Frontotemporal coordination predicts working memory performance and its local neural signatures

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Systems Neuroscience

Background:

• Neurons in sensory areas often encode working memory (WM) content via spiking activity.
• However, this neural activity rarely correlates with actual behavioral performance in WM tasks.

Purpose of the Study:

• To investigate the relationship between neural activity in the inferotemporal (IT) and Frontal Eye Field (FEF) areas and behavioral performance during an object WM task.
• To identify specific neural mechanisms, beyond simple spiking or local field potentials (LFPs), that predict successful WM maintenance.

Main Methods:

• Recorded spiking activity and LFPs from IT and FEF neurons during an object WM task.
• Analyzed object-selective activity in IT neurons and spatially selective activity in FEF neurons.
• Examined the phase-locking of IT spikes and LFPs to the beta band oscillations of FEF LFPs.

Main Results:

• Spiking activity and LFPs within IT and FEF were poor predictors of behavioral performance.
• Phase-locking between IT spikes/LFPs and the beta band of FEF LFPs robustly predicted successful WM maintenance.
• IT neurons showed enhanced object-selective persistent activity when their spikes were phase-locked to FEF LFPs.

Conclusions:

• Neural coordination, specifically phase-locking between IT and FEF, is crucial for successful visual information maintenance during WM.
• This inter-areal coordination, rather than local activity alone, underlies effective working memory performance.