SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Neuropsychological findings suggest material-specific lateralization of the medial temporal lobe's role in long-term memory, with greater left-sided involvement in verbal memory, and greater right-sided involvement in visual memory. Whether material-specific lateralization of long-term memory also extends to the anteromedial thalamus remains uncertain. We report two patients with unilateral right (OG) and left (SM) mediodorsal thalamic pathology plus probable correspondingly lateralized damage of the mammillo-thalamic tract. The lesions were mapped using high-resolution structural magnetic resonance imaging and schematically reconstructed. Mean absolute volume estimates for the mammillary bodies, hippocampus, perirhinal cortex, and ventricles are also presented. Estimates of visual and verbal recall and item recognition memory were obtained using the Doors and People, the Rey Complex Figure Test, and the Logical Memory subtests of the Wechsler Memory Scales. Each patient's performance was compared to a group of healthy volunteers matched for demographic characteristics, premorbid IQ, and current levels of functioning. A striking double dissociation was evident in material-specific long-term memory, with OG showing significant impairments in visual memory but not verbal memory, and SM showing the opposite profile of preserved visual memory and significantly impaired verbal memory. These impairments affected both recall and item recognition. The reported double dissociation provides the strongest evidence yet that material-specific lateralization of long-term memory also extends to the anteromedial thalamus. The findings are also discussed in relation to proposals that distinct anatomical regions within the medial temporal lobe, anteromedial thalamus, and associated tracts make qualitatively different contributions to recall and item recognition.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

It is well established that the anteromedial thalamus plays a critical role in post-morbid memory (e.g., Aggleton & Brown, 1999, 2006; Carlesimo, Lombardi & Caltagirone, 2011; Markowitsch, 1982; Rousseaux, 1994; Van der Werf, Witter, Uylings, & Jolles, 2000; Van der Werf et al., 2003). Aggleton and Brown's still controversial model links recall and the recollection of episodic details during recognition to an extended hippocampal pathway involving a direct projection from the hippocampus to the anterior nuclear complex and mammillary bodies via the fornix, and an additional projection from the mammillary bodies to the anterior thalamus via mammillo-thalamic tract (MTT). A separate pathway from the perirhinal cortex via the ventroamygdalofugal pathway to the mediodorsal thalamus (MDT) and prefrontal areas is believed to be critical for mediating item familiarity, which in turn provides major support for item recognition. This model allows for clear predictions to be made about the selectivity of the contributions of the anteromedial thalamic nuclei to recall and item recognition, respectively. However, it does not specify whether lateralized anteromedial thalamic damage results in the material-specific memory deficits that are frequently observed following medial temporal lobe pathology, with the right medial temporal lobe specializing in memory for hard-to-verbalize visual and visuospatial materials and the left medial temporal lobe critically involved with verbal memory (Jones-Gotman et al., 1997; Kessels, de Haan, Kappelle, & Postma, 2001; Lee, Yip, & Jones-Gotman, 2002; Milner, 1974; Moscovitch & McAndrews, 2002).

A review of the anteromedial thalamic lesion literature suggests that thalamic lesions sometimes give rise to the predicted double dissociation between verbal and visual long-term memory seen at the cortical level provided that the lesions affect the structures concerned with memory (for review see Van der Werf et al., 2000 and Carlesimo et al., 2011). For example, dissociations between deficient verbal recall and spared visual recall have been described following lesions lateralized to the left side that involve the lateral thalamic nucleus, the internal medullary lamina (iML), and midline nuclei (Mori, Yamadori & Mitani, 1986); the midline nuclei and the MDT (Mennemeier, Fennell, Valenstein, & Heilman, 1992); the MTT, ventrolateral thalamus, and the MDT (patient IG, Stuss, Guberman, Nelson, & Larochelle, 1988); the MDT and MTT (Schott, Crutch, Fox, & Warrington, 2003). A correspondence between deficient visual and visuospatial memory and preserved verbal memory have also been reported following damage lateralized to either the anterior thalamus only (Daum & Ackermann, 1994); the posterolateral thalamus (patients 4 and 7, Graff-Radford, Damasio, Yamada, Elsinger, & Damasio, 1985); the anterolateral thalamus (patients 13 and 14, Graff-Radford et al., 1985); the lateral thalamus (patients 18 and 19, Graff-Radford et al., 1985); or the MTT, MDT, ventrolateral thalamus, and midline nuclei (patient RM, Stuss et al., 1988).

There are also a number of other studies presenting the counterargument to the material-specific hypothesis, with global memory deficits following lesions lateralized the same thalamic nuclei and white matter tracts previously associated with the predicted material-specific memory deficits. So for example, both verbal and visual recall impairments have been described in patients with a unilateral right-sided lesion in the region of the posterolateral thalamus (patient 9, Graff-Radford et al., 1985); the anterolateral thalamus (patients 10 and 11, Graff-Radford et al., 1985); and the lateral thalamus (patients 21–24, Graff-Radford et al., 1985). Global memory impairments also follow right-sided thalamic lesions involving the dorsal intralaminar nuclei (Van der Werf et al., 1999); the MDT (patient 16, Graff-Radford et al., 1985) or the anterior-ventral thalamus (Summers, 2002).

There are several possible reasons why the evidence is not fully concordant. First, ‘unilateral’ lesions mapped using standard brain imaging techniques (e.g., Edelstyn, Ellis, Jenkinson, & Sawyer, 2002; Squire & Moore, 1979) may actually turn out to be bilateral when high-resolution magnetic resonance imaging is used (as in the case of Baumgartner & Regard, 1993; Edelstyn, Hunter, & Ellis, 2006; Squire, Amaral, Zola-Morgan, Kritchevsky, & Press, 1989).

Second, memory for material that is nominally verbal or visual may sometimes be supported by both verbal and visual encoding so that tasks that seem to provide pure measures of verbal or visual memory do not do so. Furthermore, it is difficult to be sure that memory performance on a task is selectively dependent on just verbal or just non-verbal visual coding. This argument was previously used by von Cramon, Hebel and Schuri (1985) to explain performance of their left thalamic lesion patient (Case 5), who performed poorly on an object paired-associate learning task and also in a visually guided stylus maze task (p. 1001).

Third and related to the above reason, deliberate encoding strategies may be employed to support memory performance. For example, Mennemeier et al. (1992) reported a patient with a left-sided medial thalamic lesion, who showed sparing of verbal memory when allowed to rehearse material in the study–test delay or to use semantic encoding strategies. However, when these strategies were prevented by use of filler tasks between study and test, her performance was impaired. The development of such strategies is more likely to be the case for patients where there is a longer time interval between lesion onset and memory assessment (von Cramon et al., 1985).

Fourth, as has recently been argued by Saling (2008) for lateralized medial temporal lesions, and probably applies by extension to lateralized anteromedial thalamic lesions, kinds of verbal memory (and possibly also visual memory) that depend strongly on semantic processing may be less well lateralized than those that depend on forming arbitrary associations. If this proves to be correct, it would, of course, require a modification of the material-specific memory hypothesis.

Finally, the material-specific memory hypothesis as it applies to the thalamus may need to be modified because Aggleton and Brown's (1999) model does not fully capture the contribution of the anteromedial thalamic nuclei to memory as suggested by the studies of Zoppelt, Koch, Schwarz, and Daum (2003) and Cipolotti et al. (2008). In Zoppelt et al.'s (2003) study of five MDT lesion patients, those with lesions predominantly in the more lateral/parvicellular subdivision of the nucleus, showed a decrease in recollection and sparing of familiarity, whereas those whose lesion was more medially placed displayed deficits in recollection and familiarity. In Cipolotti et al.'s study of two patients with asymmetric bilateral anteromedial thalamic lesions, it was found that their left-sided anteromedial damage resulted in the expected decline in recollection and familiarity of verbal memoranda. However, both patients also showed impairments in recollection and familiarity of non-verbal memoranda despite the presence of right-sided lesions that, in the case of patient 1, primarily affected the anterior thalamic nucleus (ATN), leaving the MDT relatively spared, and, in the case of patient 2, affected the MDT with relative sparing of the ATN. Questions about the contribution of the MDT to memory have already been raised by non-human primate tracing studies, where it has been reported that the perirhinal cortex only projects to the more medial (magnocellular) subdivision of the MDT as well as to the adjacent midline nucleus (Gaffan & Parker, 2000; Preuss & Goldman-Rakic, 1987; Russchen, Amaral, & Price, 1987). As the more lateral (parvicellular) MDT does not receive perirhinal cortex projections, Aggleton and Brown should predict that only damage to the more medial parts of the MDT and the midline nucleus should disrupt familiarity. So, if this modification of the Aggleton and Brown view is correct, the thalamic material-specific memory hypothesis should predict that left-sided lesions of the medial/magnocellular MDT should disrupt familiarity memory for verbal materials and right-sided lesions should disrupt familiarity memory for visual materials with the effects of parvicellular lesions remaining currently unspecified. Caution should be exercised because it is extremely difficult to be sure about the precise localization and extent of small thalamic lesions in humans. In an attempt to reconcile the sometime discordant clinical and animal lesion evidence of the contribution if different medial thalamic nuclei to recognition memory, Aggleton, Dumont, and Warburton (2011) have proposed the ‘multi-effect multi nuclei’ model that proposes that anteromedial thalamic nuclei can have both direct and indirect effects on recognition. Building on the earlier Aggleton and Brown (1999) model, direct effects on recollection are mediated via the mammillary body, MTT, and anterior thalamus, and on familiarity via the MDT. However, the major addition introduced by the multi-effect multi-nuclei model are the indirect influences that can act on both recollection and familiarity in a non-specific manner. The mechanisms by which these effects are mediated is by the modulation of arousal and attention by connections between the intralaminar and midline thalamic nuclei, the MDT, and prefrontal areas (Portas et al., 1998).

The aim of our study was to investigate material-specific lateralization of long-term memory in two patients with thalamic pathology (SM and OG). These patients were particularly well suited to the purpose, as high-resolution structural magnetic resonance imaging has shown that the SM's lesion is clearly limited to the left thalamus and OG's lesion is limited to the right thalamus. In both patients, the lesion involves the midline nuclei (central medial and paraventricular nuclei), the medial/magnocellular and part of the parviceullar subdivisions of the MDT, the intramedullary lamina, and encroached on the MTT, thereby partially disconnecting the mammillary bodies from the anterior thalamus. It should be noted that our patients’ unilateral lesions are not exact mirror images of each other. OG's lesion is centred on the medial division of the MDT whereas SM's lesion is more anterior and ventral. Importantly, in both cases, there is no evidence of pathology in structures that have been related to memory functions in the contralateral diencephalon and medial temporal lobes. However, volume measures of these structures were performed to determine if retrograde or anterograde degeneration had occurred and, therefore, may also contribute to the memory loss.

Material-specific long-term memory was assessed with item recognition, immediate and delayed recall tests, using the Doors and People Test (Baddeley, Emslie, & Nimmo-Smith, 1994), The Rey Complex Figure Test (Myers & Myers, 1995), and the Logical memory subtests from the Wechsler Memory Scales (Wechsler, 1997).

The Doors and People Test is particularly well suited for studying material-specific long-term memory as contains separate assessments of visual and verbal four-choice recognition memory (Doors and Names subtests, respectively) as well as visual and verbal recall (Shapes and People subtests). The Rey Complex Figure Test provides measures of visual recall following delays of 3 min and 20 min; and the Logical memory subtests provide measures of immediate and delayed verbal recall, and immediate verbal recognition. These standardized tests have also been used in many previous studies of amnesia, and, therefore, including them here provides a bridge with the literature.

According to the material-specific hypothesis of long-term memory, a double dissociation is predicted, with OG's right-sided lesion causing a disruption of visual memory and sparing of verbal memory, and SM's left-sided lesion causing an impairment of verbal memory and sparing of visual memory. Furthermore, we predicted material-specific memory impairments to be evident for both patients in recognition and recall because both had suffered disruption of the perirhinal-mediodorsal thalamic pathway, which subserves recognition, and of the MTT that, as part of the hippocampus-anterior thalamus circuit, subserves recall (Aggleton & Brown, 1999, 2006).

Methods

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Participants

Thalamic lesion patients

Two right-handed male patients (SM and OG) are reported.

Patient OG is a right-handed, male of high average intelligence (see Table 2). He was aged 70 at the time of testing. OG enjoys an active life-style that involves playing bridge a couple of times a week, walking up to 25 miles per week and caring for his grand-children. Prior to his stroke in 2000, he worked as an electrical engineer. The patient has a 30-year history of visual migraine and hypertension.

Table 1.  Absolute volume estimates, standard deviations, and ranges for the mammillary bodies, hippocampal-related structures, and ventricles in patients OG and SM and a group of 20 (previously published) healthy controls
StructuresSideAbsolute volume estimates
Patient OGPatient SMControls
Meanz scoret testMeanz scoret testMeanSD
  1. MB, mammillary body; Hip, hippocampus; Peri, perirhinal area; SD, standard deviation.

MBRight0.0690p= .480.070.1p= .480.0690.01
Left0.056–1.1p= .150.06–0.9p= .200.0670.01
HipRight2.970.54p= .302.840.2p= .432.770.37
Left2.750.18p= .433.161.2p= .122.680.39
PeriRight3.02–0.1p= .463.04–0.08p= .463.10.79
Left3.35–0.01p= .483.2–0.25p= .413.40.81
VentriclesRight234.3p < .0018.20.3p= .397.133.72
Left29.24p < .001151.4p < .0017.85.29
Table 2.  Age, IQ, and scores on the Doors and People for patients OG, SM, and their healthy control groups. Significant differences shown in bold
 ParticipantsStatisticParticipantsStatistic
OGControls (n= 8) (mean [SD])z scoret testSMControls (n= 8) (mean [SD])z scoret test
  1. NART, National Adult Reading Test; SD, standard deviation.

  2. Note. Significant at *p < .05, **p < .01.

    t=–0.28   t=–0.33
Age7070.75 (2.49)p= .39 5759.88 (8.17)p= .38
    t=–0.43,    
NART116118.0 (4.38)p= .34123124.0 (2.36)p= .35
Wechsler Abbreviated Scale of Intelligence      
    t=–1.40,   t= 0.00,
Performance IQ110121.0 (7.43)p= .10133133.0 (6.37)P= .50
    t=–0.1   t=–0.40,
Verbal IQ107108.13 (10.63)p= .46128132.29 (10.06)p= .35
    t=–1.03   t=–0.30,
Full Scale IQ108116.0 (8.23)p= .17134136.86 (7.52)p= .39
Doors and People
Visual recognition (Doors, max score 24) 13*17.5 (1.77)–2.54t=–2.402222.25 (5.6)–0.04t=–0.04
    p= .02   P= .48
Immediate visual recall (Shapes, max score 36) 33**35.63 (0.74)–3.55t=–3.353635.88 (0.35)0.34t= 0.32
    p < .01   p= .38
Delayed visual recall (Shapes, max score 12)1212 (0)1212 (0)
Verbal recognition (Names, max score 24)1617.0 (4.41)–0.23t=–0.21 13**19.88 (2.17)–3.17t=–2.99
    p= .42   p= .01
Immediate verbal cued recall (People, max score 36)2628.88 (3.48)–0.83t=–0.782931.20 (4.79)–0.45t=–0.43
    p= .23   P= .34
Delayed verbal cued recall (People, max score 12)       t=–1.92
 1212 (0)10*11.67 (0.82)–2.04p= .05

Patient SM is a right-handed male of superior intelligence (see Table 2). He worked as a physician prior to suffering a stroke in 2006. He has now retired from clinical duties, but continues to teach and examine medical students and junior doctors. SM has no premorbid medical history of significance.

Both patients’ initials have been changed to preserve anonymity and fully informed written consent was obtained from all participants. The study was approved by North Staffordshire NHS research ethics committee.

Healthy control groups

Because the patients varied in age and IQ, their performance on the tests of memory is compared to separate groups of healthy controls matched to each patient for gender, age, premorbid IQ (National Adult Reading Test, Nelson & Willison, 1991), and current levels of functioning (Wechsler Abbreviated Scale of Intelligence, Wechsler, 1999).

Modified t tests (Crawford & Garthwaite, 2002) and z scores have been selected in preference to the normative data provided by each of the tests to indicate impairment. The decision to collect our own control data is driven primarily by the fact that our patients are high performing (particularly true in the case of SM whose IQ performance falls in the superior range) and so the normative data provided for the tests will not have been collected from people who are IQ- and education matched to the patients. This is important as research shows that performance on The Rey Complex Figure Test (RCFT) (Fastenau, Denburg, & Hufford, 1999) and Logical memory subtests (Hawkins & Tulsky, 2001; Sass et al., 1992) is influenced by education and IQ.

There was some attrition in the membership of OG's healthy control group during the study. Of the original eight healthy controls who participated in the Doors and People Test, one was lost to follow up in the LM subtest, and a total of four controls were lost to follow up in the RCFT. Four new controls were recruited and provided missing data on the RCFT. The age and IQ scores for these newly constituted control groups are provided in Tables 3 and 4.

Table 3.  Age, IQ, and scores on the Logical Memory subtests for patients OG, SM, and their healthy control groups
 ParticipantsStatisticParticipantsStatistic
OG*Controls (n= 7) (mean [SD])z scoret testSMControls (n= 8) (mean [SD])z scoret test
  1. NART, National Adult Reading Test; SD, standard deviation.

  2. Note.*One of OG's original eight healthy controls was lost to follow-up. Significant at *p < .05, **p < .01, ***p < .001.

    t=–0.16,   t=–0.33,
Age7070.43 (2.51)p= .445759.88 (8.17)p= .38
    t=–0.41,   t=–0.22,
NART116118.0 (4.6)P= .35123124.0 (2.36)p= .35
Wechsler Abbreviated Scale of Intelligence      
    t=–1.3,   t= 0.00,
Performance IQ110121.0 (7.85)p= .12133133.0 (6.37)p= .50
    t= 0.4,   t=–0.40,
Verbal IQ107106.57 (10.45)p= .49128132.29 (10.06)p= .35
    t=–0.91,   t=–0.30,
Full Scale IQ108114.71 (7.97)p= .2134136.86 (7.52)p= .39
Logical Memory (immediate)       
Story A and B recall (max score 50)   t= 0.72,   t=–2.71,
 2925.14 (5.01)0.77p= .25 24*33.25 (4.03)–2.80p= .02
Logical Memory (delayed)       
Story A and B recall (max score 50)   t= 0.24,   t=–2.71,
 2624.29 (6.73)0.25p= .41 21*35.25 (4.95)–2.80p= .02
Recognition memory (max score 30)   t= 1.99,   t=–4.75,
 2825.0 (1.41)2.13p= .07 24***28.63 (0.92)–5.03p= .001
Recall-recognition discrepancy score       t= 2.10,
   2.73*–0.1 (1.27)2.23p= .037
Table 4.  Age, IQ, and scores on the Rey Complex Figure Test for patients OG, SM, and their healthy control groups. Significant differences shown in bold
 ParticipantsStatisticParticipantsStatistic
OGControls (n= 8) (mean [SD])z scoret testSMControls (n= 8) (mean [SD])z scoret test
  1. NART, National Adult Reading Test; SD, standard deviation.

  2. Note. Significant at *p < .05, **p < .01.

    t= 0.72   t=–0.2
Age7066.25 (4.92)p= .255758.39 (6.63)p= .43
    t=–1.49   t= 0.27
NART116121.0 (3.16)p= .09123122 (3.46)p= .4
Rey Complex Figure Test
    t= 0.32   t= 0.26
Copy3635.88 (0.35) 0.35p= .383635.9 (0.36)0.03p= .40
3-min delayed recall (max score 36)   t=–2.42   t= 0.71,
 14*23.56 (3.72)–2.20p= .023027.75 (4.32)0.52p= .25
15-min delayed recall (max score 36)   t=–2.83,   t= 0.45,
  10.5**23.0 (3.3)–3.79p= .0013027.4 (5.42)0.50p= .33

Structural Magnetic Resonance Image sMRI acquisition

All MR images were acquired using the same 3T Magnetron Trio whole-body imaging system (Siemens Medical Solutions, Siemens plc, Sir William Siemens Square, Frimley, Camberley, GU16 8QD, UK). Two coronal T1-weighted three-dimensional magnetization-prepared rapid acquisition gradient echo (MPRAGE) pulse sequences having slice thicknesses of 1 mm and 0.8 mm were acquired. Measurements of the thalamus, hippocampus, perirhinal cortex, and ventricles were performed using the 1-mm, isotropic (1 × 1 × 1) acquisition, having sequence parameters as follows: Repetition Time (TR) = 1,960 ms, Echo Time (TE) = 4.43 ms, Inversion Time (TI) = 1,100 ms, Flip Angle (FA) = 8°, FOV = 256 × 256 mm2, providing 172 contiguous coronal 1-mm thick sections having an acquisition time of 8 min, 23 s. The mammillary bodies were measured using the 0.8-mm volume scan, in which a block of 72 contiguous slices was acquired through the area of interest, this sequence provided the optimal fine resolution required for the accurate demarcation of the mammillary body landmarks. The sequence parameters for the 0.8-mm scan were TR = 2,040 ms, TE = 5.57 ms, TI = 1,100 ms, FA = 8°, Acquisition Time = 11 min, 47 s, Field of View (FOV) = 256 × 256 mm2. In addition, a sagittal T1-weighted sequence (TR = 7.92 ms and TE = 2.48 ms, having a slice thickness of 1 mm) was acquired for the qualitative visualization of the lesion.

Intracranial volume measurements were performed on a T2-weighted coronally acquired sequence, parameters as follows: TR = 3,000 ms, TE = 102 ms, FA = 150°, Slice Thickness = 3 mm, 10-mm Inter-slice Gap, FOV = 220 × 220 mm2, Acquisition Time = 2 min.

The scans were visualized using MRIcro and Brainvoyager software, and stereological volume estimation was performed using Easymeasure software.

Schematic lesion reconstruction

As no technique is regarded as highly accurate and reliable for identifying the specific thalamic nuclei damaged (Spinks et al., 2002), the MTT and the most relevant thalamic nuclei have been schematically represented for both patients following the procedure of Carlesimo et al. (2007) with reference to the brain atlases of Mai, Assheuer, and Paxinos (2004) and van Buren and Borke (1972). The schematic reconstructions are drawn onto alternate 0.8-mm coronal T1 slices and presented in Figure 1A for OG and Figure 1B for SM.

image

Figure 1. Magnification of the diencephalic region from patients OG (A) and SM (B) in normalized space that includes five T1-weighted alternate coronal slices (slice thickness 1.0 mm). For each T1-weighted brain slice, a schematic reconstruction of the lesion (shown in black) is superimposed on the corresponding slice (lower). For the purpose of illustration, only the most relevant anatomical structures are reported (the left hemisphere is shown to the left of the midline). DT = dorsal thalamus; LT = lateral thalamus; AT = anterior thalamus; VeT = external ventral thalamus; ViT = internal ventral thalamus; MTT = mammillothalamic tract; iML = internal medullary lamina; MDT = mediodorsal thalamus.

Download figure to PowerPoint

Patient OG's right thalamic lesion (shown in black) involved the medial division of the MDT (orange), iML (yellow), and caudal intralaminar nuclei, and appeared to encroach on the MTT (red), thereby partially disconnecting the indirect hippocampal projections to the anterior thalamus that run via the mammillary bodies. The ventrolateral portion of the dorsal thalamus (navy blue) was also damaged. The anterior, ventral, and lateral thalamic nuclei were spared.

Patient SM's left thalamic lesion (black) was positioned slightly more anterior and ventral to OG's medial MDT thalamic lesion, involving the ventroanterior thalamic nucleus (dark blue), the ventrolateral thalamic nucleus (pink), and the posterior ventrolateral nucleus (pVLN, green) of the ‘motor thalamus’, the iML (yellow), and the MDT (orange). The distal edge of the lesion appeared to encroach on the MTT (red).

Stereological volume estimation of medial temporal lobe structures and lateral ventricles

Mean absolute volume estimates of the mammillary bodies, hippocampus, perirhinal cortex, and lateral ventricles were obtained using the Cavalieri method of modern design stereology combined with point-counting techniques (Cruz-Orive, 1993, 1999; García-Finãna et al., 2003; see Denby et al., 2009, for a detailed account of the stereology procedure). The mammillary bodies, hippocampus, and perirhinal cortex were selected on the basis of their strong associations with anterograde amnesia and the presence of agreed reliable landmarks to provide valid volume estimates (Tsivilis et al., 2008). Finally, estimates of ventricle volume were obtained to examine potential effects of cortical shrinkage.

Previously published control data from 20 healthy volunteers (10 male, 10 female, mean age 48.1 years, age range 25–62 years) are provided for the mammillary bodies, the hippocampus, perirhinal cortex, and lateral ventricles (Denby et al., 2009; Tsivilis et al., 2008).

Material-specific memory tests

Each memory test was administered according to the instructions in its manual.

The Doors and People Test

The Doors and People Test provides separate measures of four-choice visual recognition memory, four-choice verbal recognition memory (the Doors and Names subtests, respectively), visual recall, and verbal recall (the Shapes and People subtests, respectively).

Both visual recognition and verbal recognition subtests contain 24 trials, subdivided into two equal sections (termed ‘easy’ and ‘hard’). The ‘hard’ version reflects higher inter-item similarity between each target and its three distractors at recognition.

In the visual recognition task (Doors), stimuli consist of colour photographs of different types of doors, such as plain front doors, wooden barn doors, ornate doors surrounded by porticos. High similarity between each target and its foils makes it hard for people to use verbally coded information to help performance.

The verbal recognition stimuli (Names) consist of names (first and second name). The maximum score in both recognition tests is 24.

The test of verbal (cued-) recall (People) requires participants to learn the names of four people. Each name is printed on a separate card underneath a coloured photograph of the person. At study, the task is to recall the name of the person, which is cued by the profession. For example, ‘This is the doctor. His name is…..’. At test, which immediately follows presentation of the fourth photograph, name recall is cued, for example, ‘What was the doctor's name?’ The procedure is repeated (up to a total of three times) until all of the names are successfully recalled. The maximum score in the immediate cued-recall test is 36. Delayed cued recall takes place 15–20 min later, with a maximum score of 12. Immediate and delayed cued-recall verbal scores are reported separately.

In the test of visual recall (Shapes), participants are first asked to draw four simple shapes on separate sheets of paper. Immediately after this they are asked to draw the four shapes from memory. The maximum score for immediate recall is 36, and for 15-min delayed recall the maximum score is 12.

The Rey Complex Figure Test

Participants first copy a figure consisting of 18 different elements. A surprise test of visual recall takes place following a 3-min filled delay, and then again after a further 15–20 min. Participants are not warned at the outset that this is a memory test. A maximum score of 36 is achieved, with 2 points given to each element if it is accurately drawn (1 point) and correctly positioned in relation to the other elements (1 point).

The Logical Memory test

Immediate and delayed LM subtests from the Wechsler Memory Scales (Wechsler, 1997) provide measures of immediate and delayed verbal recall. This test is more taxing than the Doors and People verbal recall task (and so less prone to ceiling effects), involving recall of two short stories (Story A and Story B). Immediate (verbatim, gist) recall of the story takes place straight after the experimenter has finished reading the story, and delayed recall follows a 20- to 25-min filled delay. The final subtest involves a yes/no recognition test.

Results

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Stereologic volume estimation

Table 1 shows the mean absolute volume estimates of the mammillary bodies, hippocampus, perirhinal areas, and lateral ventricles for OG and SM, and published control data (Tsivilis et al., 2008). Coronal MRI sections from OG and SM showing these areas of interest are presented in Figure 2.

image

Figure 2. Two consecutive horizontal T1-weighted coronal images for patient OG (left panel) and patient SM (right panel) are presented here showing the integrity of the mammillary bodies (A), hippocampus (B), perirhinal cortex (C), and ventricles (D). The left hemisphere is shown to the left of the midline. MB = mammillary bodies; hip = hippocampus; peri = perirhinal cortex; LV = lateral ventricles.

Download figure to PowerPoint

The absolute measures of OG's left and right mammillary bodies, left and right hippocampus, left and right perirhinal cortex areas were all normal range (all ps > .05), although there was evidence of dilation of both left and right lateral ventricles (both ps < .001). SM's profile was broadly similar to OG’s, with no abnormalities evident in the mammillary bodies, hippocampus, or perirhinal cortex areas on either side (all ps > .05). However, there was evidence of a unilateral enlargement of the left ventricle (p < .001).

Material-specific memory

The performance of both patients and their respective control groups on the Doors and People Test (D&P), Rey Complex Figure Test (RCFT), and Logical Memory (LM) subtests is presented in Tables 2–4, respectively, and Figure 3.

image

Figure 3. Patients’ standardized scores on tests of visual memory (upper) and verbal memory (lower), collapsed across test battery. D&P = Doors and People Test; RCFT = Rey Complex Figure Test; LM = Logical memory.

Download figure to PowerPoint

Patient OG

OG showed a dissociation between impaired memory for visual memoranda and spared memory for verbal memoranda. Visual memory decline affected both recall (D&P Shapes recall subtest, t=−3.35, p < .01; RCFT 3-min delayed recall, modified t=−2.42, p= .002; and the RCFT 15-min delayed recall, modified t=−2.83, p < .0001) and recognition (D&P Doors recognition subtest, modified t=−2.40, p= .02). Both types of retrieval exhibited comparable levels of impairment (D&P, visual recall–visual recognition discrepancy score, modified t= 0.73, p= .25). It is worth noting, at this point, that the performance of both OG and his controls on the Doors and People Memory Test (especially the cued/recall subtests) is above the average for their age range (66–75 years) according to test manual norms, and the IQ scores for a large group of age-matched healthy volunteers reported by Davis, Bradshaw, and Szabadi (1999).

Severity of impairment on the RCFT was greater following a longer retention interval compared to shorter retention interval (z=−2.57 and z=−3.79, respectively). Inspection of performance on the D&P, indicated a greater decline in recall compared to recognition (z=−3.55 and z=−2.54, respectively) (see Figure 3).

OG's verbal memory, in contrast, was spared (D&P People cued-recall subtest, modified t=−0.78, p= .23, z=−0.83; the LM immediate recall subtest, modified t= 0.72, p= .25, z= 0.77; the LM delayed recall subtest, modified t= 0.24, p= .41, z= 0.25; D&P Names recognition subtest, modified t=−0.21, p= .42, z=−0.23; and the LM delayed recognition subtest, t= 1.99, p= .07, z= 2.13).

Patient SM

SM's memory profile was characterized by a selective impairment in verbal memory, evident on tests of recall (LM immediate recall, modified t=−2.71, p= .02, z=−2.80; LM delayed recall, modified t=−4.75, p= .001, z=−2.80) and recognition (LM recognition, modified t=−4.75, p= .001, z=−5.03; D&P Names recognition, modified t=−2.99, p= .01, z=−3.17). Severity of impairment tended towards a greater decline in recognition (LM recognition, z=−5.03; D&P Names recognition, z=−3.17) compared to recall (LM 3-min recall, z=−2.80; LM 15-min recall, z=−2.80).

There was one anomalous result, in which SM's verbal recall on the People subtest was spared (modified t= 0.14, p= .45). It is possible that SM's spared immediate cued recall on the Names test is due to the anomalous features of this test where acquisition of the name is supported by provision of an occupation (used as a subsequent cue at test) and photograph of the individual at study. The combination of visual and verbal information may serve to reduce the sensitivity of this test when recall is tested immediately after study. However, introducing a delay between study and test thereby reducing the carry-over effects of a visual aide-memoire, the test becomes a more sensitive measure of verbal memory.

SM's visual memory, on the other hand, was spared (D&P Shapes recall, modified t= 0.32, p= .38, z= 0.34; RCFT 3-min recall, modified t= 0.56, p= .30, z= 0.52; RCFT 15-min recall, modified t= 0.12, p= .46, z = 0.50; and D&P Doors recognition, modified t=−0.04, p= .48, z=−0.04).

SM's verbal recall–verbal recognition discrepancy score, based on the LM immediate recall and recognition subtests, was significantly different from his controls (modified t= 2.10, p= .037) indicating a relatively greater decline in verbal recognition compared to verbal recall.

Discussion

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Our findings have direct implications for the debate regarding the relationship between material-specific deficits in long-term memory and lateralized lesions in the region of the anteromedial thalamus. In this study, we described two patients with unilateral left (SM) and right (OG) mediodorsal thalamic (MDT) pathology plus probable correspondingly lateralized damage of the MTT. The patients’ pathology was localized using high-resolution structural magnetic resonance imaging, and schematic reconstructions drawn onto alternate 0.8-mm coronal T1 slices following the procedure of Carlesimo et al. (2007). Absolute volumetric estimates of the mammillary bodies, hippocampi, perirhinal areas, and ventricles were also performed to assess the impact of damage in and around the anteromedial thalamus on efferent and afferent target sites.

The data from OG and SM showed a double dissociation in material-specific long-term memory deficits. OG's visual memory was deficient but his verbal memory was spared following a right-sided lesion, whereas SM's verbal memory was deficient and his visual memory spared following a left-sided lesion. These findings build on and extend previous studies, where dissociations between (impaired) verbal memory and (spared) visual memory following left thalamic lesions and (impaired) visual memory and (spared) verbal memory following right thalamic lesions have been reported in over 40 separate studies reporting over 50 patients (see Introduction for review of material-specific deficits in anteromedial thalamic lesions patients). However, it is of interest to note that studies reporting an association between verbal memory impairments and left anteromedial thalamic lesions are more frequently reported than a correspondence between visual and spatial memory deficits and right-sided damage (around 40 patients vs. 10 patients, respectively). As noted recently, this unequal distribution may reflect a range of (not necessarily mutually exclusive) factors such as the greater disability associated with left-sided lesions, compared to right-sided lesions, increasing the frequency with which these patients are identified (Carlesimo et al., 2011). Tests of verbal memory may be more sensitive (notwithstanding the issues described here with Names cued-recall test) than those designed to detect visual and visuo-spatial memory (e.g., the noted susceptibility of the Shapes visual recall test to ceiling effects), which may be supported by verbal encoding strategies (deliberate or unintentional).

A small number of studies describe patients who fail to conform to this pattern, with, for example, lateralized left- or right-sided lesions resulting in a global memory deficit. One important reason for such inconsistencies is the reliance on low-resolution brain imaging, which fails to identify bilateral thalamic damage. For instance, in the first of two studies on patient QX, Edelstyn et al. (2002) initially reported that his lesion was limited to the left MDT on the basis of a CT scan. However, subsequent scanning, using high-resolution MRI imaging, revealed the presence of bilateral pathology of the dorsolateral thalamic nuclei (Edelstyn et al., 2006).

The important features of the present study are, therefore, the use of high-resolution brain imaging evidence showing that our patients’ medial thalamic lesions, and presumed partial disconnection of the MTT, are clearly lateralized to the left side (SM) and right side (OG). Secondly, this is the first report of a double dissociation between visual memory and verbal memory in two thalamic lesion patients using the same battery of neuropsychological tests, thereby supporting the proposal that the material-specific lateralization of long-term memory extends to the MDT and thalamic tracts.

Before discussing some of the other implications of our study, the findings from the stereological volume estimation of medial temporal lobe memory areas and the lateral ventricles will be considered. Both patients show evidence of ventricular enlargement, for SM this is lateralized to the left ventricle on the same side as his thalamic lesion, whereas for OG both ventricles showed enlargement. SM's unilateral left ventricular enlargement is consistent with other reports of ex vacuo ventricular dilatation following various types of thalamic pathology, where there is outward movement of the ventricles to fill the lacuna. However, this compensatory enlargement is not associated with cerebrospinal fluid (CSF)-compression on proximal structures and tracts (e.g., Weisberg & Dunn, 1983; Wood & Bigler, 1995), so should not contribute to any memory deficits. The presence of bilateral ventricular enlargement in the case of OG is unlikely to be caused by a ‘hidden’ left-sided lesion either at the level of the diencephalon or the cortex since the high-resolution MR imaging employed in this study would have picked up such damage. OG's memory profile is also supportive of this view. Hydrocephalus ex-vacuo has been reported as ‘commonplace’ amongst older adults, and ‘minor degrees of this are certainly the rule with increasing age…. It is almost always symmetrical and uniform throughout the ventricular system’ (Kirkpatrick, 1978). This explanation is consistent with OG's age (70 years at the time of testing), and why generalized ventricular enlargement is absent from SM, 13 years younger than OG.

Our study also has implications for the unresolved matter regarding the importance of the extended hippocampal circuit for recognition and recall. The MTT is part of this circuit, connecting the mammillary bodies to the anterior thalamic complex, and consequently, the presumed partial disconnection of the MTT in OG and SM, will disrupt hippocampal memory processes. According to one view, the hippocampus is important for both recognition and recall, based on the assumption that it supports familiarity as well as recollection and these involve different levels of activation or degree of need for optimal functioning within the same memory system (see recent review by Wixted & Squire, 2011). The other view is that only recall is dependent on the extended hippocampal circuit, with recognition (through its dependence on familiarity memory) relying mainly on the perirhinal cortex, MDT, and connecting tracts (Aggleton & Brown, 1999, 2006; see also Yonelinas, Aly, Wang, & Koen, 2010). Both models allow for a partial dissociation between relatively preserved recognition and more impaired recall provided it is assumed that optimal recognition is usually less dependent on efficient hippocampal system functioning than is optimal recall. However, only Aggleton and Brown's model allows for a double dissociation, or a relatively greater decline in recognition compared to recall because this is not expected if familiarity is, on average, a weaker form of memory than recollection and both are mediated by the same medial temporal lobe and thalamic structures. Some evidence indicates that not only do hippocampal system lesions selectively disrupt recall, but perirhinal cortex lesions selectively disrupt familiarity memory (see Montaldi & Mayes, 2010).

It might be felt that because our patients had damage to both the perirhinal-MDT thalamic and the extended hippocampal circuits, our findings cannot have much bearing on this debate. However, when considering the patients’ performance on the recall and recognition tasks in terms of z scores and t scores, which reflect differences between the patient and the controls expressed in terms of the variance in the control group, SM's verbal memory was marked by a relatively more severe impairment in recognition compared to recall. This retrieval profile cannot easily be accommodated by the single process view. It suggests that SM's familiarity deficit was more severe than his recollection, which is the opposite of what would be expected if all the thalamic memory structures play an equal role with recollection and familiarity, and familiarity is a weaker form of memory. This view predicts that recollection should generally be more impaired because, as the stronger form of memory, it will be disrupted as soon as a structure can no longer work at its optimal level. Although OG showed this pattern of performance, SM showed the reverse pattern. Our results, therefore, suggest that this single process view does not explain some of the effects of thalamic lesions on memory. However, if the difference between recall and recognition is expressed in terms of the percentage of material retained, then SM's recall is much more impaired than recognition, which is consistent with the single process view.

Our findings also have implications for Saling's (2009) proposal that certain kinds of verbal memory, such as related-paired associates, list learning, or prose recall, which depend strongly on semantic processing, may be less well lateralized than those which depend on forming arbitrary associations, such as arbitrary-paired associates. This hypothesis predicts that OG may be impaired on the Logical Memory recall because it is semantic and is mediated bilaterally. It is clear that our findings do not fit well with this proposal, as SM showed just as clear a deficit on the Logical Memory recall test as the possibly less semantic verbal cued-recall test from the Doors and People Test. Our results indicate that left-sided lesions disrupt semantic and arbitrary verbal memory equally and that right-sided lesions do not disrupt verbal semantic memory any more than verbal arbitrary memory.

In conclusion, previous studies of patients with medial temporal lobe pathology have demonstrated lateralization of material-specific long-term memory. It has been uncertain whether material-specific lateralization also extends to the anteromedial thalamus because many studies have relied upon use of low-resolution brain imaging techniques so that lesion laterality could not be confidently identified and/or have used visual and verbal tasks performance on which may well have depended strongly on encoding stimuli verbally as well as visually. To our knowledge, this is the first study to specifically examine the issue of material-specific lateralization of memory in two patients with high-resolution magnetic resonance imaging and the same appropriate memory tests. The patients’ pathology unequivocally focused on the left medial thalamus and right medial thalamus, respectively, and, as the thalamic material-specific memory hypothesis predicts, they showed selective verbal and visual memory deficits, respectively. Unilateral damage to the medial/magnocellular part of the MDT and the MTT therefore, disrupts long-term memory in a material-specific way.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We are grateful to our participants for their assistance, and for the support from the Research Institute for Life Course Studies, Keele University. We also thank our reviewers for their insightful comments.

References

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Aggleton, J.P., & Brown, M.W. (1999). Episodic memory, amnesia, and the hippocampal-anterior thalamic axis. Behavioural and Brain Sciences, 122, 425489.
  • Aggleton, J. P., & Brown, M. W. (2006). Interleaving brain systems for episodic and recognition memory. Trends in Cognitive Science, 10, 455463.
  • Aggleton, J. P., Dumont, J. R., & Warburton, E.C. (2011). Unraveling the contributions of the diencephalon to recognition memory: A review. Learning and Memory, 18, 84400.
  • Baddeley, A., Emslie, H., & Nimmo-Smith, I. (1994). The Doors and People Test. Bury St. Edmunds , Suffolk : The Thames Valley Test Company.
  • Baumgartner, R. W., & Regard, M. (1993). Bilateral neuropsychological deficits in unilateral paramedian thalamic infarction. European Neurology, 33, 195198.
  • Carlesimo, G. A., Lombardi, M. G., & Caltagirone, C. (2011). Vascular thalamic amnesia: A reappraisal. Neuropsychologia, 45(11), 246724790.
  • Carlesimo, G. A., Serra, L., Fadda, L., Cherubini, A., Bozzali, M., & Caltagirone, C. (2007). Bilateral damage to the mammillo-thalamic tract impairs recollection but not familiarity in the recognition process: A single case investigation, Neuropsychologia, 45(11), 24672479.
  • Cipolotti, L., Husain, M., Crinion, J., Bird, C.M., Khan, S. S., Losseff, N., … Leff, J. P. (2008). The role of the thalamus in amnesia: A tractography, high-resolution MRI and neuropsychological study. Neuropsychologia, 46, 27452758.
  • Crawford, J. R., & Garthwaite, P. H. (2002). Investigation of the single case in neuropsychology: Confidence limits on the abnormality of test scores and test score differences. Neuropsychologia, 40, 11961208.
  • Cruz-Orive, L. M. (1993). Systematic sampling in stereology. Bulletin of the International Statistical Institute, Proceedings of 49th Session, Florence, 55, 451468.
  • Cruz-Orive, L. M. (1999). Precision of Cavalieri sections and slices with local errors (1999). Journal of Microscopy, 193, 182198.
  • Daum, I., & Ackerman, H. (1994). Frontal-type memory impairment associated with thalamic damage. International Journal of Neuroscience, 77, 187198.
  • Davis, C., Bradshaw, C. M., & Szabadi, E. (1999). The Doors and People Memory Test: Validation of norms and some new correction formulae. British Journal of Clinical Psychology, 38, 305314.
  • Denby, C.E., Vann, S. D., Tsivilis, D., Mayes, A. R., Montaldi, D., & Aggleton, J.P. (2009). Mammillary body volume estimation in healthy control subjects and colloid cyst surgery patients. American Journal of Neuroradiology, 30, 736743.
  • Edelstyn, N. M. J., Ellis, S. J., Jenkinson, P., & Sawyer, A. (2002). Contribution of the left dorsomedial thalamus to recognition memory: A neuropsychological case study. Neurocase, 8, 442452.
  • Edelstyn, N. M. J., Hunter, B., & Ellis, S. J. (2006). Bilateral dorsolateral thalamic lesions disrupts conscious recollection. Neuropsychologia, 44, 931938.
  • Fastenau, P. S., Denburg, N. L., & Hufford, B. J. (1999). Adult norms for the Rey-Osterrieth Complex Figure Test and for supplemental recognition matching trials from the extended complex figure test. The Clinical Neuropsychologist, 13, 3047.
  • Gaffan, D., & Parker, A. (2000). Mediodorsal thalamic function in scene memory in rhesus monkeys. Brain, 123, 816827.
  • García-Finãna, M., Cruz-Orive, L. M., MacKay, C. E., Pakkenberg, B., & Roberts, N. (2003). Comparison of MR imaging against physical sectioning to estimate the volume of human cerebral compartments. Neuroimage, 18, 505516.
  • Graff-Radford, N. R., Damasio, H., Yamada, T., Elsinger, P. J., & Damasio, A. R. (1985). Nonhaemorrhagic thalamic infarction. Brain, 108, 485516.
  • Hawkins, K.A., & Tulsky, D.S. (2001). The influence of IQ stratification on WAIS-III/WMS-III FSIQ-General memory index discrepancy base-rates in the standardization sample. Journal of the International Neuropsychological Society, 7, 875880.
  • Jones-Gotman, M., Zatorre, R. J., Olivier, A., Andermann, F., Cendes, F., Staunton, H., McMackin, D., Siegel, A. M., & Weiser, H.-G. (1997). Learning and retention of words and designs following excision from medial or lateral temporal-lobe structures. Neuropsychologia, 35, 963973.
  • Kessels, R. P., de Haan, E. H., Kappelle, L. J., & Postma, A. (2001). Varieties of human spatial memory: A meta-analysis on the effects of hippocampal lesions. Brain Research Reviews, 35, 295303.
  • Kirkpatrick, J. B. (1978). The blood-brain barrier: Its role in contrast studies. Computerized Tomography, 2(3), 189196.
  • Lee, T. M., Yip, J. T., & Jones-Gotman, M. (2002). Memory deficits have resection from left or right anterior temporal lobe in humans: A meta-analytic review. Epilepsia, 44, 339347.
  • Mai, J. K., Assheuer, J., & Paxinos, G. (2004). Atlas of the human brain. London : Elsevier.
  • Markowitsch, H. J. (1982). Thalamic mediodorsal nucleus and memory: A critical evaluation of studies in animals and man. Neuroscience and Biobehavioural Review, 6, 351380.
  • Mennemeier, M., Fennell, E., Valenstein, E., & Heilman, K. M. (1992). Contributions of the left intralaminar and medial thalamic nuclei to memory: Comparisons and report of a case. Archives of Neurology, 49, 10501058.
  • Milner, B. (1974). Hemispheric specialization: Scope and limits. In F. O. Schmitt & F. G. Worden (Eds.), The neurosciences: Third study program (pp. 7589). Cambridge , MA : MIT Press.
  • Montaldi, D., & Mayes, A .R. (2010). The role of recollection and familiarity in the functional differentiation of the medial temporal lobes. Hippocampus, 20, 12911314.
  • Mori, E., Yamadori, A., & Mitani, Y. (1986). Left thalamic infarction and disturbance of verbal memory: A clinicoanatomical study with a new method of computed tomographic stereotaxic lesion localization. Annals of Neurology, 20, 671676.
  • Moscovitch, D. A., & McAndrews, M. P. (2002). Material-specific deficits in “remembering” in patients with unilateral temporal lobe epilepsy and excisions. Neuropsychologia, 40, 13351342.
  • Myers J. E., & Myers, K. R. (1995). Rey Complex Figure Test. Odessa , FL : Psychological Assessment Resources Inc.
  • Nelson, H. E., & Willison, J. (1991). National Adult Reading Test. Windsor , Berks : NFER-Nelson.
  • Portas, C. M., Rees, G., Howseman, A. M., Josephs, O., Turner, R., & Frith, C. D. (1998). A specific role for the thalamus in mediating the interaction of attention and arousal in humans. The Journal of Neuroscience, 18, 89798989.
  • Preuss, T. M., & Goldman-Rakic, P. S. (1987). Crossed corticothlamic and thalamoccortical connections of macaque prefrontal cortex. Journal of Comparative Neurology, 257, 269281.
  • Rousseaux, M. (1994). Amnesias following limited thalamic infarctions. In J. Delacour (Ed.), The memory system of the brain. Adv Series Neurosci (Vol. 4, pp. 241277). Singapore, New Jersey, London : World Scientific Publishers.
  • Russchen, F. T., Amaral, D. G., & Price, J. L. (1987). The afferent input to the magnocellular division of the mediodorsal thalamic nucleus in the monkey, Macaca fascicularis. Journal of Comparative Neurology, 256, 175210.
  • Saling, M. M. (2009). Verbal memory in mesial temporal lobe epilepsy: Beyond material specificity. Brain, 131, 570582.
  • Sass, K. J., Sass, A., Westerveld, M., Lencz, T., Novelly, R. A., Kin, J. H., & Spencer, D. D. (1992). Specificity in the correlation of verbal memory and hippocampal neuron loss: Dissociation of memory, language and verbal intellectual ability. Journal of Clinical and Experimental Neuropsychology, 14, 662672. doi:10.1080/01688639208402854
  • Schott, J. M., Crutch, S. J., Fox, N. C., & Warrington, E. K. (2003). Development of selective verbal memory impairment secondary to a left thalamic infarct: A longitudinal case study. Journal of Neurology, Neurosurgery and Psychiatry, 74, 255257. doi:10.1080/01688639208402854
  • Spinks, R., Magnotta, V. A., Andreasen, N. C., Cretsinger, K., Ziebell, Z., & Nopolous, P. (2002). Manual and automated measurement of the whole thalamus and mediodorsal nucleus using MRI. NeuroImage, 17, 631642.
  • Squire, L. R., Amaral, D. G., Zola-Morgan, S., Kritchevsky, M., & Press. G. (1989). Description of brain injury in the amnesic patient N.A. based on magnetic resonance imaging. Experimental Neurology, 105, 2325.
  • Squire, L. R., & Moore, R. Y. (1979). Dorsal thalamic lesion in a noted case of human memory dysfunction. Annals of Neurology, 6, 503506.
  • Stuss, D. T., Guberman, A., Nelson, R., & Larochelle, S. (1988). The neuropsychology of paramedian thalamic infarction. Brain and Cognition, 8, 348378.
  • Summers, M. J. (2002). Neuropsychological consequences of right thalamic haemorrhage: Case study and review. Brain and Cognition, 50, 129138.
  • Tsivilis, D., Vann, S. D., Denby, C., Roberts, N., Mayes, A. R., Montaldi, D., & Aggleton, J. P. (2008). A disproportionate role for the fornix and mammillary bodies in recall versus recognition memory. Nature Neuroscience, 11, 834842.
  • van Buren, J. M., & Borke, R. C. (1972). Variations and connections of the human thalamus, Variations of the human diencephalons (Vol. 1). Berlin : Springer.
  • 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 following infarction in the thalamus: A study of 22 cases with localized lesions. Neuropsychologia, 41, 13301344.
  • Van der Werf, Y. D., Weerts, J. G., Jolles, J., Witter, M. P., Lindeboom, J., & Scheltens, P. (1999). Neuropsychological correlates of a right unilateral lacunar thalamic infarction. Journal of Neurology, Neurosurgery and Psychiatry, 66, 3642.
  • Van der Werf Y, Witter M, Uylings HBM, & Jolles J., (2000). Neuropsychology of infarctions in the thalamus: A review. Neuropsychologia, 38, 61327.
  • von Cramon, D. Y., Hebel, N., & Schuri, U. (1985). A contribution to the anatomical basis of thalamic amnesia. Brain, 108, 9931008.
  • Wechsler, D. (1997). Wechsler Memory Scale III. The Psychological Corporation, Harcourt Brace & Co., New York .
  • Wechsler, D. (1999). Wechsler Abbreviated Scale of Intelligence. The Psychological Corporation, Harcourt Brace & Co., New York .
  • Weisberg, L. A., & Dunn, D. (1983). Thalamic gliomas: Clinical and computed tomographic correlations. Computerised Radiology, 7, 229235.
  • Wixted, J. T., & Squire, L. R. (2011). The medial temporal lobes and attributes of memory. Trends in Cognitive Science, 15, 210217.
  • Wood, D-M. G., & Bigler, E. D. (1995). Diencephalic changes in traumatic brain injury: Relationship to sensory perceptual function. Brain Research Bulletin, 38, 545549.
  • Yonelinas, A. P., Aly, M., Wang, W.-C., & Koen, J. D. (2010). Recollection and familiarity: Examining controversial assumptions and new directions. Hippocampus, 20, 11781194.
  • Zoppelt, D., Koch, B., Schwarz, M., & Daum, I. (2003). Involvement of the mediodorsal thalamic nucleus in mediating recollection and familiarity. Neuropsychologia, 41, 11601170.