Skip to main content
. 2016 Nov 16;36(46):11601–11618. doi: 10.1523/JNEUROSCI.0516-16.2016

Figure 12.

Figure 12.

The model can learn feedback from multiple landmarks (3 here). A, LMN tuning curves show a high degree of parallelism across quadrants. B, Illustrative sketch of hypothetical distributions of stimulus intensity for 3 different stimulus attributes across the entire field of view at a given location. Each Vis population could represent a salient single landmark or the intensity of some low-level visual features (e.g., a given color, brightness, texture).