Neuronal representations of cognitive state: reward or attention?

https://doi.org/10.1016/j.tics.2004.04.003Get rights and content

Abstract

The effects of spatial or featural attention on the activity of neurons have been studied in many experiments that have used a variety of neurophysiological approaches. Other experiments have examined how expectations about reward are represented in neuronal activity in various brain regions. Although attention and reward are distinct concepts, I argue here that many neurophysiological experiments on attention and reward do not permit a clean dissociation between the two. This problem arises in part because reward contingencies are the only parameter manipulated in any of these experiments. I describe how attention and reward expectations have been confounded, giving rise to uncertainty about how signals related to attention and reward are distributed in the brain.

Section snippets

Effects of reward expectation on neuronal activity

Several studies have examined how a monkey's expectations about reward can influence the activity of individual neurons. In most such experiments, expectations are varied by having the animal do blocks of trials in which reward conditions are kept constant within blocks (allowing stable expectations to develop) but varied between blocks (allowing comparison of the effects of different expectations). Rewards are typically manipulated by changing either the amount of reward delivered for a

Reward expectation or attention?

The effect of attention on neuronal responses has been studied extensively (reviewed in 18, 19, 20, 21). Single unit studies in visual cortex have shown that increased attention to a stimulus typically increases the responses of neurons that represent that stimulus [22]. Many such studies have examined the effect of attending to a particular stimulus or location in space, but it has been shown that attention to a particular visual feature, such as a direction of motion, can also selectively

Favoring one interpretation

Most studies of attention or reward expectations have not addressed the potential for confusion between them, and have interpreted neuronal signals as either attention-related or reward-related without considering alternatives. However, a few reports did consider the problem and argued specifically for one interpretation. Although such views might be correct, the data do not provide strong support.

Kawagoe and colleagues [15] sought to distinguish their reward-related neuronal modulations from

Distinguishing reward and attention

Recent experiments do not distinguish whether neuronal signals are related to attention or to expectations about reward. Can changes in neuronal activity be assigned unequivocally to one or the other? A complete treatment of the issues involved in defining and distinguishing cognitive signals such as these is beyond the scope of this article. Nevertheless, is it helpful to consider the range of possibilities for signals related to attention and reward, and to suggest approaches that might serve

Conclusion

In summary, I have argued that some experiments examining neuronal representations of attention or reward expectation might in fact have been measuring either of those parameters. Given our limited understanding of the organization of the neuronal representations that support cognition, clarifying our terminologies and descriptions of neuronal activity is crucial for identifying the signals that are most important for different cognitive functions. Efforts to distinguish the relative

Acknowledgments

I thank John A. Assad, William H. Bosking, Raika Pancaroglu and David L. Sparks insightful discussions and helpful comments on earlier versions of this manuscript. Supported by NIH R01 EY05911. J.H.R.M. is an Investigator with the Howard Hughes Medical Institute.

References (41)

  • H.E Pashler

    The Psychology of Attention

    (1998)
  • M.L Platt et al.

    Neural correlates of decision variables in parietal cortex

    Nature

    (1999)
  • W.T Newsome

    Decision-making and the neural representation of ‘experienced value’

    Soc. Neurosci. Abstr. (Abstract Viewer)

    (2003)
  • M.S Bendiksby et al.

    Motivation focuses attention and enhances neuronal selectivity in parietal cortex

    Soc. Neurosci. Abstr. (Abstract Viewer)

    (2003)
  • M.R Roesch et al.

    Impact of expected reward on neuronal activity in prefrontal cortex, frontal and supplementary eye fields and premotor cortex

    J. Neurophysiol.

    (2003)
  • M Watanabe

    Reward expectancy in primate prefrontal neurons

    Nature

    (1996)
  • M Watanabe

    Prefrontal unit activity during associative learning in the monkey

    Exp. Brain Res.

    (1990)
  • S Kobayashi

    Influence of reward expectation on visuospatial processing in macaque lateral prefrontal cortex

    J. Neurophysiol.

    (2002)
  • M Sato et al.

    Role of primate substantia nigra pars reticulata in reward-oriented saccadic eye movement

    J. Neurosci.

    (2002)
  • K Watanabe

    Neural correlates of rewarded and unrewarded eye movements in the primate caudate nucleus

    J. Neurosci.

    (2003)
  • Cited by (331)

    • Amygdala function in emotion, cognition, and behavior

      2022, Handbook of Clinical Neurology
    View all citing articles on Scopus
    View full text