The effect of distraction on change detection in crowded acoustic scenes

Abstract

In this series of behavioural experiments we investigated the effect of distraction on the maintenance of acoustic scene information in short-term memory. Stimuli are artificial acoustic ‘scenes’ composed of several (up to twelve) concurrent tone-pip streams (‘sources’). A gap (1000 ms) is inserted partway through the ‘scene’; Changes in the form of an appearance of a new source or disappearance of an existing source, occur after the gap in 50% of the trials. Listeners were instructed to monitor the unfolding ‘soundscapes’ for these events. Distraction was measured by presenting distractor stimuli during the gap. Experiments 1 and 2 used a dual task design where listeners were required to perform a task with varying attentional demands (‘High Demand’ vs. ‘Low Demand’) on brief auditory (Experiment 1a) or visual (Experiment 1b) signals presented during the gap. Experiments 2 and 3 required participants to ignore distractor sounds and focus on the change detection task. Our results demonstrate that the maintenance of scene information in short-term memory is influenced by the availability of attentional and/or processing resources during the gap, and that this dependence appears to be modality specific. We also show that these processes are susceptible to bottom up driven distraction even in situations when the distractors are not novel, but occur on each trial. Change detection performance is systematically linked with the, independently determined, perceptual salience of the distractor sound. The findings also demonstrate that the present task may be a useful objective means for determining relative perceptual salience.

The effect of healthy aging on change detection and sensitivity to predictable structure in crowded acoustic scenes

“…Indeed, sound is often what alerts us to important changes around us -in many cases we hear a change before we see it. Accumulating evidence demonstrates that (young) listeners are sensitive to abrupt changes, such as the appearance or disappearance of a source, even in heavily populated acoustic scenes ( Cervantes Constantino et al, 2012 ;Pavani and Turatto, 2008 ;Petsas et al, 2016 ). Brain responses, recorded from naïve distracted listeners reveal that these events are often detected in the absence of directed attention Chait, 2016a , 2016b ), consistent with an, at least partially, automatic change detection process.…”

Multivoxel codes for representing and integrating acoustic features in human cortex

“…To estimate the perceived saliency of the sounds, two participants from the main fMRI experiment and four new participants (two female; mean age ​= ​29 years, SD ​= ​4) completed a short behavioral session similar in procedure to Petsas et al. (2016) .…”

Multivoxel codes for representing and integrating acoustic features in human cortex

“…To estimate the perceived saliency of the sounds, two participants from 170 the main fMRI experiment and four new participants (two female; mean age = 171 29 years, SD = 4) completed a short behavioral session similar in procedure to 172Petsas et al (2016). These participants listened to all pairwise combinations of 173 the nine sounds (eight pairs for each of the nine sounds; separated by 200 ms 174 of silence) and were asked to judge on each trial which of the two sounds was 175 more salient.…”