Figure 1.
Behavioral time courses showing the effects of a valid or invalid attentional cue on accuracy and the response time (RT) of tactile decisions.
Lines represent the average accuracy (A) and the average RT (B) for validly (solid) and invalidly (dashed) cued trials within each temporal window.
Figure 2.
Results from fitting the DDM to the data of van Ede et al.
A. The drift-diffusion model (DDM) assumes an accumulation process, until the evidence reaches a decision threshold. An increase of drift rate causes faster and more correct choices (red arrow). Non-decision time reflects the time other then the decision time (e.g. process sensory information and execute a motor response). B. Change in reaction time (RT) explained by drift rate (v) and non-decision time (Ter) for validly (blue) and invalidly (grey) cued trials compared to the baseline (uncued) RT. C. Time course for drift rate (v), showing the percentage increase from the baseline drift rate for validly (red) and invalidly (grey) cued trials. D. Time course for the percentage change in non-decision time (Ter) for validly (green) and invalidly (grey) cued trials compared to the baseline non-decision time.
Figure 3.
Quantile probability plots showing the best fitting model (see methods) for each subject.
Each graph represents the proportion correct choices and reaction time (RT) distributions for each condition (data points) and the DDM quantile probability functions describing them (lines). RT distributions are represented by five quantiles (colors), plotted along the y-axis for each condition. Conditions (neutral, validly and invalidly cued trials) are split into correct and incorrect responses and divided over the x-axis, representing response probability. Lines connecting the quantiles between conditions represent changes in RT distributions across conditions, for incorrect and correct responses.
Figure 4.
Group quantile probability plots for the validly and invalidly cued trials.
Data points represent group mean for correct (right) and incorrect (left) choices in each time course window. Lines represent the group average of the DDM predicitions.
Figure 5.
Effects of the validity of an attentional cue on non-decision time and drift rate result in changes in accuracy and RT.
For uncued trials, the process of accumulating tactile evidence (red; drift rate) is preceded by the encoding of the target and an attentional focus to the stimulated hand (green; non-decision time). For the cued trials, the encoding of the auditory cue is followed by a focused attention to the hand associated with the cue (white boxes). A target may benefit from this process when the cue is valid and the cue stimulus interval (CTI) is long enough, as the attention is already drawn to the stimulated hand. This results in shorter non-decision times and improved processing of sensory information, which in turn will lead to shorter accumulation times due to an increase in drift rate. For short CTIs, this advantage is minimal, as the processing of the auditory cue results in a delay in processing the target, resulting in cognitive slack time (empty green boxes). In contrast, invalidly cued trials lack the advantage of focused attention to the relevant hand. The onset of the target results in a re-focus of the attention to the other hand. As such, drift rate and non-decision time will be similar to those observed in the uncued trials, across different CTIs. These processes lead to divergence in decision times for validly cued and invalidly cued trials, as indicated on the timeline with red (valid) and blue (invalid) lines.