5.1 Resting-State Evidence for a Supramodal Prefrontal Architecture
Our central objective of this study was to examine whether the prefrontal hubs known to govern top-down attentional control in the visual domain extend their functional architecture into the auditory system. Our resting-state connectivity patterns are consistent with this hypothesis. Applying single-seed partial correlation analyses, we observe a functional dissociation between the FEF and the IFJa: the FEF shows preferential resting-state coupling with auditory-spatial and motion-sensitive cortical regions, whereas the IFJa couples selectively with temporal-semantic and language-related areas. This pattern supports our primary hypothesis (Section 1.3) that the auditory where-stream connects preferentially to FEF and the what-stream to IFJa. The observed dissociation is analogous to the spatial–non-spatial segregation previously established for these regions in the visual domain by Bedini & Baldauf (2021), who characterized the FEF as a spatial attention hub embedded in the dorsal attention system and the IFJa as anchoring for non-spatial, object-level processing. That our auditory connectivity profiles mirror this division suggests a prefrontal organization that may not be modality-specific but supramodal in its logic.
5.1.1 The FEF as an Auditory-Spatial Controller
The partial correlation profile of the FEF is consistent with its established role as a spatial attention controller, now extended to the auditory domain. Rather than projecting to the superior parietal lobule regions typically associated with visuospatial attention (7PC, 7Am, 7AL) — connections that vanish when shared ROI variance is partialled out — the FEF couples selectively with the inferior parietal lobule, specifically PF (left ; right ) and PFcm, as well as with the motion-sensitive area MST (bilateral, ) and multimodal convergence zones including STV and TPOJ1. Crucially, the FEF shows no significant effect-size connectivity with any temporal regions of the ventral “what” stream (e.g., TE1a, TA2, TGd), suggesting a topographically clean dissociation.
This pattern aligns with the functional logic described by Rauschecker & Scott (2009), who demonstrated that the posterior-dorsal auditory stream links posterior superior temporal cortex and parietal areas for spatial processing, with activation in regions adjacent to MT/MST specifically for auditory motion. Our resting-state data extend this framework to the prefrontal level by showing that the FEF — a node absent from classical auditory stream models — shows functional coupling with these motion-tracking areas, suggesting it may exert top-down control over auditory spatial orienting via the inferior parietal lobule. This stands in contrast to the superior parietal lobule route that characterizes visuospatial FEF connectivity Bedini & Baldauf (2021), a shift that may reflect the inherently multimodal nature of auditory spatial processing: auditory spatial localization operates within a distributed, multisensory reference frame rather than a retinotopic, egocentric one Rauschecker & Scott (2009).
5.1.2 The IFJa as a Semantic-Auditory Controller
The partial correlation profile of the IFJa reveals a complementary picture. After controlling for shared ROI variance, IFJa couples robustly with superior temporal sulcus regions central to auditory object identity processing (STSdp: left , right ; STSda: left , right ), with the Broca’s area subregions BA44 (bilateral ) and BA45 (bilateral ), and with early auditory association areas A4 and A5, while showing no substantial coupling with parietal spatial areas. In direct contrast to the FEF, the IFJa’s partial connectivity profile is predominantly language- and identity-oriented, and spatially disjoint from the parietal networks that define the dorsal where-stream.
This selectivity is consistent with the IFJa’s network membership and functional characterization. Per the resting-state fMRI parcellation of Ji et al. (2019, as reviewed in Bedini & Baldauf (2021)), the IFJa belongs to the language network, while the FEF is assigned to the cingulo-opercular network in that same partition — a classification our data directly support for the IFJa: its strongest auditory couplings are with the STS and Broca’s area. The concurrent coupling with early auditory association areas A4 and A5 suggests that IFJa may exert top-down modulation at an early stage of auditory feature extraction, biasing the processing of spectrotemporal features relevant to auditory object identity. The coupling with Broca’s area suggests coordination with the language working memory system, consistent with the IFJa’s proposed role in feature- and object-based attention encoding Bedini & Baldauf (2021). This pattern further resonates with the broader finding that top-down prefrontal control over auditory cortex is implemented via anticipatory alpha oscillations in object-based attention tasks De Vries & Baldauf (2021).
5.1.3 A Supramodal Prefrontal Architecture
Taken together, the complementary dissociation between FEF and IFJa connectivity profiles is consistent with a supramodal organization of prefrontal attentional control. The connectivity patterns suggest that the brain may not deploy dedicated, modality-specific attention controllers for each sensory stream; instead, the same two prefrontal hubs — FEF for spatial and IFJa for non-spatial control — appear to couple preferentially with corresponding auditory cortical regions in ways that parallel their known functional roles in the visual domain.
This is compatible with the view that the lateral prefrontal cortex organizes top-down attention according to a spatial/non-spatial axis that cuts across sensory modalities. Such a supramodal architecture would be computationally efficient — allowing a single prefrontal control system to flexibly coordinate attention across vision, audition, and potentially other modalities — and would be consistent with the evolutionary argument that visuospatial attention systems, which are phylogenetically older, are preserved in primates and subsequently recruited for the control of other sensory domains Bedini & Baldauf (2021). Whether these resting-state coupling patterns reflect genuine top-down control signals remains to be established with directed connectivity methods.
Notes & Scrapbook
Hier Dinge abladen, die noch keinen Platz im Text haben, damit der Schreibfluss nicht stoppt.
Alter Entwurf (verschoben am 06.04.2026)
In this project, we examined the prefrontal top-down architecture of the auditory dual-stream model using resting-state functional connectivity. By contrasting the connectivity profiles of the FEF and IFJa against a target set of ROIs, we provide additional evidence for a functional double dissociation that mirrors the We supported the overall dual-stream hypothesis for the auditory cortex with partial correlation analyses especially for the seed regions IFJa and FEF. Those regions turned out to have a significant effect on both what and where stream in the auditory domain.
Our findings align with the comprehensive characterization by Bedini & Baldauf (2021) and Soyuhos (2023) who established that both FEF and IFJa are crucial for top-down attention in two partially segregated pathways. While Bedini and Baldauf focussed on visual attention networks, our auditory connectivity profiles reveal that these same prefrontal hubs maintain direct functional coupling with auditory cortical regions, support a hypothesis forming around supramodal architecture.
5.1.1 The FEF as an Auditory-Spatial Controller (alt)
The results shown in 4.2. reflects findings from De Vries and Baldauf. We observed partial correlations between FEF, motion-sensitive areas, inferior parietal areas and auditory association areas
excludedBelt/Core regions
We intentionally excluded core acoustic regions (A1, MBelt, LBelt, 52) from the ROI set to prevent multicollinearity in the partial correlation matrix and to focus the analysis strictly on the hierarchical level where the divergence into ‘what’ and ‘where’ streams occurs (PBelt, A4, A5).
→ Diesen Punkt ggf. in 3.2 (Selection of ROIs) oder 5.4 (Anatomical Ambiguities) unterbringen.
Fact-Check-Notizen (2026-04-06)
- IFJa gehört zum language network (Ji et al. 2019, via Bedini & Baldauf 2021), NICHT zum FPN (das ist IFJp) — im neuen Text korrigiert.
- Alpha-Oszillationen aus IFJa: De Vries & Baldauf (2021) nennen IFJa nicht explizit als Source — daher nur allgemeiner Verweis auf prefrontale top-down Kontrolle via Alpha im neuen Text.
- SLF2-Claim (Bedini n.d.) ist nur für visuelles FEF→LIPd belegt, nicht für auditorisches FEF→IPL — aus dem Haupttext entfernt.
- FEF-Netzwerk: DAN (Bedini & Baldauf 2021, ^2c3952) UND cingulo-opercular (Ji et al. 2019); da MMP1-Atlas verwendet wird, wäre cingulo-opercular strenger korrekt — im Text bewusst offen gelassen.