Britni Crocker

Alumni Department Physiology of Cognitive Processes

Main Focus

Information theoretic analysis of electrophysiological signals
in the macaque prefrontal cortex

Neural Coding in the Lateral Prefrontal Cortex

Visual information from both dorsal and ventral streams of the cortex terminate in the lateral prefrontal cortex (lPFC), a cortical area thought to be involved in working memory, attentional control, and abstract higher-order thinking. In an effort to unravel how the lateral prefrontal cortex codes for visual information we use an information-theoretic approach to analyze neural activity in the lPFC during the presentation of dynamic visual stimuli. Studies in other brain areas have shown that local field potentials (LFP) convey information separate from and in parallel to the spiking activity of neurons [1,2]. Our aim is to determine how visual information is coded in the lPFC and whether that coding differs from the results reported in sensory areas.

Our Research Methods

We use multi-electrode array (MEA) recordings and functional imaging to study the fine details neural coding and spatial structure of visual information in the lPFC of the anesthetized macaque.  In particular, we record and analyze the LFP and spiking activity during the repeated presentation of dynamic visual stimuli (10 second video clips) of varying contrast.  Using the , we calculate the mutual information between these various neurophysiological signals and the visual stimuli.

Present Results and Next Steps

Our preliminary results have shown that LFP in the lPFC is modulated by the presentation of visual stimuli (see Figure 1), especially in the low frequencies (1-10 Hz).  Consistent with studies in other areas, the phase of the LFP signal carries more information than the amplitude – up to an order of magnitude in our data.  Interestingly, these two informative components of LFP have different spatial distributions; while LFP phase information is spread out relatively evenly across most channels, information in the LFP amplitude is the result of a few informative channels.  In addition to confirming these results, we are now analyzing the spiking activity and integrating our findings with results from functional imaging studies.

Figure 1:

Representative LFP signals from 2 different stimulus presentations: one with high contrast (left) and one with low contrast (right).  (Movie Contrast) The frame-by-frame movie contrast (green).  The high contrast movie has  larger fluctuations in contrast over time.  (Average Voltage) The voltage recorded by a representative channel averaged over trials (blue).  (Filtered Signal) The filtered signal from the channel in a limited frequency range, averaged over trials (red).  (Amplitude and Phase) LFP amplitude and phase in the 1-4 Hz range over time and trials for a representative channel.  Each horizontal line indicates a single trial.  For the amplitude, cool colors represent relative low amplitude and hot colors represent relative high amplitude.  For the phase, the spectrum from cold to hot colors indicates phase values between negative pi and pi.  Responses to the stimuli are more robust in the presentation of the high contrast movie than the low contrast movie and are more robust in the phase of the signal than in its amplitude.


(MPI for Biological Cybernetics, Tubingen, Germany)

(MPI for Biological Cybernetics, Tubingen, Germany)

(Italian Institute of Technology, Robotics, Brain and Cognitive Sciences Department, Genova, Italy)

(Baylor College of Medicine, Department of Neuroscience, Houston, U.S.A)

(MPI for Biological Cybernetics, Tubingen, Germany)


[1] Belitski, A, A Gretton, C Magri, Y Murayama, M A Montemurro, N K Logothetis, and S Panzeri. 2008. Low-Frequency Local Field Potentials and Spikes in Primary Visual Cortex Convey Independent Visual Information. J. Neurosci. 28, no. 22 (May 28): 5696-5709.

[2] Kayser, C, M A Montemurro, N K Logothetis, and S Panzeri. 2009. Spike-Phase Coding Boosts and Stabilizes Information Carried by Spatial and Temporal Spike Patterns. Neuron 61, no. 4 (2): 597-608.

Go to Editor View