Retinotopic coding organizes the interaction between internally and externally oriented brain networks
- Beckman Institute, University of Illinois Urbana-Champaign, United States
- Department of Psychology, University of Illinois Urbana-Champaign, United States
- Department of Psychological and Brain Sciences, Dartmouth College, United States
eLife Assessment
This study addresses a fundamental question about how large-scale brain networks interact, and specifically how the default mode network exchanges information with sensory cortex. The analyses provide solid evidence for the claims made in the paper. The findings should be of broad interest to researchers studying brain network organization and dynamics.
https://doi.org/10.7554/eLife.110234.3.sa0Significance of the findings:
Fundamental: Findings that substantially advance our understanding of major research questions
- Landmark
- Fundamental
- Important
- Valuable
- Useful
Strength of evidence:
Solid: Methods, data and analyses broadly support the claims with only minor weaknesses
- Exceptional
- Compelling
- Convincing
- Solid
- Incomplete
- Inadequate
During the peer-review process the editor and reviewers write an eLife Assessment that summarises the significance of the findings reported in the article (on a scale ranging from landmark to useful) and the strength of the evidence (on a scale ranging from exceptional to inadequate). Learn more about eLife Assessments
Abstract
The human brain seamlessly integrates internally generated thoughts with incoming sensory information, yet the large-scale networks that support these functions – the internal default network (DN) and external dorsal attention network (dATN) – are traditionally viewed as functionally independent. This raises a crucial question: how does the brain integrate information across these seemingly noninteractive systems? Here, using densely sampled 7T fMRI, individualized resting-state parcellations, and voxel-wise population-receptive-field mapping, we show that these internal/external networks are more interlocked than previously thought. Spontaneous DN and dATN activity during rest is uncorrelated at the network level. However, voxel-scale functional coupling across networks is shaped by the latent visual field preferences of individual voxels in each network, as measured during independent retinotopic mapping. Voxels that share visual field preferences exhibit stronger spontaneous coupling than those with divergent preferences. These retinotopically specific interactions are bivalent: DN voxels with negative (suppressive) visual response amplitudes are anticorrelated with matched (positive) dATN voxels, while those DN voxels with positive response amplitudes are positively correlated. Thus, distinct subpopulations of visually tuned DN voxels participate in spatially specific interactions with the dATN. Further, retinotopic coding is intrinsic to the DN, persisting even during periods when the DN signal is elevated. These findings reveal a latent, voxel-level architecture of retinotopically grounded interactions between the DN and dATN. Taken together, our results suggest that retinotopic coding underpins the dynamic coordination of perception and thought in the human brain.
Introduction
A fundamental goal of neuroscience is to understand how the coordinated activity of distributed brain networks gives rise to cognition (Yeo et al., 2011; Margulies et al., 2016; Braga and Buckner, 2017; Gordon et al., 2017; Ranganath and Ritchey, 2012; Andrews-Hanna et al., 2010; Barnett et al., 2021; Du et al., 2024; Laumann et al., 2015). Central to this aim is understanding principles that govern interactions among networks with distinct computational roles. In particular, it is unclear how networks involved in externally oriented attention (e.g., processing sensory input) interact with those that support internally oriented attention (e.g., introspection and memory) (Chun et al., 2011; Dixon et al., 2017; Fox et al., 2005; Raichle, 2015; Fox et al., 2006). Understanding this interplay is crucial because the dynamic exchange between internal and external representations underlies how we perceive, remember, and navigate our world – yet the ‘common language’ that enables communication between these systems remains unclear.
Historically, the default network (DN) and dorsal attention network (dATN) – prototypical networks for internal and external attention, respectively – have been viewed as functionally competitive (Chun et al., 2011; Dixon et al., 2017; Fox et al., 2005; Fox et al., 2006; Raichle et al., 2001). Classic neuroimaging studies show that externally oriented tasks activate the dATN and deactivate the DN (Chun et al., 2011; Raichle, 2015; Fox et al., 2006; Fedorenko et al., 2013; Vossel et al., 2014; Shulman et al., 1997), while internally oriented tasks produce the opposite pattern (Andrews-Hanna et al., 2010; Silson et al., 2019; DiNicola et al., 2020; Thakral et al., 2017; Gilmore et al., 2021; Christoff et al., 2009). This dissociation is also observed at rest, where DN and dATN activity is typically independent or anti-correlated (Dixon et al., 2017; Fox et al., 2005; Greicius et al., 2003; Menon, 2023), depending on the application of global signal regression (Dixon et al., 2017; Saad et al., 2012). Such findings have led to the prevailing view that DN/dATN antagonism supports selective attention to either internal or external information, minimizing interference between perception and memory (Chun et al., 2011; Fox et al., 2005; Murphy et al., 2019; Murphy et al., 2018). Yet, this framework leaves open a fundamental question: if these networks are functionally dissociated, how does the brain integrate perceptual and mnemonic information when both are required – for example, during anticipatory saccades or memory-guided attention?
Recent studies of the voxel-level coding properties within these large-scale networks offer insight into this question. Although the DN is traditionally associated with abstract or semantic coding (Margulies et al., 2016; Murphy et al., 2018; Bellmund et al., 2018; Popham et al., 2021), emerging evidence suggests that it also exhibits a neural code typically associated with externally oriented visual processing: coding of retinotopic stimulus position (Szinte and Knapen, 2020; Knapen, 2021; Klink et al., 2021; Steel et al., 2024). Notably, retinotopic coding in the DN differs from that observed in traditional visually responsive areas, including the dATN. While visual stimulation elicits position-dependent increases in neural activity in traditional visual areas, many DN voxels often show position-dependent decreases in activity (Szinte and Knapen, 2020; Klink et al., 2021; Steel et al., 2024). This voxel-level pattern suggests that the DN’s global deactivation during externally oriented tasks may not reflect a simple disengagement from external input, but rather a structured, spatially tuned response to visual stimulation.
What purpose might the DN’s voxel-level retinotopic code serve? We and others have proposed that the visual code in the DN might play a functional role in structuring mnemonic–perceptual interactions (Szinte and Knapen, 2020; Knapen, 2021; Steel et al., 2024; Groen et al., 2022). Recent work provides critical support for this theory, revealing that scene-selective memory regions near the DN (Steel et al., 2021) exhibit voxel-level retinotopic opponency with scene-selective visual areas (Steel et al., 2024). Specifically, the inverted retinotopic code in mnemonic regions is functionally tethered to the positive code in perceptual regions: during perception, visual stimuli at specific field locations activate perceptual voxels while suppressing memory voxels tuned to the same location, a pattern that reverses during memory retrieval (Steel et al., 2024). These results indicate that visual field location can coordinate dynamic interactions between specialized perceptual and mnemonic areas at the cortical apex, raising the intriguing hypothesis that retinotopic grounding might support perceptual–mnemonic interactions more broadly across the brain. Yet, whether retinotopic coding serves as a more general organizing principle for interactions across large-scale networks remains unknown.
Here, we tested the hypothesis that global interactions across large-scale internally and externally oriented brain networks are structured at the voxel level by a shared retinotopic code (Groen et al., 2022). This framework challenges the traditional view of internally and externally oriented brain networks as globally antagonistic, proposing instead that voxel-level retinotopic tuning enables functional coupling and shared information processing between them. To evaluate this hypothesis, we leveraged a high-resolution 7T fMRI dataset (Allen et al., 2021), individualized network identification (Braga and Buckner, 2017; Gordon et al., 2017; Du et al., 2024), and voxel-wise modeling (Dumoulin and Wandell, 2008) to test three core predictions. First, voxel-wise retinotopic coding should be robustly present within individuals’ DN and dATN networks, but with divergent tuning polarities: positive retinotopic responses should dominate the dATN, while DN should contain a significant proportion of retinotopic voxels with negative response amplitudes to visual stimulation. Second, we predicted that spontaneous DN–dATN interactions during resting-state fMRI would be structured by the retinotopic properties of individual voxels. Specifically, functional coupling should be stronger between voxels that share visual field preferences across networks, revealing that retinotopic position-specific responses form a latent architecture to scaffold cross-network interactions even in the absence of overt task demands. Third, we predicted that these retinotopically grounded interactions would support both coupling driven by the DN and coupling driven by dATN, suggesting that shared spatial tuning enables dynamic, bidirectional interactions. Together, these findings would position retinotopy as a unifying framework for brain-wide information processing and for the coordination of internal and external attentional dynamics.
Results
Retinotopic coding in internally and externally oriented networks
To investigate the role of retinotopic coding in structuring interactions between internally and externally oriented brain networks, we analyzed 7T fMRI data from the Natural Scenes Dataset (Allen et al., 2021), acquired from the same densely sampled participants during both resting-state and retinotopic mapping sessions (Figure 1A, B).
Figure 1 with 4 supplements see all
Inversion of retinotopic coding between externally and internally oriented brain networks.
(A) Population receptive field (pRF) modeling with fMRI. We established visual field preferences for each voxel using pRF modeling (Dumoulin and Wandell, 2008). Voxels with positive BOLD responses to the visual stimulus are classified as positive pRFs (+pRFs), and those with negative BOLD responses are classified as negative pRFs (−pRFs). (B) Individualized resting-state network parcellation. Resting-state fMRI was collected in all participants (N = 7; 34–102 min per participant) and used to derive individualized cortical network parcellations. Parcellations were generated using the multi-session hierarchical Bayesian modeling approach with the Yeo 15 HCP atlas as a prior (Du et al., 2024; Kong et al., 2019). (C) PRF model variance explained is greater when fit to the pRF mapping task vs. resting-state fMRI data for negative and positive pRFs in the DN and dATN. Histograms show voxel-wise variance explained for all pRFs fit to rest (dark) and the pRF modeling task fMRI data (light). All distributions were significantly shifted rightward for data fit to the pRF modeling task (dATN pRFs: K–S test: D− = 0.655, p < 0.001; negative DN pRFs: D− = 0.232, p < 0.001; positive DN pRFs: D− = 0.124, p < 0.001). Dashed lines indicate the 95th percentile of R2 within the null distribution used; we chose the threshold for visually defined voxels for subsequent analyses (R2 > 0.14) to exceed these values. (D) DN and dATN pRF amplitude and center location estimates are reliable. To assess the reliability of pRF amplitude estimates, we compared voxel-wise signed amplitude by computing the Dice coefficients of binarized pRF amplitude maps (i.e., positive or negative pRFs) between pairs of pRF modeling runs. To estimate center position, we calculated the distance between pairs of pRF centers iteratively between pRF modeling runs. For both metrics, chance was estimated by performing the same computation on a random sampling of voxels drawn from both pRFs and non-pRFs (5000 iterations for each metric). In all subjects, for dATN positive and DN positive and negative pRFs, pRF amplitude and center position exceeded what would be expected by chance (z-scored Dice/distance values >2.9). Although only dATN positive pRFs were included in subsequent analyses, the dATN pRF amplitude was not reliable above chance in two participants; this suggests that negative amplitude pRFs are not stable in all cortical networks. (E) The internally oriented DN contains a larger proportion of negative pRFs than the externally oriented dATN. The proportion of negative pRFs (of total pRFs that survive thresholding in that network) in all cortical networks is shown. Inset. DN-A and DN-B contain more negative pRFs than the dATN (t(6) = 7.94, p < 0.001), and there is no difference in the proportion of negative pRFs among subnetworks. (F) DN and dATN pRF networks show contralateral bias. Coverage plots across participants for positive dATN pRFs and positive and negative DN pRFs. Bar plots show the difference in coverage between contra- vs. ipsi-lateral visual fields for each pRF type. Bar plots show the mean across participants; connected points represent individual participant data.
We first identified individualized large-scale networks for each participant using their resting-state fMRI data (Du et al., 2024; Kong et al., 2021; Figure 1B, Figure 1—figure supplement 1). Resting-state data were preprocessed using independent components analysis (ICA) with manual noise component selection (Griffanti et al., 2014; Griffanti et al., 2017), and no global signal regression was performed (Saad et al., 2012). Networks were identified using a multi-session hierarchical Bayesian modeling approach (Du et al., 2024; Kong et al., 2021; Kong et al., 2019), which leverages repeated data runs to estimate within-participant variation and derive stable parcellations for 15 cortical networks, including the Default Networks A–B and the Dorsal Attention Networks A–B (Figure 1B). Unless otherwise noted, we combined the Default Networks A–B and the Dorsal Attention Networks A–B to constitute our internal and externally oriented networks, respectively (hereafter, DN and dATN).
Next, we determined the visual field preferences of all voxels within these individualized networks using a voxel-wise population receptive field (pRF) modeling approach (Dumoulin and Wandell, 2008; Silson et al., 2015) to fit each participant’s BOLD activity during retinotopic mapping (Dumoulin and Wandell, 2008; Figure 1A, Figure 1—figure supplement 2). We fit the data using a Gaussian model (implemented in AFNI) that allows for both positive and negative signed amplitudes.
We considered any voxel with >14% variance explained by our pRF model to be exhibiting a retinotopic code, and hereafter we refer to these voxels as ‘pRFs’. We determined this threshold by defining the noise floor of our retinotopic model in the DN and dATN by comparing task-derived model fits to a control null distribution derived from resting-state fMRI data (Figure 1C; distribution of voxel-wise variance explained by task vs. null, K–S tests: D−s > 0.124, ps < 0.001). Two pRF features were of primary interest: (1) center position, which estimates the preferred location in visual space that drives a given voxel’s response (x, y), and (2) amplitude of the voxel’s response (positive or negative) (Figure 1A). Both positive and negative pRF amplitude was reliable in the DN for all subjects (all z-scored Dice coefficients >2.9, Figure 1D). In the dATN, positive amplitude pRFs were reliable in all participants, and negative amplitude pRFs (which constituted a small proportion of the overall pRFs in this network) were reliable in 5/7 participants. For the remainder of the paper, we only consider positive pRFs in the dATN. Importantly, pRF center position was highly reproducible across runs of pRF data in the dATN and DN in all subjects (Figure 1D).
Next, we compared the retinotopic properties of voxels within the DN and dATN (Dixon et al., 2017; Fox et al., 2005; Shulman et al., 1997). Across participants, we observed a larger proportion of retinotopic voxels in the dATN as compared to the DN (43% of dATN voxels, 9% of DN voxels), consistent with their role as externally and internally oriented networks, respectively. Further, the amplitude of retinotopic responses (i.e., whether a stimulus in their preferred visual field location evoked a positive or a negative BOLD response) differed significantly between these two networks (Figure 1E). Consistent with prior reports, negative voxel-wise response amplitudes were clustered within the DN and in similar locations across participants (Szinte and Knapen, 2020; Steel et al., 2024), evidenced by a consensus map of pRF sign (Figure 1—figure supplement 3). On average, more than half (59%) of all pRFs in the DNs were inverted (i.e., had a negative BOLD response to visual stimulation in their pRF, −pRFs). In contrast, just 8% of the pRFs in the dATN were −pRFs (i.e., the vast majority of the voxels had positive BOLD responses to visual stimulation in their receptive field, +pRFs). The difference in negative pRF proportion between the dATNs and DNs was significant (t(6) = 7.94, p < 0.001; Figure 1E, inset), and there was no difference among the sub-networks of each system (dATNs: t(6) = 0.24, p = 0.82; DNs: t(6) = 0.89, p = 0.41). This distinction is particularly remarkable given the proximity of dATN and DN clusters in the posterior cerebral cortex. Like visual areas, both DN and dATN pRFs exhibited a contralateral visual field bias, as we would predict from prior literature (Szinte and Knapen, 2020; Klink et al., 2021; Steel et al., 2024) (positive dATN pRFs: t(6) = 26.0, p < 0.001; negative DN pRFs: t(6) = 4.86, p = 0.003; positive DN pRFs: t(6) = 5.20, p = 0.002) (Figure 1F; see Figure 1—figure supplement 4 for additional retinotopic properties). Together, these results show that individualized DNs have voxel-wise responses that exhibit visual coding – the necessary foundation for assessing whether the voxel-wise retinotopic code structures global interactions between the internally and externally oriented networks.
Visual responsiveness stratifies DN–dATN resting-state interactions
Next, we examined whether the retinotopic response properties of individual voxels (estimated during the retinotopic mapping sessions) predict their functional coupling across DN–dATN during task-absent, spontaneous activity (during the resting-state scans). Specifically, we tested whether voxel-level visual response classes – non-retinotopic voxels, positively tuned pRFs (+pRFs), and negatively tuned pRFs (–pRFs) – predict the strength and direction of coupling between the DN and dATN. Such a finding would reveal a stable, voxel-level organizational principle underlying the global dissociation that is hypothesized to exist between the DN and dATN (Dixon et al., 2017; Fox et al., 2005; Raichle, 2015; Raichle et al., 2001; Greicius et al., 2003; Menon, 2023).
Across resting-state runs, global activity of the DN and dATN was independent (i.e., not correlated) after accounting for variance across all cortical networks using partial correlation (Smith et al., 2013) (mean correlation = 0. 0.031 ± 0.13 SD, t(6) = 0.6287, p = 0.55; Figure 2A). Crucially, however, this global relationship masked a striking stratification based on DN voxel responsiveness (Figure 2B). Non-retinotopic DN voxels (i.e., pRF model R2 < 0.14) were not significantly correlated with the dATN (mean correlation = 0.014 ± 0.135, t(6) = 0.26, p = 0.79). In contrast, DN voxels that responded positively to visual stimulation (DN positive pRFs, +pRFs) had a positive correlation with the dATN (mean correlation = 0.284 ± 0.152, t(6) = 4.96, p = 0.0025), while DN voxels with systematic negative responses to visual stimulation (DN negative pRFs, −pRFs) were anti-correlated with the dATN (mean correlation = –0.21 ± 0.149, t(6) = –3.75, p = 0.0094). This relationship was further strengthened by adopting more conservative R2 thresholds up to 0.30 to define the pRF population, despite the overall number of included voxels decreasing, suggesting that this effect is not driven by false-positive voxels at the edge of our threshold criteria (Figure 2—figure supplement 1).
Figure 2 with 3 supplements see all
Retinotopic coding organizes spontaneous interaction between internally and externally oriented brain networks.
(A) Overall activity of the dorsal attention networks (dATN-A/B) and default networks (DN-A/B) was independent at rest. One individual’s dATN and DN are shown. Time series from all voxels in the DN and dATN subnetworks were averaged together, and we calculated the partial correlation (Smith et al., 2013) of these networks while accounting for variance of the other 11 networks within the group-prior parcellation (Yeo et al., 2011; Du et al., 2024). The dATN and DN network correlation was not significantly different from zero, suggesting the activity of these networks is largely independent. Bar plot shows the average partial correlation across participants; each data point is the average of all runs for a single participant. (B) Interaction between the DN and dATN differs by visual field preference of DN voxels. Splitting the DN voxels by visual responsiveness revealed distinct patterns of resting-state correlation. DN +pRFs had a positive correlation with the dATN (mean correlation = 0.284 ± 0.152, t(6) = 4.96, p = 0.0025), while non-retinotopic DN voxels (i.e., pRF model R2 < 0.14) were not significantly correlated with the dATN (mean correlation = 0.014 ± 0.135, t(6) = 0.26, p = 0.79). On the other hand, DN −pRFs were anti-correlated with the dATN (mean correlation = –0.21 ± 0.149, t(6) = –3.75, p = 0.0094). (C) Determining spatially matched pRFs in dATNs and DNs. We assessed the influence of retinotopic coding on the interaction between internally and externally oriented brain areas’ spontaneous activity during resting-state fMRI, by comparing the correlation in activation between pRFs in these networks that represent similar (vs. different) regions of visual space. For each DN −pRF, we established the top 10 closest positive dATN pRFs’ centers based on their x and y positions (‘matched’), and correlated the DN pRF time series with the average time series of these matched dATN pRFs. We compared these correlation values with the correlation of the DN pRF time series paired with 10 randomly selected pRFs from the 33% furthest (1000 iterations per pRF). We computed this for all DN pRFs for each all runs in each subject. Inset shows the pairwise distances between all pRFs in all participants; dotted line shows the 66%ile. (D) Negative (upper) and positive (lower) DN pRFs show retinotopic specific interaction with the dATN. Each set of bars depicts the average correlation of the matched and randomly paired pRFs between the DN and dATN for each participant, with each run shown as connected points. All participants showed greater correlation (either positive or negative) for matched compared with randomly paired pRFs. (E) Data shown averaging runs within each participant. For both positive and negative pRFs, the correlation between matched compared with randomly matched pRFs was significantly greater (consistent with their amplitude, ps < 0.01). (F, G) show the correlation between positive and negative pRFs in the DN with the dATN, split across the subnetworks of the DN-A and B. Both DN-A and B show a significant impact of matching, indicating that both subnetworks interact with the dATN via a retinotopic code (ps < 0.05). *p < 0.05, **p < 0.01, ***p < 0.001.
This result shows that voxel-level visual response profiles shape DN–dATN coupling during spontaneous resting-state activity. Specifically, the DN and dATN activation is independent during rest. However, at the voxel-level, specific subgroups of DN voxels have distinct coupling patterns with the dATN that depend on the valence of voxels’ visual responses. DN and dATN voxels with positive visual responses show a positive relationship during rest, and a notable subset of DN voxels with negative visual responses display the canonical opponency with dATN voxels. This suggests that retinotopic coding may be a mechanism that enables visual information to be exchanged between these large-scale brain systems.
Shared visual field preferences structure functional coupling among visually responsive DN and dATN voxels
We next asked whether functional coupling between DN and dATN voxels is modulated by the visual field preferences of retinotopic voxels – that is, whether their pRFs are centered on similar vs. different visual field locations (i.e., their retinotopic alignment). We hypothesized that DN–dATN interactions would be stronger for voxel pairs with shared visual field preferences. To test this, we quantified the retinotopic alignment of voxel-wise pRFs across the DN and dATN. Because we only considered positive pRFs in the dATN, hereafter we refer to these as ‘dATN pRFs’.
For each pRF in the DN (+pRFs and –pRFs), we computed the Euclidean distance between its receptive-field center (x, y) and those of all dATN pRFs (Figure 2C). We then identified the 10 closest dATN pRFs (‘Matched’ set) and averaged their resting-state time series. As a comparison group, we repeated this procedure using 10 randomly selected, poorly matched dATN pRFs (‘Random’ set; see Methods). This procedure was repeated 1000 times for each DN voxel across resting-state runs.
Matched dATN pRFs were distributed throughout the network, spanning posterior retinotopic cortex to anterior regions, including prefrontal cortex (Figure 2—figure supplement 2). Crucially, the spatial dispersion (i.e., the distance between the pRFs across the cortex) of matched and randomly matched pRFs did not significantly differ (matched vs. random – DN −pRFs × dATN pRFs: t(6) = 0.38, p = 0.71; DN +pRFs × dATN pRFs: t(6) = 0.67, p = 0.52), which suggests that spatial proximity is not a confounding factor.
If DN–dATN interactions are scaffolded by a retinotopic code, spatially matched pRFs should exhibit stronger functional coupling than randomly matched pRFs. Consistent with this prediction, after partialling out variance from DN +pRFs to isolate the specific co-fluctuation between DN −pRFs with dATN pRFs (Steel et al., 2024; Smith et al., 2013), DN −pRFs remained anticorrelated with dATN pRFs in nearly every resting-state run (Figure 2D). Critically, matched pRFs were significantly more anticorrelated compared with randomly matched pRFs (linear mixed-effects model with random intercepts for subject and run: F(1,194) = 105.7, p < 0.001; coefficient = –0.023 ± 0.002 SD). This effect was evident when run-level correlations were averaged within each participant (7/7 participants rmatched < rrandom; t(6) = 6.9, p < 0.001, Figure 2E) and could not be explained by signal quality differences, as matched and random pRF sets did not differ in tSNR (t(6) = 1.01, p = 0.353). Finally, to rule out any possible influences from vascular stealing (i.e., the shunting of blood into active tissue from nearby regions), we repeated the matching analysis after excluding any voxels within a 3-mm radius of a major vessel (Figure 2—figure supplement 3; see Methods). Both matching effects remained after excluding vascularly susceptible voxels (+DN × dATN: t(6) = 6.054; p < 0.001; −DN × dATN: t(6) = –5.0448; p < 0.01). This is strong evidence that the retinotopic code underpins the opponent interaction between negative DN and positive dATN pRFs.
Voxel-wise retinotopic alignment scaffolded the interaction between positive DN and dATN pRFs as well. After controlling for DN −pRFs, DN +pRFs were positively correlated with dATN pRFs (Figure 2D). Retinotopic alignment again strengthened the interaction: matched correlations exceeded random correlations in a mixed-effects model (F(1,194) = 406.5, p < 0.001; coefficient = +0.050 ± 0.003 SD) and for every individual participant (7/7; t(6) = 5.7, p = 0.001) (Figure 2E).
Together, these results demonstrate that shared retinotopic alignment enhances functional coupling between DN and dATN voxels. Voxels tuned to the same region of visual space show stronger spontaneous co-fluctuation – whether positively or negatively coupled – than those with non-overlapping spatial preferences. This suggests that a shared spatial code helps organize spontaneous interactions between large-scale internal and external attention networks during rest.
Subnetwork specificity within the DN
Recent work has suggested that the subnetworks of the DN, DN-A and DN-B, support different cognitive processes: episodic projection and theory of mind, respectively (Du et al., 2024; DiNicola and Buckner, 2021; DiNicola et al., 2019; DiNicola et al., 2023). Indeed, DN-A is particularly sensitive to the process of scene construction (DiNicola et al., 2023), closely linked to visual imagery, suggesting that dATN pRFs would show greater interaction with visually responsive voxels in DN-A compared with DN-B. Consistent with this hypothesis, spontaneous correlations with dATN pRFs were significantly stronger for DN-A than DN-B voxels (collapsed across positive and negative pRFs: t(6) = 4.80, p = 0.003). Crucially, however, voxel-scale retinotopic alignment structured interactions between the DN and the dATN for both DN subnetworks (Figure 2F, G; DN-A: −pRFs t(6) = –4.56, p = 0.004; +pRFs t(6) = 5.65, p = 0.0013; DN-B: −pRFs t(6) = –7.49, p < 0.001; +pRFs t(6) = 2.92, p = 0.027). The strength of this effect did not significantly differ between DN-A and DN-B (t(6) = 1.51, p = 0.182). Thus, although DN-A is more strongly globally coupled with dATN, retinotopic preferences of individual voxels structure functional interactions across both DN subnetworks equally.
In addition, using the same matched vs. random pRF procedure, we found that resting activity between the lateral place-memory area (LPMA) and the adjacent occipital place area (OPA) was significantly more correlated for retinotopically matched than random voxel pairs (t(6) = 3.01, p = 0.024; Appendix 1). This shows that retinotopic scaffolding within domain-specific systems extends beyond perceptual and memory tasks (Steel et al., 2024), organizing spontaneous activity within these systems as well. These findings build on and extend our previous work (Steel et al., 2024) and others (Scrivener and Silson, 2025).
Retinotopic coding is intrinsic to internally oriented areas
Our findings thus far highlight a role for retinotopic coding in structuring spontaneous interactions between a subset of voxels across large-scale internally and externally oriented brain networks. However, a key mechanistic question remains: is retinotopic coding an intrinsic property of the DN, or does it only emerge in response to input from sensory systems and lower-level networks like the dATN? One possibility is that DN-driven modulation of dATN activity reflects a global, nonselective suppression, with retinotopic coding in the DN arising only through input from the dATN. Alternatively, retinotopic coding may be intrinsic to the DN itself – such that signals projecting from the DN to the dATN are retinotopically specific, allowing the DN to shape fine-grained spatial representations within the dATN. Disambiguating these possibilities is essential for determining whether the retinotopic scaffold constitutes a general mechanism for cross-network communication, independent of task demands or cognitive domain (e.g., memory vs. perception).
To address this, we again turned to participants’ resting-state fMRI data. Because participants are not engaged in any task, resting-state provides a neutral testbed for assessing spontaneous DN–dATN interactions. Moreover, experience sampling during rest has shown that attention naturally alternates between internal and external focus during rest (Gonzalez-Castillo et al., 2021; Van Calster et al., 2017), and thus is likely to result in activation of both the DN and dATN, making it possible to assess whether retinotopic coding emerges primarily during periods of elevated DN or dATN activity.
We developed an event-triggered analytical approach to dissociate periods when activity of the retinotopic voxels within the dATN and DN was elevated. We refer to these periods when dATN or DN voxels’ activity was elevated as dATN-driven from DN-driven events, respectively. Specifically, we identified timepoints where spontaneous activity of individual DN or dATN pRFs exceeded the 99th percentile of their resting-state distribution. These high-activation time points were treated as neural ‘events’ (Figure 3A–C). To test whether these events produced retinotopically specific responses in the target network, we compared peri-event activation in the target network’s most 10 closely matched pRFs (i.e., those with similar visual field preferences) to that in the 10 worst-matched (anti-matched) pRFs (Figure 3A). We used the worst matched pRFs for this analysis because the large number of pRFs made the random matching procedure impractical. All analyses were conducted separately for positive and negative pRF populations within the DN. This design enabled us to test two competing predictions: if retinotopy in the DN only reflects input from more canonically perceptual networks like the dATN, we should only see retinotopically specific coupling during dATN-driven events. If, however, the DN possesses an intrinsic retinotopic scaffold, we should observe retinotopic specificity during both dATN- and DN-driven events. A windowed correlation analysis produced similar results (Figure 3—figure supplement 1).
Figure 3 with 2 supplements see all
DN- vs. dATN-driven neural events detected in spontaneous resting-state dynamics show evidence for retinotopically specific suppression.
(A) Event detection and analysis procedure and example events from a single resting-state run. To detect events, we extracted the time series from each pRF in the source regions (positive or negative pRFs in the DN; positive pRFs in the dATN) and isolated time points where the z-scored time series exceeded 2.4 SD (99th percentile). We then examined the activity of 10 best matched and 10 worst matched (anti-matched) pRFs from the target region in this peri-event time frame (6 TRs (8 s) before and after the event). (B) Example DN-driven event in the representative time series shows elevated source pRF activity (DN −pRF) and corresponding target area activity (dATN +pRF). (C) Same as B, but for dATN-driven activity with the event detected in a dATN +pRF. Evidence for retinotopically specific suppression of ongoing activity during DN-driven events detected in DN −pRFs (D) and dATN-driven events detected in dATN +pRFs (E). (D) Time series show the peri-event activity of matched and anti-matched dATN +pRFs averaged across all participants for events detected in DN −pRFs (N = 78,978 events). (E) Time series shows the peri-event activity in DN −pRFs during events detected in dATN +pRFs (N = 541,546). (F) For DN -pRFs paired with dATN +pRFs, target area shows retinotopically specific suppression of activity for both DN- and dATN-driven events. Bars show the average activation at event time of the target areas’ matched and anti-matched pRFs for each participant. Activity was lower in matched compared with anti-matched pRFs for DN- and dATN-driven events (ts > 3.8, ps < 0.009). Evidence for retinotopically specific excitation of ongoing activity during events DN-driven events detected in DN +pRFs (G) and dATN-driven events detected in dATN +pRFs (H). (G) Grand average time series for matched and anti-matched dATN +pRFs during events detected in DN +pRFs for all participants. (H) Grand average time series for matched and anti-matched DN +pRFs during events detected in dATN +pRFs. (I) For DN +pRFs paired with dATN +pRFs, target area shows retinotopically specific excitation for both DN- and dATN-driven events. Bars show the average activation at event time of the target areas’ matched and anti-matched pRFs for each participant. Activity was higher in matched compared with anti-matched pRFs for DN- and dATN-driven events (ts > 8.75, ps < 0.001). Interestingly, the activity of anti-matched pRFs was greater during DN-driven compared with dATN-driven events (t(6) = 4.92, p = 0.0026), suggesting that input from DN +pRFs to the dATN +pRFs may be more global and less retinotopically targeted compared with dATN-driven signaling between these networks. For time series plots, solid bars along the x-axis indicate periods with a significant difference between matched and anti-matched pRF activity within the target area. *p < 0.05, **p < 0.01, ***p < 0.001. −DN: negative DN pRFs; +DN: positive DN pRFs; +dATN: positive dATN pRFs.
Across all participants, we detected 78,978 events in DN −pRFs, 52,595 events in DN +pRFs, and 541,546 events in dATN pRFs. dATN-driven events were more frequent than either DN-driven type (dATN vs. negative DN: t(6) = 15.19, p < 0.001; dATN vs. positive DN: t(6) = 25.63, p < 0.001), but there was no difference between DN-driven event types (t(6) = 2.03, p = 0.18). On average, participants exhibited 2082 ± 535 (mean ± SD) negative DN, 1799 ± 400 positive DN, and 3102 ± 602 dATN events per run. Each pRF showed 0–4 events per run (Figure 3—figure supplement 2). More than 90% of time points involved <10 simultaneous events, suggesting that individual pRF-level correlation could be isolated from global fluctuations in regional activity, making this approach suitable for evaluating distinct interactions at the individual voxel level.
We found a robust effect of retinotopic alignment (i.e., matched vs. anti-matched) on pRF activation during both dATN- and DN-driven interactions (Figure 3D–I). For DN −pRFs, a linear mixed-effects model (factors: DN-/dATN-driven and retinotopic alignment; subject as random effect) revealed a main effect of match status (F(1,24) = 15.3, p = 0.0006) and no interaction between match status and direction (F(1,24) = 0.0748, p = 0.40). During DN-driven events in DN −pRFs, matched dATN pRFs showed lower activity than anti-matched (t(6) = 5.46, p = 0.001) and were significantly deactivated relative to the pre-event baseline. The same was true of DN -pRFs during dATN-driven events: we observed lower activation for matched vs. anti-matched pRFs (t(6) = 3.80, p = 0.009). These results are consistent with a retinotopically specific interaction between dATN +pRFs and DN −pRFs during both DN- and dATN-driven events detected in spontaneous activity during resting-state fMRI.
The interaction between DN +pRFs and dATN pRFs also showed evidence of retinotopic coding during both dATN- and DN-driven events, but these effects were modulated by a direction × match interaction (F(1,24) = 12.84, p = 0.002). Both DN-driven (t(6) = 8.76, p < 0.001) and dATN-driven (t(6) = 9.14, p < 0.001) events evoked strong activation in the target network (dATN or DN, respectively). However, anti-matched pRFs were more active during DN-driven compared to dATN-driven events (t(6) = 4.92, p = 0.0026), suggesting that signals from positive DN pRFs to dATN may be less retinotopically specific than incoming signals from the dATN. These results align with prior findings showing decreased spatial precision in visual representations during memory recall compared to perception (Favila et al., 2022; Breedlove et al., 2020).
Taken together, these results support the view that retinotopic coding serves as a general scaffold for DN-dATN interactions. This alignment shapes spontaneous coupling during both periods of DN- and dATN-driven activity, even in the absence of external tasks. Because the effect persists without sensory input, it likely reflects an intrinsic feature of the DN activity rather than a transient result of drive from sensory input.
Discussion
Despite extensive evidence that internally and externally oriented brain networks – including the default network and dorsal attention network – are critical for human cognition (Yeo et al., 2011; Margulies et al., 2016; Braga and Buckner, 2017; Gordon et al., 2017; Ranganath and Ritchey, 2012; Andrews-Hanna et al., 2010; Barnett et al., 2021; Du et al., 2024; Dixon et al., 2017; Fox et al., 2005; Raichle, 2015), the coding principles that enable these distributed systems to interact have remained unclear (Popham et al., 2021; Buschman et al., 2011; Libby and Buschman, 2021; Semedo et al., 2019; Gregoriou et al., 2009; Bosman et al., 2012). Here, we show that DN–dATN coupling is retinotopically grounded at the voxel scale: the visual field biases of individual voxels scaffold bidirectional interactions between these networks, even during spontaneous dynamics in the absence of visual task demands. Analyses of transient elevated activity in either the DN or dATN confirm that this retinotopic opponency shapes both dATN-driven signals to the DN and DN-driven signals to the dATN, indicating that the retinotopic code is intrinsic to both internal and external networks. These findings offer a multi-scale account of neural communication, in which interactions among sub-populations of voxels with shared tuning preferences are nested within macroscale network dynamics. Nesting multiple neural codes might enable ongoing computations within a larger brain system (e.g., attending to internal mental states within the DN during memory recall), while simultaneously allowing for the sharing of fine-grained representations across brain systems (Steel et al., 2024). Here we show this latent, voxel-scale interaction persists even without structured behavioral demands, opening the opportunity to study these fine-scale interactions across brain networks in resting-state data (Hedger et al., 2026).
Resolving the coding principles of the DN is critical given its central roles in spatial and episodic memory, social cognition, and executive function (Raichle, 2015; Margulies et al., 2016; Buckner and DiNicola, 2019; Yeshurun et al., 2021; Spreng et al., 2010; Spreng et al., 2009; Foster et al., 2023). Classic views of systems-level brain organization posit regions at the apex of the cortical hierarchy, like the DN, likely have abstract, amodal neural codes, in contrast to sensory codes – like retinotopy – that manifest in earlier areas of the cortical hierarchy (Margulies et al., 2016; Bellmund et al., 2018; Popham et al., 2021). Our findings and others challenge this view. Instead of shedding the retinotopic code at the cortical apex, the DN retains a retinotopic code (Szinte and Knapen, 2020; Klink et al., 2021) and functionally deploys this code to structure interactions with the externally oriented dATN. Thus, a code once thought to be exclusive to early visual cortex and visually oriented tasks plays a widespread role in shaping brain activity across the cortical hierarchy – even at rest.
Establishing this voxel-level structure within the DN and dATN reframes a long-standing debate that has cast these networks as globally competitive or functionally independent (Dixon et al., 2017; Christoff et al., 2009; Murphy et al., 2019; Murphy et al., 2018). These views have stemmed from macroscale observations of reciprocal activation patterns during external vs. internal task demands (Andrews-Hanna et al., 2010; Chun et al., 2011; Raichle, 2015; Fox et al., 2006; Fedorenko et al., 2013; Vossel et al., 2014; Shulman et al., 1997; Silson et al., 2019; DiNicola et al., 2020; Thakral et al., 2017; Gilmore et al., 2021; Christoff et al., 2009), and resting-state analyses that reveal zero or negative correlations between their mean time series (Dixon et al., 2017; Fox et al., 2005; Greicius et al., 2003; Menon, 2023). However, by focusing on global, network-level activity across voxels, these approaches have obscured a latent, voxel-level structure to DN-dATN functional coupling. Specifically, we find that DN–dATN coupling is not independent. Instead, two subpopulations of voxels within the DN contribute to a retinotopically grounded cross-network interaction that is stratified by their voxel-level visual selectivity. One population of DN voxels is positively tuned to visual stimulation and positively covaries with dATN voxels that share their visual field preference. In contrast, DN voxels that are negatively tuned to visual stimulation negatively covary with retinotopically matched dATN voxels as a function of visual field preference. When these voxel-level tuning preferences are not taken into account, global cross-network opponency or independence (Dixon et al., 2017; Fox et al., 2005) reinforces the misconception that the networks support mutually exclusive cognitive modes. By revealing a shared retinotopic scaffold, our findings reconcile why DN and dATN can appear independent in coarse analyses yet remain functionally integrated, offering a new framework for understanding how internal and external attention are coordinated in the human brain.
Importantly, there are key differences in the nature of retinotopic coding between the DN and visual areas, including the dATN. First, retinotopy in the DN is predominantly negative in valence – DN voxels more often show spatially specific suppression in response to visual stimulation (Szinte and Knapen, 2020; Klink et al., 2021; Steel et al., 2024), unlike the positive spatially specific response that is typically observed in visual areas. Second, the DN lacks the map-like retinotopic organization seen in early visual cortex. In visual areas, retinotopic coding is not only the foundational code but also the dominant topographic organizing principle across the cortical surface, where voxels that are proximal on the cortical surface tend to share visual selectivity for nearby regions of visual space (Wandell et al., 2007). Prior work has emphasized the visual response bias in regions where voxel-wise retinotopic responses lack a map-like organization (Groen et al., 2022); overall, the DN does exhibit this kind of bias. However, our results show that the voxel-scale activity underpinning this bias reflects the latent connectivity of those voxels. Thus, we adopt the term ‘retinotopic coding’, because this voxel-scale coding scheme exists without a map-like organization on the cortical surface. For example, rodent and bat head direction cells are not laid out in a literal ring, but the population code of these neurons forms a ring manifold (Taube, 2007; Finkelstein et al., 2015). Similarly, in humans, retrosplenial cortex shows head-direction coding during navigation tasks despite lacking topographic organization of head-direction responses (Nau et al., 2020). Analogously, the DN may implement a retinotopic code that is distributed across the network in a format that does not spatially mirror the retina. That is, the DN may use a fine-scale retinotopic code without a retinotopic map – the representational structure can be preserved even in the absence of spatially ordered cortical layouts.
What function might the bivalent (positive and negative) voxel-scale retinotopic alignment across these large-scale networks serve? Our previous work has identified a role for retinotopic-alignment in structuring the functional coupling of scene-selective mnemonic and perceptual brain areas (Steel et al., 2024; Scrivener and Silson, 2025) located near the dATN and DN, respectively (Steel et al., 2021; Steel et al., 2025). Near to the DN, scene-selective memory areas exhibit a negative retinotopic code, while functionally paired scene-selective perceptual regions exhibit a positive retinotopic code (Steel et al., 2024). This bivalent retinotopic code structures the mutual activity of these mnemonic and perceptual areas: scene-recall activates scene-memory voxels, but suppresses voxels in scene-perception areas that are tuned to matched visual locations, a pattern that reverses during scene perception (Steel et al., 2024). The current work significantly extends this observation beyond perceptual and mnemonic task states – identifying widespread retinotopic alignment across large-scale networks with presumed different orientations (internal vs. external) even during spontaneous activity at rest, in the absence of tasks.
As to the functional role of the bivalent retinotopic alignment across internal and externally oriented large-scale networks, we propose two complementary possibilities. One possibility is that the bivalent (positive and negative) subpopulations of DN pRFs may serve as a neural mechanism for maintaining separate channels for perceptual and mnemonic information (Libby and Buschman, 2021), preventing mnemonic signals from being interpreted as or interfering with sensory signals. A second possibility is that bivalent coding across the DN and dATN could be a mechanism for signaling mnemonic predictions via top-down inhibitory processes (Steel et al., 2024; Bastos et al., 2012). Specifically, when the DN generates predictions about incoming sensory information based on memory, the inverted retinotopic code could act as a suppressive signal that sharpens tuning in perceptual areas and enhances signal-to-noise for expected stimuli (Albright, 2012; Barron et al., 2017). These mechanisms may work in concert: the inverted code simultaneously maintains separate channels for memory/perception within the DN, while enabling precise top-down predictive signaling across DN–dATN. Future work that combines high-resolution recordings with behavior will be essential for testing how these mechanisms jointly support constructive operations such as memory-based prediction during visually guided navigation.
Relatedly, the DN is considered a transmodal hub for cortical processing, where disparate sensory and motor processes converge (Margulies et al., 2016; Huntenburg et al., 2018). The DN’s position at the cortical apex implies connections with and influence over unimodal cortical areas. However, the mechanism linking unimodal and transmodal networks had been unknown. Prior work posited that sensory coding in transmodal areas might serve this function (Szinte and Knapen, 2020; Groen et al., 2022), and our data provide direct empirical support for this account: specific visually responsive voxels provide an input/output interface bridging perceptual and memory systems. This complements work delineating specific affective and effective subregions within the DN that link the DN to other brain areas (Zhang et al., 2026). Thus, while the DN may be ‘distant from input’ (Murphy et al., 2018), these data suggest that it is not disengaged from sensory processing.
Retinotopic alignment between the DN and dATN was clear in both subnetworks of the DN, DN-A and DN-B (Braga and Buckner, 2017; Du et al., 2024; DiNicola et al., 2020). This is surprising given their distinct functional profiles: DN-A has been more directly linked to visually grounded memory processes, such as episodic projection and scene construction (Du et al., 2024; DiNicola et al., 2020; DiNicola et al., 2023), and spatially overlaps with regions previously shown to exhibit retinotopy and engage in scene memory and perception (Steel et al., 2025). In contrast, DN-B is more often associated with processes that are not obviously tied to vision: theory of mind and affective tasks, such as interpreting false beliefs or representing physical and emotional pain (Du et al., 2024; DiNicola et al., 2020). Why might the DN-B show retinotopically grounded coupling with perceptual networks? Retinotopic coding might still be a useful format for transforming abstract social representations into sensory-grounded signals to inform perceptual predictions (e.g., anticipating changes in facial expression) or behavioral decisions (e.g., making eye movements). Future work might consider direct manipulation of social perceptual processes to assess the impact of social tasks on retinotopic coding between DN-B and dATN.
Moments of high spontaneous activity originating in the DN co-occurred with suppression of retinotopically matched dATN pRFs, suggesting that retinotopy functions as a native coding format in the DN for targeted modulation of perceptual cortex. This result aligns with other recent studies suggesting an important role of the DN in shaping visual processing. For example, ongoing pre-stimulus DN activity has been shown to influence sensitivity during near-threshold visual object recognition (Wu et al., 2024). Other work has shown that subareas of the DN encode kinematics of spatial exploration in rodents (Nghiem et al., 2024), and represent the visuospatial context that is associated in memory with a perceived scene (Steel et al., 2024; Steel et al., 2021; Steel et al., 2025; Steel et al., 2019; Steel et al., 2023). Relatedly, DN activity reflects semantic-level attentional priorities during visual search (Çukur et al., 2013). Taken together, these findings show that, rather than being disengaged during visual tasks, the DN actively shapes responses in perceptually oriented cortex in a spatially precise manner. In this sense, retinotopic coding is part of the DN’s ‘native language’. Taken together, these findings prompt a reevaluation of the DN’s role in perceptual processing, and the extent to which roles in ‘internally and externally oriented attention’ adequately capture the DN and dATN’s role in cognition.
Is the DN a ‘visual’ network? One interpretation of retinotopic coding in the DN is that the DN directly and obligatorily represents retinal position in addition to more abstract features (Dumoulin and Wandell, 2008; Wandell et al., 2007). Under this hypothesis, the retinotopic code may indicate that the information represented in the DN is, in part, sensory. Alternatively, the apparent visual coding may simply represent an underlying connectivity structure (Arcaro and Livingstone, 2024), which enables associative and semantic information that the DN directly represents (Popham et al., 2021; Peer et al., 2021; Huth et al., 2016) to be effectively transferred to sensory areas of the brain. Under this alternative hypothesis, the information represented in the DN is not itself sensory in nature; the retinotopic code in the DN simply reflects the connectivity ‘highway’ between the abstract representations in the DN and sensory representations in the dATN. Notably, in either case the retinotopic scaffolding serves as a latent structure linking high- and low-level areas through consistent, spatially organized interactions (Arcaro and Livingstone, 2024). This provides more evidence that low-level sensory codes are distributed across the brain, joining recent work showing latent somatotopic codes in regions of the brain that are not typically associated with somatosensation, including visual cortex (Hedger et al., 2024). The wide and overlapping distribution of sensory codes across the brain may allow for multiplexing of complex information to structure cross-network interactions (Groen et al., 2022; Hedger et al., 2024). Further studies investigating this framework with diverse tasks and stimuli could reshape our understanding of how the brain processes and integrates information across different levels of cognitive complexity.
Our study and other recent studies of voxel-wise visual responses in traditionally non-visual areas (Szinte and Knapen, 2020; Klink et al., 2021; Steel et al., 2024; Scrivener and Silson, 2025; Pizka et al., 2026; van Es et al., 2019; Angeli et al., 2024) raise important questions about model selection and thresholding criteria. In canonical visual cortex, pRF analyses often rely on variance-explained thresholds to identify reliable voxel-wise model fits. However, it is less clear what R² criterion should be applied in non-visual regions, where responses may be weaker or more heterogeneous. To address this issue, we designed a method to determine a threshold for voxel-wise model fits (R2) empirically by leveraging resting-state fMRI data as a noise model. Crucially, pRF parameter estimates (center position and amplitude) for voxels that passed this empirical threshold were reliable across runs, suggesting this method robustly selects voxels with stable pRF properties. This empirical method may be used to evaluate retinotopic responses across the brain in future studies (Angeli et al., 2024).
We used the Natural Scenes Dataset (NSD), which included pRF modeling data to characterize visually responsive cortex in a small group of individual participants. One limitation of this dataset is the small number of pRF task runs, which might lead to noisier model estimates. Our robust thresholding approach mitigates this concern. Furthermore, our primary results were based on voxel-wise pRF alignment within independently acquired resting-state fMRI data; noisy estimates of pRF parameters would obscure this relationship and make our results less likely. Relatedly, here we used a single Gaussian model, consistent with prior work on negative visual responses in memory systems (Szinte and Knapen, 2020; Klink et al., 2021; Steel et al., 2024). However, other models of visual receptive fields might offer further insight into the DN’s visual responsiveness, such as double Gaussian models of surround suppression (Kay et al., 2013) or compressive summation (Zuiderbaan et al., 2012). Future studies might directly compare different visual models to further refine the computations underpinning visual responses in the DN.
In summary, our results show that a voxel-level retinotopic code organizes the spontaneous interactions between large-scale internally and externally oriented networks in the human brain – the DN and dATN. Even when global activity appears uncorrelated across DN–dATN, voxel-level visual-field tuning reveals a tightly interlocked, bidirectional functional dynamic that persists in the absence of external stimulation. This latent scaffold reconciles the apparent macroscale independence of internally and externally oriented systems with their evident functional need to cooperate, revealing that the DN is not divorced from perceptual networks, but modulates perceptual voxels through a common visuospatial reference frame. Taken together, by grounding large-scale network interactions in a shared, bivalent spatial code, retinotopy may offer a simple organizing principle for the brain’s coordination of thought and perception.
Materials and methods
Dataset and participants
Request a detailed protocolThe data analyzed here are part of the NSD, a large-scale precision 7T fMRI dataset consisting of high-resolution anatomical scans, retinotopic mapping, functional localizers for visual areas, and both task and resting-state fMRI data for eight adult participants. A full description of the dataset can be found in the original manuscript (Allen et al., 2021). Here, we detail the data, processing, and analysis steps relevant to the present work.
Participants (N = 8; 2 male, 6 female; ages 19–32) were scanned at the University of Minnesota over the course of 33–43 sessions. All participants had normal or correct to normal vision and no known neurological impairments. Informed consent was collected from all participants, and the study was approved by the University of Minnesota institutional review board. One subject (subj03) was excluded from resting-state analyses for having insufficient resting-state runs that passed our quality metrics (see below).
For the present study, we focused on two components of the NSD protocol: retinotopic mapping and resting-state fMRI. All participants completed three runs of a sweeping bar retinotopic mapping task (5 min each). Retinotopic mapping was conducted during Session 1, along with the functional localizer scans. This task is referred to as ‘RETBAR’ in NSD documentation.
All participants completed at least 14 runs of a resting-state fixation task (range: 14–34 runs; 5 min each). Resting-state data runs were included in all NSD scan sessions after session 20 (i.e., sessions 21, 22, etc.). Each of these sessions included two 5-min resting-state runs – one at the start and one at the end of the session – yielding 14–34 total resting-state runs per participant. During the second resting-state run, participants performed a deep breath in the first 5 s of scanning. To account for this, and to ensure the two resting-state runs were equated in length, we excluded the first 30 s of the resting-state data from all runs. One participant (subj03) was excluded from resting-state analyses due to having an insufficient number of usable runs after quality control. All remaining participants had at least eight high-quality resting-state runs (mean: 14 ± 6.13 runs).
MRI acquisition and processing
For this study we made use of the following data from the NSD: anatomical (T1 and FreeSurfer segmentation/reconstruction; Dale et al., 1999; Fischl et al., 2002), functional regions of interest (ROIs), and minimally preprocessed retinotopy and resting-state time series. All analyses except network identification were conducted in original subject volume space, and data were projected on the surface for visualization purposes only. Network identification was performed on the fsaverage6 surface, then projected into subject specific volume space for quantification.
Anatomical data
Request a detailed protocolAnatomical data was collected using a 3T Siemens Prisma scanner and a 32-channel head coil. We used the T1-weighted anatomical images provided at 0.8 mm resolution, as well as the registered output from FreeSurfer recon-all, aligned to the 1.8 mm functional data. For venography, we used the provided time-of-flight (TOF) MR venography images with a resolution of 0.5 mm3. For visualization purposes, we projected statistical and retinotopy data to the cortical surface using SUMA (Saad and Reynolds, 2012) from the AFNI software package (Cox, 1996).
Functional MRI data acquisition and processing
Request a detailed protocolAll analyses were conducted on 1.8 mm isotropic resolution, minimally processed runs of a sweeping-bar retinotopy task and resting-state fixation provided in the NSD.
Quality assessment
Request a detailed protocolTo ensure only high-quality resting-state data were included, we trimmed the first 25 TRs (40 s) from each run (Steel et al., 2019), which left 4.25 min of resting-state data per run. We then used AFNI’s quality control assessment tool (APQC; Taylor et al., 2024) on the raw trimmed resting-state time series to assess the degree of motion in the resting-state scans. We excluded runs with greater than 1.8 mm maximum displacement per run or 0.12 mm framewise displacement from analysis, and assessed runs with greater than 1 mm displacement or 0.10 mm framewise displacement on a case-by-case basis (Du et al., 2024). After exclusions, one participant (subj03) had only a single run that survived our criteria (4.25 min of data), so we excluded them from our analyses. The remaining seven participants had at least eight resting-state runs (>34 min of data; mean number of runs: 14 ± 6.13 (SD), range: 8–24 runs).
ICA denoising
Request a detailed protocolTo further denoise the retinotopy and remaining resting-state data, we used manual ICA classification of signal and noise on the minimally preprocessed time series (Steel et al., 2021; Griffanti et al., 2014; Griffanti et al., 2017). We used manual classification because automated tools perform poorly on the high spatial and temporal resolution data of the NSD (Beckers et al., 2023). For each retinotopy and resting-state run, we decomposed the data into independent spatial components and their associated temporal signals using ICA (FSL’s melodic; Jenkinson et al., 2012; Smith et al., 2004). We then manually classified each component as signal or noise using the criteria established in Griffanti et al., 2017. Noise signals were projected out of the data using fsl_regfilt (Griffanti et al., 2014).
We did not perform global signal regression (Saad et al., 2012). Data were normalized to percent signal change. A 2.5-mm FWHM smooth was applied on the surface to resting-state fixation data used in functional connectivity network identification. The analysis of voxel-scale interactions was performed on unsmoothed data.
Data analysis – identifying resting-state networks
Individual-specific cerebral network estimation
Request a detailed protocolIn each participant, we identified a set of 15 distributed networks on the cerebral cortical surface using a multi-session hierarchical Bayesian model (MS-HBM) functional connectivity approach (Du et al., 2024; Kong et al., 2021). Briefly, the MS-HBM approach calculates a connectivity profile for every vertex on the cortical surface derived from the top 10% of correlations from that vertex to 1175 ROIs uniformly distributed across the cortical surface (Yeo et al., 2011). We applied an expectation-maximization algorithm, initialized with a 15-network group-level spatial prior created by applying previous methods of network identification (Yeo et al., 2011) on a portion of the HCP S900 data, to all the connectivity profiles from all participants. This algorithm then estimated each network’s group-level connectivity profile, along with parameters for subject-, session/run-, and region-level variability, which together were used to generate individual-specific, labeled parcellations.
The MS-HBM approach is well-suited to derive parcellations for the participants in the NSD, where resting-state data were collected across multiple days, because this method uses repeated resting-state runs for each participant to estimate within-participant variance. This leverages a strength of the NSD (the relatively large number of resting-state runs for each individual) and can account for differences in connectivity both within an individual (likely the result of confounding variables such as scanner variability and time of day), and between participants (reflecting potentially meaningful individual differences). Moreover, the MS-HBM incorporates a group spatial prior, further stabilizing the resulting network parcellations for each participant. This also ensures that all networks will be identified in all participants, while allowing for idiosyncratic topographic differences.
Data analysis – retinotopic mapping
pRF modeling
Request a detailed protocolThe NSD retinotopy stimuli consist of a mosaic of faces, houses, and objects superimposed on pink noise that are revealed through a continuously drifting aperture. For our analysis, we considered only the bar stimulus time series, which is consistent with other studies investigating negative pRFs in high-level cortical areas (Szinte and Knapen, 2020; Klink et al., 2021; Steel et al., 2024). We did not consider the wedge/ring stimulus for any analyses.
After denoising the retinotopy data, we averaged the three retinotopy runs with the bar aperture together to form the final retinotopy time series. We performed pRF modeling using AFNI following the procedure described in Silson et al., 2015. First, because the pRF stimulus in the NSD is continuous, we resampled the stimulus time series to the fMRI temporal resolution (TR = 1.333 s). Next, we implemented AFNI’s pRF mapping procedure (3dNLfim). Given the position of the stimulus in the visual field at every time point, the model estimates the pRF parameters that yield the best fit to the data: pRF amplitude (positive, negative), pRF center location (x, y), and size (diameter of the pRF). Both Simplex and Powell optimization algorithms are used simultaneously to find the best time series/parameter sets (amplitude, x, y, size) by minimizing the least-squares error of the predicted time series with the acquired time series for each voxel. Relevant to the present work, the amplitude measure refers to the signed (positive or negative) degree of linear scaling applied to the pRF model, which reflects the sign of the neural response to visual stimulation of its receptive field.
Identifying retinotopic voxels – empirically derived R² thresholds
Request a detailed protocolWe developed a rigorous, data-driven approach to determine whether our pRF model fits truly reflect visual coding rather than physiological noise. To do this, we leveraged the resting-state fMRI data – where there is no dynamic visual stimulation – to create a control (null) distribution for retinotopic model fits and establish a ‘noise floor’ for each network (i.e., the DN and dATN). Then, we compared task-derived model fits to this rest-based null distribution to establish an empirical R² threshold for pRFs in each network.
Specifically, we fit the pRF model to the average time series across all resting-state runs for each participant. We then compared the resulting voxel-level R² distribution to that from the pRF mapping task data, pooled across participants. To test whether the model fits were significantly higher for the task vs. null distribution, we performed a one-tailed Kolmogorov–Smirnov test for each network on the respective cumulative distribution functions (H0: CDFrest(R2) ≤ CDFtask(R2), Ha: CDFrest(R2) > CDFtask(R2)). These analyses were performed separately for the dATN and positive and negatively responsive voxels within the DN.
We defined a voxel as exhibiting a retinotopic code (hereafter referred to as a ‘pRF’) if its model fit exceeded 14% variance explained. This threshold was based on the 95th percentile of the null R² distributions, which yielded noise floors of R² = 0.124 (dATN), R² = 0.132 (DN-positive), and R2 = 0.137 (DN-negative). We therefore set our global threshold for visually responsive voxels across all networks at R² > 0.14, as this value exceeded the noise floor calculated in each network separately.
Quantifying visual-field coverage
Request a detailed protocolVisual-field coverage (VFC) plots represent the sensitivity of an ROI across the visual field. We followed the procedure in Steel et al., 2024, which we have reproduced here. Individual participant VFC plots were first derived based on pRF center, as well as pRF size and R2. These plots combine the best Gaussian receptive field model for each suprathreshold voxel within each ROI by calculating the max operator at each point in the visual field, defined as the maximum value from all pRFs within the ROI. The resulting coverage plot thus represents the maximum envelope of sensitivity across the visual field. Individual participant VFC plots were averaged across participants to create group-level coverage plots.
To compute the elevation biases, we calculated the mean pRF value (defined as the mean of the average max operator values in a specific portion of the VFC plot) in the contralateral upper visual field (UVF) and contralateral lower visual field (LVF) and computed the difference (UVF–LVF) for each participant, ROI, and amplitude (+/−) separately. A positive value thus represents an upper visual-field bias, whereas a negative value represents a lower visual-field bias.
Reliability of pRF amplitude and center position estimates
Request a detailed protocolWe tested whether key pRF features of interest – amplitude (i.e., positive vs. negative) and center position (i.e., x, y) – were reliable across retinotopy runs. To assess the reliability of pRF amplitude, we iteratively compared the amplitude of significant pRFs from individual runs of pRF data. Specifically, for each retinotopy run, we fit the pRF model and then binarized any significant voxels (R² > 0.14) by amplitude. When testing positive-pRF reliability (in dATN and DN), positive voxels = 1; all others = 0. For negative-pRF reliability (in DN), significant negative voxels = 1. We then computed a three-run Dice coefficient: (Dice ). We compared this value against 5000 null Dice scores derived from the same number of voxels randomly sampled from all voxels, both significant and non-significant, in the ROI. For each participant, this resulted in 1 ‘observed’ Dice coefficient along with 5000 null values representing the distribution of Dice coefficients expected by chance. The observed coefficient was z-scored against this null distribution for each participant.
We also evaluated center position (x, y) reliability. For every significant voxel, we calculated the Euclidean distance between its center estimates across the three run pairs (i.e., 1–2, 2–3, and 1–3). As with amplitude, we generated a null distribution by repeating this computation on 5000 random voxel sets (significant + non-significant). To establish significance of the true estimate, the observed estimate was z-scored against this null distribution for each participant.
Data analysis – retinotopically grounded resting-state analyses
Evaluating global functional coupling between cortical networks at rest
Request a detailed protocolTo assess global functional coupling between the default network and the dorsal attention network at rest, we used a partial correlation analysis that accounted for variation associated with other cortical networks (Smith et al., 2013). For this and all subsequent analyses, each individual resting-state run was z-scored in time. Prior to calculating the partial correlation between them, we first averaged the time series within each DN and dATN subnetwork.
Evaluating the contributions of voxel-level visual response valence to cross-network coupling
Request a detailed protocolTo test whether DN–dATN coupling is modulated by the visual responsiveness (i.e., response valence) of DN voxels, we repeated this analysis separately using only: (1) sub-threshold/non-retinotopic voxels, (2) positive pRFs, or (3) negative pRFs from the DN – while still partialing out the remaining networks. This allowed us to assess whether specific classes of visually responsive voxels in the DN differentially contribute to functional coupling with positive pRFs in the dATN.
Evaluating voxel-level retinotopic alignment in cross-network coupling
Request a detailed protocolTo test whether functional coupling is shaped by the shared visual field preferences of DN–dATN voxels (i.e., retinotopic alignment), we compared resting-state correlations between pRF pairs across the DN and dATN that had matched (vs. unmatched) pRF centers.
To do this, for each pRF in the DN (positive or negative), we computed the Euclidean distance between its receptive-field center (x, y) and those of all dATN +pRFs. We focused on center position instead of pRF size because (1) size estimates are relatively unstable compared with center position (Alvarez et al., 2015), and (2) we lacked a clear prediction about how pRF size would be related between the two networks. For each DN voxel, we identified the 10 nearest dATN +pRFs based on center position – the ‘matched’ set – and averaged their resting-state time series. As a null comparison, we defined a ‘random’ set by sampling with replacement 10 dATN +pRFs from the most distant third of the match distribution (i.e., the 1/3 furthest pRF pairs). We considered the bottom 33% because the overwhelming majority of pRFs constituted relatively equal ‘good’ matches. This random sampling was repeated 1000 times per DN voxel, generating a distribution of 1000 random correlation values, which we then averaged to yield a single ‘random’ correlation for each DN source pRF.
We calculated the correlation between each DN pRF’s resting-state time series and that of its 10 best-matched dATN pRFs (‘matched’ correlation). We averaged across the 10 best matches to (1) reflect the likely distributed input from multiple dATN sources and (2) improve signal-to-noise. To isolate pRF-specific effects, we partialed out the mean time series of the complementary DN pRF population from the signals of interest. For example, when analyzing correlations between –DN and +dATN, we treated +DN activity as a nuisance regressor (and vice versa) (Steel et al., 2024; Smith et al., 2013). This provided a network-level analog to the whole-brain partial correlation described above (see Correlation among cortical networks section), allowing us to isolate the unique contribution of the pRFs of interest from overall activity in the region as well as to control for global fluctuations that cause widespread positive correlations like motion and attentional state (Steel et al., 2024; Smith et al., 2013). All correlation values were Fisher-transformed, and we took the median value across voxels within each network to yield a single value per run. This procedure was repeated for all runs and all participants.
We then statistically compared Fisher-transformed correlations for matched vs. random pRF pairs using a linear mixed effects model, with match status (matched vs. random) as a fixed effect and subject and run as random effects. This model accounted for both between- and within-subject variation across sessions. For follow-up tests, we compared the average matched and random values across participants using a paired t-test.
Masking vascularly sensitive voxels
Request a detailed protocolTo ensure that the above retinotopic alignment analyses were not the result of large-scale vascular features (e.g., vascular stealing), we created a whole brain mask to remove any voxels within 3 mm of a significant vessel. We identified and enhanced vessels within the 0.5 mm TOF MR venography image using a Hessian-based multiscale vesselness filter (Matlab command fibermetric), then binarized the result based on a threshold calculated from a fit 2-Gaussian Mixture (to isolate vessels from background noise). Finally, we dilated the vessels using a spherical structuring element with a radius of 3 mm (6 voxels), and inverted the image to produce a mask excluding vessel-adjacent voxels.
Resting-state event detection analyses
Request a detailed protocolTo dissociate the influence of DN vs. dATN drive on spontaneous interactions, we developed an event-triggered analysis of resting-state fMRI data. Because participants were not engaged in any task, resting-state data provide a neutral testbed for evaluating the directionality and spatial specificity of DN–dATN interactions. Building on prior findings that attention during rest naturally alternates between internally and externally oriented states (Christoff et al., 2009; Gonzalez-Castillo et al., 2021; Van Calster et al., 2017), thought to be supported by the DN and dATN, respectively, we reasoned that transient ‘neural events’ in these networks could reveal spontaneous DN- and dATN-driven communication.
We defined DN-driven events as transient high-activation periods in DN pRFs, and dATN-driven events as high-activation periods in positive-polarity dATN pRFs (Gordon et al., 2023; Mitra et al., 2015). Event detection was performed independently for each source pRF by taking each pRF’s z-scored resting-state time series and identifying TRs with unusually high activity. Specifically, we considered any time point with a z-value greater than 2.4 (i.e., 99.18 percentile) as a neural event. Results were comparable with a range of thresholds (2.1 < z < 2.9).
At each detected event, we extracted the resting-state time series of the 10 most retinotopically aligned pRFs in the target network – that is, the 10 with the smallest Euclidean distance in receptive-field center (x, y) – over a 13-TR peri-event window (–6 to +6 TRs). These time series were normalized to the mean of the first four pre-event TRs to allow comparisons across events and conditions. To evaluate the spatial specificity of these interactions, we compared responses in the matched pRFs to those in an ‘anti-matched’ pRFs set: the 10 most distant pRFs in the target network in terms of visual field preference (i.e., pRF center positions). Unlike prior analyses that used random sampling to define unmatched control pRFs, here we used a deterministic anti-match procedure to avoid the computational burden of repeated resampling, given the large number of events.
This procedure was repeated across all relevant source–target network pairs: DN-driven (with dATN as target), dATN-driven (with DN as target), and for both positive and negative DN pRFs. For each event type and match status, we computed the average response in the target pRFs at time zero (event peak). These values were entered into a linear mixed effects model with fixed effects of match status (matched vs. anti-matched) and event direction (DN- vs. dATN-driven), and with subject as a random effect. This allowed us to test whether retinotopic alignment modulates DN-dATN interactions during spontaneous events and whether such modulation differs for DN- vs. dATN-driven events.
Statistical tests
Request a detailed protocolStatistical analyses were implemented in MATLAB (Mathworks, Inc). Given the relatively small number of participants and large number of runs per subject, we used linear mixed effects models to account for both within- and between-subject variability, as described in the relevant sections above. For follow-up tests, we used paired-sample t-tests and corrected for multiple comparisons where appropriate. Corrected p-values are reported where appropriate.
Appendix 1
Results
Our main results establish that retinotopic coding structures spontaneous interactions between internally and externally oriented neural systems in the absence of task demands. However, high-level cortical areas generally associate into networks based on their functional domain. For example, within the visual system, brain areas with differing retinotopic preferences (e.g., the scene-selective areas on the lateral and ventral surfaces of the brain) (Silson et al., 2015; Silson et al., 2016a) nevertheless form functional networks based on their apparent preference for specific visual categories (e.g., faces, objects, or scenes) (Hutchison et al., 2014; Stevens et al., 2015). This raises a question: do retinotopic and domain-specific organizational principles interact to facilitate or constrain information flow across internally and externally oriented networks? Addressing this question would shed light on mechanisms that enable the brain to integrate information while maintaining functional specialization.
To address this question, we focused on the functional interplay between a set of areas in posterior cerebral cortex that are established models for mnemonic and visual processing in the domains of scene and face perception. Specifically, we considered the mnemonic lateral place-memory area (LPMA; Steel et al., 2021; Steel et al., 2023), an area on the brain’s lateral surface that is implicated in processing mnemonic information relevant to visual scenes at the border between DN-A and the dATNs (Figure 2A). We examined OPA and LPMA because these regions are maximally spatially dissociable (Steel et al., 2021; Steel et al., 2025; Steel et al., 2023) and can be localized independently using resting-state data (Silson et al., 2016b; Baldassano et al., 2016). We examined how LPMA activation co-fluctuates with the adjacent, scene-perception area ‘occipital place area’ on the brain’s lateral surface (OPA; Dilks et al., 2013; Hasson et al., 2003) compared to two face-perception regions on the lateral and ventral surfaces, the occipital and fusiform face areas (OFA and FFA; Kanwisher et al., 1997; Halgren et al., 1999). At a group level, these perceptual regions are situated within the dATNs and are at the same level of the visual hierarchy (Appendix 1—figure 1A), but they are differentially associated with the domains of scene (OPA) and face (OFA and FFA) processing, making them ideal model systems to examine the impact of domain specificity and retinotopic coding in organizing neural activity.
We first defined the mnemonic area LPMA in the NSD participants. To do this, we contrasted functional connectivity between the anterior and posterior halves of the parahippocampal place area using each participant’s resting-state data (Silson et al., 2016b; Baldassano et al., 2016; Appendix 1—figure 1B), which revealed a cluster in lateral parietal cortex with a similar topographic profile as LPMA based on a group analysis from our prior work (Steel et al., 2021; Appendix 1—figure 1C). Replicating our previous findings (Steel et al., 2024), this connectivity-defined LPMA had a higher concentration of robust negative pRFs compared to OPA (t(7) = 5.26, p = 0.002) (Appendix 1—figure 1D) and exhibited a similar lower visual field bias as OPA (OPA: 8/8, t(7) = 3.13, p = 0.016; LPMA: 7/8 participants, t(7) = 2.11, p = 0.07; OPA vs. LPMA: t(7)=0.441, p = 0.67) (Appendix 1—figure 1E).
We considered whether the retinotopic opponent dynamic we have previously shown in the domain of scenes (i.e., between −LPMA and +OPA pRFs) during perceptual and mnemonic tasks (Steel et al., 2024) was also present at rest (Appendix 1—figure 1F). As we observed for the overall negative DN and positive dATN pRFs, we found that −LPMA and +OPA pRFs are interlocked in a retinotopically grounded opponent interaction. We used the same matching and randomization procedure as we used for the DN.
Remarkably, we found that LPMA and OPA were interlocked in a retinotopically grounded opponent interaction (Appendix 1—figure 1G). Resting-state activity of −LPMA and +OPA pRFs was reliably negatively correlated after accounting for variance associated with +LPMA pRFs, and, critically, this negative correlation was stronger for matched compared with randomly matched pRFs (t(6) = 3.012, p = 0.024). We next tested whether this opponent dynamic was modified by functional domain (i.e., the scene memory area LPMA paired with the face perception areas FFA and OFA), and we observed no significant difference between the distribution of correlation values for matched and randomly matched pRFs between −LPMA and OFA (t(6) = 0.59, p = 0.57) and −LPMA and FFA (t(6) = 2.01, p = 0.09). However, when we compared the difference of matched and randomly matched pRFs across domains, we found no significant interaction (F(1,19) = 0.99, p = 0.33).
One possibility is that the quality of the retinotopic matching between pRFs in different areas might contribute to the within vs. across domain distinction in retinotopic coding. We checked whether there was a difference in the match quality between the top 10 best matched pRFs between connectivity-defined LPMA and OPA compared with the face perception areas FFA and OFA. For this analysis, we computed the Euclidean distance between all pRFs centers between regions, and we took the median of the best pRF center matches. For example, if a participant had 162 negative pRFs in LPMA and 90 pRFs in OPA, we found the top 10 best matched OPA pRFs for each LPMA pRF, and then took the median distance of these 1620 matches. This gave us the most representative distance for matched pRFs for each participant.
Overall, we found that the quality of pRF matching was better for within domain (LPMA to OPA) compared with across domains (LPMA to OFA and FFA) (negative pRFs: t(6) = 5.86, p = 0.001; positive pRFs: t(6) = 6.79, p < 0.001). This suggests that the quality of the matches does depend on the stimulus domain. This fits with the notion that visual response properties are inherited by upstream areas, and that these response properties depend in part on the nature of the stimuli within a domain.
These results suggest that retinotopic coding structures the interaction between brain areas if they share a functional domain, but more studies are needed to confirm whether this is specific to within-domain matching.
Methods
Defining functional regions of interest (PPA, OPA, OFA, and FFA)
We used the volumetric functional regions of interest provided with the NSD at 1.8 mm resolution. Specifically, we used the parahippocampal place area (PPA), OPA, iog-faces (referred to here as OFA), and pfus-faces (fusiform face area, FFA1) regions of interest. In brief, these regions were defined in the NSD using within-subject data collected from 6 runs of a multi-category visual localizer paradigm (Allen et al., 2021).
Defining functional region LPMA
Because the NSD did not include mnemonic localizers (Steel et al., 2021), we used the resting-state data to define the LPMA. Briefly, the LPMA is a region anterior to OPA and near the caudal inferior parietal lobule (Silson et al., 2016b; Baldassano et al., 2016) that selectively responds during recall of personally familiar places compared to other stimulus types. We have previously shown that OPA and LPMA are functionally linked and work jointly to process knowledge of visuospatial context out of view during scene perception.
Prior work has suggested that a mnemonic area linked to scenes on the lateral surface can be localized by comparing resting-state co-fluctuations of anterior vs. posterior PPA (aPPA and pPPA, respectively) (Silson et al., 2019; Silson et al., 2016b; Baldassano et al., 2016), and we adopted that approach here. We preprocessed the resting-state fixation fMRI data, runs 1 and 14 from NSD sessions 22 and 23, in all participants (prior to data exclusion) and extracted the average time series of aPPA, pPPA, aFFA, and pFFA. We used these time series as regressors in a general linear model, and we compared the beta-values from the aPPA and pPPA. We considered any voxels with a t-statistic >5 within the posterior parietal-occipital cortex on the lateral surface as an LPMA ROI. Across subjects, there was considerable overlap between this connectivity-defined area and a group-level LPMA defined based on our prior work (Figure 2).
Appendix 1—figure 1
Retinotopic coding structures the spontaneous interaction between functionally coupled mnemonic and perceptual areas during resting-state fMRI.
(A) Isolating functionally coupled internally and externally oriented brain areas within the DNs and dATNs. We identified brain areas that were specialized in two domains: scenes and faces processing. Specifically, we focused on the lateral place-memory area (LPMA; from Steel et al., 2021), white, a memory area in the domain of scene perception at the posterior edge of the DN-A (purple; from Du et al., 2024). We examined LPMA’s relationship to three different category-selective visual areas in the dATN (green; from Du et al., 2024), (1) the occipital place area (OPA; from Steel et al., 2021), an area within the domain of scene perception, along with (2) the occipital face area (OFA) and (3) the fusiform face area (FFA), two areas involved in the domain of face perception (white; from Rosenke et al., 2021). (B, C) We localized LPMA in all participants by contrasting the correlation in resting-state activity between anterior and posterior parahippocampal place area (PPA) (B). This yielded a region in lateral occipital–parietal cortex that overlapped with the LPMA defined in an independent group of participants (C). Consistent with prior work, the connectivity-defined LPMA had a greater concentration of negative pRFs compared to OPA (D), and exhibited a lower visual field bias to OPA (E), consistent with an opponent interaction between these areas during perception. (F) We assessed the influence of retinotopic coding on the interaction between negative pRFs in mnemonic and positive pRFs in perceptual areas using the same pRF matching and correlation procedure described above. We compared pRFs within functional domain (scene memory × perception – LPMA to OPA) as well as across domains (scene memory × face perception – LPMA to the OFA and FFA). (G) We found that LPMA and OPA were interlocked in a retinotopically grounded opponent interaction. Resting-state activity of −LPMA and +OPA pRFs was reliably negatively correlated after accounting for variance associated with +LPMA pRFs, and, critically, this negative correlation was stronger for matched compared with randomly matched pRFs (t(6) = 3.012, p = 0.024). We next tested whether this opponent dynamic was modified by functional domain (i.e., the scene memory area LPMA paired with the face perception areas FFA and OFA), and we observed no significant difference between the distribution of correlation values for matched and randomly matched pRFs between −LPMA and OFA (t(6) = 0.59, p = 0.57) and −LPMA and FFA (t(6) = 2.01, p = 0.09). However, when we compared the difference of matched and randomly matched pRFs across domains, we found no significant interaction (F(1,19) = 0.99, p = 0.33).
Data availability
All data is made publicly available via the Natural Scenes Dataset (https://naturalscenesdataset.org/). All code required to reproduce the primary results of this paper areis available in an Open Science Framework repository at https://doi.org/10.17605/OSF.IO/X2R4H.
-
Open Science FrameworkRetinotopic coding organizes the interaction between internally and externally oriented brain networks.https://doi.org/10.17605/OSF.IO/X2R4H
-
University of MinnesotaID naturalscenesdataset. Natural Scenes Dataset.
References
-
Comparing different stimulus configurations for population receptive field mapping in human fMRIFrontiers in Human Neuroscience 9:96.https://doi.org/10.3389/fnhum.2015.00096
-
A whole-brain topographic ontologyAnnual Review of Neuroscience 47:21–40.https://doi.org/10.1146/annurev-neuro-082823-073701
-
Two distinct scene-processing networks connecting vision and memoryeNeuro 3:ENEURO.0178-16.2016.https://doi.org/10.1523/ENEURO.0178-16.2016
-
The brain’s default network: updated anatomy, physiology and evolving insightsNature Reviews. Neuroscience 20:593–608.https://doi.org/10.1038/s41583-019-0212-7
-
A taxonomy of external and internal attentionAnnual Review of Psychology 62:73–101.https://doi.org/10.1146/annurev.psych.093008.100427
-
AFNI: software for analysis and visualization of functional magnetic resonance neuroimagesComputers and Biomedical Research, an International Journal 29:162–173.https://doi.org/10.1006/cbmr.1996.0014
-
Attention during natural vision warps semantic representation across the human brainNature Neuroscience 16:763–770.https://doi.org/10.1038/nn.3381
-
The occipital place area is causally and selectively involved in scene perceptionThe Journal of Neuroscience 33:1331–6a.https://doi.org/10.1523/JNEUROSCI.4081-12.2013
-
Parallel distributed networks dissociate episodic and social functions within the individualJournal of Neurophysiology 123:1144–1179.https://doi.org/10.1152/jn.00529.2019
-
Precision estimates of parallel distributed association networks: evidence for domain specialization and implications for evolution and developmentCurrent Opinion in Behavioral Sciences 40:120–129.https://doi.org/10.1016/j.cobeha.2021.03.029
-
Organization of the human cerebral cortex estimated within individuals: networks, global topography, and functionJournal of Neurophysiology 131:1014–1082.https://doi.org/10.1152/jn.00308.2023
-
A tripartite view of the posterior cingulate cortexNature Reviews. Neuroscience 24:173–189.https://doi.org/10.1038/s41583-022-00661-x
-
Dynamic content reactivation supports naturalistic autobiographical recall in humansThe Journal of Neuroscience 41:153–166.https://doi.org/10.1523/JNEUROSCI.1490-20.2020
-
How to interpret resting-state fMRI: ask your participantsThe Journal of Neuroscience 41:1130–1141.https://doi.org/10.1523/JNEUROSCI.1786-20.2020
-
Visuospatial coding as ubiquitous scaffolding for human cognitionTrends in Cognitive Sciences 26:81–96.https://doi.org/10.1016/j.tics.2021.10.011
-
Large-scale gradients in human cortical organizationTrends in Cognitive Sciences 22:21–31.https://doi.org/10.1016/j.tics.2017.11.002
-
The fusiform face area: a module in human extrastriate cortex specialized for face perceptionJournal of Neuroscience 17:4302–4311.
-
Compressive spatial summation in human visual cortexJournal of Neurophysiology 110:481–494.https://doi.org/10.1152/jn.00105.2013
-
Structuring knowledge with cognitive maps and cognitive graphsTrends in Cognitive Sciences 25:37–54.https://doi.org/10.1016/j.tics.2020.10.004
-
The brain’s default mode networkAnnual Review of Neuroscience 38:433–447.https://doi.org/10.1146/annurev-neuro-071013-014030
-
Two cortical systems for memory-guided behaviourNature Reviews. Neuroscience 13:713–726.https://doi.org/10.1038/nrn3338
-
Common blood flow changes across visual tasks: II. decreases in cerebral cortexJournal of Cognitive Neuroscience 9:648–663.https://doi.org/10.1162/jocn.1997.9.5.648
-
Scene-selectivity and retinotopy in medial parietal cortexFrontiers in Human Neuroscience 10:412.https://doi.org/10.3389/fnhum.2016.00412
-
Functional connectomics from resting-state fMRITrends in Cognitive Sciences 17:666–682.https://doi.org/10.1016/j.tics.2013.09.016
-
Scene perception and visuospatial memory converge at the anterior edge of visually responsive cortexThe Journal of Neuroscience 43:5723–5737.https://doi.org/10.1523/JNEUROSCI.2043-22.2023
-
Visual organization of the default networkCerebral Cortex 30:3518–3527.https://doi.org/10.1093/cercor/bhz323
-
The head direction signal: origins and sensory-motor integrationAnnual Review of Neuroscience 30:181–207.https://doi.org/10.1146/annurev.neuro.29.051605.112854
-
A role for the left angular gyrus in episodic simulation and memoryThe Journal of Neuroscience 37:8142–8149.https://doi.org/10.1523/JNEUROSCI.1319-17.2017
-
Topographic maps of visual space in the human cerebellumCurrent Biology 29:1689–1694.https://doi.org/10.1016/j.cub.2019.04.012
-
The organization of the human cerebral cortex estimated by intrinsic functional connectivityJournal of Neurophysiology 106:1125–1165.https://doi.org/10.1152/jn.00338.2011
-
The default mode network: where the idiosyncratic self meets the shared social worldNature Reviews. Neuroscience 22:181–192.https://doi.org/10.1038/s41583-020-00420-w
Article and author information
Author details
Caroline Robertson
Funding
National Institutes of Health (R01MH130529)
- Caroline Robertson
Neukom Institute for Computational Science, Dartmouth College
- Adam Steel
The funders had no role in study design, data collection, and interpretation, or the decision to submit the work for publication.
Acknowledgements
The authors would like to thank the authors of the Natural Scenes Dataset for making these data publicly available. This work was supported by an award from the National Institutes of Mental Health (R01MH130529) to CER. AS was supported by the Neukom Institute for Computational Sciences.
Ethics
All data used is from the Natural Scenes Dataset. Informed consent was acquired from participants in the Natural Scenes Dataset in accordance with protocols approved by the University of Minnesota Institutional Review Board. See Allen et al., 2022 for additional details.
Version history
- Preprint posted:
- Sent for peer review:
- Reviewed Preprint version 1:
- Reviewed Preprint version 2:
- Version of Record published:
Cite all versions
You can cite all versions using the DOI https://doi.org/10.7554/eLife.110234. This DOI represents all versions, and will always resolve to the latest one.
Copyright
© 2026, Steel, Angeli et al.
This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 408
- views
-
- 16
- downloads
-
- 0
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Retinotopic coding organizes the interaction between internally and externally oriented brain networks
eLife 15:RP110234.
https://doi.org/10.7554/eLife.110234.3
