Selective attention allows listeners to focus on relevant sounds within complex acoustic scenes, yet how it reshapes neural population codes remains unclear. Using fMRI and representational geometry analyses, we show that attention reorganizes auditory scene representations by attracting attended sound manifolds toward their object-alone templates. This effect emerged when attended and distractor sounds belonged to different categories. These findings demonstrate that selective attention reshapes neural population geometry in a context- and category-dependent manner, with geometric changes closely reflecting behavior.
Using fMRI and surface-based searchlight multivariate pattern analysis (pairwise SVM decoding), we show that selective attention enhances within-category differentiation of natural auditory objects in the human auditory cortex. Speech identity differentiation engaged higher-level auditory regions, while animal and instrument differentiation relied on lower-level acoustic feature processing. These results indicate that attention engages category-specific neural mechanisms to sharpen auditory object representations.
In everyday life, many sounds—voices, traffic, birds, instruments—overlap at the same time. Our brains can separate this noisy mix into distinct “objects” (a voice, an animal call, a musical instrument); attention is thought to aid this, but the underlying brain mechanisms have been unclear. We investigated this with fMRI while people listened to diverse range of natural scenes with three overlapping sounds. When listeners focused on a sound in a scene, their brain patterns shifted to more closely match the patterns produced by that sound than the other sounds of the scene, producing contrast between the attended sound and the rest of the scene.
Selective attention enables separation of overlapping speech in noisy environments. This study using EEG-fMRI fusion with continuous audiovisual cocktail party speech, by Patrik Wikman, Viljami Salmela and co-authors, shows that attention acts by routing neural processing through recurrent feedback-feedforward loops between nodes of the speech network.
Attention requires both selecting the relevant stimuli and maintaining the focus on it while filtering out distracting stimuli. In this study, we used hyperalignment and Procrustes-based clustering to map and compare the profiles that different brain areas elicit during selective attention of speech. We found parietal areas to have a distinct profile related to the directing of attention to the relevant stimuli.
Here, we examined the neural effects associated with performing arithmetic cognitive tasks in conditions with various auditory distractions. Our results indicate that task-irrelevant meaningful speech, a naturalistic and ecologically relevant distractor type, is associated with distinct neural effects compared to meaningless sounds. Our results provide insights on how our brains achieve cognitive control in realistic noisy conditions.
Studies on how moderate to mild early life stress affect functional brain development have yielded contradictory results. In this study, we utilized an inventive combination of fMRI, intersubject representational similarity analysis, statistical shape analysis, and a 20-year spanning longitudinal dataset to leverage how similarity in early life stress maps onto representational similarity in different nodes of cortical regions.
In this study, we investigated the relationship between dyslexia-related genetic variants and brain activity during language tasks in fluent readers. We found significant associations between variations in the DNAAF4, DCDC2, and NRSN1 genes and activation in key language areas, the inferior frontal gyrus, intraparietal sulcus and middle temporal gyrus. These findings contribute to our understanding of the neurobiological underpinnings of language processing.
Selective attention to speech is often studied using highly simplified and artificial settings. However, this fMRI study by Artturi Ylinen and colleagues demonstrates that properties of the task performed with regard to the attended speech (e.g., focus on meaning or focus on phonological content) affect neural activations in widespread regions of the brain. This indicates that naturalistic tasks are likely required, if one wishes to understand neural processing as it occurs in everyday life.
Individuals with attention deficit hyperactivity disorder (ADHD) often have working memory difficulties. Our study revealed that brain activation was less synchronised across several cortical and subcortical areas in those with ADHD during working memory tasks. Functional connectivity was also associated with attention skills regardless of the participants' diagnostic status.