How decisions evolve: The temporal dynamics of action selection

Abstract

To study the process of decision-making under conflict, researchers typically analyze response latency and accuracy. However, these tools provide little evidence regarding how the resolution of conflict unfolds over time. Here, we analyzed the trajectories of mouse movements while participants performed a continuous version of a spatial conflict task (the Simon task). We applied a novel combination of multiple regression analysis and distribution analysis to determine how conflict on the present trial and carry-over from the previous trial affect responding. Response on the previous trial and the degree of conflict on the previous and the current trial all influenced performance, but they did so differently: The previous response influenced the early part of the mouse trajectory, but the degree of previous and current conflict influenced later parts. This suggests that in this task experiencing conflict may not proactively ready the system to handle conflict on the next trial; rather, when conflict is experienced on the subsequent trial the previous compensatory processing may be re-activated more efficiently.

Introduction

In everyday life, we frequently have to decide between competing action tendencies. Like attractors in a dynamic system, alternative choice options drag the mind into one direction at one moment and another direction at the next moment. When deciders contemplate between alternatives, we assume that they are in conflict. How such conflicts between different action tendencies are resolved by the brain has been studied extensively in the laboratory with tasks inducing response conflicts (cf. Botvinick, 2007) such as the Simon task. In this task, participants have to respond with a left or right key press, for example, to the direction of an arrow pointing to the left or to the right. Since the arrow is presented to the left or to the right of a central fixation point, the irrelevant stimulus location is assumed to automatically activate the spatially corresponding response. This leads to congruent trials (no conflict) when the location of the arrow (e.g. left) corresponds to the direction of the arrow (e.g. left) and to incongruent trials (response conflict) when the location of the arrow (e.g. left) does activate a different response than required by the direction of the arrow (e.g. right). The response conflict is reflected in behavioral indicators such as increased response times (RT) or error rates on incongruent relative to congruent trials (e.g. Simon, 1969), indicating that conflicting information between stimulus identity and stimulus location prolonged decision-making and response selection processes.

While the degree of conflict in the current trial is one influence on the processes determining the final decision in a Simon task, cognitive psychologists have also highlighted the role of other influences such as response and stimulus repetitions (e.g. Wühr & Ansorge, 2005) or previous experience of response conflict as additional factors that determine response selection processes in the current trial (e.g. Egner, 2007, Stürmer et al., 2002). Especially the latter one, conflict in the previous trial, has been shown to reduce the influence of conflicting information in the current trial, an effect that has been termed conflict-adaptation (e.g. Egner, 2007).

Although the study of manifold influences on response selection yielded major insights into the process of making the final decision under conflict (Proctor & Vu, 2006), the most common outcome measures such as mean RT or accuracy bear some fundamental limitations, as they provide only indirect access to the temporal dynamics of information-processing (Scherbaum et al., 2008, Spivey, 2006) by manipulating the task itself (cf. Notebaert & Verguts, 2007).

Whereas previous attempts to uncover the temporal dynamics of response decisions primarily used EEG measures, e.g. the lateralized readiness potential (e.g. Stürmer et al., 2002), and EMG measures (e.g. Burle et al., 2002, Coles et al., 1985), more recently, attempts were made to use continuously recorded mouse (McKinstry et al., 2008, Spivey et al., 2005) or hand movement trajectories (Song and Nakayama, 2006, Song and Nakayama, 2008) to obtain behavioral indicators of the underlying process dynamics. These studies aimed at providing general evidence for the continuous nature of cognitive processing (as opposed to stage-like processing), and indicated that investigating cognitive processing continuously could indeed be a worthwhile extension of the behavioral measures RT and accuracy (Song and Nakayama, 2009, Spivey, 2006).

While movement trajectories have been used before in the Simon task (Buetti and Kerzel, 2008, Buetti and Kerzel, 2009), we aim to exploit the potential of this behavioral approach by providing detailed analyses of the time-course and dynamic interactions of concurrent influences that have been found to determine the ongoing decision. These are, on the one hand, inherent properties of the current trial_N (i.e. stimulus location and stimulus direction), and, on the other hand, typical influences of the previous trial_N−1 (i.e. previous response and previous trial congruency). For this, we report two experiments in which we combined for the first time, the tracing of mouse trajectories in a continuous version of the Simon task with a multiple regression (Notebaert & Verguts, 2007) and distribution analytic approach (De Jong, Liang, & Lauber, 1994) to dissect the temporal dynamics of those four competing influences determining the selection of the required response in the Simon task.