Trends in Neurosciences
Volume 22, Issue 7, 1 July 1999, Pages 310-316
Journal home page for Trends in Neurosciences

Review
The retrosplenial cortex and emotion: new insights from functional neuroimaging of the human brain

https://doi.org/10.1016/S0166-2236(98)01374-5Get rights and content

Abstract

Little is known about the function of the retrosplenial cortex and until recently, there was no evidence that it had any involvement in emotional processes. Surprisingly, recent functional neuroimaging studies show that the retrosplenial cortex is consistently activated by emotionally salient words. A review of the functional neuroimaging literature reveals a previously overlooked pattern of observations: the retrosplenial cortex is the cortical region most consistently activated by emotionally salient stimuli. Evidence that this region is also involved in episodic memory suggests that it might have a role in the interaction between emotion and episodic memory. Recognition that the retrosplenial cortex has a prominent role in the processing of emotionally salient stimuli invites further studies to define its specific functions and its interactions with other emotion-related brain regions.

Section snippets

Where is the retrosplenial cortex?

The cingulate cortex arches around the genu, body and splenium of the corpus callosum. The posterior cingulate cortex (Fig. 1) is the part that is posterior to the transition from Brodmann's area 24 to area 23 (Talairach y coordinate ∼ − 12), and includes the isocortex of areas 23 and 31. The posterior cingulate cortex also includes the retrosplenial cortex, which comprises area 29, which lies within the callosal sulcus, and area 30, which extends from the callosal sulcus onto the convexity of

What is the evidence for a specific function of the retrosplenial cortex?

Although a variety of affective, cognitive, nociceptive and motor functions have been associated with specific subregions of the anterior cingulate cortex, the functions associated with the posterior cingulate cortex have been less well studied2, 11, 12. The pattern of cortical afferent connections to different regions of the cingulate cortex varies primarily with respect to the rostrocaudal axis. Retrosplenial cortex, the most caudal region of the cingulate cortex, is largely devoid of primary

The retrosplenial cortex and memory

Converging evidence from neuroanatomical studies, clinical lesion studies and functional neuroimaging studies suggests that the retrosplenial cortex has a memory-related function, although its specific function has not been defined. The retrosplenial cortex provides the largest projection to areas TH and TF of the parahippocampal cortex in the monkey and is second only to parahippocampal and perirhinal cortices as a source of input to entorhinal cortex16. The posterior cingulate and

Visuospatial and proprioceptive functions

The posterior cingulate cortex in the monkey is connected to cortical areas that are responsive to visual stimuli, eye movements, spatial orientation and body movements2. Olson et al.28 showed that some monkey posterior-cingulate neurons have responses that are correlated with eye-gaze angle, and the size and direction of eye movements. These neurons were located predominantly anterior to the splenium, although some retrosplenial neurons demonstrated similar responses. In contrast,

A review of functional-neuroimaging studies of emotional processes

In reviewing functional-neuroimaging studies of emotion, it is useful to discuss what is meant by the term ‘emotion’. Although emotion is sometimes defined narrowly as the conscious experience of one of several specific feeling states, such as happiness or fear, the neuroimaging studies of emotion reviewed in this article have generally relied on a broader conceptualization of the term. An important idea within this broader conceptualization of emotion is that emotional phenomena include both

Studies included in this article

This article reviews only peer-reviewed English-language studies that used functional neuroimaging methods to investigate mental processes explicitly conceptualized as ‘emotional’ and that employed repeated-measures designs, thus allowing a comparison of emotional and non-emotional conditions in the same subjects. Studies were excluded from this literature review if they involved patient populations or drug-induced emotional states, or if the analysis did not include most cortical regions. A

Do functional-imaging studies of emotion find activation in the retrosplenial cortex?

Table 1 shows the brain regions that are most commonly activated in these experiments. The posterior cingulate cortex was frequently activated by emotional conditions and was second only to the inferior frontal cortex in frequency of activation. Within the posterior cingulate cortex, activation was typically observed posterior to the splenium, in the retrosplenial cortex and in the adjacent caudal region of the posterior cingulate cortex (Fig. 2). As the boundary between the retrosplenial

Is retrosplenial-cortex activation associated specifically with emotionally salient stimuli?

Many of the studies summarized in Table 1 did not involve comparisons with well-matched emotionally neutral conditions. Thus, it is possible that other, non-emotional, characteristics of the experimental conditions were responsible for the observed activation of the retrosplenial cortex. Among the 51 experiments, 20 experiments from 11 published studies6, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45 compared specifically a task that involved an emotional condition to the same task involving a

Is the retrosplenial cortex frequently activated in non-emotional conditions?

A possible alternative explanation for these results is that the retrosplenial cortex could become activated in response to a variety of experimental conditions in addition to emotion-related conditions. However, recent reviews have shown that retrosplenial cortex is not activated by a variety of experimental conditions and is more typically reported to show ‘negative activation’ (less activation during the experimental condition than the control condition). When reviewing PET studies of

Neuroimaging studies of the retrosplenial cortex in clinical populations

Do studies of patients with neuropsychiatric disorders support the idea that this region has functions related to emotion and memory? In addition to the cases of ‘retrosplenial amnesia’ described above, there is increasing evidence for an association between Alzheimer's disease and hypofunction of the retrosplenial cortex. Several PET studies have demonstrated prominent reduction of metabolic rate in the posterior cingulate and retrosplenial cortices in Alzheimer's disease46, 47. The most

What specific emotion-related functions might be mediated by the retrosplenial cortex?

This question has not yet been addressed experimentally; however, some hypotheses have arisen from the examination of functional-neuroimaging studies of emotion. For example, it appears that the function served by the retrosplenial cortex is not restricted to a single sensory modality. Among the studies summarized in Table 2, the retrosplenial cortex was activated both by studies that used auditory stimuli alone6 and by studies that used visual stimuli alone38, 39, 44. In addition, there is

Does retrosplenial activation reflect emotional evaluation processes?

In general, emotionally salient stimuli are automatically perceived and evaluated as such, whether or not an overt emotional response occurs34, 58. The studies reviewed in this article suggest that the evaluation of emotionally salient stimuli is sufficient to activate the retrosplenial cortex, and that the provocation of an overt emotional response is not a necessary condition for such activation. Activation of the retrosplenial cortex was observed in two out of nine experiments that combined

Does the retrosplenial cortex have a role in the interaction between emotion and episodic memory?

As the retrosplenial cortex appears to have functions that are related to episodic memory, one hypothesis about its role in emotion is that it mediates an interaction between emotion and episodic memory. The retrosplenial cortex is well positioned anatomically to have a role in the interaction between emotion and memory. Two possible ‘circuits’ that mediate such an interaction are suggested by the connectivity of this region. Its strong afferent inputs from the anterior cingulate, the orbital

Concluding remarks

This review of functional neuroimaging studies of emotional processes demonstrates that the retrosplenial cortex is the cortical region that is most consistently activated by the emotional salience of experimental stimuli. These observations suggest that the retrosplenial cortex has a prominent role in the processing of emotionally salient information. The evidence that this region also has episodic-memory functions raises the possibility that it has a role in the interaction between emotion

Acknowledgements

The author thanks Dr David Amaral for his invaluable comments and Trish Foley for her technical assistance in the preparation of this manuscript.

References (65)

  • R.J. Maddock et al.

    Psychiatry Res. Neuroimaging

    (1997)
  • R.J. Maddock et al.

    Biol. Psychiatry

    (1996)
  • P.S. Goldman-Rakic et al.

    Neuroscience

    (1984)
  • R. Kawashima et al.

    Brain Res

    (1994)
  • M.S. George

    Biol. Psychiatry

    (1996)
  • R.D. Lane

    Neuropsychologia

    (1997)
  • S. Minoshima et al.

    Lancet

    (1994)
  • N.C. Andreasen

    Lancet

    (1997)
  • J.F. Demonet

    Neurosci. Lett

    (1994)
  • F. Pratto
  • L. Cahill et al.

    Trends Neurosci

    (1998)
  • E. Valenstein

    Brain

    (1987)
  • B.A. Vogt et al.

    Cereb. Cortex

    (1992)
  • P.M. Grasby

    Brain

    (1993)
  • D.G. Amaral
  • J.E. LeDoux

    Annu. Rev. Psychol

    (1995)
  • R.J. Maddock et al.

    Biol. Psychiatry

    (1998)
  • H. Braak

    Cell Tissue Res

    (1979)
  • B.R. Vogt

    J. Comp. Neurol

    (1995)
  • O. Devinsky et al.

    Brain

    (1995)
  • R. Cabeza et al.

    J. Cogn. Neurosci

    (1997)
  • S.Y. Musil et al.
  • G.W. Van Hoesen et al.
  • W.A. Suzuki et al.

    J. Comp. Neurology

    (1994)
  • T. Shallice

    Nature

    (1994)
  • J. Katai

    Neurology

    (1992)
  • N. Kasahata

    Jpn. J. Stroke

    (1994)
  • N. Takahashi

    Neurology

    (1997)
  • N.C. Andreasen

    Am. J. Psychiatry

    (1995)
  • G.R. Fink

    J. Neurosci

    (1996)
  • L. Nyberg

    NeuroReport

    (1995)
  • A.M. Owen

    J. Cogn. Neurosci

    (1996)
  • Cited by (632)

    View all citing articles on Scopus
    View full text