Fig. 1.

Possible relationships between selectivity across shape space. Consider 2 inferior temporal (IT) neurons: the first neuron is a sparse firing neuron that responds to 2 distinct stimuli (left). The second neuron is a distributed firing neuron that responds to 4 distinct stimuli (middle). Each row represents a scenario that shows how each neuron might respond to small variations in the neighborhood of its preferred stimuli. Top row: Scenario 1: selectivity is heterogeneous. Neuron 1 is sharply tuned to changes in stimulus 1 (stim1) and broadly tuned to variations around stimulus 2, whereas neuron 2 is sharply tuned around both stimuli. According to this scenario, the tuning of a neuron depends on how well stimuli match the preferred features of the neuron and is therefore heterogeneous with no overall constraint. This predicts no correlation across neurons between their tuning widths in the neighborhood of the 2 stimuli. Bottom row: Scenario 2: selectivity has an intrinsic component. Neuron 1 shows consistently sharp tuning to variations around all stimuli, whereas neuron 2 shows consistently broad tuning to variations around both stimuli. In other words, selective neurons respond to fewer stimuli and are narrowly tuned in the local neighborhood of each stimulus, whereas less-selective neurons respond to many stimuli and are broadly tuned to local variations of each stimulus. This predicts a positive correlation across neurons between their tuning widths across stimuli. This possibility imposes no constraint on the features preferred by each neuron but rather, constrains the sharpness of tuning in the neighborhood of each feature.