About

Dr. Adam Steel joined the Department of Psychology and the Beckman Institute at the University of Illinois Urbana-Champaign as an Assistant Professor in January 2025. His research investigates how the human brain represents and processes visual information, with a particular focus on scene perception, spatial memory, and the interaction between these cognitive processes. He employs a combination of behavioral studies, neuroimaging, and computational approaches to explore these topics. Dr. Steel completed his postdoctoral training at Dartmouth College, where he studied the neural mechanisms of scene perception and memory with Dr. Caroline Robertson. His research has contributed to fundamental principles about how the brain organizes visual information and has been published in leading journals including Nature Neuroscience, Nature Communications, and Current Biology. Prior to his postdoctoral work, he earned his D.Phil. in Biomedical Sciences from the University of Oxford as an NIH-Oxford/Cambridge Scholar, working with Dr. Chris Baker and Professor Charlotte Stagg. He also holds a BA with honors in Neuroscience and English composition from Vassar College.

Research topics

• Computer Science
• Neuroscience
• Psychology
• Computer vision
• Chemistry
• Cognitive psychology
• Endocrinology
• Internal medicine
• Biochemistry
• Biology
• Medicine

Selected publications

• Relating Scene Memory and Perception Activity to Functional Properties, Networks, and Landmarks of Posterior Cerebral Cortex—A Probabilistic Atlas

  Journal of Neuroscience · 2025-04-29 · 8 citations

  articleOpen access1st authorCorresponding

  Adaptive behavior in complex environments requires integrating visual perception with memory of our spatial environment. Recent work has implicated three brain areas in posterior cerebral cortex-the place memory areas (PMAs) that are anterior to the three visual scene perception areas (SPAs)-in this function. However, PMAs' relationship to the broader cortical hierarchy remains unclear due to limited group-level characterization. Here, we examined the PMA and SPA locations across three fMRI datasets (44 participants, 29 female). SPAs were identified using a standard visual localizer where participants viewed scenes versus faces. PMAs were identified by contrasting activity when participants recalled personally familiar places versus familiar faces (Datasets 1-2) or places versus multiple categories (familiar faces, bodies, and objects, and famous faces; Dataset 3). Across datasets, the PMAs were located anterior to the SPAs on the ventral and lateral cortical surfaces. The anterior displacement between PMAs and SPAs was highly reproducible. Compared with public atlases, the PMAs fell at the boundary between externally oriented networks (dorsal attention) and internally oriented networks (default mode). Additionally, while SPAs overlapped with retinotopic maps, the PMAs were consistently located anterior to mapped visual cortex. These results establish the anatomical position of the PMAs at inflection points along the cortical hierarchy between unimodal sensory and transmodal, apical regions, which informs broader theories of how the brain integrates perception and memory for scenes. We have released probabilistic parcels of these regions to facilitate future research into their roles in spatial cognition.

  PublisherOA PDFDOI

• Anterior shift for visual recall vs. perception is specific to scenes

  Journal of Vision · 2025-07-15

  articleOpen access

  Although prior work proposed that visual recall reactivates visual category-selective brain regions, recent research shows that recalling familiar scenes recruits areas anterior and adjacent to visual scene-selective regions (Steel et al., 2021). Here we examined whether the anterior shift for visual recall vs. perception is unique to scenes, or if this occurs for other visual categories (e.g., faces, bodies, objects) using fMRI. We scanned participants (N=18) performing a visual recall and a visual perception paradigm. Before the visual recall task, participants provided a list of personally familiar faces, places, objects, and body parts (5 examples per category), which they vividly recalled during the scan. During the visual perception task, participants viewed dynamic videos from each category. For both paradigms, we contrasted activity during place-, face-, and body trials against object-activity to identify category-specific activation clusters and intersected this activity with locations of eight functional areas defined from publicly available atlases: parahippocampal (PPA), occipital (OPA) place areas, occipital (OFA) and fusiform face areas (FFA1, FFA2), and extrastriate body area subregions (ITG, LOS, MTG). Scene areas had a higher percentage of significant voxels during recall compared to body and face areas (all p-values ≤ 0.001); these recall activity clusters were larger and more contiguous in scene areas than in other areas. As predicted, all scene areas showed a significant anterior shift in peak activity for recall versus perception (all p-values < 0.001). Crucially, no anterior shift was observed for any face or body areas (all p-values ≥ 0.05). Taken together this suggests that scenes have distinct yet adjacent areas for perception and memory, but other categories do not. The unique anterior shift for scenes may reflect the need to represent contextual information outside the current field of view during navigation.

  PublisherDOI

• Relating scene memory and perception activity to functional properties, networks, and landmarks of posterior cerebral cortex - a probabilistic atlas

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-01-06 · 2 citations

  preprintOpen access1st authorCorresponding

  Abstract Adaptive behavior in complex environments requires integrating visual perception with memory of our spatial environment. Recent work has implicated three brain areas in posterior cerebral cortex — the place memory areas (PMAs) that are anterior to the three visual scene perception areas (SPAs) – in this function. However, PMAs’ relationship to the broader cortical hierarchy remains unclear due to limited group-level characterization. Here, we examined the PMA and SPA locations across three fMRI datasets (44 participants, 29 female). SPAs were identified using a standard visual localizer where participants viewed scenes versus faces. PMAs were identified by contrasting activity when participants recalled personally familiar places versus familiar faces (Datasets 1-2) or places versus multiple categories (familiar faces, bodies, and objects, and famous faces; Dataset 3). Across datasets, the PMAs were located anterior to the SPAs on the ventral and lateral cortical surfaces. The anterior displacement between PMAs and SPAs was highly reproducible. Compared to public atlases, the PMAs fell at the boundary between externally-oriented networks (dorsal attention) and internally-oriented networks (default mode). Additionally, while SPAs overlapped with retinotopic maps, the PMAs were consistently located anterior to mapped visual cortex. These results establish the anatomical position of the PMAs at inflection points along the cortical hierarchy between unimodal sensory and transmodal, apical regions, which informs broader theories of how the brain integrates perception and memory for scenes. We have released probabilistic parcels of these regions to facilitate future research into their roles in spatial cognition. Significance statement Complex behavior requires the dynamic interplay between mnemonic and perceptual information. For example, navigation requires representation of the current visual scene and its relationship to the surrounding visuospatial context. We have suggested that the place memory areas, three brain areas located anterior to the scene perception areas in visual cortex, are well-positioned to serve this role. Here, in a large group of participants, we show that the place memory areas are robustly localizable, and that their position at the interface of multiple distributed brain networks is uniquely suited to mnemonic-perceptual integration. We have released probabilistic regions-of-interest and localization procedure so that others can identify these areas in their own participants.

  PublisherOA PDFDOI

• Characterizing the probable location of scene perception and place memory areas along the cortical hierarchy – a publicly available resource

  Journal of Vision · 2025-07-15

  articleOpen access1st authorCorresponding

  The human brain must integrate visual perception with spatial memory for effective navigation. Recent work has identified place memory areas (PMAs) that process remembered spatial information, positioned anterior to scene perception areas (SPAs) that analyze visual scenes. However, the relationship of the PMAs to the broader cortical hierarchy remains unclear due to limited group-level characterization. We examined PMA and SPA locations across three fMRI datasets (44 participants) that used different acquisition parameters. We identified SPAs in all participants using a standard visual localizer where participants viewed scenes versus faces, and we localized PMAs using a memory task where participants recalled personally familiar places versus familiar faces (Datasets 1-2) or places versus multiple categories (familiar faces, bodies, and objects, and famous faces; Dataset 3). We found PMAs could be consistently localized across datasets and maintained a systematic anterior position relative to SPAs. The relative displacement between PMAs and SPAs was highly reproducible, suggesting a fundamental feature of cortical organization. Group analyses revealed PMAs fall at the boundary between externally-oriented networks (dorsal attention) and internally-oriented networks (default mode), and at an inflection point along the cortical hierarchy between unimodal sensory and amodal, apical regions. Additionally, while SPAs overlapped with retinotopic maps, PMAs were consistently located anterior to mapped visual cortex. These results suggest PMAs represent a transition zone between perceptual and mnemonic systems. We have released probabilistic parcels of these regions to facilitate future research into their roles in spatial cognition.

  PublisherDOI

• Examining the retinotopic organization of the visual hierarchy in autistic and neurotypical individuals

  Journal of Vision · 2024-09-15

  articleOpen access

  As information ascends the visual hierarchy, smaller receptive fields that are selective for simple visual features combine to form larger ones that are selective for more complex features. Previous research demonstrates different receptive field sizes in lower levels of the autism visual hierarchy, which vary with symptom severity (Schwarzkopf et al., 2014). In this study, we examined whether such group differences propagate to the category-selective levels atop the visual hierarchy. We used population receptive field (pRF) analysis to test whether pRF sizes are atypical in both lower (early visual cortex, EVC) and higher levels (parahippocampal place area, PPA; occipital place area, OPA) of the visual hierarchy in individuals with autism spectrum conditions (ASC; n=6) compared to neurotypical controls (n=17). fMRI participants viewed traversing bar stimuli of scene fragments (4 orientations x 2 directions, each 36s, diameter=11.4°) and a traditional category localizer rapidly presenting blocks of scene, face, and object images. In both groups, pRF sizes grew with eccentricity in the EVC (one-way ANOVA, ASC p<0.001, Con p<0.001) with no interaction (Diagnosis*Eccentricity p>0.05). From EVC to category-selective areas, pRF sizes increased in both groups. In both groups, the size of pRFs in PPA was larger than those in EVC (two-sample t-test, ASC t(10)=-4.43, p<0.01, Con t(32)=-9.06, p<0.001). However, in both PPA and OPA (permutation-test, p<0.05) the ASC group had a smaller pRF size than that of controls, when comparable in EVC. These results show that the fine-grained visual architecture in autism differs from that of controls in category-selective areas. Specifically, we observe smaller pRFs in category-selective scene areas. Future work should explore pRF sizes in other category-selective areas, and how these relate to autistic sensory and cognitive traits.

  PublisherDOI

• Retinotopic coding organizes the interaction between internally and externally oriented brain networks

  bioRxiv (Cold Spring Harbor Laboratory) · 2024-09-27

  preprintOpen access1st authorCorresponding

  The human brain seamlessly integrates internally generated thoughts with incoming sensory information, yet the large-scale networks that support these functions -- the (Default Network, DN) and external (Dorsal Attention Network, dATN) -- are traditionally viewed as functionally antagonistic. This raises a crucial question: how does the brain integrate information across these seemingly opposed systems? Here, using densely sampled 7 T 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. Although spontaneous DN and dATN activity during rest is uncorrelated at the network level, 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 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 of elevated top-down drive from the DN to the dATN. These findings challenge the prevailing view of global DN-dATN antagonism, revealing a latent, voxel-level architecture of retinotopically-grounded interactions. Taken together, our results suggest that retinotopic coding underpins the dynamic coordination of perception and thought in the human brain.

  PublisherDOI

• Motor hand area GABA could be a physiological switch between motor network and default mode network connectivity

  Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2024-11-26

  article

  Motivation: The functional role of Inhibitory tone in the motor cortex is not completely understood. Previous work has been limited by the coarse spatial resolution in single-voxel spectroscopy. Goal(s): We applied a novel high spatial resolution MR spectroscopic imaging technique to test the relationship between inhibitory tone and motor network (MN) connectivity. Approach: We performed voxel-wise analysis of neurochemical data to correlate measures of inhibitory and excitatory tone with age as a confounder, MN and default mode network (DMN). Results: In the motor hand areas, we demonstrated a reciprocal correlation of inhibitory tone with MN and DMN connectivity. Inhibitory tone could &amp;ldquo;switch&amp;rdquo; node connectivity. Impact: The connectivity of key motor network nodes could be influenced by their inhibitory tone and the excitation&amp;ndash;inhibition difference. This finding advances the understanding of motor network function and could be a target for modulation in clinical settings.

  PublisherDOI

• Memory-based predictions prime perceptual judgments across head turns in immersive, real-world scenes

  Current Biology · 2024-12-17 · 4 citations

  articleOpen access
  PublisherOA PDFDOI

• Inverted visual coding across category-selective visual areas

  Journal of Vision · 2024-09-15

  articleOpen access1st authorCorresponding

  Traditional models of brain function propose a shift from retinotopic to amodal, abstract coding as visual information progresses anteriorly from visual cortex towards memory structures. However, recent evidence challenges this conception, suggesting that memory-related brain areas implement a retinotopic code characterized by spatially-selective negative responses during population receptive field modeling (-pRFs), and this code structures interactions among category-selective brain areas involved in scene perception and memory (Steel*, Silson* et al., 2023, Nat. Neuro.). Here, we investigated whether -pRFs are present within or anterior to other visual-category preferring areas (beyond scene areas), or whether -pRFs uniquely appear in regions specialized for scene memory. We computed pRFs for all subjects in the Natural Scenes Dataset (Allen et al., 2022) and compared -/+ pRF concentrations in category-preferring regions in ventral temporal cortex for scenes (anterior & posterior PPA), faces (iOG-, pFus-, mFus-faces), bodies (FBA-1 & 2), and words (OWFA, VWFA-1 & 2). Importantly, for scenes, we replicated our previous observation: -pRFs were preferentially concentrated in anterior versus posterior PPA (p<0.005), and the lateral place memory area (individually localized using resting-state fMRI) compared to the scene perception area OPA (p<0.001). Next, we examined the prevalence of -pRFs in and anterior to other category-selective visual areas. Face- and body-selective areas exhibited no differences in -pRF concentration between posterior and anterior functional regions (all ps > 0.09). Interestingly, for word-selective areas, the concentration of -pRFs increased up the processing hierarchy from OWFA to VWFA-1 and 2 (ps < 0.05). Crucially, the visual field preferences of -/+ pRFs in word-preferring areas were well-matched, supporting the notion of functional linkage. We propose that the -pRFs associated with visual areas may serve visual functions that demand perceptual-mnemonic interaction across the visual field such as navigation and reading.

  PublisherDOI

• A retinotopic code structures the interaction between perception and memory systems

  Nature Neuroscience · 2024-01-02 · 50 citations

  articleOpen access1st authorCorresponding
  PublisherOA PDFDOI

Frequent coauthors

• Edward H. Silson

  University of Edinburgh

  69 shared

• Caroline E. Robertson

  Dartmouth Hospital

  66 shared

• Chris I. Baker

  National Institutes of Health

  64 shared

• Charlotte J. Stagg

  MRC Brain Network Dynamics Unit

  44 shared

• Uzay E. Emir

  38 shared

• Brenda D. Garcia

  Dartmouth College

  34 shared

• Anna Mynick

  Dartmouth Hospital

  24 shared

• Stamatios N. Sotiropoulos

  University of Nottingham

  24 shared

Education

• DPhil Biomedical Sciences, Nuffield Department of Medicine

  University of Oxford

  2019

• B.A., Behavioral Neuroscience

  Vassar College

  2012