The DLPFC

With the onset of modern neuroscience a common question often discussed is: where in the brain are self-regulation, executive control, free will, volition, selection, short-term memory, attention, planning, and overall consciousness located and by what neurological processes do they occur. The prime brain areas that is thought to be responsible are the frontal lobes. However a problem is presented when studying this area in humans: comparison to monkey's brains has been difficult as the PFC, prefrontal cortex, is structured somewhat differently (Petrides, M., Pandya, D.N., 1999 and Fuster, J. M., 2008. pg. 21) and thus comparisons can be complicated, and invasive techniques are largely limited to non-humans. Lesion studies have sometimes led to seemingly contradicting results that are hotly debated. However researchers agree that the prefrontal cortex as a whole is the essential area responsible for higher behavioral functions. One of the essential brain areas associated with these psychological processes is the DLPFC, dorsolateral prefrontal cortex. This area appears to be critical for working memory, planning, selective attention, temporal integration and volition.

Location and Structure

The DLPFC is located in the middle frontal gyrus and is typically defined cytoarchitectonicaly as Brodmann areas 46 and 9 (Berntson, G. G., & Cacioppo, J. T., 2009, pg. 586). BA8 is also sometimes considered the posterior DLPFC or simply adjacent to it (Miller, Earl K., Cohen, Jonathan D., 2001, pg. 169). The DLPFC is one of the last areas in humans to develop and myelinate as the brain myelinates back to front (Fuster, J. M., 2008. pg. 15).

Connections

A critical aspect of the DLPFC is that it is extraordinarily interconnected with other brain areas and is the most connected of all the PFC areas. It is connected to sensory and motor cortexes as well as interconnected with other PFC areas.

Sensory Cortex Connections

BA46 and BA9 have connections to the visual, somatosensory and auditory projections via the occipital, parietal and temporal cortices. Both areas are also connected to the multimodal rostral superior temporal sulcus with neurons responding to multi-sensory inputs. (Miller, Earl K., Cohen, Jonathan D., 2001). The posterior parietal cortex in particular plays a large role in that it is both a mutli-modal area and is related to spatial working memory, attention. It has also been implicated in having a role in intention (Libet et al, 1999).

Motor Cortex Connections

BA46 is particularly connected to pre-motor areas; it has connections to the SMA, the pSMA, the rostral cingulate, the lateral frontal pre-motor cortex and to the cerebellum and superior colliculus ( Miller, E. K., Cohen, J. D., 2001. pg. 175). BA46 also sends projections to the BA8 frontal eye fields and indirectly receives inputs, along with the all of the prefrontal cortex, from the basal ganglia (pg. 175).

Prefrontal Interconnections

All of the prefrontal areas are interconnected to a large degree. BA46 and BA9 are both connected to the orbital and medial prefrontal areas (BA 10, 11,13,14) as well as to each other. BA9 is connected to the ventrolateral areas (BA 12,45). BA8 connects to BA46. These interconnections not only allow indirect communication to other systems that are connected to other prefrontal areas (such as the limbic system through the medial PFC areas), but also provides a way for disperse systems to be indirectly wired together through this central area. The convergence of cortico-cortical pathways suggests that part of the DLPFC's operation is as a cross-modal area of association (Fuster, J. M., 2008. pg. 34).

Experimental Connections

An experiment that supports this prefrontal interconnectedness theory (Tomita et al, 2001) studied top-down communication from the prefrontal cortex to the inferior temporal cortex (which has been shown to store long-term visual memories (Miyashita, Y., 1993)). The inferior temporal cortex receives visual input contra-laterally from the occipital lobe and the two sides are sub- cortically connected. In this experiment two monkeys were given split visual stimuli and were trained to associate certain cues with categories of stimuli. Then the connection between the inferior temporal cortices was severed -- to theoretically prevent a cue presented to the ipsilateral side (the side that did not see the choice) from activating a category recall on the contralateral side. The bottom-up condition was when both the cue and the choice were offered to the same contralateral vision field, and the top-down condition was when the cue was offered on the ipsilateral side while the choice was offered on the contralateral side. The results, measured via single neuron recordings, showed that when the direct connection between the sides of the inferior temporal lobe was severed there was a delay in recall but that the ipsilateral side never-the-less showed correct activation. This experiment suggests that the visual information travelled through the prefrontal cortex connections. To confirm this finding the connections to the prefrontal cortex from the inferior temporal cortex lobes were severed and only then did the monkeys fail the task.

Functionality

Given the strong interconnectedness of all the prefrontal areas, it is no surprise that a compartmentalization of each cytoarchitectonic area would be an oversimplification (Fuster, J. M., 2008. pg. 6). All these areas work together to achieve its various executive functions. However, these areas do have discernable specializations and the dorsolateral prefrontal cortex in particular has a number of fortes including basic working memory, cross- temporal/modal integration, temporal ordering, planning, acting and rule encoding, selective attention, will and volition.

Working Memory

It is commonly agreed that the lateral prefrontal cortex plays a large role in working memory (Curtis, C. E, D'Esposito, M., 2004). However, precisely how it works and what role the DLPFC plays in it is disputed. Special types of neurons called memory neurons are certainly necessary if not sufficient. Memory neurons are neurons that in a delay task appear to activate during the delay period between a presentation and a recall of a stimuli. Some activate during the stimulus, others right after, and even others stay active for a while. The DLPFC has more memory neurons than any other prefrontal area (Fuster, J. M., 2008. pg. 247). As a side note, the Frontal Eye Fields -- area 8 -- contains memory cells that are particularly attuned for eye saccades to cued locations, whereas in areas 46 and 9 there is some evidence for memory neuron specialization toward spatial tasks. There are however memory neurons in many other areas of the brain, including the thalamus, inferotemporal cortex and basal ganglia, but those memory neurons appear to be specialized toward the sensory task of the particular area, whereas in the prefrontal cortex the memory neurons are more multi-modal. The simple existence of memory neurons is not enough to show the DLPFC's role in working memory.

Working memory as a function of the DLPFC has been primarily shown via delay tasks with either lesion patients or appropriately trained animals with ablations/lesions or reversible lesions via cooling. Commonly a delay task will begin with the representation of a cue. The cue informs the participant about a choice that is to be made later (usually the location of a salient item such as food). Then the cue is covered up for some amount of time: the delay period. Following this a choice is presented -- the correct choice being informed by the cue presented prior to the delay. In many studies of animals, when the delay is non-existent or very short, the DLPFC lesioned animals perform normally, but as the delay increases so success at the task drops (Fuster, J. M., 2008. pg. 144). Similar tasks however that do not have any spatial element have shown no difference in some experiments.

These lesion studies lead to the theory that the DLPFC is particularly important for spatial working memory. However there remain questions as to what role the DLPFC plays in working memory. The interconnectedness of the DLPFC plays a vital role in its working memory functionality. A study of the interaction of the DLPFC and other brain areas (Fuster, J. M., Bauer, R. H., & Jervey, J. P., 1985) shows working memory to be more than a few memory neurons but a distributed representational network encompassing many brain areas. In particular it reveals the DLPFC as a top-down moderator of "online" working memory tasks. The study used a Peltier chip to cool down the DLPFC and the inferotemporal cortex in turn. Single neuron recording was used to record the activity of memory neurons sensitive to certain colors in the inferotemporal cortex. The monkeys were trained in a color match delay task. As they performed the task the color sensitive neurons in the inferotemporal cortex would activate on the working memory of the cue color. However, when the DLPFC was cooled the firing rate of the inferotemporal neurons related to the working memory of the task would attenuate and the animal, additionally, would not be able to perform the task. This suggests a modulation effect from the DLPFC to the inferotemporal cortex, where the sensory memory for this particular task was represented, at least in part. This experiment was also performed in reverse: the inferotemporal cortex cooled and the DLPFC activity measured during the same delay task. The results were very similar: there was a modulation effect in addition to lowered task performance. Fuster makes the supposition that the model for working memory is that of,

"a frontal substrate of executive memory cooperating with other brain structures, cortical and subcortical, for the maintenance of working memory. The function of working memory would be distributed, much as its neural substrate, albeit under a degree of prefrontal executive -- or 'top-down' --control." (Fuster, J. M., 2008. pg. 251)

Reverberating reentry is the method by which these cortico-cortical/cortico-subcortical memory loops are maintained neurologically (pg. 252). Projections to and from posterior sensory cortices and the DLPFC reverberate back and forth, each moderating and resubmitting the memory back into the loop as needed. The theory outlined by Fuster (pg. 296) is that the DLPFC is involved in the selection and maintenance of working memory and that the representational aspect of any particular memory requires the brain areas which are specialized to the sensory features of that memory.

There have been many PET and fMRI studies that have localized spatial working memory to the DLPFC especially in monkeys, however in human studies not only has it been harder to pinpoint some studies seem to suggest that the DLPF's role is quite more elaborate and involved.

Cross-Temporal Integration

It has already been demonstrated that the DLPFC has a particular importance in temporal integration. Simple recognition tasks without a delay do not appear to trigger the DLPFC. Additionally, as already mentioned, the greater the delay the more a DLPFC lesion causes impairment. Several studies show DLPFC lesioned monkeys

"deficient at tasks in which temporal frequency, temporal order, or temporal sequence is of the essence. [..] The dorsolateral prefrontal cortex seems most important for the mediation of the cross-temporal contingencies, as delay tasks require" (Fuster, J. M., 2008. pg. 145).

This is not surprising as working memory is an essential component of temporally organizing cognition and behavior.

Cross-Modal Integration

Much of the DLPFC's functionality can be attributed to how interconnected it is. This interconnectedness however not only enables cross-temporal integration but also cross-modal integration -- neurons that receive projections from multiple sensory areas are the mechanism behind this.

Fuster demonstrated cross-modal integration with a single neuron recording study performed on monkeys (Fuster, J. M., 2008. pg. 226). The monkeys were trained to associate an auditory tone with a particular visual stimuli -- in this case a color. The experiment first presented the animal with the auditory cue and then after a delay presented two colors buttons -- the animal would need to choose the correct one that corresponded with the learned association. Single cell neurons were recorded in the DLPFC that responded both to the auditory cue and to the color. Thus the sensory rule for the experimental condition was encoded in a multi-modal association. This is a precursor to the rule encoding functionality that the DLPFC is implicated in as well.

Planning, Acting and Rule Encoding

The DLPFC is essential to the "ability to maintain any memory, recent or remote, in the active state for the prospective performance of a goal-directed act" (Fuster, J. M., 2008. pg. 146). Performance of a goal directed act requires planning for the act and acting on in the plan. Related to this is the ability to encode the rules that may be required for the task. The beginnings of rule encoding are in the cross-modal associative abilities of the DLPFC.

A recent study showed that even abstract mathematical rules can be encoded in the DLPFC (Bongard, S., & Nieder, A., 2010). The monkeys were trained to distinguish numerosities with greater and less than rules. The stimuli were varied and so the monkeys were unable to build concrete associative relations between specific numbers. Using single cell recordings the study found that there were neurons that fired specifically when a rule-indicating cue appeared and during the application of that specific rule. The suggestion here is not that numerosity itself is encoded in the DLPFC rather only the selection of choosing greater or less than as rules for executive control. Numerosity itself is a function of the posterior parietal cortex. It is also important to note that the DLPFC abstracts rules only when the task surpasses some basic difficulty.

White and Wise in a 1999 study showed that DLPFC neurons were involved in rules involving both spatial and non-spatial cues with specific eye movements and hand gestures. As in many of these sorts of studies, using single cell recording, they found a significant number of neurons that were selective for changing rules and applying them.

Planning is a task that is commonly associated with the prefrontal cortex in general, but also with the DLPC in particular. A study by Morris et al in 1993 showed DLPFC activation during the tower of london task. The tower of london is a task design to test logical planning ability and is similar to the towers of hanoi puzzle. A common lay association with planning is internal silent speech. A study by Ryding et al in 1996 showed DLPFC activation during silent counting but not during counting aloud. The study attributes this activation to the attentional component of counting internally, although it is not obvious why there is a difference between the two in terms of attention. Slightly more modern theories of the DLPFC may attribute this activation to the selective goal of speaking internally or to the monitoring needs of the feedback loop given the lack of audible feedback.

Selective Attention and Monitoring

If the DLPFC helps to maintain goal related working memory, does it also participate in the selection and attentional aspects related to the goal as well? An fMRI study by D'Esposito et al in 1995 seems to indicate that this may be the case. The experiment consisted of two tasks, a spatial-rotation task and a semantic-judgement task. When performed singly neither task required working memory and thus did not show any DLPFC activation. However when the tasks were to be performed simultaneously (sequentially but rapidly) the DLPFC showed significant activation.

Hare et al (2009) showed that goal related decisions require DLPFC modulation of a value signal provided/computed by the VMPFC. DLPFC activity increased in subjects during successful self-control trails involving choosing healthy over unhealthy-but-liked food. The experimenters suggest that this result helps confirm the theory that the DLPFC acts via its connections to other brain areas to promote goal-oriented behavior.

The DLPFC has also been shown to be activated when a task requires higher levels of monitoring. The silent counting experiment by Ryding shows this to some degree. Another set of tasks that shows DLPFC involvement are self- ordered tasks. In these tasks the subject is shown a set of stimuli and then is asked self-order them. This requires the subject to keep the memory of which items have been selected and which have not been selected yet in working memory and requires them to monitor those memories as they choose objects. Petrides showed that self-ordered tasks involved DLPFC activation while externally-ordered tasks did not (2000). Additionally he showed that the more items were involved in the task the higher the activation regardless of whether the stimuli were spatial or not. This result seems to indicate that monitoring is an additional feature of the DLPFC.

Will and Volition

Given the DLPFC's role in so many aspects of "executive control" perhaps it also has a role in the voluntary choice of actions that is so tightly associated with the concept of consciousness.

"Damage in the dorsolateral prefrontal cortex in humans leads to a lack of spontaneous activity, distractibility by environmental cues, and the repetitive, stereotypic use of inappropriate behavioral responses (perseveration). These phenomena may indicate an inability to choose or initiate the correct course of action." (Freeman, A., Libet, B., & Sutherland, K., 1999 pg. 16).

Frith defines the actual act of such choosing as "a deliberate selection [that] is subjectively experienced as willed and occurs when we have a choice of action. [These] spontaneous or self-generated actions are not specified by an external trigger stimulus, but are internally driven." (Frith et al, 1991). One of the main findings on volition and the DLPFC is his PET study of word generation and finger movements.

The word generation experiment consists of 3 conditions: in the first the subject hears a list of random words and repeats each word. In the second the subject hears words that have unambiguous opposites and must say out loud the opposite words. The final third condition the subject hears the word "next" and must choose a word that starts with a preset letter. The idea is that the first two conditions are "externally specified" (Frith et al, 1991); the word to repeat is unambiguous and no choice exists. In the third condition however the subject must pick a word. The PET scans showed no difference in rCBF between the first two "externally specified" conditions while the third condition showed increased left DLPFC rCBF.

The finger movement part of the experiment had similar results (although rCBF was increased bilaterally in the finger condition). The conditions were similar: in the first the subject had one of two fingers tapped and they had to raise that same finger in response, in the second they had to raise the opposite, and in the third free-willed condition they had to choose which finger to raise.

One problem with the word generation part of this experiment is that there is a small working memory component -- that of keeping the already stated previous word in mind. However, a later study by Desmond et al in 1998, changed the condition to involve no working memory by simply giving the participant different word stems that required choice of completions on each trail. This study had similar results except with fMRI. Further studies by Libet and Jahanshahi showed that temporal choice about when to make the movement also activated the DLPFC (Freeman, A., Libet, B., & Sutherland, K., 1999 pg. 19).

However the DLPFC, as has been previously mentioned, does not act alone in any of this. In each of these tasks the DLPFC worked with other brain areas by virtue of its interconnectedness (Freeman, A., Libet, B., & Sutherland, K., 1999 pg. 20). The conclusion may be that the DLPFC "plays a role in the selection of action, [but] the performance of the action itself is facilitated by 'lower' motor regions, such as SMA and basal ganglia" (pg. 21). And furthermore, the "DLPFC seems to be involved in keeping possible action in mind before they are executed, and selecting which one will be performed" (pg. 27).

Disorders and Lesions

What role does the DLPFC play in human disorders? Lesions of the DLPFC contribute to a myriad of symptoms including: lack of drive and awareness (selective attention suppressed), visuospatial neglect (FEF damage), "frontal dynamic aphasia": reduced verbal production quality, apathy and "dysexecutive syndrome" leading to being "incapacitated in initiating spontaneous and deliberate action" (Fuster, J. M., 2008. pg. 198). Working memory deficits contribute to a myriad of failures upstream including in action selection, temporal integration, and goal oriented planning. The DLPFC is also implicated in several neurological disorders.

Schizophrenia

Schizophrenia is commonly marked by such symptoms as poor speech, temporal integration problems, hallucinations and lack of volitional control. A study by Weinberger et al in 1986 showed via a PET scan study that there was a lack of appropriate increase in DLPFC activity in schizophrenia patients while performing the Wisconsin card sorting task, which has been shown many times to elicit DLPFC activation. This study was later expanded on in 2002 by Meyer- Lindenberg et al to not only using PET to study rCBF during the WCST but also presynaptic dopamine. Their study showed an inverse correlation with reduced DLPFC activation and increased striatal 6-FD uptake, leading to the conclusion that dopaminergic transmission dysfunction in schizophrenia may be DLPFC related. Since schizophrenia is characterized by, "the failure to construct logically coherent temporal configurations (gestalts) of thought -- and consequently of speech and behavior" (Fuster, J. M., 2008. pg. 313) -- that precisely implicates the DLPFC in terms of working memory and temporal integration problems.

Conclusions

A summary of DLPFC working memory involvement is provided by D'Esposito et al:

When the amount of to-be-remembered information presented at the beginning of a delayed-response trial approaches or exceeds short-term memory capacity [...], dorsolateral PFC is preferentially engaged. Dorsolateral PFC-supported processes may facilitate the efficient encoding of information. During the subsequent delay interval, when no information is accessible to the subject, both ventro- and dorsolateral PFC are recruited. If manipulation of this information is additionally required during the delay period, dorsolateral PFC is recruited to an additional extent. Upon the presentation of the probe stimulus, when a subject is required to make a response based on what was presented at the beginning of the trial, dorsolateral PFC is again engaged, presumably as the subject scans the information that was retained across the trial and chooses an appropriate motor response. [...] Together, the results of these studies highlight the temporal dynamics of PFC function during working memory task performance. (2000)

However it is obvious from the DLPFC's involvement in selection and planning and volition that its function cannot be limited to only working memory. Selective attention is at the most basic level attention to internal representations. These internal representations are maintained by working memory and thus these two are closely related. Additionally planning involves attention to future possibilities, what Ingvar (1985) called "memory of the future", and is thus also closely related to working memory. It appears that although there is no cohesive theory for the DLPFC, a general sense could be that all things DLPFC are related to temporal integration across multiple sensory modes. The interconnectedness of the DLPFC is its defining characteristic.

References:

Boller, F., Grafman, J., & Holyoak, K. J. (1995). Structure and functions of the human prefrontal cortex. New York: New York Academy of Sciences. Bongard, S., & Nieder, A. (2010).

Basic mathematical rules are encoded by primate prefrontal cortex neurons. Proceedings of the National Academy of Sciences, 107(5), 2277-2282. doi:10.1073/pnas.0909180107 Curtis, C. E., & D'Esposito, M. (2004).

The effects of prefrontal lesions on working memory performance and theory. Cognitive, Affective, & Behavioral Neuroscience, 4(4), 528-539. Retrieved from http://cabn.psychonomic-journals.org/content/4/4/528.abstract Desmond, J. E., Gabrieli, J. D. E., & Glover, G. H. (1998).

Dissociation of frontal and cerebellar activity in a cognitive task: Evidence for a distinction between selection and search. NeuroImage, 7(4), 368-376. Retrieved from http://www.sciencedirect.com/science/article/B6WNP-45M2XYH-W/2/bf49e4719a18fb7a65ba16cf0b5d3547 D'Esposito, M., Detre, J. A., Alsop, D. C., Shin, R. K., Atlas, S., & Grossman, M. (1995).

The neural basis of the central executive system of working memory. Nature, 378(6554), 279-281. Retrieved from http://dx.doi.org/10.1038/378279a0 D'Esposito, M., Postle, B. R., & Rypma, B. (2000).

Prefrontal cortical contributions to working memory: Evidence from event-related fMRI studies.. Experimental Brain Research, 133(1), 3-11. Retrieved from http://dx.doi.org/10.1007/s002210000395 D'Esposito, M., (2003).

Neurological foundations of cognitive neuroscience. Cambridge, Mass.: MIT Press. Retrieved from http://www.netLibrary.com/urlapi.asp?action=summary&v=1&bookid=78182 Freeman, A., Libet, B., & Sutherland, K. (1999).

The volitional brain :Towards a neuroscience of free will. Thorverton: Imprint Academic. Freeman, A., Libet, B., & Sutherland, K. (1999).

The volitional brain :Towards a neuroscience of free will. Thorverton: Imprint Academic. Frith, C. D., Friston, K., Liddle, P. F., & Frackowiak, R. S. J. (1991).

Willed action and the prefrontal cortex in man: A study with PET. Proceedings: Biological Sciences, 244(1311), 241-246. Retrieved from http://www.jstor.org/stable/76606 Fuster, J. M. (2008).

The prefrontal cortex (4th ed.). Amsterdam; Boston: Academic Press/Elsevier. Fuster, J. M., Bauer, R. H., & Jervey, J. P. (1985).

Functional interactions between inferotemporal and prefrontal cortex in a cognitive task. Brain Research, 330(2), 299-307. doi:DOI: 10.1016/0006-8993(85)90689-4 Gathercole, S. E. (1999). Cognitive approaches to the development of short-term memory. Trends in Cognitive Sciences, 3(11), 410-419. doi:DOI: 10.1016/S1364-6613(99)01388-1 Hare, T. A., Camerer, C. F., & Rangel, A. (2009).

Self-control in decision-making involves modulation of the vmPFC valuation system. Science, 324(5927), 646-648. doi:10.1126/science.1168450 Hoshi, E. (2006).

Functional specialization within the dorsolateral prefrontal cortex: A review of anatomical and physiological studies of non-human primates. Neuroscience Research, 54(2), 73-84. doi:DOI: 10.1016/j.neures.2005.10.013 Meyer-Lindenberg, A., Miletich, R. S., Kohn, P. D., Esposito, G., Carson, R. E., Quarantelli, M., et al. (2002).

Reduced prefrontal activity predicts exaggerated striatal dopaminergic function in schizophrenia. Nature Neuroscience, 5(3), 267-271. Retrieved from http://dx.doi.org/10.1038/nn804 Miller, E. K., & Cohen, J. D. (2001).

An integrative theory of prefrontal cortex function. Annual Review of Neuroscience, 24(1), 167-202. Retrieved from http://dx.doi.org/10.1146/annurev.neuro.24.1.167 Miyashita, Y. (1993).

Inferior temporal cortex: Where visual perception meets memory. Annual Review of Neuroscience, 16(1), 245-263. Retrieved from http://dx.doi.org/10.1146/annurev.ne.16.030193.001333 Morris, R. G., Ahmed, S., Syed, G. M., & Toone, B. K. (1993).

Neural correlates of planning ability: Frontal lobe activation during the tower of london test. Neuropsychologia, 31(12), 1367-1378. doi:DOI: 10.1016/0028-3932(93)90104-8 Petrides, M., & Pandya, D. N. (1999).

Dorsolateral prefrontal cortex: Comparative cytoarchitectonic analysis in the human and the macaque brain and corticocortical connection patterns. European Journal of Neuroscience, 11(3), 1011. doi:10.1046/j.1460-9568.1999.00518.x Petrides, M. (2000).

The role of the mid-dorsolateral prefrontal cortex in working memory. Experimental Brain Research, 133(1), 44-54. Retrieved from http://dx.doi.org/10.1007/s002210000399 Rossi, A., Pessoa, L., Desimone, R., & Ungerleider, L. (2009).

The prefrontal cortex and the executive control of attention. Experimental Brain Research, 192(3), 489-497. Retrieved from http://dx.doi.org/10.1007/s00221-008-1642-z Ryding, E., BraÅdvik, B., & Ingvar, D. H. (1996).

Silent speech activates prefrontal cortical regions asymmetrically, as well as speech-related areas in the dominant hemisphere. Brain and Language, 52(3), 435-451. doi:DOI: 10.1006/brln.1996.0023 Singh, J., & Knight, R. T. (1990).

Frontal lobe contribution to voluntary movements in humans. Brain Research, 531(1-2), 45-54. doi:DOI: 10.1016/0006-8993(90)90756-2 Tanji, J., & Hoshi, E. (2008).

Role of the lateral prefrontal cortex in executive behavioral control. Physiological Reviews, 88(1), 37-57. doi:10.1152/physrev.00014.2007 Tomita, H., Ohbayashi, M., Nakahara, K., Hasegawa, I., & Miyashita, Y. (1999).

Top-down signal from prefrontal cortex in executive control of memory retrieval. Nature, 401(6754), 699-703. Retrieved from http://dx.doi.org/10.1038/44372 Weinberger, D. R., Berman, K. F., Zec, R. F., & National Institute of Mental Health. (1986; 1986).

Physiologic dysfunction of dorsolateral prefrontal cortex in schizophrenia. Washington, D.C.: National Institute of Mental Health. White, I. M., & Wise, S. P. (1999).

Rule-dependent neuronal activity in the prefrontal cortex. Experimental Brain Research, 126(3), 315-335. Retrieved from http://dx.doi.org/10.1007/s002210050740