Stable numerosity representations irrespective of magnitude context in macaque prefrontal cortex

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Primate Studies

Background:

• Prefrontal cortex (PFC) neurons encode information for cognitively demanding tasks.
• Tuning functions of PFC neurons adapt to context in learned discrimination tasks.

Purpose of the Study:

• Investigate if PFC neuron selectivity to visual numerosities changes with magnitude context.
• Determine how neuronal representations of 'natural' categories are affected by context.

Main Methods:

• Single-cell recordings from the lateral PFC of two monkeys performing a numerosity discrimination task.
• Comparison of neuronal activity during isolated numerosity tasks versus mixed-modality tasks (including line length and color).

Main Results:

• Numerosity tuning curves of selective PFC neurons were stable across different contexts.
• Preferred numerosity and tuning sharpness did not change whether tasks were pure or mixed.

Conclusions:

• Numerosity detectors in the PFC do not adapt to changing magnitude context.
• Neuronal representations of numerosity appear to use a stable 'labeled line' code.
• Natural categories like numerosity may have privileged representations unaffected by context, unlike learned categories.