Valuation and decision-making in frontal cortex: one or many serial or parallel systems?

https://doi.org/10.1016/j.conb.2012.04.011Get rights and content

We evaluate the merits of different conceptualizations of frontal cortex function in value-guided decision-making. According to one view each frontal cortical region is concerned with a different aspect of the process of learning about and evaluating choices and then selecting actions. An alternative view, however, sees sets of decision-making circuits working in parallel within the frontal lobes in order to make different types of decisions. While there is a neural circuit for making choices between pairs of simultaneously presented items in the manner that is frequently assessed in the laboratory, there is also evidence that other frontal lobe circuits have evolved to make other types of choices such as those made during the course of foraging.

Highlights

► We review three different accounts of frontal cortical function in reward-guided decision-making. ► A serial model proposes that value expectation is represented in ventromedial prefrontal cortex/medial orbitofrontal cortex (vmPFC/mOFC) while value comparison takes place in anterior cingulate cortex (ACC). ► A parallel model envisages parallel comparison processes occurring in both vmPFC/mOFC and ACC. ► vmPFC/mOFC and ACC may be concerned with decision-making and foraging, respectively.

Section snippets

Value assignment and prefrontal cortex

An emerging and influential account of frontal brain mechanisms of decision-making holds that what we and other animals do when we make a choice is to decide between two different goods on the basis of their independently computed reward values [1]. The lateral orbitofrontal cortex (lOFC) plays a central role in learning the values associated with different goods. LOFC lesions disrupt the assignment of precise values to stimuli [2••]. The representations of values in this area are appropriate

A possible serial circuit in frontal cortex for making a value-guided choice

In contrast to lOFC, vmPFC/mOFC appears more intimately concerned with the use of reward representations to guide behaviour (Figure 1b). Little is known about neurons in vmPFC/mOFC but it is clear that while they represent rewards they differ from lOFC neurons because they encode little about the stimuli that are associated with the rewards [14]. Several human neuroimaging studies have reported vmPFC/mOFC blood oxygen level dependent (BOLD) signals that are proportional to reward expectations

Different decisions in frontal cortex: decisions about rewards and about actions

An alternative interpretation is that vmPFC/mOFC is an important determinant of value-guided decisions (Figure 1c). Not only does vmPFC/mOFC activity level reflect the value of the goal that is currently being pursued [6, 15] but it also represents the values of both options that are being considered during the course of a decision [24, 25••, 26]. There is a positive relationship between vmPFC/mOFC BOLD and the value of the choice taken, and a negative relationship with the value of the choice

Parallel decision-making mechanisms

Perhaps the most intriguing question is whether there is a single valuation system or many valuation systems (Figure 1d). For example, a distinction has been drawn between habitual and goal-based decision-making [34••, 35] and there is evidence for differences between social and non-social decision-making [36]. There may, however, be other types of fundamental distinctions to be drawn between types of value-guided choice.

It is clear that we and other primates can make the binary comparative

Conclusions

In summary, a goods-based account of decision-making provides a persuasive description of several frontal cortical brain regions as well as of the decisions of greatest interest to economists. We may not, however, have evolved to make only such decisions. Other regions such as aPFC and ACC, that encode the best alternative action to the one that is currently being taken or the average richness of the foraging environment, also influence choice (Figure 1, Figure 3, Figure 4). Their existence

References and recommended reading

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

  • •• of outstanding interest

Acknowledgements

Funded by the MRC and Wellcome Trust.

References (52)

  • S. Kobayashi et al.

    Adaptation of reward sensitivity in orbitofrontal neurons

    J Neurosci

    (2010)
  • S.W. Kennerley et al.

    Double dissociation of value computations in orbitofrontal and anterior cingulate neurons

    Nat Neurosci

    (2011)
  • M.F. Rushworth et al.

    Frontal cortex and reward-guided learning and decision-making

    Neuron

    (2011)
  • M.R. Milad et al.

    Obsessive–compulsive disorder: beyond segregated cortico-striatal pathways

    Trends Cogn Sci

    (2012)
  • S.E. Morrison et al.

    The convergence of information about rewarding and aversive stimuli in single neurons

    J Neurosci

    (2009)
  • M. Noonan et al.

    Re-evaluating the role of orbitofrontal cortex in reward and reinforcement

    Eur J Neurosci

    (2012)
  • M.P. Noonan et al.

    Distinct roles of three frontal cortical areas in reward-guided behavior

    J Neurosci

    (2011)
  • P.H. Rudebeck et al.

    Dissociable effects of subtotal lesions within the macaque orbital prefrontal cortex on reward-guided behavior

    J Neurosci

    (2011)
  • S. Bouret et al.

    Ventromedial and orbital prefrontal neurons differentially encode internally and externally driven motivational values in monkeys

    J Neurosci

    (2010)
  • M. Lebreton et al.

    An automatic valuation system in the human brain: evidence from functional neuroimaging

    Neuron

    (2009)
  • D.V. Smith et al.

    Distinct value signals in anterior and posterior ventromedial prefrontal cortex

    J Neurosci

    (2010)
  • A. Litt et al.

    Dissociating valuation and saliency signals during decision-making

    Cereb Cortex

    (2011)
  • S.J. Gershman et al.

    Human reinforcement learning subdivides structured action spaces by learning effector-specific values

    J Neurosci

    (2009)
  • U. Basten et al.

    How the brain integrates costs and benefits during decision making

    Proc Natl Acad Sci U S A

    (2010)
  • M.G. Philiastides et al.

    A mechanistic account of value computation in the human brain

    Proc Natl Acad Sci U S A

    (2010)
  • T.A. Hare et al.

    Transformation of stimulus value signals into motor commands during simple choice

    Proc Natl Acad Sci U S A

    (2011)
  • Cited by (227)

    View all citing articles on Scopus
    View full text