Distraction driven by reward history: Attentional capture and sequential effects

Abstract

Values learned through previous experiences of reward can later modulate attentional capture if associated with a distractor in singleton search tasks (value-driven attentional capture; VDAC). Moreover, it has been shown that re-encountering distractor features can facilitate performance or reduce attentional capture (sequential effects). However, little is known about how sequential effects and attentional capture are jointly modulated by learned distractor value. Here, we examined the role of learned reward in sequential modulation of attentional capture. In two experiments we used a VDAC paradigm, varying the type of reward (monetary vs. sustainability-related). After associating letter colors with a high or low reward, or none at all, in a flanker task (learning phase), in a subsequent singleton task (test phase) we manipulated the effects of distractor value of the present and of the previous trial on attentional capture. In both experiments repetition of the same distractor value from trial N-1 to trial N was associated with faster responses, and reward value did not modulate this facilitation. In addition, attentional capture by rewarded, compared with unrewarded, distractors was observed when the preceding trial was unrewarded. Value-signaling distractors, if re-encountered, reduced attentional capture in the current trial, and this happened even for rewarded distractors of different values (e.g., high value followed by low value, and vice versa). These results suggest that, for different forms of incentives, repetition of previously rewarded distractors and attentional capture by the current reward interact in modulating the processing of learned values.

Keywords: Attentional capture; Attentional suppression; Binding-retrieval; Distractor repetition; Intertrial priming; Reward learning.

Fig. 1

Trial structure of the value-driven attentional capture (VDAC) paradigm. (a) In the learning phase (flanker task), participants made a two-alternative forced-choice response on the central letter (target) identity according to the response mapping explained at the beginning of each block. Only the feedback screens differed between experiments: The feedback display in Experiment 1 indicated the monetary gain for the trial and the cumulative total amount earned, while the feedback display in Experiment 2 indicated the point gain (drawn as bins of plastic collected from the sea) for each accurate trial, which then translated into a donation to an association fighting pollution. (b) In the test phase (singleton search task), participants were asked whether the different number (target) in the search display was an odd or an even number. One of the five remaining numbers (distractors) was colored (red, green, or blue), and only corrective feedback was provided

Fig. 2

Response times in Experiment 1. Error bars represent within-participants standard error of the mean (O’Brien & Cousineau, 2014)

Fig. 3

Response times in Experiment 2. Error bars represent within-participants standard error of the mean (O’Brien & Cousineau, 2014)

Fig. 4

The effects of feature change (from a valued preceding distractor to different distractor conditions in the actual trial). The distractor condition (high-value, low-value, unrewarded) in the N-1 trial in this analysis is the same across the three conditions (value-same, value-different, value-to-control)

Fig. 5

The effects of attentional capture (from a preceding trial in different conditions to a valued distractor). The distractor value (high-value, low-value, unrewarded) in the N trial in this analysis is the same across the three conditions (value-same, value-different, control-to-value)