Corticolimbic Oscillations in Fear Learning Open Access

Madsen, Teresa (2014)

Permanent URL:


Neural oscillations are thought to mediate efficient communication of related information across distant brain regions and contribute to the long term synaptic alterations that underlie memory. Recently, many neuropsychiatric disorders have been correlated with unique "spectral fingerprints," i.e., patterns of disruption in EEG or MEG recordings of neural oscillations. However, little is known about the role of these oscillations in normal emotion.

The functional anatomy of the corticolimbic system - the collection of brain regions most implicated in emotion and neuropsychiatric disorders - has been thoroughly studied within the context of Pavlovian fear conditioning in rodents. Here, the basolateral complex of the amygdala (BLA) and medial prefrontal cortex (mPFC) have been identified as critical nodes with distinct roles in the acquisition and extinction of fear memory.

Using multielectrode recording of LFPs in the mPFC and BLA of freely moving rats, we investigated functional interactions between these two regions during fear conditioning and extinction. Both regions displayed significantly increased power, and coherence between regions, in a sharply tuned delta/theta (2-6 Hz) band during successful fear acquisition and recall, as compared to baseline (before habituation tones). Throughout fear acquisition, the mid-gamma (45-60 Hz) range was significantly elevated over baseline in terms of mPFC power, BLA power, and coherence between the two regions. After the shock, there were dramatic increases in high gamma (60-90 Hz) power for the mPFC and in low gamma (30-45 Hz) power for the BLA, as well as mid-gamma within and between both regions.

Furthermore, our analysis of Cross-Frequency Coupling (CFC) in the mPFC demonstrates the presence of at least two distinct pairs of frequency bands for which the amplitude of the higher frequency oscillation (mid- or high gamma) is significantly modulated by the phase of the lower frequency oscillation (delta or theta).

Combined with previous evidence, these results suggest unique mechanisms and functions for each feature of this complex interplay of oscillations in fear learning, retrieval, and extinction. Future experiments are being designed to demonstrate causal links between patterned neural activity and emotional memory, and to translate the analytic approach for clinical use.

Table of Contents

1. Introduction

1.1. Motivation: Neuropsychiatric Disorders

1.2. Corticolimbic Processing of Emotion

1.2.1. Anatomy and Connectivity
1.2.2. Roles in Fear Conditioning and Extinction
1.2.3. Microcircuitry of the BLA

1.3. Neural Oscillations

1.3.1. Gamma Mechanisms & Functions
1.3.2. Theta Mechanisms & Functions
1.3.3. Cross-Frequency Coupling Mechanisms & Functions
1.3.4. Motivation and Emotion in Human EEG/MEG

1.4. Dysrhythmias / Oscillopathies

1.4.1. Schizophrenia
1.4.2. Anxiety Disorders

1.5. Corticolimbic Oscillations in Emotion

1.5.1. Delta (1.5-4 Hz) / Theta (4-10 Hz) Frequencies
1.5.2. Gamma (30-80 Hz) Frequencies
1.5.3. Cross-Frequency Coupling Knowledge Gap

2. Materials and Methods

2.1. Animals and Pre-training
2.2. Surgery
2.3. Fear Conditioning
2.4. Recording
2.5. Histology
2.6. Behavioral Analysis
2.7. Spectral Analysis of Local Field Potentials
2.8. Neural Correlates of Behavior
2.9. Cross-Frequency Coupling

3. Results

3.1. Histology
3.2. Behavior
3.3. Spectral Analysis

3.3.1. Habituation
3.3.2. Acquisition
3.3.3. Recall
3.3.4. Extinction

3.4. Neural correlates of behavior
3.5. Cross-frequency coupling

4. Discussion

4.1. Low Frequency (<30 Hz) Oscillations
4.2. Gamma (30-100 Hz) Oscillations
4.3. Cross-Frequency Coupling
4.4. Conclusions and Future Directions

5. References
6. Figures and Legends

Figure 1. Experimental protocol, histology, and fear behavior in response to each tone.
Figure 2. Cross-Frequency Coupling analysis flowchart.
Figure 3. Single trial example shows sharply tuned significant increase in delta power and significant decrease in theta power.
Figure 4. Normalized spectrograms show increases and decreases in power over time and across phases of fear learning - coherent between mPFC and BLA.
Figure 5. Significant spectral changes in response to fear conditioned tones.
Figure 6. Spectral correlates of freezing behavior during fear recall.
Figure 7. Magnitude and significance of cross-frequency coupling.
Figure 8. Phase preference of significant cross-frequency coupling.

About this Dissertation

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
  • English
Research field
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files