Object-based attention in complex, naturalistic auditory streams

Marinato, Giorgio; Baldauf, Daniel · 2019 · Crossref

DOI: 10.1038/s41598-019-39166-6

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates whether object-based attention, a mechanism well-established in visual perception, operates similarly in the auditory domain within complex, naturalistic scenes. While auditory research has historically focused on sound segregation and spatial or feature-based attention, the interaction between top-down selective attention and the formation of auditory objects remains poorly understood. The authors aim to determine if attention can facilitate the processing of segregated auditory objects—defined as temporally confined, bound entities like speech streams or environmental soundscapes—independent of spatial or low-level feature cues. To address this, the researchers developed a novel auditory repetition detection task. Participants listened to two spatially and temporally overlapping auditory streams presented from a single central speaker, thereby ruling out spatial attention confounds. Experiment 1 paired a foreign language speech signal with an environmental soundscape (e.g., a railway station), while Experiment 2 paired two distinct speech signals. In each trial, a 750 ms segment of one stream was repeated immediately after its original occurrence. Participants were cued to attend to one stream, the other, or neither (neutral), and were required to detect the repetition as quickly and accurately as possible. This design ensured that successful performance required processing the acoustic stream to the level of cognitive object recognition rather than relying on simple feature detection. The results demonstrated a robust cue-validity effect in both experiments. Participants were significantly faster and more accurate in detecting repetitions when the cue validly indicated the stream containing the target compared to invalid or neutral conditions. In Experiment 1, accuracy for validly cued speech targets reached 85.6%, while invalidly cued targets dropped to 57.3%. Reaction times were approximately 100 ms faster for valid cues compared to neutral ones. Signal detection theory analyses confirmed that these performance improvements were driven by increased sensitivity ($d'$) rather than changes in response bias. The attentional modulation effect was consistent across both speech and environmental streams, indicating that top-down attention effectively weighted incoming acoustic information at the level of segregated auditory objects. These findings provide empirical evidence for object-based attention in the auditory domain, paralleling mechanisms observed in vision. By demonstrating that attention can facilitate the processing of high-level auditory objects independent of spatial location, the study bridges a gap in understanding auditory scene analysis. This suggests that selective attention operates on grouped auditory entities, facilitating relevant information while suppressing irrelevant streams. The results support the biased competition model, implying that attentional mechanisms play a central role in parsing complex auditory environments, such as the "cocktail party" scenario, by biasing processing toward behaviorally relevant auditory objects.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• attention selective divided
• auditory