Orbitofrontal cortex
The orbitofrontal cortex is a prefrontal cortex region in the frontal lobes of the brain which is involved in the cognitive process of decision-making. In non-human primates it consists of the association cortex areas Brodmann area 11, 12 and 13; in humans it consists of Brodmann area 10, 11 and 47.
The OFC is functionally related to the ventromedial prefrontal cortex. Therefore, the region is distinguished due to the distinct neural connections and the distinct functions it performs. It is defined as the part of the prefrontal cortex that receives projections from the medial dorsal nucleus of the thalamus, and is thought to represent emotion, taste, smell and reward in decision-making. It gets its name from its position immediately above the orbits in which the eyes are located. Considerable individual variability has been found in the OFC of humans. A related area is found in rodents.
Structure
The OFC is divided into multiple broad regions distinguished by cytoarchitecture, including Brodmann area 47/12, Brodmann area 11, Brodmann area 14, Brodmann area 13, and Brodmann area 10. Four gyri are split by a complex of sulci that most frequently resembles a "H" or a "K" pattern. Extending along the rostro-caudal axis, two sulci, the lateral and orbital sulci, are usually connected by the transverse orbital sulcus, which extends along a medial-lateral axis. Most medially, the medial orbital gyrus is separated from the gyrus rectus by the olfactory sulcus. Anteriorly, both the gyrus rectus and the medial part of the medial orbital gyrus consist of area 11, and posteriorly, area 14. The posterior orbital gyrus consists mostly of area 13, and is bordered medially and laterally by the anterior limbs of the medial and lateral orbital sulci. Area 11 makes up a large part of the OFC involving both the lateral parts of the medial orbital gyrus and the anterior orbital gyrus. The lateral orbital gyrus consists mostly of area 47/12. Most of the OFC is granular, although the caudal parts of area 13 and area 14 are agranular. These caudal regions, which sometimes include parts of the insular cortex, respond primarily to unprocessed sensory cues.Connections
The connectivity of the OFC varies somewhat along a rostral-caudal axis. The caudal OFC is more heavily interconnected with sensory regions, notably receiving direct input from the pyriform cortex. The caudal OFC is also the most heavily interconnected with the amygdala. Rostrally, the OFC receives fewer direct sensory projections, and is less connected with the amygdala, but it is interconnected with the lateral prefrontal cortex and parahippocampus. The connectivity of the OFC has also been conceptualized as being composed of two networks; an orbital network composed of most of the central parts of the OFC, including most of the areas 47/12, 13, and 11; a medial network composed of the medial most and caudolateral regions of the OFC, as well as areas 24, 25 and 32 of the medial prefrontal cortex. The medial and orbital networks are sometimes referred to as the "visceromotor network" and the "sensory network", respectively.Afferents
The OFC receives projections from multiple sensory modalities. The primary olfactory cortex, gustatory cortex, secondary somatosensory cortex, superior and inferior temporal gyrus all project to the OFC. Evidence for auditory inputs is weak, although some neurons respond to auditory stimuli, indicating an indirect projection may exist. The OFC also receives input from the medial dorsal nucleus, insular cortex, entorhinal cortex, perirhinal cortex, hypothalamus, and amygdala.Efferents
The orbitofrontal cortex is reciprocally connected with the perirhinal and entorhinal cortices, the amygdala, the hypothalamus, and parts of the medial temporal lobe. In addition to these outputs, the OFC also projects to the striatum, including the nucleus accumbens, caudate nucleus, and ventral putamen, as well as regions of the midbrain including the periaqueductal gray, and ventral tegmental area. OFC inputs to the amygdala synapse on multiple targets, including two robust pathways to the basolateral amygdala and intercalated cells of the amygdala, as well as a weaker direct projection to the central nucleus of the amygdala.Function
Multiple functions have been ascribed to the OFC, including mediating context-specific responses, encoding contingencies in a flexible manner, encoding value, encoding inferred value, inhibiting responses, learning changes in contingency, emotional appraisal, altering behavior through somatic markers, driving social behavior, and representing state spaces.While most of these theories explain certain aspects of electrophysiological observations and lesion related changes in behavior, they often fail to explain, or are contradicted by other findings.
One proposal that explains the variety of OFC functions is that the OFC encodes state spaces, or the discrete configuration of internal and external characteristics associated with a situation and its contingencies. For example, the proposal that the OFC encodes economic value may be a reflection of the OFC encoding task state value. The representation of task states could also explain the proposal that the OFC acts as a flexible map of contingencies, as a switch in task state would enable the encoding of new contingencies in one state, with the preservation of old contingencies in a separate state, enabling switching contingencies when the old task state becomes relevant again. The representation of task states is supported by electrophysiological evidence demonstrating that the OFC responds to a diverse array of task features, and is capable of rapidly remapping during contingency shifts. The representation of task states may influence behavior through multiple potential mechanisms. For example, the OFC is necessary for ventral tegmental area neurons to produce a dopaminergic reward prediction error, and the OFC may encode expectations for computation of RPEs in the VTA.
Specific functions have been ascribed to subregions of the OFC. The lateral OFC has been proposed to reflect potential choice value, enabling fictive prediction errors to potentially mediate switching choices during reversal, extinction and devaluation. Optogenetic activation of the lOFC enhances goal directed over habitual behavior, possibly reflecting increased sensitivity to potential choices and therefore increased switching. The mOFC, on the other hand, has been proposed to reflect relative subjective value. In rodents, a similar function has been ascribed to the mOFC, encoding action value in a graded fashion, while the lOFC has been proposed to encode specific sensory features of outcomes. The lOFC has also been proposed to encode stimulus outcome associations, which are then compared by value in the mOFC. Meta analysis of neuroimaging studies in humans reveals that a medial-lateral valence gradient exists, with the medial OFC responding most often to reward, and the lateral OFC responding most often to punishment. A posterior-anterior abstractness gradient was also found, with the posterior OFC responding to more simple rewards, and the anterior OFC responding more to abstract rewards. Similar results were reported in a meta-analysis of studies on primary versus secondary rewards.
The OFC and basolateral amygdala are highly interconnected, and their connectivity is necessary for devaluation tasks. Damage to either the BLA or the OFC before, but only the OFC after devaluation, impairs performance. While the BLA only responds to cues predicting salient outcomes in a graded fashion in accordance with value, the OFC responds to both value and the specific sensory attributes of cue-outcome associations. While OFC neurons that, early in learning, respond to outcome receipt normally transfer their response to the onset of cues that predict the outcome, damage to the BLA impairs this form of learning.
The posterior orbitofrontal cortex is connected to the amygdala via multiple paths that are capable of both upregulating and downregulating autonomic nervous system activity. Tentative evidence suggests that the neuromodulator dopamine plays a role in mediating the balance between the inhibitory and excitatory pathways, with a high dopamine state driving autonomic activity.
It has been suggested that the medial OFC is involved in making stimulus-reward associations and the reinforcement of behavior, while the lateral OFC is involved in stimulus-outcome associations and the evaluation and possibly reversal of behavior. Activity in the lateral OFC is found, for example, when subjects encode new expectations about punishment and social reprisal.
The mid-anterior OFC has been found to consistently track subjective pleasure in neuroimaging studies. A hedonic hotspot has been discovered in the anterior OFC, which is capable of enhancing liking response to sucrose. The OFC is also capable of biasing the affective responses induced by α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid antagonism in the nucleus accumbens towards appetitive responses.
The OFC is capable of modulating aggressive behavior via projections to interneurons in the amygdala that inhibit glutaminergic projections to the ventromedial hypothalamus.
Electrophysiology
Neurons in the OFC respond both to primary reinforcers, and to cues that predict rewards across multiple sensory domains. The evidence for responses to visual, gustatory, somatosensory, and olfactory stimuli is robust, but evidence for auditory responses is weaker. In a subset of OFC neurons, neural responses to rewards or reward cues are modulated by individual preference and by internal motivational states such as hunger. A fraction of neurons that respond to sensory cues predicting a reward are selective for reward, and exhibit reversal behavior when cue outcome relationships are swapped. Neurons in the OFC also exhibit responses to the absence of an expected reward and punishment. Another population of neurons exhibits responses to novel stimuli and can "remember" familiar stimuli for up to a day.During cued reward or cued instrumental reward tasks, neurons in the OFC exhibit three general patterns of firing; firing in response to cues: firing before reward receipt; firing in response to reward receipt. In contrast to the medial prefrontal cortex and striatum, OFC neurons do not exhibit firing mediated by movement. Their reward-predictive responses are, however, shaped by attention: when shifting attention between two alternatives, the same OFC population will represent positively the value of a currently attended item, but negatively the value of the unattended item. The encoding of reward magnitude is also flexible, and takes into account the relative values of present rewards.