Large-scale dissociations between views of objects, scenes, and reachable-scale environments in visual cortex

Emilie Josephs, Talia Konkle

TL;DR

This study identifies a distinct set of brain regions in the visual cortex that are specifically tuned to "reachable-scale" environments, like countertops or workbenches. These regions show unique activation patterns compared to those that process large navigable scenes or isolated objects, suggesting a specialized neural representation for our immediate, interactive surroundings.

What Problem Does This Paper Address?

We know our brains process large-scale environments (like navigating a city) differently from small, isolated objects (like a single cup). However, much of our daily visual experience falls in between – think of your kitchen counter or a desk. These "reachable-scale" environments are not navigable like a scene, but they contain multiple objects and are crucial for everyday actions. The question is: how does the brain represent these intermediate environments? Do they simply fall under existing scene or object processing, or do they require a dedicated neural system? This paper aims to resolve this by investigating the neural basis of reachable-scale environment perception.

Methods

The researchers employed functional magnetic resonance imaging (fMRI) to observe brain activity in human participants. They designed experiments to present participants with three types of visual stimuli:

  • Objects: Close-up views of single objects.
  • Reachspaces: Views of environments within arm's reach, typically containing multiple objects.
  • Scenes: Views of large, navigable environments.

They used a "preference mapping" technique to identify brain regions that responded more strongly to one category of stimuli over others. Additionally, they conducted control experiments to ensure their findings weren't due to simple low-level visual features like luminance or contrast. They also performed "fingerprint profile" analyses to understand the broader response characteristics of these newly identified regions.

Datasets

The study utilized custom-created image datasets for each stimulus category (objects, reachspaces, scenes). These datasets were carefully curated and controlled for various factors, including semantic categories and image properties. No new benchmarks were introduced; rather, existing categories of visual stimuli were used and compared.

Key Insights and Ideas

  • A "Third Way" of Visual Processing: The most significant finding is the identification of distinct brain regions that are not simply subdivisions of scene or object processing areas, but rather represent a separate system for processing reachable-scale environments.
  • Sensitivity to Multiple Objects: The newly identified "reachspace-preferring" regions show a particular sensitivity to the presence of multiple, individuated objects within the environment, more so than to spatial layout alone. This is a key differentiator from scene-processing regions.
  • Intermediate Activation in Classic Regions: While dedicated reachspace regions exist, classic scene- and object-processing areas (like PPA and LO) show intermediate activation levels for reachspace stimuli, suggesting some degree of integration or overlap.
  • Peripheral Bias: Reachspace-preferring regions, much like scene-preferring regions, exhibit a preference for stimuli presented in the visual periphery, aligning with their role in processing larger spatial contexts.

Key Results

  • Three distinct brain regions consistently showed a preference for reachable-scale environments: the posterior collateral sulcus (ventral ROI), the occipito-parietal cortex (opROI), and the superior parietal lobule (spROI).
  • These reachspace-preferring regions showed significantly higher activation to reachspace images compared to both scene and object images.
  • Control experiments confirmed that these responses were not driven by low-level image features like luminance or contrast.
  • In classic scene- and object-preferring regions, reachspaces elicited intermediate levels of activation, falling between responses to scenes and objects.
  • Reachspace-preferring regions showed a strong response to stimuli featuring multiple objects, a pattern distinct from both scene and object regions.

Limitations and Future Work

The authors acknowledge that the precise functional role of these reachspace-preferring regions is still being elucidated. While they are sensitive to multiple objects, they also process spatial layout. Future work is needed to disentangle the contributions of object content versus spatial configuration. They also suggest extending theoretical frameworks to incorporate this new category of visual information processing. Further research will explore how these regions interact with action-based processing and how they contribute to our understanding of the physical world.

Summary updated

Abstract

Space-related processing recruits a network of brain regions separate from those recruited in object processing. This dissociation has largely been explored by contrasting views of navigable-scale spaces to views of close-up, isolated objects. However, in naturalistic visual experience, we encounter spaces intermediate to these extremes, like the tops of desks and kitchen counters, which are not navigable but typically contain multiple objects. How are such reachable-scale views represented in the brain? In three human functional neuroimaging experiments, we find evidence for a large-scale dissociation of reachable-scale views from both navigable scene views and close-up object views. Three brain regions were identified that showed a systematic response preference to reachable views, located in the posterior collateral sulcus, the inferior parietal sulcus, and superior parietal lobule. Subsequent analyses suggest that these three regions may be especially sensitive to the presence of multiple objects. Further, in all classic scene and object regions, reachable-scale views dissociated from both objects and scenes with an intermediate response magnitude. Taken together, these results establish that reachable-scale environments have a distinct representational signature from both scene and object views in visual cortex.
BibTeX
@inproceedings{josephs2020large,
  author = {Emilie Josephs and Talia Konkle},
  title = {{Large-scale dissociations between views of objects, scenes, and reachable-scale environments in visual cortex}},
  booktitle = {Proceedings of the National Academy of Sciences},
  year = {2020},
  doi = {10.1073/pnas.1912333117},
}
Figure