Texture Segregation Causes Early Figure Enhancement and Later Ground Suppression in Areas V1 and V4 of Visual Cortex

Jasper Poort, Matthew W. Self, Bram van Vugt, Hemi Malkki, Pieter R. RoelfsemaCerebral 2016 · DOI

TL;DR

This study reveals that our visual system doesn't just highlight what's important (figures) but actively suppresses what's not (backgrounds) to achieve figure-ground segregation. They found that figure enhancement happens first, followed by background suppression, and these processes appear to be distinct and rely on feedback from higher brain areas.

What Problem Does This Paper Address?

Understanding how our brains distinguish between a "figure" (the object of interest) and the "ground" (the background) is fundamental to visual perception. While we know neurons in visual cortex respond more strongly to figures, it's been unclear how this happens. Is it solely by boosting the signal from the figure, or does it also involve actively dampening the signal from the background? This paper aims to disentangle these two mechanisms, figure enhancement and ground suppression, and understand their timing and neural basis in early visual areas (V1 and V4).

Methods

The researchers used electrophysiology to record neuronal activity (multi-unit activity, MUA) from areas V1 and V4 in monkeys performing a texture segregation task. They presented stimuli where a texture-defined figure was either in the neuron's receptive field (RF), outside the RF (background), or a uniform texture (control). They also used current-source density (CSD) analysis to investigate synaptic inputs across different cortical layers. To analyze the timing of neural events, they fitted response curves and calculated latencies. They also used a "d-prime" measure to quantify how well neurons discriminated between stimulus conditions.

Datasets

Not discussed in the paper. The study used custom-designed visual stimuli presented to monkeys.

Key Insights and Ideas

  • Two distinct processes: Figure enhancement and ground suppression are not just two sides of the same coin; they appear to be independent processes with different temporal dynamics.
  • Timing is key: Figure enhancement consistently precedes ground suppression, suggesting a sequential processing stream.
  • Feedback is crucial: The laminar analysis of synaptic inputs (CSD) points towards a significant role for feedback connections from higher visual areas in both figure enhancement and ground suppression, particularly in superficial and deep cortical layers.
  • Layer-specific processing: While figure enhancement and ground suppression are strongest in superficial and deep layers of V1, they are weaker in the input layer 4, further supporting the role of feedback.
  • No simple correlation: The strength of figure enhancement in a neuron did not predict the strength of its ground suppression, reinforcing their independence.

Key Results

  • Figure enhancement in V1 and V4: Neuronal activity in both V1 and V4 showed an early enhancement when the RF was on a figure compared to a uniform texture. This enhancement occurred around 82 ms in V1 and 57 ms in V4 during the first stimulus period.
  • Ground suppression in V1 and V4: Following figure enhancement, neuronal activity was suppressed when the RF was on the background compared to a uniform texture. This suppression occurred around 137 ms in V1 and 133 ms in V4.
  • Layer-specific differences in V1: Both figure enhancement and ground suppression were significantly stronger in superficial and deep layers of V1 compared to layer 4.
  • CSD evidence for feedback: Figure enhancement was associated with increased synaptic sinks in feedback-recipient layers (1, 2, and 5), while ground suppression showed weaker inputs in these layers, suggesting feedback modulation.
  • Independence of processes: Correlations between the strength of figure enhancement and ground suppression were not significant in V1 and V4.

Limitations and Future Work

The authors acknowledge that while their findings strongly suggest feedback from higher visual areas, the exact pathways (e.g., V4 directly to V1 vs. via V2) and the specific higher areas involved in ground suppression remain to be fully elucidated. They also note that the study did not explicitly test the role of attention, though their results are compatible with attention shifting away from the ground. Future work could involve using more advanced techniques to directly manipulate specific neural projections and investigate the roles of different cell types (like interneurons) in these processes, particularly in primate models. They also suggest that future research could explore how these mechanisms integrate with attentional selection.

Summary updated

Abstract

Segregation of images into figures and background is fundamental for visual perception. Cortical neurons respond more strongly to figural image elements than to background elements, but the mechanisms of figure-ground modulation (FGM) are only partially understood. It is unclear whether FGM in early and mid-level visual cortex is caused by an enhanced response to the figure, a suppressed response to the background, or both.We studied neuronal activity in areas V1 and V4 in monkeys performing a texture segregation task. We compared texture-defined figures with homogeneous textures and found an early enhancement of the figure representation, and a later suppression of the background. Across neurons, the strength of figure enhancement was independent of the strength of background suppression.We also examined activity in the different V1 layers. Both figure enhancement and ground suppression were strongest in superficial and deep layers and weaker in layer 4. The current-source density profiles suggested that figure enhancement was caused by stronger synaptic inputs in feedback-recipient layers 1, 2, and 5 and ground suppression by weaker inputs in these layers, suggesting an important role for feedback connections from higher level areas. These results provide new insights into the mechanisms for figure-ground organization.
BibTeX
@inproceedings{poort2016texture,
  author = {Jasper Poort and Matthew W. Self and Bram van Vugt and Hemi Malkki and Pieter R. Roelfsema},
  title = {{Texture Segregation Causes Early Figure Enhancement and Later Ground Suppression in Areas V1 and V4 of Visual Cortex}},
  booktitle = {Cerebral},
  year = {2016},
  doi = {10.1093/cercor/bhw235},
}
Figure