4.4 Resolving Ambiguities

Outline 4.4 Resolving Ambiguities

• Core Argument dieses Kapitels definieren

To-Do’s

• check wording
• check sources

4.4 Resolving Ambiguities

During the analysis and literature review, areas A4, A5, PBelt, PSL and STV demonstrated an ambiguous connectivity pattern. This chapter will dissolve ambiguities found in the literature and results of the RSFC analysis.

A4 and A5. We placed areas A4 in the ‘where’-stream and A5 in the ‘what’-stream (Section 3.2.5), given the conflicting evidence in the literature (Rolls et al. (2023) - Cerebral Cortex, Glasser et al. (2016) - Nature). Partial correlation analysis resolves this conflict empirically by showing robust and bilateral connectivity of both areas A4 and A5 with IFJa (A4: left $z=8.25$, $\overline{r}=0.01$; right $z=8.76$, $\overline{r}=0.01$; A5: left $z=11.63$, $\overline{r}=0.01$; right $z=6.96$, $\overline{r}=0.01$), while showing no meaningful coupling with FEF. The only exception is a weak, left-lateralized A4-FEF connection ($z=4.06$, $\overline{r}=0.01$) in the single seed analysis with FEF. This pattern positions both A4 and A5 as ‘what’-stream gateways under potential IFJa top-down control.

PBelt. PBelt was grouped alongside A4 as a potential dorsal gateway in the pre-analysis classification (Section 3.2.5). In the single-seed IFJa analysis, PBelt shows only weak coupling (right: $z=4.79$, $\overline{r}=0.01$). Partial correlation reduces this further to near-equivalent values for both prefrontal seeds (IFJa: right $z=3.05$, $\overline{r}=0.01$; FEF: left $z=2.92$, $\overline{r}=0.00$). The absence of a dominant prefrontal partner, combined with near-zero mean connectivity, indicates that PBelt’s functional coupling is insufficient for reliable stream assignment.

PSL. Rolls et al. (2023) - Cerebral Cortex classify PSL within the ventral semantic network, while Dureux (2024) reports complete acoustic unresponsiveness across all stimulus categories. Our analysis reveals a strongly asymmetric connectivity profile. The IFJa single-seed analysis shows bilateral PSL-IFJa coupling (left $z=9.65$, right $z=13.06$, $\overline{r}=0.01$). The FEF single-seed analysis reveals a right-lateralized PSL-FEF connection ($z=21.37$, $\overline{r}=0.03$), entirely absent in the left hemisphere. No other ROI in this analysis displays this pattern of bilateral ‘what’-stream coupling alongside exclusive right-lateralized ‘where’-stream coupling.

STV. STV was classified as part of the ventral language network by Rolls et al. (2023) - Cerebral Cortex and couples directly with both prefrontal seeds in separate single-seed analyses. In full correlation, STV couples with FEF stronger in both hempsheres, but when it comes to partial correlation, IFJa wins in both. In the IFJa analysis, STV couples bilaterally with IFJa (left $z=18.90$, $\overline{r}=0.02$; right $z=18.55$, $\overline{r}=0.02$), after partialling out FEF. In the FEF analysis, STV couples bilaterally with FEF (left $z=19.50$, $\overline{r}=0.02$; right $z=17.82$, $\overline{r}=0.02$), after partialling out IFJa. In both cases, the coupling is direct, not mediated through indirect connections.

---