Re-evaluating the role of the mammillary bodies in memory
ABSTRACT Although the mammillary bodies were among the first brain regions to be implicated in amnesia, the functional importance of this structure for memory has been questioned over the intervening years. Recent patient studies have, however, re-established the mammillary bodies, and their projections to the anterior thalamus via the mammillothalamic tract, as being crucial for recollective memory. Complementary animal research has also made substantial advances in recent years by determining the electrophysiological, neurochemical, anatomical and functional properties of the mammillary bodies. Mammillary body and mammillothalamic tract lesions in rats impair performance on a number of spatial memory tasks and these deficits are consistent with impoverished spatial encoding. The mammillary bodies have traditionally been considered a hippocampal relay which is consistent with the equivalent deficits seen following lesions of the mammillary bodies or their major efferents, the mammillothalamic tract. However, recent findings suggest that the mammillary bodies may have a role in memory that is independent of their hippocampal formation afferents; instead, the ventral tegmental nucleus of Gudden could be providing critical mammillary body inputs needed to support mnemonic processes. Finally, it is now apparent that the medial and lateral mammillary nuclei should be considered separately and initial research indicates that the medial mammillary nucleus is predominantly responsible for the spatial memory deficits following mammillary body lesions in rats.
- SourceAvailable from: R. Nathan Spreng[Show abstract] [Hide abstract]
ABSTRACT: Sex differences in episodic and spatial memory are frequently observed, suggesting there may be sex-related structural differences in the hippocampus (HC). Earlier findings are inconsistent, possibly due to known variability along the hippocampal longitudinal axis. Here, we assessed potential sex differences in hippocampal volume and structural covariance with the rest of the brain in young men and women (N=76), considering the anterior (aHC) and posterior (pHC) hippocampus separately. Women exhibited a larger pHC than men adjusted for brain size. Using partial least squares, we identified two significant patterns of structural covariance of aHC and pHC. The first included brain areas that covaried positively and negatively in volume with both aHC and pHC in men, but showed greater covariance with aHC than pHC in women. The second pattern revealed distinct structural covariance of aHC and pHC that showed a clear difference between men and women: in men pHC showed reliable structural covariance with the medial and lateral parietal lobes and the prefrontal cortex, whereas in women aHC showed reliable structural covariance with the anterior temporal lobe bilaterally. This pattern converges with resting state functional connectivity of aHC and pHC and suggests that these hippocampal sections interact with different brain regions, consistent with a division of labor with regards to episodic and spatial memory. Our findings lend support to a division of the HC into an anterior and posterior part and identify sex as a potential moderating factor when investigating hippocampal structure and connectivity.NeuroImage 05/2014; · 6.25 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: To better understand serotonin function in the primate brain, we examined the mRNA expression patterns of all the 13 members of the serotonin receptor (5HTR) family, by in situ hybridization (ISH) and the distribution of serotonergic terminations by serotonin transporter (SERT) protein immunohistochemical analysis. Ten of the 13 5HTRs showed significant mRNA expressions in the marmoset brain. Our study shows several new features of the organization of serotonergic systems in the marmoset brain. (1) The thalamus expressed only a limited number of receptor subtypes compared with the cortex, hippocampus, and other subcortical regions. (2) In the cortex, there are layer-selective and area-selective mRNA expressions of 5HTRs. (3) Highly localized mRNA expressions of 5HT1F and 5HT3A were observed. (4) There was a conspicuous overlap of the mRNA expressions of receptor subtypes known to have somatodendritic localization of receptor proteins with dense serotonergic terminations in the visual cortex, the central lateral (CL) nucleus of the thalamus, the presubiculum, and the medial mammillary nucleus of the hypothalamus. This suggests a high correlation between serotonin availability and receptor expression at these locations. (5) The 5HTRs show differences in mRNA expression pattern between the marmoset and mouse cortices whereas the patterns of both the species were much similar in the hippocampus. We discuss the possible roles of 5HTRs in the marmoset brain revealed by the analysis of their overall mRNA expression patterns.Frontiers in Neural Circuits 05/2014; 8(52). · 3.33 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Although anterograde amnesia can occur after damage in various brain sites, hippocampal dysfunction is usually seen as the ultimate cause of the failure to learn new episodic information. This assumption is supported by anatomical evidence showing direct hippocampal connections with all other sites implicated in causing anterograde amnesia. Likewise, behavioural and clinical evidence would seem to strengthen the established notion of an episodic memory system emanating from the hippocampus. There is, however, growing evidence that key, interconnected sites may also regulate the hippocampus, reflecting a more balanced, integrated network that enables learning. Recent behavioural evidence strongly suggests that medial diencephalic structures have some mnemonic functions independent of the hippocampus, which can then act upon the hippocampus. Anatomical findings now reveal that nucleus reuniens and the retrosplenial cortex provide parallel, disynaptic routes for prefrontal control of hippocampal activity. There is also growing clinical evidence that retrosplenial cortex dysfunctions contribute to both anterograde amnesia and the earliest stages of Alzheimer's disease, revealing the potential significance of this area for clinical studies. This array of findings underlines the importance of redressing the balance and the value of looking beyond the hippocampus when seeking to explain failures in learning new episodic information.Proceedings of the Royal Society B: Biological Sciences 07/2014; 281(1786). · 5.68 Impact Factor
Neuropsychologia 48 (2010) 2316–2327
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/neuropsychologia
Re-evaluating the role of the mammillary bodies in memory
Seralynne D. Vann∗
School of Psychology, Cardiff University, Tower Building, 70 Park Place, Cardiff CF10 3AT, UK
a r t i c l ei n f o
Received 2 June 2009
Received in revised form 20 October 2009
Accepted 21 October 2009
Available online 30 October 2009
Tegmental nucleus of Gudden
a b s t r a c t
tional importance of this structure for memory has been questioned over the intervening years. Recent
patient studies have, however, re-established the mammillary bodies, and their projections to the ante-
rior thalamus via the mammillothalamic tract, as being crucial for recollective memory. Complementary
ical, neurochemical, anatomical and functional properties of the mammillary bodies. Mammillary body
these deficits are consistent with impoverished spatial encoding. The mammillary bodies have tradition-
ally been considered a hippocampal relay which is consistent with the equivalent deficits seen following
lesions of the mammillary bodies or their major efferents, the mammillothalamic tract. However, recent
findings suggest that the mammillary bodies may have a role in memory that is independent of their
hippocampal formation afferents; instead, the ventral tegmental nucleus of Gudden could be provid-
ing critical mammillary body inputs needed to support mnemonic processes. Finally, it is now apparent
that the medial and lateral mammillary nuclei should be considered separately and initial research indi-
cates that the medial mammillary nucleus is predominantly responsible for the spatial memory deficits
following mammillary body lesions in rats.
© 2009 Elsevier Ltd. All rights reserved.
The mammillary bodies have a number of features that sin-
gle them out as prime targets for research into episodic memory
in humans and episodic-like memory in rodents. The first is his-
torical: it is over one hundred years since Gudden (1896) first
arguably the very first brain region implicated in amnesia. Despite
this fact, remarkably little research has been conducted on this
(medial and lateral), with a limited array of cell types in each. The
third is connectional: the mammillary bodies have major connec-
tions with a limited number of structures. These connections are
making it possible to make selective disconnections of mammil-
lary body inputs and outputs. The presence of mammillary body
atrophy in a number of conditions, including Korsakoff’s syndrome
(Kopelman, 1995; Victor, Adams, & Collins, 1989), colloid cysts
in the third ventricle (Denby et al., 2009), Alzheimer’s disease
∗Tel.: +44 2920 876253; fax: +44 2920 874858.
E-mail address: email@example.com.
(Callen, Black, Gao, Caldwell, & Szalai, 2001; Copenhaver et al.,
2006; Grossi, Lopez, & Martinez, 1989), schizophrenia (Bernstein
et al., 2007; Briess, Cotter, Doshi, & Everall, 1998), heart failure
(Kumar et al., 2009), and sleep apnea (Kumar et al., 2008), empha-
sizes the growing need to clarify the functional importance of this
Until recently, uncertainty regarding the importance of the
region. Most obviously, neuropathological studies of Korsakoff’s
syndrome, one of the most common causes of amnesia, had failed
to establish the extent to which mammillary body atrophy con-
tributes to the memory loss in this condition (Harding, Halliday,
et al., 1989). This uncertain picture has changed dramatically in
recent years. One key discovery was that damage to the mammil-
lothalamic tract provides the sole, consistent predictor of whether
a thalamic stroke will cause anterograde amnesia (Carlesimo et
al., 2007; Clarke et al., 1994; Graff-Radford, Tranel, Van Hoesen,
& Brandt, 1990; Van der Werf, Jolles, Witter, & Uylings, 2003; Van
der Werf, Scheltens, et al., 2003; Van der Werf, Witter, Uylings, &
Jolles, 2000; von Cramon, Hebel, & Schuri, 1985; Yoneoka et al.,
2004). The mammillothalamic tract is a major fiber tract formed
by the unidirectional projections that arise from every mammil-
lary body neuron to terminate in the anterior thalamus (Guillery,
1955; Vann, Saunders, & Aggleton, 2007). While it is most likely
0028-3932/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
cephalic amnesia (e.g. intralaminar nuclei/medial dorsal nucleus),
the pathway from the mammillary bodies to the anterior thalamus
seems to be of paramount importance. More recently, a study of
revealed the mammillary bodies to be the only site consistently
linked to recollective memory impairments (Tsivilis et al., 2008;
Vann et al., 2009). Patients, who were matched on all factors other
than the degree of mammillary body atrophy, differed significantly
on measures of recollection but not familiarity-based recognition
(Tsivilis et al., 2008; Vann et al., 2009), consistent with some dual-
process models of memory (e.g. Aggleton & Brown, 1999). While
all these patient studies have implicated the mammillary bodies in
human cognitive processes, pathology outside of the mammillary
interpretations. Due to the size and position of the mammillary
bodies, current functional imaging techniques are unsuitable for
investigating this structure in healthy controls. These limitations
in human mammillary body research make animal models all the
bodies in memory emphasize the importance of the hippocam-
pal inputs to the region; indeed the mammillary bodies are often
referred to as part of an “extended hippocampal system” (Aggleton
& Brown, 1999; Delay & Brion, 1969; Gaffan, 1992; Gaffan, 2001).
Remarkably, the mammillary bodies appear to lack interneurons
(Veazey, Amaral, & Cowan, 1982b) and every mammillary neuron
is thought to project to the thalamus (e.g. Guillery, 1955; Vann et
al., 2007); properties that have reinforced the idea of the struc-
ture simply acting as a relay. Within this extended hippocampal
system, the proposed functional importance of the mammillary
bodies and mammillothalamic tract is to relay hippocampal inputs
to the anterior thalamic nuclei and from there to the cingulate cor-
tex (Barbizet, 1963; Delay & Brion, 1969) or the prefrontal cortex
(Warrington & Weiskrantz, 1982). Memory impairments following
mammillary body lesions would, therefore, reflect the disconnec-
tion of either the cingulate cortex or prefrontal cortex from these
hippocampal inputs. An alternative model combines these two
accounts of mammillary body function by proposing that dien-
to a loss of hippocampal inputs, and this is responsible for sub-
sequent memory impairment (Mair, Warrington, & Weiskrantz,
1979; Paller, 1997; Vann & Aggleton, 2004). However, because
the hippocampus projects directly to the anterior thalamic nuclei
(Aggleton, Desimone, & Mishkin, 1986; Poletti & Creswell, 1977),
prefrontal cortex (Jay, Glowinski, & Thierry, 1989) and cingulate
cortex (Meibach & Siegel, 1977b), it must be assumed that the hip-
pocampal formation–mammillary body projections are providing
unique information or these indirect pathways would seem redun-
dant; this is possible as the majority of the hippocampal formation
efferents arise from different populations of cells within the subic-
ular complex and CA1 (Aggleton, Vann, & Saunders, 2005; Jay et
al., 1989; Namura, Takada, Kikuchi, & Mizuno, 1994; Saunders,
Mishkin, & Aggleton, 2005; Witter, Groenewegen, Lopes da Silva, &
The similarity in mammillary body structure and connections
across species make animal studies of this brain region partic-
ularly relevant. However, the size and/or type of mammillary
body lesions in some studies have made it, on occasion, difficult
to interpret lesion effects. For example, electrolytic or radiofre-
quency lesions (e.g. Harper, McLean, & Dalrymple-Alford, 1994;
Santin, Rubio, Begega, & Arias, 1999; Saravis, Sziklas, & Petrides,
1990; Sharp & Koester, 2008a, 2008b; Sziklas & Petrides, 1993;
Tonkiss & Rawlins, 1992) will also damage the large number of
fiber tracts in the vicinity of the mammillary bodies (Nauta &
Haymaker, 1969). If the lesions are too large they will encroach
upon adjacent structures, most likely the supramammillary nuclei
which are situated immediately dorsal to the mammillary bod-
ies (e.g. Saravis et al., 1990; Sharp & Koester, 2008a, 2008b;
implicated in hippocampal theta and may also contribute to spa-
tial memory (e.g. Aranda, Santin, Begega, Aguirre, & Arias, 2006;
Pan & McNaughton, 1997,2002,2004; Vann, Brown, & Aggleton,
2000), additional damage to this structure makes the subsequent
interpretation of mammillary body lesion effects problematic.
Finally, if insufficient detail is given regarding the size and loca-
tion of the lesions (e.g. Sharp & Koester, 2008a, 2008b) and the
extent of disconnection following mammillothalamic tract lesions
is not determined (e.g. Sharp & Koester, 2008a, 2008b; Vann
& Aggleton, 2003; Vann, Honey, & Aggleton, 2003) it becomes
increasingly difficult to attribute subsequent effects or lack of
effects with any confidence. These issues highlight the need to
make the best use of animal models by making discrete lesions,
determining the extent of any fiber disconnections, and provid-
ing sufficient detail regarding the extent of pathology. Despite
some of these limitations, animal studies have been invaluable
in advancing our knowledge about the mammillary bodies. In
the following sections the anatomical, electrophysiological, and
functional properties of the mammillary bodies, as revealed by
these studies, will be detailed. Current models of mammillary body
function will then be re-evaluated in light of more recent find-
The mammillary bodies comprise two main nuclei: medial and
lateral. The medial mammillary nucleus is the larger of the two
nuclei with the lateral mammillary nuclei accounting for about
6% of the entire structure across species (Guillery, 1955; Rose,
1939). The medial mammillary nucleus has been divided fur-
ther into between one and six subregions depending on species
and anatomist (Allen & Hopkins, 1988; Rose, 1939). However, the
most commonly used distinctions across species are pars lateralis,
pars medialis, and pars basalis (Fig. 1). While the medial and lat-
eral mammillary nuclei differ in terms of their cell morphology,
within each nucleus there appears to be only one cell type (Cajal,
1911; Veazey et al., 1982b). The neurons in the lateral mammil-
lary nucleus are much larger than the very small neurons found in
the pars lateralis of the medial nucleus and the intermediate size
Allen, & Hopkins, 1985) with no apparent interneurons (Veazey et
The mammillary bodies have major connections with only a
limited number of structures (Fig. 2). The principal mammillary
body inputs are from the hippocampal formation (via the post-
commissural fornix) and from the tegmental nuclei of Gudden
(via the mammillary peduncle). Their main outputs are to the
anterior thalamic nuclei (via the mammillothalamic tract) and
to the tegmental nuclei of Gudden (via the mammillotegmen-
tal tract); some of these projections arise from collateral axons
are connected to the same overall structures but different subre-
gions within those structures, thus forming two parallel systems
(Vann & Aggleton, 2004). With regards to the hippocampal for-
mation afferents, the medial mammillary nucleus receives inputs
from the dorsal, ventral and intermediate subiculum and medial
entorhinal cortex while the lateral mammillary nucleus is inner-
vated by projections from the presubiculum, postsubiculum, and
parasubiculum (Allen & Hopkins, 1989; Meibach & Siegel, 1977a;
Shibata, 1988; Swanson & Cowan, 1977; Van Groen & Wyss, 1990).
In terms of the anterior thalamic projections, the medial mammil-
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
lary nuclei project unilaterally to the anterior medial and anterior
ventral thalamic nuclei whereas the lateral mammillary nuclei
Seki & Zyo, 1984; Vann et al., 2007). The medial mammillary
Fig. 1. a. Magnetic resonance scan showing human mammillary bodies in the coro-
nal plane. b. Nissl-stained coronal section showing the mammillary nuclei in the
cynomolgus monkey (Macaca fascicularis). c. Coronal section showing mammillary
nuclei in the rat (Fluorogold). d. High-power photomicrograh showing lateral mam-
millary nucleus and pars lateralis of the medial mammillary nucleus. B, pars basalis,
L, pars lateralis, LMN, lateral mammillary nucleus; M, pars medialis; MMN, medial
mammillary nucleus; MTT, mammillothalamic tract. Scale bar for b and c, 1mm;
Scale bar for d, 0.25mm.
nucleus has reciprocal connections with the ventral tegmental
nucleus of Gudden and the lateral mammillary nucleus has recip-
rocal connections with the dorsal tegmental nucleus of Gudden
(Cruce, 1977; Hayakawa & Zyo, 1984,1985,1989; Veazey, Amaral,
& Cowan, 1982a). Both medial and lateral mammillary nuclei are
innervated by the supramammillary nucleus, the tuberomammil-
lary nucleus and the septal region (Cajal, 1911; Fry & Cowan, 1972;
Shibata, 1989) and both medial and lateral mammillary nuclei
project to separate but adjacent parts of the nucleus reticularis
tegmenti ponti and pontine nuclei (Cruce, 1977; Takeuchi et al.,
1985); these projections are of interest as they provide a mecha-
nism for the mammillary bodies to influence visual and vestibular
processes (Allen & Hopkins, 1990; Hopkins, 2005). The prefrontal
cortex appears to be the only region where the parallel lateral and
mammillary connections are not upheld as the prefrontal cortex
Although the mammillary bodies receive both excitatory and
inhibitory inputs, their major outputs appear to be solely exci-
tatory. The mammillary bodies receive excitatory inputs from
the hippocampal formation; these projection use either glu-
tamate or aspartate (Storm-Mathisen & Woxen Opsahl, 1978)
as well as neurotensin (Kiyama et al., 1986; Sakamoto et al.,
1986). The inputs from the prefrontal cortex are also excita-
tory (Allen & Hopkins, 1989). The projections from the tegmental
nuclei are inhibitory, using GABA and leu-enkephalin (Allen &
Hopkins, 1989; Gonzalo-Ruiz, Romero, Sanz, & Morte, 1999;
Gonzalo-Ruiz, Sanz-Anquela, & Spencer, 1993; Hayakawa & Zyo,
1991; Wirtshafter & Stratford, 1993). The mammillary bodies
also receive a dopaminergic input from the nearby supramam-
millary nuclei (Gonzalo-Ruiz, Alonso, Sanz, & Llinas, 1992b). The
mammillary body efferents to both the tegmental nuclei and the
anterior thalamic nuclei are excitatory (Allen & Hopkins, 1990;
Gonzalo-Ruiz, Morte, & Sanz, 1998). The projections to the ante-
rior thalamic nuclei use glutamate, aspartate (Gonzalo-Ruiz et
al., 1998), enkephalins (Fujii, Senba, Kiyama, Ueda, & Tohyama,
1987; Gonzalo-Ruiz et al., 1998) and cholecystokinin (Kiyama et
Consistent with their different anatomical properties and con-
nections, the lateral and medial mammillary nuclei also have very
both in vitro and in vivo. Recent electrophysiological discoveries
have been extremely instrumental in developing models of mam-
millary body function.
3.1. In vitro
Using an in vitro slice preparation, Alonso and Llinas found
maker properties and these are mediated by calcium-dependent
mechanisms (Alonso & Llinas, 1992). Although there only appears
to be one cell type within the medial mammillary nucleus (Cajal,
distinct cell populations within this structure which differ in terms
also able to switch from tonic repetitive firing to a low threshold-
bursting pattern and, as with the medial mammillary neurons, this
response is calcium-dependent (Llinas & Alonso, 1992). Unlike the
cells in the medial mammillary nuclei, the lateral mammillary neu-
rons appear electrophysiologically homogenous (Llinas & Alonso,
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
Fig. 2. The main direct connections of the medial and lateral mammillary nuclei.
3.2. In vivo
3.2.1. Lateral mammillary nucleus
Both head-direction cells and angular velocity cells have been
reported in the lateral mammillary nucleus (Blair, Cho, & Sharp,
1998; Stackman & Taube, 1998). Head-direction cells and angular
velocity cells fire differentially depending on the rat’s head-
direction (Taube, Muller, & Ranck, 1990) or velocity of head
movements (Stackman & Taube, 1998), respectively. The lateral
mammillary nuclei are connected with a number of other head-
direction regions and they are instrumental in the generation and
maintenance of the head-direction signal throughout the head-
direction circuit. The lateral mammillary nucleus requires the
inputs from dorsal tegmental nucleus of Gudden to generate the
head-direction signal (Bassett, Tullman, & Taube, 2007). In turn,
(Bassett et al., 2007) and postsubiculum (Sharp & Koester, 2008b)
are dependent on the integrity of the lateral mammillary nuclei.
mammillary nuclei, these cells only account for a small propor-
tion of the total cell numbers in this structure (Blair et al., 1998;
Stackman & Taube, 1998).
3.2.2. Medial mammillary nucleus
While there are no head-direction cells in the medial mammil-
lary nucleus, approximately one third of cells respond to angular
head velocity (Sharp & Turner-Williams, 2005). However, unlike
the angular head velocity cells in the lateral mammillary nuclei,
which fire irrespective of the direction in which the animal’s head
is turning (Stackman & Taube, 1998), the cells in the medial mam-
millary nuclei fire differentially for clockwise and anticlockwise
movements (Sharp & Turner-Williams, 2005). In addition, over half
of the cells in the medial mammillary nucleus correlate with run-
velocity (Sharp & Turner-Williams, 2005). Finally, nearly all cells in
the medial mammillary nucleus modulate their firing rate at a fre-
quency of theta (Bland, Konopacki, Kirk, Oddie, & Dickson, 1995;
Kirk, 1998; Kirk, Oddie, Konopacki, & Bland, 1996; Kocsis & Vertes,
1994; Sharp & Turner-Williams, 2005). The current view is that
medial mammillary nucleus theta is driven by the CA1 field of
the hippocampus as theta-related cells in the medial mammillary
nuclei show a strong correlation with CA1 theta (Kocsis & Vertes,
1994). In addition, septal procaine infusion that attenuates hip-
providing further evidence that medial mammillary body theta is
driven by descending projections from the septo-hippocamapal
system (Kirk et al., 1996). One proposal is that the medial mam-
millary bodies act as a relay of hippocampal theta, passing it along
the diencephalon and back to the hippocampus, thus forming a
re-entrant loop which is necessary for successful encoding (Kirk &
More recently, the ventral tegmental nucleus of Gudden has
been linked to mammillary body theta. As described earlier, the
medial mammillary nucleus has reciprocal connections with the
ventral tegmental nucleus of Gudden, and all cells in this struc-
ture fire rhythmically and highly coherently with hippocampal
theta (Kocsis, Di Prisco, & Vertes, 2001). One account is that the
ventral tegmental nucleus of Gudden moderates the hippocampal-
driven rhythmic firing in the medial mammillary (Kocsis et al.,
2001; Vertes, Hoover, & Viana Di Prisco, 2004). A more radical
account is that the Gudden’s ventral tegmental nucleus is a genera-
tegmental nucleus occur 1 or 2s prior to the onset of hippocampal
theta (Bassant & Poindessous-Jazat, 2001). In addition, Gudden’s
ventral tegmental nucleus recordings show similar properties to
the rhythmic discharges seen in the medial septum/diagonal band
(Apartis, Poindessous-Jazat, Lamour, & Bassant, 1998). However,
lesions of the supramammillary nuclei, which may also involve
the mammillary bodies or mammillothalamic tract, can disrupt
some aspects of hippocampal theta but do not eliminate it com-
pletely (Sharp & Koester, 2008a; Thinschmidt, Kinney, & Kocsis,
1995). As Gudden’s ventral tegmental nucleus would presumably
act on hippocampal theta via the mammillary bodies and/or supra-
mammillary nuclei, it seems unlikely that this tegmental nucleus
generates hippocampal theta, although it could act as a modulator.
Finally, the projections from the mammillary bodies to the
anterior thalamic nuclei are necessary for the excitatory training-
induced activity in the anteroventral thalamic nucleus that occurs
when rabbits learn a conditional avoidance discrimination (Gabriel
et al., 1995). Not only are the mammillary body projections to
the anterior thalamic nuclei necessary for these behavior-induced
changes in anteroventral nucleus activity, but they are also needed
for the spontaneous baseline unit activity normally seen in the
anteroventral thalamic nucleus (Gabriel et al., 1995). This finding
demonstrates the importance of mammillary body efferents for
normal anterior thalamic function.
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
4. Behavioral lesion studies
As current models of mammillary body function emphasize
their hippocampal formation inputs, the majority of behavioral
lesion studies in rodents have focused on spatial memory. Mam-
millary body lesions, in both mice and rats, disrupt performance on
reinforced and spontaneous T-maze alternation (Aggleton, Neave,
Nagle, & Hunt, 1995; Beracochea & Jaffard, 1987,1990; Gaffan,
Bannerman, Warburton, & Aggleton, 2001; Rosenstock, Field, &
during the standard task, where animals are given a separate sam-
ple and test trial, and also during continuous alternation which
increases task demands (Aggleton et al., 1995; Field, Rosenstock,
King, & Greene, 1978; Vann & Aggleton, 2003). Lesions of the main
millotegmental tract, also disrupt T-maze performance (Field et al.,
1978; Vann & Aggleton, 2003).
Tasks in the water-maze have also been used to assess the
importance of the mammillary bodies for allocentric spatial mem-
ory. These tasks prevent the use of intra-maze cues, such as odor
trails, which can in some instances mask impairments. For the ref-
erence memory task in the water-maze, the platform remains in
the same place in the water-maze throughout testing thus result-
ing in very low levels of proactive interference. Sutherland and
Rodriguez found rats with mammillary body lesions to be slower
at learning a platform position, although this impairment disap-
peared with training (Sutherland & Rodriguez, 1989). However, a
later study reported no effect of mammillary body lesions on the
reference memory task (Santin et al., 1999). The working memory
task (delayed matching-to-place) in the water-maze differs in that
the platform is in a different position during each session so that
animals have to rapidly learn the new spatial location within a ses-
impairments on the working memory task in the water-maze
(Santin et al., 1999; Vann & Aggleton, 2003). In addition, lesions of
the mammillary bodies or mammillothalamic tract produce equiv-
alent impairments on this task (Vann & Aggleton, 2003). The size of
on prior experience as pre-surgical training can improve perfor-
mance on reinforced alternation (Rosenstock et al., 1977) and the
Another widely used paradigm to assess spatial memory is the
radial-arm-maze task. The “working memory” version of this task
requires animals to enter all arms to retrieve a reward, such as a
previously entered arms. Animals, therefore, have to keep track of
the arms they have entered within a session. Rats with mammil-
lary body lesions and mammillothalamic tract lesions are impaired
centric cues (Vann & Aggleton, 2003). Using a modified task in
the radial-arm-maze, designed to assess memory for “lists”, mam-
millary body lesions disrupt both the primacy and recency effects
McLean, 1993). In contrast to the frequently reported spatial mem-
ory impairments, mammillary body lesions in rats do not affect
object recognition (Aggleton et al., 1995), again consistent with
some dual-models of recognition memory (e.g. Aggleton & Brown,
1999) and findings from patient studies (Kapur et al., 1994; Tsivilis
et al., 2008; Vann et al., 2009).
The standard tasks in the T-maze, water-maze and radial-
arm-maze require animals to use distal spatial cues to navigate
in the environment; however, mammillary body lesion effects
are not restricted to tasks that involve navigation. Mammillary
body lesions facilitate rats’ performance on a visual discrimination
task where animals are required to discriminate simultaneously
between two scenes that contain different combinations of the
same objects and positions (Gaffan et al., 2001). Mammillary body
lesions also facilitate performance when animals are required to
discriminate between familiar and less familiar objects, and famil-
iar and less familiar object-position compounds, but have no effect
when animals are required to discriminate familiar positions from
less familiar positions (Gaffan et al., 2001). A possible explanation
put forward for this facilitation was that animals with mammillary
process one object at a time which would give them an advantage
on certain trial types (Gaffan et al., 2001). Mammillothalamic tract
lesions in rats impair acquisition of a visuo-spatial but not non-
spatial conditional discrimination task in an operant box (Vann
et al., 2003). For this task, animals have to respond differentially
to stimuli (light/sound) depending on which visual–spatial con-
text (spotted/striped walls) or non-spatial context (hot/cold) they
are in, in order to receive a reward. A similar contextual impair-
ment was found in mice with mammillary body lesions; they were
impaired on contextual fear conditioning, but not auditory con-
ditioning, in a comparable manner to dorsal hippocampal lesions
(Celerier, Pierard, & Beracochea, 2004). Mice were presented with
ditioned stimulus) in a conditioning chamber. Mammillary body
lesioned mice exhibited the freezing response when given the con-
ditioned stimulus in a neutral environment but not when placed
back in the previously experienced conditioning chamber, that is,
they responded to the auditory cue but not the context (Celerier
et al., 2004). Finally, mammillothalamic tracts lesions in rabbits
impair the acquisition of a discriminative avoidance task where
animals have to learn to step into a wheel on hearing one tone
to avoid a shock whilst ignoring a different tone which does not
predict a shock (Gabriel et al., 1995).
Although mammillothalamic tract lesions impair visuo-spatial
contextual discrimination (Vann et al., 2003), mammillary body
lesions do not affect all conditional tasks. For example, mammil-
lary body lesions do not affect the acquisition of a visuo-spatial
associations between visual stimuli and spatial locations (Sziklas &
Petrides, 1993; Sziklas, Petrides, & Leri, 1996). The lack of impair-
ment on these tasks could be due to animals requiring extensive
training in order to acquire the task. In the Vann et al. study (2003),
One of the first theories of mammillary body function was that
they form part of a circuit that underlies emotion (MacLean, 1949;
Papez, 1937). There continues to be support for this theory and
it has been proposed that emotional disturbances, resulting from
mammillary body pathology, may contribute to subsequent mem-
ory impairments (Beracochea, 2005). Mice with mammillary body
lesions spend more time in the open arms of an elevated plus
than control animals suggesting that mammillary body lesions
reduce anxiety levels (Beracochea & Krazem, 1991). In addition,
mammillary body lesions increase the rate of responding during
the punished, but not unpunished, period in a continuous rein-
forcement conflict schedule (Shibata, Kataoka, Yamashita, & Ueki,
1986). A similar effect was found when benzopiazepines were
injected directly into the mammillary bodies (Kataoka, Shibata,
Gomita, & Ueki, 1982) suggesting that in both manipulations the
animals were less affected by aversive stimuli; this effect appears
to be moderated by noradrenaline (Kataoka, Shibata, Yamashita, &
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
Ueki, 1987). It is possible that the anxiolytic effects of mammil-
lary body lesions are due to the high density of benzodiazepine
receptors in this structure (Young & Kuhar, 1980). A proposal
put forward by Beracochea (2005) is that animals with mammil-
lary body lesions are less anxious and, therefore, less emotionally
aroused; this results in the animals processing relevant stimuli
less well. This proposal is consistent with pharmacological stud-
ies that have shown that administering a benzodiazepine receptor
inverse agonist to mice with mammillary body lesions increases
fear reactivity (so that they now behave similarly to controls) and
reverses observed memory impairments on an alternation task
(Beracochea, Krazem, & Jaffard, 1995). In addition, administering
diazepam reduces neuronal activity in the mammillary bodies as
measured by glucose utilization (Ableitner, Wuster, & Herz, 1985).
However, mammillary body lesions do not disrupt the acquisition
of all emotion-related learning as they do not impair conditioned
taste aversion (Sziklas & Petrides, 1993) or basic fear-avoidance
(e.g. Celerier et al., 2004).
Finally, there is some evidence that the mammillary bodies are
important for temporal processing in addition to spatial process-
ing. Tonkiss and Rawlins (1992) showed that mammillary body
lesions impair performance on tasks that require animals to delay
their response to a stimulus for a minimum time (DRL tasks). DRL
tasks are hippocampal-dependent (Clark & Isaacson, 1965; Sinden,
Rawlins, Gray, & Jarrard, 1986) and the ability to delay a response
is disrupted in amnesics (Oscar-Berman, Zola-Morgan, Oberg, &
Bonner, 1982) and monkeys with hippocampal lesions (Jackson
& Gergen, 1970). However, it is possible that deficits observed on
DRL tasks could reflect impairments in either temporal judgment
or response inhibition.
4.2. Medial vs lateral mammillary body lesions
Lesions of the mammillary bodies typically involve the medial
mammillary nuclei more than the lateral mammillary nuclei (e.g.
Beracochea & Jaffard, 1987, 1995; Field et al., 1978; Rosenstock et
al., 1977; Santin et al., 1999; Sziklas & Petrides, 1993). In addition,
discrete lesions of the mammillothalamic tract can selectively dis-
connect medial mammillary nucleus efferents (Vann & Aggleton,
2003, 2004) and it is the medial mammillary nucleus that is always
affected in Korsakoff syndrome (Kopelman, 1995; Victor et al.,
1989). While it is, therefore, apparent that the medial mammil-
lary nucleus is implicated in memory, the importance of the lateral
mammillary nucleus is less clear. There are very few studies that
have assessed lesions restricted to the lateral mammillary nuclei
and there is only one study that has assessed the behavioral effects
restricted to the lateral mammillary nuclei were unimpaired on a
spatial alternation task in the T-maze which is sensitive to com-
plete mammillary body lesions (Vann & Aggleton, 2003). While
the lateral mammillary nucleus lesions impaired performance on
the working memory task in the water-maze this effect was only
transient (Vann, 2005) unlike the robust deficits seen following
mammillary body and mammillothalamic tract lesions (Vann &
Aggleton, 2003). From these findings, it is evident that the loss of
lowing complete mammillary nucleus lesions. In addition, it would
appear that spatial memory impairments seen following mammil-
lary body lesions cannot simply be explained in terms of loss of
4.3. Non-human primates
To date, the majority of mammillary body lesion studies have
been carried out on rodents with very few studies in monkeys;
this is in part due to the size and position of the mammillary
body lesions in primates which make them a difficult target for
surgery. However, from the few studies that are available, the
impairments across species appear consistent. Mammillary body
performance on a spatial discrimination task (Aggleton & Mishkin,
1985). Similarly, Holmes et al. reported a mammillary body lesion
induced impairment on spatial reversal learning but not object
mammillary body lesions impair monkeys ability to learn new
object-in-place scenes; the mammillary body lesion effects were
comparable to those seen following fornix lesions (Parker & Gaffan,
4.4. Accounts of mammillary body lesion deficits
ory deficits resulting from mammillary body lesions: increased
sensitivity to proactive interference, that is, difficulty in sepa-
rating events (Aggleton et al., 1995; Beracochea & Jaffard, 1990;
Jaffard, Beracochea, & Cho, 1991); increased rate of forgetting
(Beracochea & Jaffard, 1987; Rosenstock et al., 1977; Saravis et
al., 1990; Sziklas & Petrides, 1998; Tako, Beracochea, & Jaffard,
1988); impaired encoding (Butters, 1985; Vann & Aggleton, 2003;
Vann et al., 2003); impaired retrieval (Lhermitte & Signoret, 1972;
Warrington & Weiskrantz, 1974); and impaired emotion–memory
interactions (Beracochea, 2005). These explanations are not, how-
ever, mutually exclusive. While there is evidence supporting each
of these accounts, there are cases where animals with mammillary
body lesion are not differentially affected by increased interfer-
ence (Aggleton, Keith, & Sahgal, 1991; Harper et al., 1994; Vann
& Aggleton, 2003) or increased retention delays (Aggleton et al.,
1991; Harper et al., 1994; Santin et al., 1999; Vann & Aggleton,
2003). These findings make it difficult to provide a simple explana-
tion based on interference or retention. Instead, mammillary body
lesion effects are most consistent with impaired rapid allocentric
encoding, such that deficits are most clearly seen during initial
stages of learning or when animals have to perform a spatial work-
ing memory tasks that preclude the use of non-allocentric spatial
strategies. This interpretation would account for deficits on tasks
such as the working memory task in the water-maze and would
There is also evidence that mammillary body lesion deficits reflect
impaired retrieval processes. For example, Bereacochea et al., have
shown that the performance of mice with mammillary body dam-
the context increases arousal levels in the animal which facilitates
retrieval processes (Beracochea & Jaffard, 1987; Tako et al., 1988).
In addition, rats trained on the DRL procedure and subsequently
given a mammillary body lesion became less efficient on the task,
although their performance was superior to rats given mammil-
lary body lesions before any DRL training (Tonkiss & Rawlins,
1992). However, this decline in DRL performance following mam-
millary body lesions may reflect a decreased ability to inhibit
responses rather than a retrieval deficit per se. Finally, as described
earlier, mammillary body lesions have been reported to have an
anxiolytic effect so Beracochea (2005) proposed that mammillary
body lesions result in animals being less emotionally aroused; this
reduced arousal would result in animals processing relevant stim-
uli less well which would affect subsequent memory (Beracochea,
2005). It is unlikely that mammillary body lesion effects reflect
the disruption of a single process, instead, the mammillary bodies
may contribute to several processes required to support memory
and this may be dependent on specific task demands and output
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
4.5. Animal models of Korsakoff’s syndrome
Animal models of Kosakoff’s syndrome have provided further
insights into mammillary body function. These animal models
are the result of either chronically administered ethanol or treat-
ment with the thiamine anatagonist, pyrithiamine. In the same
way as patients with Korsakoff’s syndrome, the pathology in the
pyrathiamine-induced thiamine deficiency (PTD) rat model is dif-
fuse and, in addition to diencephalic damage, there is widespread
cortical and thalamic damage as well as damage to major white
matter tracts including the corpus callosum and internal capsule
amine are impaired on spatial alternation (Irle & Markowitsch,
1982; Mair, Anderson, Langlais, & McEntee, 1985) and the ref-
erence memory task in the water-maze (Langlais et al., 1992).
pocampal and cortical acetylcholine and noradrenalin levels (Mair
et al., 1985; Pires, Pereira, Oliveira-Silva, Franco, & Ribeiro, 2005;
Roland & Savage, 2007; Savage, Chang, & Gold, 2003). In addition,
chronic administration of alcohol in mice produces medial mam-
millary atrophy and impairs spatial memory (Tako, Beracochea,
Lescaudron, & Jaffard, 1991). The spatial memory impairments and
hippocampal dysfunction reported in these Korsakoff models are
consistent with the effects of more discrete mammillary body or
mammillothalamic tract lesions. Therefore, despite the additional
pathology in these models, the pattern of deficits is likely to be
attributable, in the most part, to the medial diencephalic atro-
4.6. Functional gene imaging in normal animals
There are very few studies that have used functional gene imag-
ing to assess specific contributions of the mammillary bodies. This
genes c-fos and zif268 in the mammillary bodies, both at baseline
and following appetitive learning tasks (e.g. Amin, Pearce, Brown,
& Aggleton, 2006; Jenkins, Amin, Brown, & Aggleton, 2006; Vann,
In one study, where changes in c-Fos levels could be assessed, the
authors reported increases in immediate-early gene expression in
the lateral, but not medial, mammillary nucleus in animals that
had undergone contextual and auditory fear conditioning (Conejo,
Gonzalez-Pardo, Lopez, Cantora, & Arias, 2007). An earlier study
used 2-deoxyglucose as a marker of activity and found increased
metabolic activity in the mammillary bodies of monkeys that per-
formed a spatial working memory task compared to monkeys
that performed a control task (Friedman, Janas, & Goldman-Rakic,
4.7. Distal effects of mammillary body lesions
As traditional theories consider the mammillary bodies to form
part of a hippocampal relay, it would be expected that discon-
necting this relay would affect distal brain sites. Consistent with
this prediction, the integrity of the mammillary bodies appears
to be essential for the normal functioning of other key structures
hypoactivity are regularly seen in Korsakoff’s syndrome patients
(Caulo et al., 2005; Joyce et al., 1994; Reed et al., 2003). Reduced
hippocampal activity was also reported in patient BJ; this patient
suffered mammillary damage following an intranasal penetration
injury and, therefore, had more restricted pathology than that seen
in Korsakoff’s syndrome (Kapur et al., 1994). These findings are
consistent with the distal hypoactivity seen in animal models of
diencephalic amnesia. Mammillary body lesions in mice disrupt
cholinergic activity, as measured by sodium-dependent high affin-
ity choline uptake, in both the hippocampus and frontal cortex;
this reduction in cholinergic levels is found irrespective of whether
animals were taken from the home-cage or actively exploring a
maze (Beracochea et al., 1995b). Similar results were found in rats
with mammillothalamic tract lesions which selectively disrupted
the efferents from the medial mammillary nucleus to the anterior
thalamic nuclei (Vann & Albasser, 2009). These lesions resulted in
hypoactivity, as measured by the immediate-early gene c-fos, in
the hippocampus, retrosplenial cortex and prefrontal cortex (Vann
selective, indirect effects upon multiple regions thought to be criti-
have been problems dissociating involvement of the frontal lobe
in diencephalic amnesia as additional damage to the mediodorsal
thalamus would result in a loss of frontal projections and direct
damage to the frontal cortex can occur in cases of Korsakoff’s
syndrome. This immediate-early gene study shows that prelimbic
dysfunction can occur in a model of diencephalic amnesia without
direct deafferentation or damage (Vann & Albasser, 2009).
5. Lesions of major mammillary body afferents and
5.1. Mammillary body, anterior thalamic and fornix lesions
fornix–mammillary body–thalamic system. As mammillothalamic
bodies exert their role on spatial memory via their projections to
the anterior thalamic nuclei. Within the mammillary bodies’ role
as a relay the comparative effects of mammillary body, anterior
thalamic and fornix lesions become particularly relevant. When
anterior thalamic lesions produce greater effects than mammil-
lary body lesion effects, it is presumed that the direct projections
from the subicular complex to the anterior thalamic nuclei are suf-
ficient to support this task, although this would not preclude an
additional contribution from the mammillary body projections in
There are occasions where mammillary body lesions have had
no effect on tasks that are sensitive to either anterior thalamic,
hippocampal or fornix lesions. For example, mammillary body
lesions do not affect performance on an automated delayed non-
matching-to-position task, where animals are required to select
the lever in the choice stage that was not presented in the sam-
ple phase (Aggleton et al., 1991; Harper et al., 1994), despite this
task being sensitive to both fornix and anterior thalamic nuclei
lesions (Aggleton et al., 1991; Harper et al., 1994). In addition,
while mammillary body lesions disrupt the acquisition of the ref-
erence memory task in the water-maze, animals were able to
learn the platform position and perform normally on a probe trial
(Sutherland & Rodriguez, 1989); this is in contrast to rats with
either anterior thalamic nuclei or fornix lesions where they were
impaired throughout training and on the probe trial (Sutherland
& Rodriguez, 1989). Mammillary body lesions are also less dis-
ruptive on the standard T-maze task than anterior thalamic or
fornix lesions (Aggleton et al., 1995; Gaffan et al., 2001). Finally,
mammillary body lesions do not disrupt performance on a task
that requires animals to form associations between visual stim-
uli and spatial locations (Sziklas & Petrides, 2000) even though this
task is sensitive to hippocampal (Sziklas, Lebel, & Petrides, 1998;
Sziklas et al., 1996) and anterior thalamic nuclei lesions (Sziklas
& Petrides, 1999). However, these tasks are not sensitive to fornix
lesions (Dumont, Petrides, & Sziklas, 2007) or retrosplenial cortex
functional circuit is necessary for this task.
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
In contrast, there are instances where mammillary body, ante-
rior thalamic and fornix lesions have equivalent effects, consistent
with a fornix–mammillary body–anterior thalamic nucleus path-
increases proactive interference and task difficulty, mammillary
but less so than fornix lesion rats (Aggleton et al., 1995). With
subsequent delays, all lesion groups show an equivalent impair-
ment (Aggleton et al., 1995). This suggests mammillary body lesion
effects are only milder than anterior thalamic and fornix lesions
when there are fewer tasks demands. Mammillary body, anterior
thalamic and fornix lesions result in equivalent levels of facilita-
Likewise, mammillary body, anterior thalamic, and fornix lesions
result in equivalent levels of performance on a scene-learning and
object-in-place task in monkeys (Parker & Gaffan, 1997a, 1997b).
Comparisons can also be made between the effects of these
lesions on immediate-early gene expression in distal brain sites.
prefrontal cortex following mammillothalamic tract lesions (Vann
rior thalamic nuclei lesions (Jenkins, Dias, Amin, & Aggleton, 2002;
Jenkins et al., 2002b) and this is consistent with these anterior tha-
lamic effects being driven by the loss of their mammillary body
afferents. However, the striking decrease in Fos in the retrosple-
tract lesions is in contrast to the much smaller effects following
fornix lesions (Vann, Brown, Erichsen, et al., 2000). The implication
mammillary body efferents cannot solely be explained in terms of
an indirect loss of hippocampal/fornical inputs, but that the mam-
millary bodies have an additional, independent contribution.
5.2. Descending postcommissural fornix lesions
Although fornix lesions disconnect the mammillary bodies
from their hippocampal formation inputs, they also disconnect
a large number of additional hippocampal efferents and affer-
ents (Nauta, 1956; Poletti & Creswell, 1977; Saunders & Aggleton,
2007; Swanson & Cowan, 1977; Vann, Brown, Erichsen, et al.,
2000). It is, therefore, difficult to use findings from complete fornix
lesion studies to specifically assess the importance of the hip-
pocampal formation–mammillary projections. While there have
been a couple of studies that have targeted the postcommis-
sural fornix these have all been at the level of the septum which
would disconnect a number of other sites (Henderson & Greene,
1977; Thomas, 1978; Tonkiss, Feldon, & Rawlins, 1990). Recently,
the hippocampal formation–mammillary projections have been
targeted selectively by making a lesion of the descending post-
commissural fornix at a level caudal to the anterior thalamus;
retrograde tracers were subsequently used to confirm the com-
pleteness of the intended disconnection and the preservation of
the hippocampal formation–anterior thalamic projections (Vann,
2009a). On standard tests of spatial memory, these descending
postcommissural fornix lesions produce either no effect or only
mild impairments and these lesions appear much less disruptive
than lesions of the mammillary bodies or mammillothalamic tract
(Vann, 2009a). This discrepancy between postcommissural and
mammillary body effects is consistent with earlier findings where
mammillary body lesions were significantly more disruptive than
lesions of the descending postcommissural fornix on a DRL task
(Tonkiss & Rawlins, 1992).
5.3. Gudden’s ventral tegmental nucleus lesions
sural fornix lesion study are not consistent with the mammillary
bodies simply acting as a hippocampal relay. Attention must,
therefore, be directed towards the remaining mammillary body
inputs, and the largest of these inputs comes from the tegmen-
tal nuclei of Gudden. As the medial mammillary nuclei seem to
be predominantly responsible for mammillary body effects on
memory, the ventral tegmental nucleus of Gudden becomes of par-
ticular interest. The functions of this brain structure have been
largely overlooked but a recent study found that selective exci-
totoxic ventral tegmental nucleus lesions result in robust deficits
on various spatial memory tasks, including working memory in
the water-maze, T-maze alternation, and working memory in the
radial-arm-maze (Vann, 2009b); these tasks are all sensitive to
mammillary body and mammillothalamic tract lesions (Vann &
Aggleton, 2003). In contrast, rats with ventral tegmental nucleus
of Gudden lesions performed normally on a visually cued task in
the water-maze and on the acquisition and reversal of an egocen-
deficits were not a reflection of sensori-motor disturbances, moti-
vational or gross learning impairments (Vann, 2009b). In addition,
neurochemical assessments confirmed that the lesion effects were
not a result of the loss of cholinergic projections from the adjacent
laterodorsal tegmental nucleus or the loss of serotonergic raphe
nuclei neurons (Vann, 2009b). These lesion effects are consistent
that have unintentionally included the ventral tegmental nucleus
radial-arm-maze performance (Asin & Fibiger, 1984; Wirtshafter
& Asin, 1983), although there was additional damage to fibers of
passage and adjacent fiber bundles. A study of excitotoxic median
raphe nuclei lesions reported a significant correlation between the
extent of ventral tegmental nucleus damage and the number of
errors made on a reinforced T-maze task (Asin & Fibiger, 1984)
although, again, there was extensive raphe cell loss. Finally, there
is a report of a man with amnesia that was attributed to pathology
in the ventral tegmental nucleus of Gudden area (Goldberg et al.,
1981). One suggestion is that Gudden’s ventral tegmental nucleus
acts as an inhibitory feedback loop, with the mammillary bodies,
and controls the transfer of information from the hippocampal for-
mation to the anterior thalamic nuclei (Wirtshafter & Stratford,
1993). However, this account is not consistent with the descending
postcommissural fornix lesions having such a small effect (see pre-
vious section) compared to mammillary body or Gudden’s ventral
6. Re-evaluating mammillary models of memory
Despite some previous uncertainty it is now apparent, from
both patient and animal studies, that the mammillary bodies are
important for memory. The mammillary bodies comprise two
main nuclei, medial and lateral, which differ in terms of their cell
morphology, electrophysiology and connections. The lateral mam-
millary nucleus forms part of the head-direction system while the
medial mammillary nucleus is situated within a “theta-related”
system. It is likely that these two nuclei are also functionally dis-
tinct. The medial mammillary bodies, and their projections to the
anterior thalamus, are necessary for spatial memory and normal
hippocampal, retrosplenial and prefrontal function. Although lat-
eral mammillary body lesions have only mild effects on standard
spatial tasks they may contribute to additional aspects of spatial
memory and/or navigation in normal animals.
Mammillary body lesion-induced deficits appear largely con-
sistent across species and seem to reflect impoverished spatial
encoding, although this does not preclude the mammillary
bodies supporting other aspects of memory. Current theo-
ries of mammillary body function emphasize the hippocampal
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
formation–fornix–mammillary body pathway (Aggleton & Brown,
ies being considered part of the “extended hippocampal system”.
However, recent findings are inconsistent with this traditional
view, that the hippocampal formation drives the medial dien-
cephalon via the fornix and, in fact, the reverse may be true. It is
possible that the diencephalon has a role in memory that is largely
independent of its hippocampal formation inputs and instead pro-
vides critical indirect hippocampal inputs that are required for
lary body function the inputs from the tegmental nuclei of Gudden
may prove critical.
SDV is funded by a Biotechnology and Biological Sci-
ences Research Council (BBSRC) UK David Phillips fellowship
utilization in the rat brain induced by acute and chronic diazepam. Brain Res,
Aggleton, J. P., & Brown, M. W. (1999). Episodic memory, amnesia, and the
hippocampal-anterior thalamic axis. Behav Brain Sci, 22(3), 425–444.
Aggleton, J. P., & Mishkin, M. (1985). Mamillary-body lesions and visual recognition
in monkeys. Exp Brain Res, 58(1), 190–197.
Aggleton, J. P., Desimone, R., & Mishkin, M. (1986). The origin, course, and termi-
nation of the hippocampothalamic projections in the macaque. J Comp Neurol,
Aggleton, J. P., Keith, A. B., & Sahgal, A. (1991). Both fornix and anterior thalamic, but
not mammillary, lesions disrupt delayed non-matching-to-position memory in
rats. Behav Brain Res, 44(2), 151–161.
Aggleton, J. P., Neave, N., Nagle, S., & Hunt, P. R. (1995). A comparison of the effects
of anterior thalamic, mamillary body and fornix lesions on reinforced spatial
alternation. Behav Brain Res, 68(1), 91–101.
region to the mammillary bodies in macaque monkeys. Eur J Neurosci, 22(10),
Allen, G. V., & Hopkins, D. A. (1988). Mamillary body in the rat: A cytoarchitectonic,
Golgi, and ultrastructural study. J Comp Neurol, 275(1), 39–64.
Allen, G. V., & Hopkins, D. A. (1989). Mamillary body in the rat: Topography and
synaptology of projections from the subicular complex, prefrontal cortex, and
midbrain tegmentum. J Comp Neurol, 286(3), 311–336.
Allen, G. V., & Hopkins, D. A. (1990). Topography and synaptology of mamillary body
projections to the mesencephalon and pons in the rat. J Comp Neurol, 301(2),
Alonso, A., & Llinas, R. R. (1992). Electrophysiology of the mammillary complex in
vitro 2. Medial mammillary neurons. J Neurophysiol, 68(4), 1321–1331.
rations of stimuli produce discrete changes in immediate-early gene expression
in the rat hippocampus. Eur J Neurosci, 24(9), 2611–2621.
Apartis, E., Poindessous-Jazat, F. R., Lamour, Y. A., & Bassant, M. H. (1998).
Loss of rhythmically bursting neurons in rat medial septum following selec-
tive lesion of septohippocampal cholinergic system. J Neurophysiol, 79(4),
Aranda, L., Santin, L. J., Begega, A., Aguirre, J. A., & Arias, J. L. (2006). Supramam-
millary and adjacent nuclei lesions impair spatial working memory and induce
anxiolitic-like behavior. Behav Brain Res, 167(1), 156–164.
Asin, K. E., & Fibiger, H. C. (1984). Spontaneous and delayed spatial alternation
following damage to specific neuronal elements within the nucleus medianus
raphe. Behav Brain Res, 13(3), 241–250.
Barbizet, J. (1963). Defect of memorizing of hippocampal-mammillary origin: A
review. J Neurol Neurosurg Psychiatry, 26, 127–135.
A pontine hippocampal theta generator? Hippocampus, 11(6), 809–813.
circuit in the head direction system disrupt the head direction signal in the
anterior thalamus. J Neurosci, 27(28), 7564–7577.
Beracochea, D. (2005). Interaction between emotion and memory: Importance of
mammillary bodies damage in a mouse model of the alcoholic Korsakoff syn-
drome. Neural Plast, 12(4), 275–287.
Beracochea, D. J., & Jaffard, R. (1987). Impairment of spontaneous alternation behav-
ior in sequential test procedures following mammillary body lesions in mice:
Evidence for time-dependent interference-related memory deficits. Behav Neu-
rosci, 101(2), 187–197.
on spontaneous and rewarded spatial alternation in mice. Journal of Cognitive
Neuroscience, 2(2), 133–140.
Beracochea, D. J., & Jaffard, R. (1995). The effects of mammillary body lesions on
delayed matching and delayed non-matching to place tasks in the mice. Behav
Brain Res, 68(1), 45–52.
Beracochea, D. J., & Krazem, A. (1991). Effects of mammillary body and mediodor-
sal thalamic lesions on elevated plus maze exploration. Neuroreport, 2(12),
Beracochea, D., Lescaudron, L., Tako, A., Verna, A., & Jaffard, R. (1987). Build-up
and release from proactive interference during chronic ethanol consumption
in mice: A behavioral and neuroanatomical study. Behav Brain Res, 25(1), 63–74.
reverses the working-memory deficits induced either by chronic alcohol-
consumption or mammillary body lesions in balb/c mice. Psychobiology, 23(1),
Beracochea, D. J., Micheau, J., & Jaffard, R. (1995). Alteration of cortical and hip-
pocampal cholinergic activities following lesion of the mammillary bodies in
mice. Brain Res, 670(1), 53–58.
Bernstein, H. G., Krause, S., Krell, D., Dobrowolny, H., Wolter, M., Stauch, R., et al.
(2007). Strongly reduced number of parvalbumin-immunoreactive projection
neuropathology. Ann N Y Acad Sci, 1096, 120–127.
Blair, H. T., Cho, J. W., & Sharp, P. E. (1998). Role of the lateral mammillary nucleus
in the rat head direction circuit: A combined single unit recording and lesion
study. Neuron, 21(6), 1387–1397.
Bland, B. H., Konopacki, J., Kirk, I. J., Oddie, S. D., & Dickson, C. T. (1995). Discharge
patterns of hippocampal theta-related cells in the caudal diencephalon of the
urethane-anesthetized rat. J Neurophysiol, 74(1), 322–333.
Briess, D., Cotter, D., Doshi, R., & Everall, I. (1998). Mamillary body abnormalities in
schizophrenia. Lancet, 352(9130), 789–790.
Butters, N. (1985). Alcoholic Korsakoff’s syndrome: Some unresolved issues con-
cerning etiology, neuropathology, and cognitive deficits. J Clin Exp Neuropsychol,
Cajal, S. R. (1911). Histologie du systeme nerveux de l’homme et des vertebres. Paris:
Callen, D. J., Black, S. E., Gao, F., Caldwell, C. B., & Szalai, J. P. (2001). Beyond the hip-
Bilateral damage to the mammillo-thalamic tract impairs recollection but not
familiarity in the recognition process: A single case investigation. Neuropsy-
chologia, 45(11), 2467–2479.
Caulo, M., Van Hecke, J., Toma, L., Ferretti, A., Tartaro, A., Colosimo, C., et al. (2005).
Functional MRI study of diencephalic amnesia in Wernicke–Korsakoff syn-
drome. Brain, 128(Pt 7), 1584–1594.
Celerier, A., Pierard, C., & Beracochea, D. (2004). Effects of ibotenic acid lesions of the
dorsal hippocampus on contextual fear conditioning in mice: Comparison with
mammillary body lesions. Behav Brain Res, 151(1–2), 65–72.
Clark, C. V., & Isaacson, R. L. (1965). Effect of bilateral hippocampal ablation on DRL
performance. J Comp Physiol Psychol, 59, 137–140.
Clarke, S., Assal, G., Bogousslavsky, J., Regli, F., Townsend, D. W., Leenders, K. L., et
al. (1994). Pure amnesia after unilateral left polar thalamic infarct: Topographic
and sequential neuropsychological and metabolic (pet) correlations. J Neurol
Neurosurg Psychiatry, 57(1), 27–34.
Conejo, N. M., Gonzalez-Pardo, H., Lopez, M., Cantora, R., & Arias, J. L. (2007). Induc-
tion of c-fos expression in the mammillary bodies, anterior thalamus and dorsal
hippocampus after fear conditioning. Brain Res Bull, 74(1–3), 172–177.
Copenhaver, B. R., Rabin, L. A., Saykin, A. J., Roth, R. M., Wishart, H. A., Flashman, L. A.,
et al. (2006). The fornix and mammillary bodies in older adults with Alzheimer’s
study. Psychiatry Res, 147(2–3), 93–103.
Cruce, J. A. (1975). An autoradiographic study of the projections of the mammil-
lothalamic tract in the rat. Brain Res, 85(2), 211–219.
Cruce, J. A. (1977). An autoradiographic study of the descending connections of the
mammillary nuclei of the rat. J Comp Neurol, 176(4), 631–644.
Delay, J., & Brion, S. (1969). Le syndrome de Korsakoff. Paris: Mason.
Denby, C. E., Vann, S. D., Tsivilis, D., Aggleton, J. P., Montaldi, D., Roberts, N., et al.
(2009). The frequency and extent of mammillary body atrophy associated with
surgical removal of a colloid cyst. AJNR Am J Neuroradiol.
Dumont, J., Petrides, M., & Sziklas, V. (2007). Functional dissociation between
fornix and hippocampus in spatial conditional learning. Hippocampus, 17(12),
Field, T. D., Rosenstock, J., King, E. C., & Greene, E. (1978). Behavioral role of the
mammillary efferent system. Brain Res Bull, 3(5), 451–456.
activity in the diencephalon of monkeys performing working memory task: A
2-deoxyglucose study in behaving rhesus monkeys. Journal of Cognitive Neuro-
science, 2(1), 18–31.
Fry, F. J., & Cowan, W. M. (1972). A study of retrograde cell degeneration in the
branching in the preservation of the cell. J Comp Neurol, 144(1), 1–23.
Fujii, S., Senba, E., Kiyama, H., Ueda, Y., & Tohyama, M. (1987). Mammillothalamic
enkephalinergic pathway in the rat: An immunocytochemical analysis. Brain
Res, 401(1), 1–8.
Gabriel, M., Cuppernell, C., Shenker, J. I., Kubota, Y., Henzi, V., & Swanson, D. (1995).
Mamillothalamic tract transection blocks anterior thalamic training-induced
neuronal plasticity and impairs discriminative offidance behavior in rabbits. J
Neurosci, 15(2), 1437–1445.
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
Gaffan, D. (1992). The role of the hippocampus–fornix–mammillary system in
episodic memory. In L. R. Squire, & N. Butters (Eds.), Neuropsychology of memory
(Second ed., pp. 336–346). New York: Guildford Press.
Gaffan, D. (2001). What is a memory system? Horel’s critique revisited. Behav Brain
Res, 127(1–2), 5–11.
Gaffan, E. A., Bannerman, D. M., Warburton, E. C., & Aggleton, J. P. (2001). Rats’ pro-
nuclei or the retrohippocampal region. Behav Brain Res, 121(1–2), 103–117.
Goldberg, E., Antin, S. P., Bilder, R. M., Jr., Gerstman, L. J., Hughes, J. E., & Mattis, S.
(1981). Retrograde amnesia: Possible role of mesencephalic reticular activation
in long-term memory. Science, 213(4514), 1392–1394.
Gonzalo-Ruiz, A., Alonso, A., Sanz, J. M., & Llinas, R. R. (1992a). Afferent-projections
to the mammillary complex of the rat, with special reference to those from
surrounding hypothalamic regions. Journal of Comparative Neurology, 321(2),
Gonzalo-Ruiz, A., Alonso, A., Sanz, J. M., & Llinas, R. R. (1992b). A dopamin-
ergic projection to the rat mammillary nuclei demonstrated by retrograde
transport of wheat-germ-agglutinin horseradish-peroxidase and tyrosine-
hydroxylase immunohistochemistry. Journal of Comparative Neurology, 321(2),
Gonzalo-Ruiz, A., Sanz-Anquela, J. M., & Spencer, R. F. (1993). Immunohistochemical
localization of Gaba in the mammillary complex of the rat. Neuroscience, 54(1),
Gonzalo-Ruiz, A., Morte, L., & Sanz, J. M. (1998). Glutamate/aspartate and leu-
enkephalin immunoreactivity in mammillothalamic projection neurons of the
rat. Brain Res Bull, 47(6), 565–574.
Gonzalo-Ruiz, A., Romero, J. C., Sanz, J. M., & Morte, L. (1999). Localization of amino
acids, neuropeptides and cholinergic neurotransmitter markers in identified
projections from the mesencephalic tegmentum to the mammillary nuclei of
the rat. J Chem Neuroanat, 16(2), 117–133.
amnesia. Brain, 113(Pt 1), 1–25.
Grossi, D., Lopez, O. L., & Martinez, A. J. (1989). Mamillary bodies in Alzheimer’s
disease. Acta Neurol Scand, 80(1), 41–45.
Gudden, H. (1896). Klinische und anatommische beitrage zur kenntnis der mul-
tiplen alkoholneuritis nebst bemerzungen uber die regenerationsvorgange im
peripheren nervensystem. Archiv fur Psychiatrie, 28, 643–741.
Guillery, R. W. (1955). A quantitative study of the mamillary bodies and their con-
nexions. J Anat, 89(1), 19–32.
nuclei differentiates alcoholics with amnesia. Brain, 123, 141–154.
Harper, D. N., Dalrymple-Alford, J. C., & McLean, A. P. (1993). The effect of medial
septal and mammillary body lesions on the serial position curve in rats. Psy-
chobiology, 21(2), 130–138.
Harper, D. N., McLean, A. P., & Dalrymple-Alford, J. C. (1994). Forgetting in rats fol-
lowing medial septum or mammillary body damage. Behav Neurosci, 108(4),
millary projections in some mammals—A horseradish-peroxidase study. Brain
Res, 300(2), 335–349.
Hayakawa, T., & Zyo, K. (1985). Afferent connections of Guddens tegmental nuclei
in the rabbit. Journal of Comparative Neurology, 235(2), 169–181.
Hayakawa, T., & Zyo, K. (1989). Retrograde double-labeling study of the mammil-
lothalamic and mammillotegmental projections in the rat Journal of Comparative
Neurology, 284(1), 1–11.
Hayakawa, T., & Zyo, K. (1991). Qualitative and ultrastructural-study of ascending
projections to the medial mammillary nucleus in the rat. Anatomy and Embryol-
ogy, 184(6), 611–622.
Henderson, J., & Greene, E. (1977). Behavioral effects of lesions of precommissural
and postcommissural fornix. Brain Res Bull, 2(2), 123–129.
Holmes, E. J., Jacobson, S., Stein, B. M., & Butters, N. (1983). Ablations of the mam-
millary nuclei in monkeys: Effects on postoperative memory. Exp Neurol, 81(1),
Hopkins, D. A. (2005). Neuroanatomy of head direction cell circuits. In S. I. Wiener,
& J. S. Taube (Eds.), Head direction cells and the neural mechanisms of spatial
orientation. Cambridge, Mass.; London: MIT (pp. xxii, 480 p.,  leaves of
Irle, E., & Markowitsch, H. J. (1982). Thiamine deficiency in the cat leads to severe
learning deficits and to widespread neuroanatomical damage. Exp Brain Res,
Jackson, F. B., & Gergen, J. A. (1970). Acquisition of operant schedules by squirrel
monkeys lesioned in the hippocampal area. Physiol Behav, 5(5), 543–547.
Jaffard, R., Beracochea, D., & Cho, Y. (1991). The hippocampal-mamillary system:
Anterograde and retrograde amnesia. Hippocampus, 1(3), 275–278.
Jarrard, L. E., Okaichi, H., Steward, O., & Goldschmidt, R. B. (1984). On the role of hip-
pocampal connections in the performance of place and cue tasks: Comparisons
with damage to hippocampus. Behav Neurosci, 98(6), 946–954.
Jay, T. M., Glowinski, J., & Thierry, A. M. (1989). Selectivity of the hippocampal pro-
jection to the prelimbic area of the prefrontal cortex in the rat. Brain Res, 505(2),
Jenkins, T. A., Dias, R., Amin, E., & Aggleton, J. P. (2002). Changes in fos expression in
Jenkins, T. A., Dias, R., Amin, E., Brown, M. W., & Aggleton, J. P. (2002). Fos imaging
reveals that lesions of the anterior thalamic nuclei produce widespread limbic
hypoactivity in rats. J Neurosci, 22(12), 5230–5238.
Jenkins, T. A., Amin, E., Brown, M. W., & Aggleton, J. P. (2006). Changes in immediate
Neuroscience, 137(3), 747–759.
Korsakoff’s syndrome. Psychiatry Res, 54(3), 225–239.
Kapur, N., Scholey, K., Moore, E., Barker, S., Brice, J., Mayes, A., et al. (1994). The
mammillary bodies revisited: Their role in human memory functioning. In L. S.
Cermak (Ed.), Neuropsychological explorations of memory and cognition: Essays in
honor of nelson butters (pp. 159–190). New York: Plenum Press.
Kataoka, Y., Shibata, K., Gomita, Y., & Ueki, S. (1982). The mammillary body is
a potential site of antianxiety action of benzodiazepines. Brain Res, 241(2),
Kataoka, Y., Shibata, K., Yamashita, K., & Ueki, S. (1987). Differential mechanisms
involved in the anticonflict action of benzodiazepines injected into the central
amygdala and mammillary body. Brain Res, 416(2), 243–247.
Kirk, I. J. (1998). Frequency modulation of hippocampal theta by the supramammil-
lary nucleus, and other hypothalamo-hippocampal interactions: Mechanisms
and functional implications. Neurosci Biobehav Rev, 22(2), 291–302.
Kirk, I. J., & Mackay, J. C. (2003). The role of theta-range oscillations in synchronis-
ing and integrating activity in distributed mnemonic networks. Cortex, 39(4–5),
Kirk, I. J., Oddie, S. D., Konopacki, J., & Bland, B. H. (1996). Evidence for differential
control of posterior hypothalamic, supramammillary, and medial mammillary
theta-related cellular discharge by ascending and descending pathways. J Neu-
rosci, 16(17), 5547–5554.
Kiyama, H., Shiosaka, S., Takami, K., Tateishi, K., Hashimura, E., Hamaoka, T., et
al. (1984). Cck pathway from supramammillary region to the nucleus anterior
ventralis thalami of the young rats. Peptides, 5(5), 889–893.
body in the rat. Brain Res, 375(2), 357–359.
Kocsis, B., & Vertes, R. P. (1994). Characterization of neurons of the supramam-
millary nucleus and mammillary body that discharge rhythmically with the
hippocampal theta rhythm in the rat. J Exp Psych Anim Behav Proc, 14(11 Pt 2),
Kocsis, B., Di Prisco, G. V., & Vertes, R. P. (2001). Theta synchronization in the limbic
system: The role of Gudden’s tegmental nuclei. Eur J Neurosci, 13(2), 381–388.
Kopelman, M. D. (1995). The Korsakoff syndrome. The British Journal of Psychiatry,
Kumar, R., Birrer, B. V. X., Macey, P. M., Woo, M. A., Gupta, R. K., Yan-Go, F. L., et al.
(2008). Reduced mammillary body volume in patients with obstructive sleep
apnea. Neurosci Lett, 438(3), 330–334.
Kumar, R., Woo, M. A., Birrer, B. V. X., Macey, P. M., Fonarow, G. C., Hamilton, M. A.,
et al. (2009). Mammillary bodies and fornix fibers are injured in heart failure.
Neurobiology of Disease, 33(2), 236–242.
rat. Behav Brain Res, 48(2), 177–185.
Lhermitte, F., & Signoret, J. L. (1972). Neuropsychologic analysis and differentiation
of amnesia syndromes]. Rev Neurol (Paris), 126(3), 161–178.
Llinas, R. R., & Alonso, A. (1992). Electrophysiology of the mammillary complex in
vitro.1. Tuberomammillary and lateral mammillary neurons. Journal of Neuro-
physiology, 68(4), 1307–1320.
MacLean, P. (1949). Psychosomatic disease and the visceral brain; recent devel-
opments bearing on the papez theory of emotion. Psychosom Med, 11(6),
Mair, W. G., Warrington, E. K., & Weiskrantz, L. (1979). Memory disorder in Kor-
sakoff’s psychosis: A neuropathological and neuropsychological investigation
of two cases. Brain, 102(4), 749–783.
Res, 360(1–2), 273–284.
Meibach, R. C., & Siegel, A. (1977a). Efferent connections of the hippocampal forma-
tion in the rat. Brain Res, 124(2), 197–224.
Meibach, R. C., & Siegel, A. (1977b). Subicular projections to the posterior cingulate
cortex in rats. Exp Neurol, 57(1), 264–274.
Namura, S., Takada, M., Kikuchi, H., & Mizuno, N. (1994). Topographical organiza-
tion of subicular neurons projecting to subcortical regions. Brain Res Bull, 35(3),
Nauta, W. J. (1956). An experimental study of the fornix system in the rat. J Comp
Neurol, 104(2), 247–271.
In W. E. A. E. Haymaker, & W. J. H. Nauta (Eds.), The hypothalamus (p. 819).
Springfield, Ill: Thomas.
Neave, N., Nagle, S., & Aggleton, J. P. (1997). Evidence for the involvement of the
mammillary bodies and cingulum bundle in allocentric spatial processing by
rats. Eur J Neurosci, 9(5), 941–955.
Oscar-Berman, M., Zola-Morgan, S. M., Oberg, R. G., & Bonner, R. T. (1982). Compara-
tive neuropsychology and Korsakoff’s syndrome III—Delayed response, delayed
alternation and drl performance. Neuropsychologia, 20(2), 187–202.
Paller, K. A. (1997). Consolidating dispersed neocortical memories: The missing link
in amnesia. Memory, 5(1–2), 73–88.
Pan, W. X., & McNaughton, N. (1997). The medial supramammillary nucleus, spatial
learning and the frequency of hippocampal theta activity. Brain Res, 764(1–2),
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
Pan, W. X., & McNaughton, N. (2002). The role of the medial supramammillary
nucleus in the control of hippocampal theta activity and behaviour in rats. Eur J
Neurosci, 16(9), 1797–1809.
Pan, W. X., & McNaughton, N. (2004). The supramammillary area: Its organization,
functions and relationship to the hippocampus. Prog Neurobiol, 74(3), 127–166.
Papez, J. W. (1937). A proposed mechanism of emotion. Arch Neurol Psychiatry, 38,
Parker, A., & Gaffan, D. (1997a). The effect of anterior thalamic and cingulate cor-
tex lesions on object-in-place memory in monkeys. Neuropsychologia, 35(8),
Pires, R. G., Pereira, S. R., Oliveira-Silva, I. F., Franco, G. C., & Ribeiro, A. M. (2005).
mation: A study using a model of Wernicke–Korsakoff syndrome. Behav Brain
Res, 162(1), 11–21.
Poletti, C. E., & Creswell, G. (1977). Fornix system efferent projections in the squirrel
monkey: An experimental degeneration study. J Comp Neurol, 175(1), 101–128.
Powell, T. P., & Cowan, W. M. (1954). The origin of the mamillo-thalamic tract in the
rat. J Anat, 88(4), 489–497.
Reed, L. J., Lasserson, D., Marsden, P., Stanhope, N., Stevens, T., Bello, F., et al.
(2003). Fdg-pet findings in the Wernicke–Korsakoff syndrome. Cortex, 39(4–5),
Roland, J. J., & Savage, L. M. (2007). Blunted hippocampal, but not striatal, acetyl-
choline efflux parallels learning impairment in diencephalic-lesioned rats.
Neurobiol Learn Mem, 87(1), 123–132.
J Anat, 74(Pt 1), 91–115.
Rosenstock, J., Field, T. D., & Greene, E. (1977). The role of mammillary bodies in
spatial memory. Exp Neurol, 55(2), 340–352.
Sakamoto, N., Michel, J. P., Kiyama, H., Tohyama, M., Kopp, N., & Pearson, J. (1986).
Neurotensin immunoreactivity in the human cingulate gyrus, hippocampal
Res, 375(2), 351–356.
Santin, L. J., Rubio, S., Begega, A., & Arias, J. L. (1999). Effects of mammillary
body lesions on spatial reference and working memory tasks. Behav Brain Res,
Saravis, S., Sziklas, V., & Petrides, M. (1990). Memory for places and the region of the
mamillary bodies in rats. Eur J Neurosci, 2(6), 556–564.
Saunders, R. C., & Aggleton, J. P. (2007). Origin and topography of fibers contributing
to the fornix in macaque monkeys. Hippocampus, 17(5), 396–411.
Saunders, R. C., Mishkin, M., & Aggleton, J. P. (2005). Projections from the entorhinal
cortex, perirhinal cortex, presubiculum, and parasubiculum to the medial thala-
mus in macaque monkeys: Identifying different pathways using disconnection
techniques. Exp Brain Res, 167(1), 1–16.
Savage, L. M., Chang, Q., & Gold, P. E. (2003). Diencephalic damage decreases hip-
pocampal acetylcholine release during spontaneous alternation testing. Learn
Mem, 10(4), 242–246.
Seki, M., & Zyo, K. (1984). Anterior thalamic afferents from the mamillary body and
the limbic cortex in the rat. J Comp Neurol, 229(2), 242–256.
Sharp, P. E., & Koester, K. (2008a). Lesions of the mammillary body region alter
gration models. Hippocampus, 18(9), 862–878.
Sharp, P. E., & Koester, K. (2008b). Lesions of the mammillary body region severely
disrupt the cortical head direction, but not place cell signal. Hippocampus, 18(8),
(vol 94, pg 1920, 2005). Journal of Neurophysiology, 94(6), 4554–14554.
Shibata, H. (1988). A direct projection from the entorhinal cortex to the mammillary
nuclei in the rat. Neurosci Lett, 90(1–2), 6–10.
Shibata, H. (1989). Descending projections to the mammillary nuclei in the rat,
as studied by retrograde and anterograde transport of wheat germ agglutinin-
horseradish peroxidase. J Comp Neurol, 285(4), 436–452.
Shibata, K., Kataoka, Y., Yamashita, K., & Ueki, S. (1986). An important role of the
central amygdaloid nucleus and mammillary body in the mediation of conflict
behavior in rats. Brain Res, 372(1), 159–162.
Sinden, J. D., Rawlins, J. N., Gray, J. A., & Jarrard, L. E. (1986). Selective cytotoxic
lesions of the hippocampal formation and drl performance in rats. Behav Neu-
rosci, 100(3), 320–329.
Stackman, R. W., & Taube, J. S. (1998). Firing properties of rat lateral mammillary
single units: Head direction, head pitch, and angular head velocity. J Neurosci,
St-Laurent, M., Petrides, M., & Sziklas, V. (2009). Does the cingulate cortex con-
Storm-Mathisen, J., & Woxen Opsahl, M. (1978). Asparate and/or glutamate may
be transmitters in hippocampal efferents to septum and hypothalamus. Neuro-
science Letters, 9(1), 65–70.
Sutherland, R. J., & Rodriguez, A. J. (1989). The role of the fornix/fimbria and some
related subcortical structures in place learning and memory. Behav Brain Res,
Swanson, L. W., & Cowan, W. M. (1977). An autoradiographic study of the organiza-
tion of the efferent connections of the hippocampal formation in the rat. J Comp
Neurol, 172(1), 49–84.
millary region of the rat. Eur J Neurosci, 5, 525–540.
Sziklas, V., & Petrides, M. (1998). Memory and the region of the mammillary bodies.
Prog Neurobiol, 54(1), 55–70.
on object-place associations in rats. Eur J Neurosci, 11(2), 559–566.
of the mammillary region in rats. Hippocampus, 10(3), 325–328.
Sziklas, V., Petrides, M., & Leri, F. (1996). The effects of lesions to the mammillary
region and the hippocampus on conditional associative learning by rats. Eur J
Neurosci, 8(1), 106–115.
Sziklas, V., Lebel, S., & Petrides, M. (1998). Conditional associative learning and the
hippocampal system. Hippocampus, 8(2), 131–137.
Takeuchi, Y., Allen, G. V., & Hopkins, D. A. (1985). Transnuclear transport and axon
collateral projections of the mamillary nuclei in the rat. Brain Res Bull, 14(5),
Tako, A. N., Beracochea, D. J., & Jaffard, R. (1988). Accelatered rate of for-
getting of spatial information following mammillary-body lesions in mice:
Effects of context change on retention-test performance. Psychobiology, 16(1),
Tako, A. N., Beracochea, D. J., Lescaudron, L., & Jaffard, R. (1991). Differential effects
of chronic ethanol consumptions or thiamine deficiency on spatial working
memory in balb/c mice: A behavioral and neuroanatomical study. Neurosci Lett,
Taube, J. S., Muller, R. U., & Ranck, J. B., Jr. (1990). Head-direction cells recorded from
J Neurosci, 10(2), 420–435.
Thinschmidt, J. S., Kinney, G. G., & Kocsis, B. (1995). The supramammillary nucleus:
Is it necessary for the mediation of hippocampal theta rhythm? Neuroscience,
Thomas, G. J. (1978). Delayed alternation in rats after pre- or postcommissural for-
nicotomy. J Comp Physiol Psychol, 92(6), 1128–1136.
Tonkiss, J., & Rawlins, J. N. (1992). Mammillary body lesions and restricted subicular
output lesions produce long-lasting drl performance impairments in rats. Exp
Brain Res, 90(3), 572–582.
fornix produces delay- and interference-dependent working memory deficits.
Behav Brain Res, 36(1–2), 113–126.
Tsivilis, D., Vann, S. D., Denby, C., Roberts, N., Mayes, A. R., Montaldi, D., et al. (2008).
A disproportionate role for the fornix and mammillary bodies in recall versus
recognition memory. Nat Neurosci, 11(7), 834–842.
Van der Werf, Y. D., Witter, M. P., Uylings, H. B., & Jolles, J. (2000). Neuropsy-
chology of infarctions in the thalamus: A review. Neuropsychologia, 38(5),
Van der Werf, Y. D., Jolles, J., Witter, M. P., & Uylings, H. B. (2003). Contribu-
tions of thalamic nuclei to declarative memory functioning. Cortex, 39(4–5),
Van der Werf, Y. D., Scheltens, P., Lindeboom, J., Witter, M. P., Uylings, H. B., &
Jolles, J. (2003). Deficits of memory, executive functioning and attention fol-
lowing infarction in the thalamus; a study of 22 cases with localised lesions.
Neuropsychologia, 41(10), 1330–1344.
Van Groen, T., & Wyss, J. M. (1990). The postsubicular cortex in the rat: Characteri-
zation of the fourth region of the subicular cortex and its connections. Brain Res,
Vann, S. D. (2005). Transient spatial deficit associated with bilateral lesions of the
lateral mammillary nuclei. Eur J Neurosci, 21(3), 820–824.
Vann, S. D. (2009a). Fornical inputs to the mammilliary bodies: Implica-
tions for diencephalic amnesia. British Neuroscience Association Abstracts, 20,
Vann, S. D. (2009b). Gudden’s ventral tegmental nucleus is vital for memory: Re-
evaluating diencephalic inputs for amnesia. Brain, 132, 2372–2384.
Vann, S. D., & Aggleton, J. P. (2003). Evidence of a spatial encoding deficit in rats with
lesions of the mammillary bodies or mammillothalamic tract. J Neurosci, 23(8),
Vann, S. D., & Aggleton, J. P. (2004). The mammillary bodies: Two memory systems
in one? Nat Rev Neurosci, 5(1), 35–44.
Vann, S. D., & Albasser, M. M. (2009). Hippocampal, retrosplenial, and pre-
frontal hypoactivity in a model of diencephalic amnesia: Evidence towards
an interdependent subcortical-cortical memory network. Hippocampus, 19(11),
Vann, S. D., Brown, M. W., & Aggleton, J. P. (2000). Fos expression in the rostral
thalamic nuclei and associated cortical regions in response to different spatial
memory tests. Neuroscience, 101(4), 983–991.
Vann, S. D., Brown, M. W., Erichsen, J. T., & Aggleton, J. P. (2000). Using fos imaging in
the rat to reveal the anatomical extent of the disruptive effects of fornix lesions.
J Neurosci, 20(21), 8144–8152.
Vann, S. D., Honey, R. C., & Aggleton, J. P. (2003). Lesions of the mammillothalamic
tract impair the acquisition of spatial but not nonspatial contextual conditional
discriminations. Eur J Neurosci, 18(8), 2413–2416.
Vann, S. D., Saunders, R. C., & Aggleton, J. P. (2007). Distinct, parallel pathways link
the medial mammillary bodies to the anterior thalamus in macaque monkeys.
Eur J Neurosci, 26(6), 1575–1586.
(2009). Impaired recollection but spared familiarity in patients with extended
hippocampal system damage revealed by 3 convergent methods. Proc Natl Acad
Sci U S A, 106(13), 5442–5447.
S.D. Vann / Neuropsychologia 48 (2010) 2316–2327
Veazey, R. B., Amaral, D. G., & Cowan, W. M. (1982a). The morphology and
fascicularis). 2. Efferent connections. Journal of Comparative Neurology, 207(2),
Veazey, R. B., Amaral, D. G., & Cowan, W. M. (1982b). The morphology and
Vertes, R. P., Hoover, W. B., & Viana Di Prisco, G. (2004). Theta rhythm of the hip-
pocampus: Subcortical control and functional significance. Behav Cogn Neurosci
Rev, 3(3), 173–200.
related neurological disorders due to alcoholism and malnutrition. Philadelphia: F.
A. Davis Company.
von Cramon, D. Y., Hebel, N., & Schuri, U. (1985). A contribution to the anatomical
basis of thalamic amnesia. Brain, 108(Pt 4), 993–1008.
retention in amnesic patients. Neuropsychologia, 12(4), 419–428.
Warrington, E. K., & Weiskrantz, L. (1982). Amnesia: A disconnection syndrome?
Neuropsychologia, 20(3), 233–248.
Wirtshafter, D., & Asin, K. E. (1983). Impaired radial maze performance in rats with
electrolytic median raphe lesions. Exp Neurol, 79(2), 412–421.
Wirtshafter, D., & Stratford, T. R. (1993). Evidence for gabaergic projections from
the tegmental nuclei of Gudden to the mammillary body in the rat. Brain Res,
Witter, M. P., Groenewegen, H. J., Lopes da Silva, F. H., & Lohman, A. H. (1989).
Functional organization of the extrinsic and intrinsic circuitry of the parahip-
pocampal region. Prog Neurobiol, 33(3), 161–253.
Yoneoka, Y., Takeda, N., Inoue, A., Ibuchi, Y., Kumagai, T., Sugai, T., et al. (2004).
Acute Korsakoff syndrome following mammillothalamic tract infarction. AJNR
Am J Neuroradiol, 25(6), 964–968.
Young, W. S., 3rd, & Kuhar, M. J. (1980). Radiohistochemical localization of benzodi-
azepine receptors in rat brain. J Pharmacol Exp Ther, 212(2), 337–346.