Divyan Bavan
Introduction
The prefrontal cortex (PFC) is responsible for executive decisions in the brain. It enables the control of goals, working memory, and emotion. This enables higher functions such as understanding social cues, problem solving, and decision making. With such a broad scope for executive control, several studies have aimed to determine if there is functional specialization within the PFC. After many experiments, there are three main views that take form—information, emotion, and network centric models. These models focus on the type of information being processed, how emotion is integrated, and overall networks that form, respectively. Several experiments support these theories, making the topic still under debate. Furthermore, some evidence also points to integrated models of processing. By understanding these theories and the experiments behind them, it is possible to gain an appreciation for the PFC’s complexity.
The Role of the Prefrontal Cortex
As mentioned previously, the PFC is critical for executive control over the brain. This role was discovered through several studies. The most famous of these cases was Phineas Gage, a railroad worker who had a tamping iron go through his frontal lobe. After this incident, Gage could still move and speak, but he had a much more irritable personality and did not conform to social norms. This was the first evidence that the frontal lobe played a role in controlling the brain.
The effects of frontal lobe lesions were further studied by Lhermitte and colleagues in several studies. In 1993, Lhermitte showed that patients with frontal lobe lesions displayed utilisation behaviour. For example, one patient was given a box of cigarettes and a lighter. Despite being a nonsmoker, he proceeded to take out the cigarette and light it. In 1996, this was extended to imitation behaviour—the tendency to imitate actions despite having no instructions to do so. Patients with frontal lobe lesions were observed to imitate the behaviour of the examiners. These two experiments illustrate the PFC’s role in inhibiting unwanted actions (Lhermitte, 1983; Lhermitte et al., 1986).
Information-Centric Specialization
This first model of PFC functional specialization focuses on a dissociation between processing spatial information and object information. It mirrors the dorsoventral dissociation in visual processing. In the PFC, the dorsolateral PFC (DLPFC) is theorized to handle spatial and action-centric information; the ventrolateral PFC (VLPFC) is suggested to handle object-oriented information.
The first study to show this specialization was from Goldman and Rosvold in 1970. In their experiment they had monkeys observe food being placed into one of two wells. Then, after a delay, they were tasked with deciding which well had the food. This was done in two sets of monkeys: one control group and one group with lesions to the principal sulcus. The principal sulcus is in the DLPFC. When the experiment was done with the control monkeys, they did not have trouble remembering which well the food was in. However, lesioned monkeys did not choose the correct well most of the time. This suggests that the DLPFC is critical for spatial memory (Goldman and Rosvold, 1970).
Although this result localized spatial memory to the DLPFC, a double dissociation was not yet made between the VLPFC and the DLPFC. This changed with Passingham’s 1975 study. Similar to Goldman and Rosvold, Passingham lesioned monkeys in the DLPFC and tested their spatial memory. However, he also lesioned monkeys in the inferior convexity—a part of the VLPFC. He found that when monkeys were tasked with remembering a colour over a delay period, monkeys lesioned in the VLPFC were impaired. He also discovered that if these tasks were swapped between the two groups, the monkeys performed on par with controls. This was evidence for a double dissociation between spatial memory and object memory in the PFC (Passingham, 1975).
To explain this phenomenon, Funahashi and colleagues performed electrophysiology recordings on monkeys. The method they used was the oculomotor delayed response (ODR) task. This experiment has monkeys focused on a central point in the screen, with another point appearing in the periphery. The peripheral point is then hidden, forcing the monkey to keep it in its working memory. Finally, the central point disappears, and the monkey must make a saccade to where the peripheral point was located. This determines if the monkey can remember the spatial location of the point. The researchers found that specific neurons in the DLPFC would fire during the delay period, effectively encoding the position of the peripheral point. This was evidence for the mechanism of DLPFC functional specialization. A similar experiment was done for the VLPFC by Wilson and colleagues. After testing object memory in the monkeys, they found that specific neurons in the VLPFC fired during the delay period. Even more interestingly, several neurons encoded the identity of faces specifically. This suggests that the VLPFC is highly specialized for object-related working memory (Funahashi et al., 1989; Wilson et al., 1993).
Although this model has lots of evidence, another model that has taken form is manipulation versus maintenance. These are thought to be the roles of the DLPFC and the VLPFC, respectively. The evidence for this comes from many experiments: alphabet ordering, self-ordering, and others. Thus, this also provides a view for how information is processed in the PFC (Petrides, 1995; Petrides 2000).
Emotion-Centric Specialization
Another functional specialization is that of the split between cognitive and emotional control. This theory suggests that the ventromedial PFC (VMPFC) is implicated in the integration of reward prediction and emotion into executive decisions. The key experiment showing this distinction was the Iowa Gambling Task.
In 1993, Bechara and colleagues developed a task where subjects had to choose a deck of cards from four choices. Two of these decks were disadvantageous while two were advantageous. While the advantageous decks provided long-term benefit, the short-term benefit was smaller. The disadvantageous deck provided larger benefits but would eventually lead to a net loss. The goal of the study was to see how patients with damage to the VMPFC acted differently than control patients. In the latter group, subjects originally sampled all decks but eventually started choosing advantageous decks once they developed an understanding for the task. However, subjects with VMPFC lesions did not do this; they kept on choosing disadvantageous decks. An interesting feature of these patients is that they also did not show a skin conductance response when choosing a bad deck. This contrasts with control patients, where a conductance response was present. This suggests that the VMPFC is important for integrating reward prediction into conscious choices and physiological responses (Bechara et al., 1998).
Network-Centric Specialization
A third way to look at functional specialization in the PFC is to see the path information takes during processing. This has resulted in several different models for PFC control, each of which has experimental evidence supporting it. The first example of this is the rostral-caudal axis of organization. This theory suggests that the timing of a response can be mapped to specific areas of the PFC. A past episode is linked to the rostral PFC, contextual signals are linked to the caudal PFC, and stimulus is linked to the premotor cortex.
The evidence for this was from a study in 2003. Subjects were asked to click left if a letter was a vowel or right if it was a consonant. As the experiment went on, the tasks increased in complexity; later tasks involved rule-based selection and multitasking. From fMRI measurements in the subjects, the researchers realized that as the tasks increased in complexity, more rostral activity was being detected. After further analysis, they ended up with the rostral-caudal axis of organization (Koechlin et al., 2003).
Another theory for functional specialization PFC is that there is no organisation. This theory suggests that inputs are processed mass action. The evidence for this theory comes from a meta-analysis by Duncan and Owen in 2000. In their analysis, they compared the MRIs of subjects doing various tasks and tried to find any functional distinctions. However, they were unable to do this and concluded that processing appears to be distributed (Duncan and Owen, 2000). When expanding the search, however, a network of frontal-parietal circuits appears to be heavily involved in general cognition. This is the multiple demand network theory for PFC organization, and again, distributes functional characteristics (Duncan, 2010).
Conclusion
It is evident that the PFC’s complexity has provided a challenge for psychologists to understand its functional specializations, if they exist. Early theories suggest that there is a dissociation between the DLPFC and VLPFC. This was explained as spatial versus object memory, further described by experiments showing consistent firing of specific neurons encoding information. Other scientists proposed a different dissociation for the DLPFC and VLPFC: maintenance versus modification of working memory. However, these experiments did not explain how emotion and reward are integrated into executive control. This was explained by a dissociation between the VMPFC and DLPFC, where lesions in the former lead to impaired reward prediction. These models suggested where specific information was processed in the PFC, but not how it was dissected. The rostral-caudal axis of organization proposed a solution to this problem; as the complexity of a problem shifted from stimulus to episode-based, activity increased in the rostral PFC. However, despite all of this, analysis still showed how the circuits are still unclear. Although the multiple demand network theory proposed a broader view of executive control, it is still debated how exactly this works. Thus, it is critical to keep exploring this area for further understanding of our higher functions.
*Works Cited*
Bechara, Antoine, et al. “Dissociation of Working Memory from Decision Making within the Human Prefrontal Cortex.” The Journal of Neuroscience, vol. 18, no. 1, 1 Jan. 1998, pp. 428–437, https://doi.org/10.1523/jneurosci.18-01-00428.1998.
Duncan, John. “The Multiple-Demand (MD) System of the Primate Brain: Mental Programs for Intelligent Behaviour.” Trends in Cognitive Sciences, vol. 14, no. 4, Apr. 2010, pp. 172–179, https://doi.org/10.1016/j.tics.2010.01.004.
Duncan, John, and Adrian M Owen. “Common Regions of the Human Frontal Lobe Recruited by Diverse Cognitive Demands.” Trends in Neurosciences, vol. 23, no. 10, Oct. 2000, pp. 475–483, psych.colorado.edu/~munakata/teaching/prosem07/DuncanOwen00.pdf, https://doi.org/10.1016/s0166-2236(00)01633-7.
Funahashi, S., et al. “Mnemonic Coding of Visual Space in the Monkey’s Dorsolateral Prefrontal Cortex.” Journal of Neurophysiology, vol. 61, no. 2, 1 Feb. 1989, pp. 331–349, https://doi.org/10.1152/jn.1989.61.2.331.
Goldman, Patricia S., and H.Enger Rosvold. “Localization of Function within the Dorsolateral Prefrontal Cortex of the Rhesus Monkey.” Experimental Neurology, vol. 27, no. 2, May 1970, pp. 291–304, https://doi.org/10.1016/0014-4886(70)90222-0. Accessed 10 Dec. 2019.
Koechlin, E. “The Architecture of Cognitive Control in the Human Prefrontal Cortex.” Science, vol. 302, no. 5648, 14 Nov. 2003, pp. 1181–1185, https://doi.org/10.1126/science.1088545.
LHERMITTE, F. ““UTILIZATION BEHAVIOUR” and ITS RELATION to LESIONS of the FRONTAL LOBES.” Brain, vol. 106, no. 2, 1983, pp. 237–255, https://doi.org/10.1093/brain/106.2.237.
Lhermitte, F., et al. “Human Autonomy and the Frontal Lobes. Part I: Imitation and Utilization Behavior: A Neuropsychological Study of 75 Patients.” Annals of Neurology, vol. 19, no. 4, Apr. 1986, pp. 326–334, https://doi.org/10.1002/ana.410190404. Accessed 19 Aug. 2020.
Passingham, Richard. “Delayed Matching after Selective Prefrontal Lesions in Monkeys (Macaca Mulatta).” Brain Research, vol. 92, no. 1, July 1975, pp. 89–102, https://doi.org/10.1016/0006-8993(75)90529-6. Accessed 28 Mar. 2022.
Petrides, M. “Impairments on Nonspatial Self-Ordered and Externally Ordered Working Memory Tasks after Lesions of the Mid-Dorsal Part of the Lateral Frontal Cortex in the Monkey.” The Journal of Neuroscience, vol. 15, no. 1, 1 Jan. 1995, pp. 359–375, https://doi.org/10.1523/jneurosci.15-01-00359.1995. Accessed 11 Apr. 2021.
Petrides, Michael. “The Role of the Mid-Dorsolateral Prefrontal Cortex in Working Memory.” Experimental Brain Research, vol. 133, no. 1, 20 May 2000, pp. 44–54, https://doi.org/10.1007/s002210000399.
Wilson, F., et al. “Dissociation of Object and Spatial Processing Domains in Primate Prefrontal Cortex.” Science, vol. 260, no. 5116, 25 June 1993, pp. 1955–1958, https://doi.org/10.1126/science.8316836. Accessed 6 Apr. 2020.