(RT) to also analyze RT distributions (e.g., de Jong, Liang,
& Lauber, 1994). Since collapsing data to mean RTs always
means losing information, experimental manipulations that
do not infuence mean RTs might still afect the shape of
RT distributions (Balota et al., 2008; Heathcote, Popiel, &
Mewhort, 1991; Townsend & Ashby, 1983). Also, model
predictions can be compared against observed data on the
level of RT distributions (Miller, 1982; Ruthruf, 1996;
Sigman & Dehaene, 2005). The goal of the present inves-
tigation was to examine distribution-level predictions moti-
vated by the Difusion Model for Confict Tasks (DMC),
a novel model of the mechanisms underlying the Simon
efect recently proposed by Ulrich et al. (2015) in a man-
ner that is consistent with recent neurophysiological results
(Servant, White, Montagnini, & Burle, 2016).
An elucidating approach for investigating the efects of
experimental manipulations on RT distributions in confict
tasks is the delta plot (DP), developed by de Jong et al.
(1994). Consider an experiment in which RTs are meas-
ured in two diferent conditions (e.g., congruent and incon-
gruent trials). For each condition, RTs are rank ordered
and divided into equal-sized bins. Then, conditions can
Abstract The Simon efect (prolonged RT when the
task-irrelevant stimulus location is incongruent with the
response side) has been reported to decrease at longer
RTs, which is refected in negative-going delta functions.
This fnding has been attributed to gradual dissipation of
the response automatically activated by the task-irrelevant
location information. The Difusion Model for Confict
Tasks (DMC, Ulrich, Schröter, Leuthold, & Birngruber,
Cognitive Psychology 78:148–174, 2015) formally speci-
fes the time-course of this automatic activation process as
a pulse-like function. In contrast to alternative views, DMC
is consistent with the notion that this time-course is unaf-
fected by the presentation duration of the target stimulus.
Therefore, we expected that delta functions are invariant
against changes of stimulus duration. This prediction was
verifed in two Simon task experiments. Consistent with
this general result, DMC’s parameter τ which defnes the
time-course of the automatic response activation was esti-
mated to not meaningfully difer between short and long
durations. We argue that our results are coherent with pro-
cessing architectures that assume a transient automatic pro-
cess that is virtually unafected by stimulus duration.
Introduction
The mechanisms underlying performance in cognitive con-
fict situations are often investigated in the Simon paradigm
(Simon, 1969; for reviews see Hommel, 2011 and Lu &
* Ruben Ellinghaus
ruben.ellinghaus@uni-tuebingen.de
1
Psychological Institute, University of Tübingen,
Schleichstrasse 4, 72076 Tübingen, Germany