Resilience of adolescents and young adults towards climate-related stressors - a systematic review.

Human listeners can selectively attend to a single speaker amid a mixture of competing voices—a phenomenon known as the cocktail party effect. This ability can be guided by multiple cues, including acoustic, temporal, semantic, and spatial information. Although the cocktail party effect has long been a central topic in auditory neuroscience, the neural mechanisms specifically shaped by spatial cues have received far less attention and remain insufficiently characterized. Accumulating evidence indicates that the cocktail party effect arises from a hierarchical neural processing architecture in which unattended speech is represented primarily within auditory cortex, whereas the attended stream is selectively propagated to higher-order cortical regions for further cognitive processing. The most direct support for such mechanisms comes from electrocorticography (ECoG). ECoG recordings reveal a clear hierarchical pattern: primary auditory cortex encodes both the attended and unattended speech streams, whereas non-primary auditory regions predominantly represent only the attended stream, as if the unattended input were scarcely present in the neural signal. Consistent with this organization, primary auditory cortex maintains distinct representations of individual speakers irrespective of attentional focus, whereas non-primary regions collapse these distinctions and selectively track the attended stream. In further support of this hierarchy, speech tracking—captured in low-frequency phase and high-frequency amplitude neural dynamics—persists for both streams within auditory cortex but aligns exclusively with the attended stream in higher cortical areas. This hierarchical architecture is further reinforced by converging evidence across single-unit recordings and noninvasive modalities, including magnetoencephalography (MEG), functional magnetic resonance imaging (fMRI), and functional near-infrared spectroscopy (fNIRS). However, because these studies have largely employed non-spatial stimulus designs, it remains unclear how spatial cues shape or modulate this hierarchical neural mechanism. Although substantial work has underscored the pivotal role of spatial cues in auditory scene analysis, research examining spatial guided cocktail-party listening remains comparatively underexplored. Neuroimaging and lesion studies have demonstrated that regions including the planum temporale, superior temporal gyrus (STG), inferior parietal lobule, and precuneus contribute critically to the localization of target sounds among competing distractors in cluttered environments. Electrophysiological approaches further suggest that dynamic cortical microstates track the temporal evolution of spatially selective attention. Moreover, frequency-tagging studies have shown that spatial attention enhances steady-state responses to attended streams in contralateral auditory cortex and recruits control-related frontal regions, such as the left precentral sulcus, in a manner specific to attended inputs. Importantly, several cocktail-party studies have incorporated spatialized stimulation or more naturalistic settings, including scenarios involving real communication partners. For example, spatial cues have been embedded within cocktail-party paradigms in combined electroencephalography (EEG)–fMRI investigations, examined in the context of sound localization deficits in complex auditory scenes, and naturalistic real world like communication settings. Nevertheless, in most of these approaches, spatial separation primarily functions to disambiguate target and competing streams. The direction of attention itself is seldom treated as a parametrically controlled variable, systematically manipulated to compare neural mechanisms across different attended spatial locations. Consequently, these studies do not clarify whether the spatially guided cocktail party effect relies on the same hierarchical neural processing architecture observed during non-spatial selection, nor do they delineate how spatial cues reshape the neural flow of attended versus unattended speech information in multi-talker environments.