The role of voluntary(intentional) decision making in the Serial Reaction Time Task, Version 2

Perceptual decision making is an action to choose an option among a list of available choices, typically from two choices over a short time period (e.g. recognising if a traffic light is red or green during driving). Computational, neural, cognitive and behavioural theories have merged to describe that this dynamic behaviour of perceptual decision making is a consequence of competitions between accumulating sensory evidence of available stimuli (e.g. gradually recognising that a traffic light is red because perceptual information of red light reaches a decisional threshold earlier than that of green light) [for a review, see 1-4]. Similar processes have been described by those theories in a voluntary condition where one is given a full control to decide to choose one stimulus over a list of available options (or even given a choice to choose none), competing between accumulating voluntary decisions within one's mind, which is referred to as an intentional (and also referred to as a voluntary) decision making [e.g. 5-7]. In our first study, we used the perceptual decision making task of the Serial Reaction Time Task or SRTT with an involuntary (i.e. choosing a highlighted cue while avoiding three other unhighlighted cues) and a voluntary or freely-selectable (i.e. choosing one cue among all of four highlighted cues) conditions. This study demonstrated that the voluntary condition showed larger RT improvement on tapping a keyboard key spatially corresponding to a visual cue presented on a computer screen after a series of the tapping training than in the involuntary condition. This indicates that cognitive processes of choosing a choice through one’s mind alone leads to faster learning in such perceptual decision making task than inducing an action through perceiving a highlighted cue that guides which key to choose. Based on the initial finding, our second study reduced a number of free choices from fully selectable four options to two freely available and two inhibiting options in the voluntary condition. In this new study setting, participants’ reaction time became slower at the beginning of training, but the voluntary condition still showed larger RT improvement across repetitions of training than in the involuntary condition. In addition to that, we set one of the two freely selectable choices following a same sequential pattern as the fixed sequence pattern in the involuntary condition. This design increased a likelihood of choosing the free choice more frequently and speeded up participants’ reaction time faster than the other free choice. This suggests that trained or familiar sequential pattern leads to a preferred or biased choice within such intentional decision making task, and participants’ reaction speed might be boosted by skipping the additional decision making processes of inhibiting an action to certain choices and directly moving into the decision processes of choosing a choice underlying the trained sequential pattern in the involuntary condition. The finding in the second study raises a new question: how important are the trained sequential pattern (i.e. trained sequential pattern in the involuntary or fully fixed sequential condition) and participants’ intentional choices (i.e. choices among free choices in the voluntary condition) in this dynamic decision making task? Specifically, which factor produces a stronger effect on the dynamic decision making task? Do those effects interact each other (e.g. effects are additive or competitive)? Do those effects link to and predict subsequent decision making patterns? If so, which factor leads to a better link and prediction? Is the link/prediction better when combined, or separated? Here, our third study aims to examine these questions, by assigning the two free choice - two inhibiting choice setting to all trials. In this new study design, the fully fixed sequential condition (i.e. previously classified as an involuntary condition) follows a fixed sequential pattern with a pair of two freely selectable choices so that participants will learn a fixed sequential pattern of pairs, but they also need to make a further intentional decision between the two choices. We will test if the trained pattern of pairs of two free choices or participants’ intentional choices between the pair lead to a better performance (e.g. faster RTs or accuracy) individually or their effects are interactive. Also, we will examine if these effects link to and predict subsequent decision making patterns.