Value Associations Modulate Visual Attention and Response Selection

Abstract

Every day, we are confronted with a vast amount of information that all competes for our attention. Some of this information might be associated with rewards (e.g., gambling) or losses (e.g., insurances). To what extent such information, even if irrelevant for our current task, not only attracts attention but also affects our actions is still a topic under examination. To address this issue, we applied a new experimental paradigm that combines visual search and a spatial compatibility task. Although colored stimuli did not modulate the spatial compatibility effect more than gray stimuli, we found clear evidence that reward and loss associations attenuated this effect, presumably by affecting attention and response selection. Moreover, there are hints that differences in these associations are also reflected in a modulation of the spatial compatibility effect. We discuss theoretical implications of our results with respect to the influences of color, reward, and loss association on selective attention and response selection.

Keywords: Simon task; associative learning; attention; monetary loss; monetary reward; response selection.

FIGURE 1

Illustration of the congruent (A) and incongruent (B) condition in the Simon search task, as well as an example trial sequence in Experiment 1 (C).

FIGURE 2

Experiment 1. The left panel displays the RTs and error rates for the different congruency and distractor conditions. The right panel displays differences in the Simon congruency effects for respective pairs of distractor conditions. Displayed are only data from trials in which the deadline was met. Error bars correspond to the within-subject confidence intervals (Morey, 2008). *Significant on p < 0.05, **significant on p < 0.01, ***significant on p < 0.001, n.s. = not significant.

FIGURE 3

RTs averaged across congruent and incongruent conditions for successive blocks in Experiment 1. The graph in the upper panel shows the complete data, irrespective of timeout errors. The graph in the lower panel displays only data in which the deadline was met. For clarity, error bars are not included.

FIGURE 4

RTs averaged across congruent and incongruent conditions for successive blocks in Experiment 2. The graph in the upper panel shows the complete data, irrespective of timeout errors. The graph in the lower panel displays only data, where the deadline was met. For clarity, error bars are not included.

FIGURE 5

Experiment 2. The left panel displays the RTs and error rates for the different congruency and distractor conditions. The right panel displays differences in the Simon congruency effects for respective pairs of distractor conditions. Displayed are only data from trials in which the deadline was met. Error bars correspond to the within-subject confidence intervals (Morey, 2008). *Significant on p < 0.05, **significant on p < 0.01, ***significant on p < 0.001, n.s. = not significant.