Tame the fear or lure for extra reward: Amygdala holds the central key
Fear conditioning is a valuable behavioral paradigm for studying the neural basis of emotional learning and memory. Given that the neural circuit underlying fear conditioning as well as aversive emotional memory has been implicated in emotional disorders in humans, the molecular mechanisms of fear conditioning are potential targets of psychotherapeutic drug development. Amygdala is involved in processing emotional memory, particularly aversive emotional memories of fear. Is it amygdala exclusively process memories of fear and reward related behaviors? Studies on memory suggest that multiple cellular mechanisms are required to integrate information after a new experience is perceived. The idea of distributed circuits, where neural control of one behavior is accomplished by multiple circuits in different regions of the brain suggests that there are specific brain networks that sub serve motivation and emotions and that both function and adaptation (plasticity) within the networks are enabled by extra cellular and intracellular molecular signaling. Fear conditioning is a form of associative learning where neutral conditioned stimulus (CS) such as auditory tone is paired with an aversive unconditioned stimulus (US) such as foot shock, causing the animals to illicit expression of fear by freezing while it listens the neutral auditory tone. Amygdala consists of multiple interconnected nuclei, which can be further divided into basolateral complex, BLA (Cell groups can be classified into lateral, basal and basal medial lobes) and Central nucleus of amygdala, CeA (lateral, CeL and medial, CeM). Information from thalamus and sensory cortices projects to the lateral amygdala (LA), which in turn projects to the basal, medial as well as to the CeA. BLA is reciprocally connected with prefrontal cortices and hippocampus however CeA also receives directly information from sensory cortices. Using the combination of optogenetics with cell type and projection type manipulations it has been shown that LA has been involved in the acquisition of fear memory, though there is evidence that CeL is also a locus for fear acquisition whilst CeA has been implicated in expression of fear memories. Due to space constraints, I will not be able to enlist the cell type specific projections from the different lobes of amygdala as well as their output projections from distinct nuclei of amygdala to hippocampus, entorhinal cortex, prefrontal cortex, midbrain dopaminergic regions and the causality of translating their activities to different behavioral output ranges from feeding, anxiety, fear, reward based behavior to name a few. You can find some excellent review articles on the distributed circuits controlling fear acquisition, expression and extinction from the literature.
Recently one of the emerging themes in amygdala research is that it can encode both positive (reward) and negative (aversive) valence, thereby informing the situation of the animal at a given moment. In essence, there exists distinct class of amygdalar neurons, which responds exclusively to reward or fear. For instance, it has shown that the stimulation of the projections from BLA to nucleus accumbens causes positive reinforcement learning for a reward while the projections from BLA to CeM causes negative reinforcement learning for fear conditioning. Different regions in amygdala are proposed to elicit different function such as discriminating among multiple outcomes of similar valences or involved in the significance of motivation to reach an outcome. To switch between different behavioral states, one of the best ways to operate will be through inhibition between parallel competing networks. Long-range projections into amygdala can suppress or activate opposing networks (this can be changing the synaptic weights) involved in distinct behavioral states.
I will try to outline the interesting questions that lurk around the immediate corner to answer to understand the structural and functional organization of the neuronal circuits of fear learning in the amygdala and the circuits in the CNS that governs the control of fear learning and expression. In a general perspective the question arises are a) How neurons integrate information from functionally different pathways? b) How individual representations were co-related temporally from functionally distinct pathways impinging on individual neurons, which is part of the microcircuits?
Some of the specific questions, which I believe, will be interesting to address to learn the functional organization of fear or reward circuits are enumerated below:
1. One of the emerging concepts is that both CS and US causedisinhibition of projection neurons by
enhancing associativelearning.This is by recruiting VIP interneurons, which provides phasic
inhibition to paravalbumin (PV) and stomatostatin (SOM) neurons, causing the disinhibition of
the somatodendritic regions of the projection neurons in the auditory cortex thereby honing the
stimulus induced activity. This partially explains the plasticity rules governed by fear
learning and how it changes the function ofmicrocircuits however we still don’t know how
and what kind of restructuring or re-wiring of the auditory circuit does occur during
learning. To note: PV also inhibits SOMs thus by creating disinhibition of projection neurons
during CS. (In my next blogs, I will describe how different types of interneurons modulate
information processing in projection neurons keeping the excitation/inhibition in balance which
causes the generation and maintenance of network oscillations culminating in transferring
information between different brain regions.
2. What exactly are the US pathways to the amygdala, especially to the LA (Lateral nucleus of the
amygdala) and BA (Basal nucleus of the amygdala) as most studies in the past have focused only
on the CS pathways?
3. How do prediction error signals fit together with fear conditioning /extinction? It had been
shown that the neurons which are involved in fear acquisition in CeL can send projections
directly to the periaqueductal gray (PAG). Does PAG acts as a prediction errorcenter? PAG is
involved in transmission in cerebellar and spinal- motor reflex pathways coordinating sensory
input and motor output.It might be possible that PAG prepare the animals to react to aversive
stimuli and might encode error prediction signal thereby thwarting the need to be alert. In
addition to it, there is also a dopamine prediction error in reward learning and addiction.
4. How does activity in the amygdala drive changes upstream, e.g. in thalamus? On the same line,
what are the down stream interactions between Amygdala – Hippocampus – Prefrontal cortex?
This makes sense in the light of recent findings proposing that the associative learning
on decision-making and the experience dependent interaction-entraining oscillations between the
above mentioned regions during fear learning and memory.
5. Are there any convergent inputs from LA efferents and contextual inputs on to the same neurons in
BA? Quantifying convergent – divergent connectivity is a crucial factor that determines the
optimal for information transfer between two networks. In general, in neural
systems there is a dynamic interaction between network structure and network activity as shaped
by the activity dependent plasticity rules.
6. What are the effects of interneurons on fear and extinction cells to understand the evolving
firing patterns of neurons and the acquiring activity in population level?
7. BLA and CeA neurons in amygdala encode reward and fear. These distinct functional modalities can
be attributed to their specific connectivity both in terms of specific input to this neurons and
the output to their targets. How do structural features of neuronal wiring influence or determine
the emerging neuronal function, both at the level of single – cells and of microcircuits?
What is the functional mportance of each input, with reference to the input – specific
plasticity in individual synapses? Besides functional mapping of the projection to and fro from
amygdala will elucidate circuits involved in reward-based learning.
Manipulating specific neuron types and afferent/efferent projections using optogenetics combining with viral tracings puts us in a vantage position to understand the functional basis of the connectivity in fear circuits in different sub nuclei in amygdala and how expression of fear or reward changes the activity distribution of the population of neurons in different regions of the brain as well as behavior of the animal. Taming the fear and carve for the extra reward is not that far reach.
To note: The header image was downloaded from google images of smithsonian magazine.
1. A circuit mechanism for differentiating positive and negative assosciations – Nature, 2015.
2. Encoding of fear learning and memory in distributed neuronal circuits – Nat Neuroscience,
3. Amygdala interneuron subtypes control fear learning through disinhibition – Nature, 2014.
4. Basomedial amygdala mediates top-down control of anxiety and fear – Nature, 2015.
5. A causal link between prediction errors, dopamine neurons and learning – Nat Neuroscience,
6. Context dependent encoding of fear and extinction memories in a large-scale network model of
the basal amygdala – Plos Computational Biology, 2011.
7. A disinhibitory microcircuit for associative fear learning in the auditory cortex – Nature,