Functional organization of the medial frontal cortex

https://doi.org/10.1016/j.conb.2007.03.001Get rights and content

The anterior cingulate cortex (ACC) and adjacent areas of the medial frontal cortex (MFC) have been implicated in monitoring behaviour and in detecting errors. Recent evidence, however, suggests that the ACC not only registers the occurrence of errors but also represents other aspects of the reinforcement history that are crucial for guiding behaviour. Other studies raise the possibility that dorsal MFC areas not only monitor behaviour but also actually control response selection, particularly when the task in hand is changing. Many decisions are made in social contexts and their chances of success depend on what other individuals are doing. Evaluation of other individuals is therefore crucial for effective action selection, and some ACC regions are implicated in this process.

Introduction

The medial frontal cortex (MFC) is a collective term used to describe cortex dorsal and rostral to the corpus callosum (Figure 1a). In the primate brain, it is usually taken to include the anterior cingulate cortex (ACC) as well as several areas in the more dorsal superior frontal gyrus, such as the pre-supplementary motor area (pre-SMA), the supplementary motor area (SMA) and the supplementary eye fields (SEF). In this review, we focus on the studies from the past two years that have attempted to understand the role of the MFC in executive control, action selection and social cognition. Changes in activity, as indexed by the blood oxygenation-level-dependent (BOLD) signal measured using functional magnetic resonance imaging (fMRI), are prominent in the MFC when human subjects start a new cognitive task or switch between performing two different tasks [1••, 2••, 3••, 4••]. An influential account of such findings has emphasized the importance of the MFC in monitoring brain response systems to ascertain whether different, and therefore conflicting, responses are being prepared [5]. This might occur if the stimulus guiding behaviour is ambiguous, as is the case in the Stroop task, or because a change in context means that a different set of stimulus–response associations should be called into play, as occurs when switching from one task to another.

A recent fMRI study [6] reported activation of the MFC when stimuli afforded competing and conflicting responses, but found that there was no effect on MFC activity when conflict was manipulated at a perceptual level by altering the salience of irrelevant stimulus information. According to the conflict monitoring hypothesis, the increase in MFC activity leads to activation of the dorsolateral prefrontal cortex on the next trial of the task and this, in turn, ensures careful control of the process of response selection. Kerns et al. [7] used fMRI to scan subjects while they were performing the colour-naming Stroop task. There was increased MFC activity in trials where the ink colour (which subjects were to name) spelled out an incongruent colour word that afforded an alternative naming response (e.g. the word blue printed in red ink). MFC activity on the conflict trial was correlated both with dorsolateral prefrontal activity on the subsequent trial and with the extra control exerted over response selection, as indexed by slower response times.

Section snippets

Conflict detection or context-dependent control of action

There has been uncertainty about which MFC region is the key region concerned with conflict. A meta-analysis of fMRI studies emphasized the variable location of conflict-related activity within the MFC [8]. Single-neuron recording studies in macaques, in which the location of activity can be precisely determined, have failed to find evidence for conflict-related activity in the ACC (Figure 2) [9]. Such findings are consistent with demonstrations that ACC lesions do not impair task switching or

Monitoring performance for errors: the role of the ACC

There has been considerable interest in the neural mechanisms that monitor behaviour for errors. It is argued that the detection of errors is important because errors are highly informative of how best to adjust future behaviour [8, 18•, 19]. An event-related potential called the event-related negativity (ERN) can be recorded from the scalp when human subjects make mistakes or when their gambles lead to an undesired outcome [20]. The ERN is present not just when subjects made mistakes

Errors and other important outcomes

Despite the emphasis on errors, there is also evidence that single neurons in the ACC are active when positive reinforcement is delivered [27]. Although errors are indeed often useful sources of information about how to improve performance, so are correct outcomes. This is especially true when the correct way to perform the task is unclear. Walton et al. [28] asked subjects to perform a task-switching paradigm in which the switch cues did not tell subjects which of two alternative task sets was

Detecting errors or representing the reinforcement history

For an animal foraging in an uncertain environment, the outcomes of actions are unlikely to be either categorically correct or erroneous, and both positive and negative outcomes are important sources of information about the potential value of actions. Although the ACC can register when an error occurs, its function might not be error detection per se and instead it might represent aspects of a more extended choice–outcome history, based on both positive and negative feedback, that can be used

Making decisions about the costs and benefits of actions

It is important to know the reinforcement history of an action before a choice can be made. However, in addition to the potential benefits that the action might entail, it is also important to know what costs the action might entail. Optimal foraging theories have, for some time, emphasized that costs, as well as rewards, are important determinants of which options animals will pursue outside the laboratory. When animals are foraging, their choices are influenced not only by their reinforcement

Social information and the ACC

In many situations, the best decision to make will depend both on the reward histories associated with each possible course of action and on what choices other individuals are taking [39]. Some parts of the ACC, particularly more ventral and rostral parts, seem to be particularly active when subjects are acquiring or using information about other individuals [40, 41].

When the acquisition and encoding of social information is emphasized, there is activity in ACC and adjacent MFC areas [42, 43••

Conclusions

The dorsal MFC has a central role in decision making and action selection. The ACC is important when decisions are guided by the history of reinforcement and when consideration is given to the costs associated with the action. More dorsal regions, such as the pre-SMA, become more important when subjects are changing or initiating a new task that entails distinct ways of selecting actions. Importantly, many decisions are made by individuals that are within a social context, not in isolation. The

References and recommended reading

Papers of particular interest, published within the period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

References (54)

  • C. Liston et al.

    Anterior cingulate and posterior parietal cortices are sensitive to dissociable forms of conflict in a task-switching paradigm

    Neuron

    (2006)
  • J.G. Kerns et al.

    Anterior cingulate conflict monitoring and adjustments in control

    Science

    (2004)
  • K.R. Ridderinkhof et al.

    The role of the medial frontal cortex in cognitive control

    Science

    (2004)
  • L.K. Fellows et al.

    Is anterior cingulate cortex necessary for cognitive control?

    Brain

    (2005)
  • S. de Wit et al.

    Dorsomedial prefrontal cortex resolves response conflict in rats

    J Neurosci

    (2006)
  • V. Stuphorn et al.

    Executive control of countermanding saccades by the supplementary eye field

    Nat Neurosci

    (2006)
  • P. Nachev et al.

    Volition and conflict in human medial frontal cortex

    Curr Biol

    (2005)
  • E.A. Crone et al.

    Brain regions mediating flexible rule use during development

    J Neurosci

    (2006)
  • C.S. Li et al.

    Imaging response inhibition in a stop-signal task: neural correlates independent of signal monitoring and post-response processing

    J Neurosci

    (2006)
  • M.F.S. Rushworth et al.

    Role of the human medial frontal cortex in task switching: a combined fMRI and TMS study

    J Neurophysiol

    (2002)
  • S. Debener et al.

    Trial-by-trial coupling of concurrent electroencephalogram and functional magnetic resonance imaging identifies the dynamics of performance monitoring

    J Neurosci

    (2005)
  • H.T. Van Schie et al.

    Modulation of activity in medial frontal and motor cortices during error observation

    Nat Neurosci

    (2004)
  • M. Ullsperger et al.

    The role of intact frontostriatal circuits in error processing

    J Cogn Neurosci

    (2006)
  • A.M. Hogan et al.

    Impact of frontal white matter lesions on performance monitoring: ERP evidence for cortical disconnection

    Brain

    (2006)
  • E. Magno et al.

    The anterior cingulate and error avoidance

    J Neurosci

    (2006)
  • R.B. Mars et al.

    Neural dynamics of error processing in medial frontal cortex

    Neuroimage

    (2005)
  • C.B. Holroyd et al.

    Dorsal anterior cingulate cortex shows fMRI response to internal and external error signals

    Nat Neurosci

    (2004)
  • Cited by (299)

    View all citing articles on Scopus
    View full text