Auditory context-dependent distraction by unexpected visual stimuli

Abstract

Research findings indicate that when a task-irrelevant stimulus feature deviates from an otherwise predictable pattern, participants performing a categorization task exhibit slower responses (deviance distraction). This deviance distraction effect reflects the violation of the sensory predictions generated by the cognitive system. In this study, we sought to examine for the first time whether these predictions can be incidentally modulated by the auditory environment. Participants categorized the duration (short vs long) of a colored shape (red square or blue circle) while instructed to disregard the stimulus' visual features and the sound played in the background (two distinct chords played by different instruments). While the two visual stimuli shapes were equiprobable across the task, one was highly likely (p=.882) and the other rare (p=.118) in one auditory context and vice versa in the other context. Our results showed that participants were significantly slower in the duration judgement task whenever the stimulus was unexpected within a given auditory context (context-dependent distraction), and that the reset of their sensory predictions was completed upon the trial following a change of context. We conclude that object features and environmental context are processed in relation to each other and that sensory predictions are produced in relation to the environmental context, evidencing the first demonstration of auditory context-dependent modulation of attention.

Keywords: Attention; Distraction; Environmental context; Oddball.

Fig. 1

Schematic illustration of the task. The bottom part of the figure illustrates successive runs of 3 to 5 trials sharing the same auditory context. The upper part illustrates the timeline of the specific individual trials signaled by black arrows (see the Method section for a detailed description of the design and procedure). Participants were required to discriminate between the shorter and longer durations by pressing corresponding response keys (counterbalanced across participants) and ignoring the visual features of the target stimulus. They were also asked to ignore the auditory context (A or B). Runs of 3 to 5 trials of the same auditory context were presented in sequence (note that to avoid visual overcrowding of this figure, we display 3 pictures per run). (Color figure online)

Fig. 2

(A) Mean response times as a function of the auditory environment (Strings, WarmPad) and the visual stimulus (A, B). In the Strings context, stimuli A and B were presented with probabilities of.118 and .882, respectively. These probabilities were reversed in the WarmPad context. (B) Mean response times for context-specific standard and deviant stimuli as a function of the position of the trial within a run of trials sharing the same context. Error bars represent one standard error of the mean