Abstract

Alzheimer's disease (AD) is regarded a human-specific condition, and it has been suggested that brain regions highly expanded in humans compared with other primates are selectively targeted. We calculated shared and unique variance in the distribution of AD atrophy accounted for by cortical expansion between macaque and human, affiliation to the default mode network (DMN), ontogenetic development and normal aging. Cortical expansion was moderately related to atrophy, but a critical discrepancy was seen in the medial temporo-parietal episodic memory network. Identification of “hotspots” and “coldspots” of expansion across several primate species did not yield compelling evidence for the hypothesis that highly expanded regions are specifically targeted. Controlling for distribution of atrophy in aging substantially attenuated the expansion–AD relationship. A path model showed that all variables explained unique variance in AD atrophy but were generally mediated through aging. This supports a systems-vulnerability model, where critical networks are subject to various negative impacts, aging in particular, rather than being selectively targeted in AD. An alternative approach is suggested, focused on the interplay of the phylogenetically old and preserved medial temporal lobe areas with more highly expanded association cortices governed by different principles of plasticity and stability.

Introduction

Alzheimer's disease (AD) is associated with heterogeneous brain atrophy, with accelerated atrophy in the temporal lobe in initial phases, spreading to medial parietal and then most of the cortex in later stages (McDonald et al. 2009). Fundamental to the understanding of disease progression is knowledge about the basic factors governing the distribution of atrophy. Several complementary theories or principles have been proposed. First, AD exists almost exclusively in humans, being very uncommon in other primates (Finch and Austad 2012; Bufill et al. 2013). On this basis, it has been suggested that human-specific adaptations during evolution of the brain could be the cause of age-related neurodegenerative diseases, and specifically AD (Neill 1995, 2012; Rapoport and Nelson 2011; Buckner 2012; Bufill et al. 2013). Especially, high-expanding regions of the human cortex have been hypothesized to be preferentially targeted (Rapoport and Nelson 2011). However, the extent to which prime target areas of AD, including medial temporal lobes (MTLs), are relatively more expanded in the human cortex compared with the cortices of other primates is debatable (Van Essen and Dierker 2007). Critically, the degree of correspondence between cortical expansion from monkeys to humans and distribution of atrophy in AD has not been formally quantified.

Further, it has previously been suggested that what we now conceptualize as the default mode network (DMN) is supported by cortical areas that are particularly enlarged in humans, relative to other primates (Andreasen et al. 1995). Intriguingly, several have pointed to an overlap between DMN and distribution of amyloid plaques (Sperling et al. 2009) and speculated that this may be causally related to overall high levels of DMN activity (Jagust and Mormino 2011; Buckner 2012), thereby creating a link between evolutionary change, DMN activity and AD. Parts of the DMN, especially the medial parietal area, are characterized by hypometabolism (Gusnard et al. 2001), disturbed task-induced deactivations (Lustig et al. 2003), and reduced functional connectivity (Greicius et al. 2004) in AD. Hence, DMN function yields a complementary account to the selective cortical expansion hypothesis of distribution of atrophy in AD.

A third view is based on ontogenetic development, according to which brain regions with the most protracted development tend to be more vulnerable to degeneration in AD (Rapoport and Nelson 2011; Buckner 2012; Bufill et al. 2013) as well as in aging (Tamnes et al. 2013). For instance, according to this view, the posterior-to-anterior gradient of cortical maturation will be expected to be reversed in aging-related decline. This is often referred to as the theory of retrogenesis, in which degenerative mechanisms are thought to reverse the order of acquisition in normal development (Reisberg et al. 2002; Ewers et al. 2011). Several causes for this phenomenon have been proposed, including similarities in the sequence of myelin acquisition in development and the pattern of myelin loss, cell loss and neurometabolic change in AD (Reisberg et al. 2002). Interestingly, similarities between cortical development and the degree of cortical expansion in monkeys versus humans have also been demonstrated, with high-expanding regions with more complex cellular architecture showing more complex and protracted developmental trajectories (Rosa and Tweedale 2005; Shaw et al. 2008; Hill et al. 2010). Accordingly, regions that are selectively larger in humans than in other primates and that show relatively late maturation may be more vulnerable to AD pathology.

An alternative view to a disease-specific model where high-expanding regions in humans are preferentially targeted in AD is a systems-vulnerability model. According to such a view, brain systems affected in AD may also be vulnerable to a range of insults and conditions, including normal aging (Jagust 2013). Thus, a link between differential cortical expansion and AD can be interpreted in terms of general vulnerability of brain networks, where AD may represent one extreme case of negative impact. If so, one would expect that high-expanding brain regions are as preferentially targeted in normal aging in the absence of neurodegenerative disease and that the relationship between expansion and AD to a substantial degree would be mediated by brain changes in normal aging. The same line of reasoning applies to the relationship between the DMN system breakdown, expansion, and AD versus aging.

Lack of formal quantification and comparison of the explanatory power of each of these theories is hampering progress in understanding distribution of atrophy in AD. On this background, we had 3 major aims: (1) to contrast rate of atrophy in AD between “hotspots” and “coldspots” of expansion of the cerebral cortex across primates of different sizes. We hypothesize that rate of atrophy in AD will not strictly follow degree of expansion, with especially the MTL with its known vulnerability to AD as a candidate for deviation. (2) To test to what extent the possible accelerated atrophy in AD in high-expanding regions can be explained by atrophy that also occurs in normal aging. This contrasts the principle of high-expanding areas being selectively targeted in AD versus being more generally sensitive to negative impacts, in accordance with a systems-vulnerability view. (3) To test whether and how the affiliation of a cortical region to the DMN, degree of cortical expansion, aging, and development are interrelated or independent in accounting for regional distribution of AD atrophy. These questions directly address the degree of specificity in the vulnerability of critical brain systems in AD.

In the present work, we take advantage of well-validated cortical expansion maps between macaque and humans (Bardet et al. 2007; Hill et al. 2010), across several primate species (Chaplin et al. 2013) and from children to adults (Fjell, Westlye et al. 2015), as well as longitudinal atrophy maps from healthy older adults and AD patients from the Alzheimer's Disease Neuroimaging Initiative (ADNI) (Fjell, McEvoy et al. 2013), and of intrinsic connectivity maps based on 1000 participants (Yeo et al. 2011). These different data sets were re-analyzed and combined in a joint model. Comparing cortical expansion across primate species with grossly different brain sizes is most appropriately referred to as allometric scaling. Here, cortical expansion from other primates to humans is used as a proxy for cortical areas more or less unique to the human brain.

Materials and Methods

An overview of the different analysis steps taken is presented in Table 1.

Table 1

Overview of the main analysis and reconstruction steps

Input Processing Output 
Human atrophy maps 
1-year longitudinal MRI in 122 AD patients Quarc, surface-based analysis (Holland and Dale 2011), z-transformation AD atrophy maps (z-scores) 
1-year longitudinal MRI in 132 healthy elderly Quarc, surface-based analysis, z-transformation Aging atrophy maps (z-scores) 
1-year longitudinal MRI in 37 healthy elderly with very low AD risk Quarc, surface-based analysis, z-transformation Low-risk aging atrophy maps (z-scores) 
Cortical expansion maps between primates of different size 
MRI of 1 macaque and 12 young humans Surface-based registrations, based on inter-species homologies, by CARET (Orban et al. 2004; Van Essen and Dierker 2007), z-transformation Macaque-versus-human expansion maps (z-scores) 
MRI of 1 marmoset, 1 capuchin, and 1 macaque Surface-based registrations, based on inter-species homologies by CARET (Chaplin et al. 2013; Van Essen et al. 2012), z-transformation Marmoset-versus-capuchin expansion map, marmoset-versus-macaque expansion map (z-scores) 
Normalized (0–1) expansion maps of marmoset versus capuchin, marmoset versus macaque, macaque versus human Average of normalized expansion maps across all species (Chaplin et al. 2013), z-transformed Average expansion map across primates (z-scores) 
Average expansion map (z-scores) Thresholded at 1 < Z or Z < −1 “Hotspots” and “Coldspots” of expansion 
Human developmental expansion 
MRI of 331 healthy children, 4–20 years FreeSurfer reconstructions and area calculations, surface-based smoothing spline (Fjell et al. 2010), z-transformation Estimated mean cortical expansion (z-scores) 
DMN 
Resting-state BOLD scans from 1000 healthy participants Surface-based alignment, clustering approach (Yeo et al. 2011), thresholded at 0.15 confidence Thresholded DMN confidence maps 
Input Processing Output 
Human atrophy maps 
1-year longitudinal MRI in 122 AD patients Quarc, surface-based analysis (Holland and Dale 2011), z-transformation AD atrophy maps (z-scores) 
1-year longitudinal MRI in 132 healthy elderly Quarc, surface-based analysis, z-transformation Aging atrophy maps (z-scores) 
1-year longitudinal MRI in 37 healthy elderly with very low AD risk Quarc, surface-based analysis, z-transformation Low-risk aging atrophy maps (z-scores) 
Cortical expansion maps between primates of different size 
MRI of 1 macaque and 12 young humans Surface-based registrations, based on inter-species homologies, by CARET (Orban et al. 2004; Van Essen and Dierker 2007), z-transformation Macaque-versus-human expansion maps (z-scores) 
MRI of 1 marmoset, 1 capuchin, and 1 macaque Surface-based registrations, based on inter-species homologies by CARET (Chaplin et al. 2013; Van Essen et al. 2012), z-transformation Marmoset-versus-capuchin expansion map, marmoset-versus-macaque expansion map (z-scores) 
Normalized (0–1) expansion maps of marmoset versus capuchin, marmoset versus macaque, macaque versus human Average of normalized expansion maps across all species (Chaplin et al. 2013), z-transformed Average expansion map across primates (z-scores) 
Average expansion map (z-scores) Thresholded at 1 < Z or Z < −1 “Hotspots” and “Coldspots” of expansion 
Human developmental expansion 
MRI of 331 healthy children, 4–20 years FreeSurfer reconstructions and area calculations, surface-based smoothing spline (Fjell et al. 2010), z-transformation Estimated mean cortical expansion (z-scores) 
DMN 
Resting-state BOLD scans from 1000 healthy participants Surface-based alignment, clustering approach (Yeo et al. 2011), thresholded at 0.15 confidence Thresholded DMN confidence maps 

Atrophy and DMN Maps

Samples were drawn from ADNI (www.adni-info.org) and are identical to those in Fjell, McEvoy et al. (2013). Participants were 55–91 years of age at baseline. Healthy controls scored 24–30 on Mini-Mental State Examination (MMSE) (Folstein et al. 1975), Clinical Dementia Rating (Morris 1993) of 0, and were non-depressed. A total of 132 healthy older adults with no conversion to mild cognitive impairment over at least 3 year were available (60–90 years, mean 75.4, SD = 5.1, 63 females/69 males), as were 122 AD patients with no change of diagnostic status over the 3-year interval (55–89 years, mean 74.4, SD = 7.6, 57 females/65 males). From this group, a subsample (n = 37) of very low AD risk was created based on normal levels of CSF Aβ1–42 (>192 pg/mL, corresponding to an established criterion from the ADNI, see Shaw et al. 2009) and homozygoticy for apolipoprotein (APOE) ε3. This biomarker-defined very low-risk subsample was used in an additional test of common effect of aging and AD in the final path analysis (mentioned later). All participants had 1-year longitudinal MRI passing internal quality control. 1.5-T scanners were used (see http://adni.loni.ucla.edu/research/protocols/mri-protocols/), and percentage volumetric change over 1 year was calculated for each point on the cortical surface with no anatomical constraints using Quarc (Holland and Dale 2011; Holland et al. 2012). DMN connectivity maps were drawn from FreeSurfer version 5.3 (http://surfer.nmr.mgh.harvard.edu/), based on resting-state data from 1000 subjects (Yeo et al. 2011) and thresholded by a vertex-wise confidence of 0.15.

Expansion Maps

Cortical expansion maps between the macaque monkey and 12 young adult humans (Van Essen and Dierker 2007; Hill et al. 2010), and between marmoset and capuchin and marmoset and macaque (Chaplin et al. 2013), were used for the calculations. Maps were originally created by surface-based registration methods, computed based on a combination of functional and structural homologies, described in detail elsewhere (Orban et al. 2004; Van Essen and Dierker 2007; Hill et al. 2010; Van Essen et al. 2012; Chaplin et al. 2013). Only the right hemispheres were available. Computation of expansion maps for human development is described in Fjell, Westlye et al. (2015), based on a sample of 331 healthy children 4–20 years of age by use of a surface-based smoothing spline approach (Fjell et al. 2010) applied to FreeSurfer-generated cortical surface maps (Dale et al. 1999; Fischl, Sereno and Dale 1999; Fischl, Sereno and Tootell et al. 1999).

Statistics

All maps were z-transformed to remove scaling differences (mean = 0, SD = 1). The maps were correlated vertex-by-vertex by Pearson correlations to yield a global measure of anatomical overlap (with the medial wall masked out before analyses). Significance was decided by permutation testing (10 000 permutations), and P-values were Bonferroni corrected for multiple comparisons. Partial correlations were run to control for the effect of aging. Regions of correlation versus anti-correlation between AD atrophy and cortical expansion were identified by thresholding the z-transformed atrophy and macaque-to-human expansion maps. Regions of correlation were characterized by Z > 0.5 or Z < −0.5 for both cortical expansion and AD, and areas of anti-correlation by combinations of Z > 0.5 with Z < −0.5. A similar procedure was used for DMN versus AD atrophy. “Hotspots” and “coldspots” of differential cortical expansion were identified by thresholding average expansion across all included species at z > 1 (hotspots) or z < −1 (coldspots). These regions were then used to extract rate of atrophy for all AD patients, tested against mean cortical atrophy by one-sample t-tests. Next, atrophy in AD and aging, as well as cortical expansion, were correlated with the confidence value of each vertex of belonging to the DMN network, thresholded at 0.15. To show the distribution of variance across the different variables, a path model was constructed and the significance of each path was decided by permutation testing. This model was also run with the biomarker-defined low-risk subsample of healthy elderly.

Results

Cortical Expansion

The degree to which the surface model of the macaque cortex had to be expanded to reach the size of the surface model of human cortex, that is, cortical expansion, as well as annual percentage volume change in AD and normal aging, was calculated for each points on the brain surface and related by Pearson correlations (Fig. 1). All reported correlations were significant at P < 0.05 (two-tailed, corrected for multiple comparisons) as evidenced by permutation testing. Expansion correlated with regional distribution of atrophy in AD (r = 0.35) and normal aging (r = 0.34). Scatterplot (Fig. 2) of the relationship between expansion and atrophy in 34 cortical regions revealed great variability, with some showing high expansion and atrophy, for example, lateral temporal regions, whereas others deviated from this trend, for example, MTL regions. Atrophy in AD and aging showed a stronger correlation (r = 0.74). The AD-expansion correlation dropped significantly to partial r = 0.15 when atrophy in aging was controlled for.

Figure 1.

Regional distribution of atrophy and cortical expansion. (A) Mean annual atrophy in terms of volume reduction in AD, z-transformed to remove scaling effects. (B) z-transformed cortical expansion between macaque and human. (C) z-transformed maps of atrophy in terms of volume reductions in normal aging. For all panels, red–yellow indicates higher than mean annual atrophy/cortical expansion, whereas blue–cyan indicates lower than mean atrophy/cortical expansion. Scales go from 2 (red–yellow) to −2 (blue–cyan). The brains are semi-inflated to allow better visualization of effects within sulci.

Figure 1.

Regional distribution of atrophy and cortical expansion. (A) Mean annual atrophy in terms of volume reduction in AD, z-transformed to remove scaling effects. (B) z-transformed cortical expansion between macaque and human. (C) z-transformed maps of atrophy in terms of volume reductions in normal aging. For all panels, red–yellow indicates higher than mean annual atrophy/cortical expansion, whereas blue–cyan indicates lower than mean atrophy/cortical expansion. Scales go from 2 (red–yellow) to −2 (blue–cyan). The brains are semi-inflated to allow better visualization of effects within sulci.

Figure 2.

AD atrophy, cortical expansion and DMN. (A) Mean AD atrophy and expansion (z-scores) were quantified within 34 cortical regions of interest to illustrate variability across regions. The x-axis denotes degree of cortical expansion from relatively lower (blue–cyan axis color) to higher (red–yellow axis color). The transition point between blue and red axis color denotes mean expansion across the surface. The y-axis denotes percentage annual cortical volume decline in the AD patients. The color of the dots indicates from which gross region of the brain they are taken. Although there is a correlation between how much cortical expansion is seen within a region and the amount of AD atrophy, there are also regions not following this tendency, for example, the medial temporal regions characterized by relatively low degree of expansion but high levels of atrophy in AD. (B) The z-transformed cortical expansion and AD atrophy maps from Fig. 1 were thresholded at 0.5 < Z < −0.5. Areas of correlation were characterized by Z > 0.5 (green) or Z < −0.5 (yellow) for both cortical expansion and AD, whereas areas of anti-correlation (pink) were characterized by combinations of Z > 0.5 with Z < −0.5. The medial surface is slightly tilted to allow inspection of inferior temporal and fusiform cortex. (C) A similar procedure was used to show areas of high atrophy within DMN (green), low atrophy outside DMN (yellow), and high atrophy outside DMN (pink).

Figure 2.

AD atrophy, cortical expansion and DMN. (A) Mean AD atrophy and expansion (z-scores) were quantified within 34 cortical regions of interest to illustrate variability across regions. The x-axis denotes degree of cortical expansion from relatively lower (blue–cyan axis color) to higher (red–yellow axis color). The transition point between blue and red axis color denotes mean expansion across the surface. The y-axis denotes percentage annual cortical volume decline in the AD patients. The color of the dots indicates from which gross region of the brain they are taken. Although there is a correlation between how much cortical expansion is seen within a region and the amount of AD atrophy, there are also regions not following this tendency, for example, the medial temporal regions characterized by relatively low degree of expansion but high levels of atrophy in AD. (B) The z-transformed cortical expansion and AD atrophy maps from Fig. 1 were thresholded at 0.5 < Z < −0.5. Areas of correlation were characterized by Z > 0.5 (green) or Z < −0.5 (yellow) for both cortical expansion and AD, whereas areas of anti-correlation (pink) were characterized by combinations of Z > 0.5 with Z < −0.5. The medial surface is slightly tilted to allow inspection of inferior temporal and fusiform cortex. (C) A similar procedure was used to show areas of high atrophy within DMN (green), low atrophy outside DMN (yellow), and high atrophy outside DMN (pink).

Color-coded conjunction maps of correlations and anti-correlations between AD and differential cortical expansion were computed by thresholding the z-transformed maps as described earlier (Fig. 2). High expansion and high rate of atrophy could be seen in the entire lateral temporal cortex, extending into the temporo-parietal junction (TPJ) and inferior parietal cortices. Low expansion and low rate of atrophy in AD characterized the central sulcus, cuneus, calcarine sulcus, and adjacent extrastriate cortex and the lingual gyrus. Anti-correlations were seen in the MTL, including entorhinal, parahippocampal, and fusiform cortices, as well as posterior cingulate/retrosplenial cortices extending towards the precuneus and supplementary motor cortex, which have marked decline both in AD and in normal aging despite relatively low cortical expansion in the comparison between macaque and human.

Next, expansion maps across several Simian primate species (see Chaplin et al. 2013) were compared, showing a conserved pattern of differential cortical expansion (Fig. 3). MTL showed the same relative preservation across species that was seen between macaque and human. “Hotspots” and “coldspots” of expansion were identified by thresholding the average expansion map across all species at z > 1 (hotspots) or z < −1 (coldspots) (Fig. 4). Two previously identified major hotspots were situated in the TPJ extending somewhat anteriorly into the lateral temporal cortex (H1 in Fig. 4), and in the lateral middle and inferior prefrontal cortex (H2) (Chaplin et al. 2013). Two adjacent coldspots were identified in the MTL extending inferiorly to the fusiform gyrus (C1) and in the medial occipital lobe (C2). In AD, H1 (2.41%) and C1 (2.22%) showed more decline than the mean cortical atrophy of 1.62% (H1 t = 4.60; C1 t = 5.13, df = 121, P < 0.0001), as evidenced by a one-sample t-test. C2 (0.57%) showed less decline than the mean (t = 12.96, P < 0.0001) whereas H2 (1.57%) did not differ from mean atrophy (t = 0.34, n.s.).

Figure 3.

A conserved pattern of differential expansion across primate species. Cortical expansion (z-scores) from comparisons between cortical area of smaller versus larger brained Simian primates. From left to right are shown marmoset versus capuchin, marmoset versus macaque, and macaque versus humans, as well as the mean expansion across all species. General patterns of high and relatively low expansion are well preserved across species. The medial temporal cortex (lower panel) showed high degree of preservation across all comparisons (thresholded by z < −1).

Figure 3.

A conserved pattern of differential expansion across primate species. Cortical expansion (z-scores) from comparisons between cortical area of smaller versus larger brained Simian primates. From left to right are shown marmoset versus capuchin, marmoset versus macaque, and macaque versus humans, as well as the mean expansion across all species. General patterns of high and relatively low expansion are well preserved across species. The medial temporal cortex (lower panel) showed high degree of preservation across all comparisons (thresholded by z < −1).

Figure 4.

Hotspots and coldspots of cortical expansion. Upper panel: the average expansion maps across marmoset, capuchin, macaque, and human were thresholded at z > 1 or z < −1 to reveal regions of consistently high (“hotspots”—green-yellow) and consistently low (“coldspots”—blue-pink) cortical expansion through across primates. Lower panel: mean atrophy in AD was quantified within the 2 hotspot regions (H1 and H2) and the 2 coldspot regions (C1 and C2). H1 and C1 showed elevated rate of atrophy, whereas C2 showed reduced decline.

Figure 4.

Hotspots and coldspots of cortical expansion. Upper panel: the average expansion maps across marmoset, capuchin, macaque, and human were thresholded at z > 1 or z < −1 to reveal regions of consistently high (“hotspots”—green-yellow) and consistently low (“coldspots”—blue-pink) cortical expansion through across primates. Lower panel: mean atrophy in AD was quantified within the 2 hotspot regions (H1 and H2) and the 2 coldspot regions (C1 and C2). H1 and C1 showed elevated rate of atrophy, whereas C2 showed reduced decline.

Default Mode Network Vulnerability

AD correlated 0.27, aging 0.19, and cortical expansion 0.15 with the DMN network (Yeo et al. 2011) (P < 0.05, corrected) (Fig. 5). More detailed inspections revealed overlap between DMN and AD in parts of lateral temporal cortex and around TPJ and in the medial parietal (precuneus) and posterior cingulate/retrosplenial cortex (Fig. 2), whereas high atrophy regions outside DMN was seen in the rest of the lateral and medial/inferior temporal cortex. MTL had lower probabilities of being part of the DMN according to this parcellation scheme, although by use of a lower confidence threshold for inclusion, some vertices in the parahippocampal gyrus, mainly in the left hemisphere, would have been included in the DMN (Yeo et al. 2011). Interestingly, within DMN, cortical expansion and AD correlated in the lateral temporal cortex and TPJ but were anti-correlated in posterior cingulate/retrosplenial cortex.

Figure 5.

Default mode network (DMN) and cortical expansion. Upper panel: delineation of the DMN based on Yeo et al. (2011). Colors denote the probability for each vertex being a part of the DMN. The scale goes from 0.15 (red) to 1 (yellow). Lower panel: atrophy and expansion from the maps in Fig. 1 were extracted for vertices inside versus outside DMN and plotted in terms of z-scores. Atrophy and expansion are larger for DMN vertices than vertices outside DMN. All contrasts were significant (P < 0.05, corrected).

Figure 5.

Default mode network (DMN) and cortical expansion. Upper panel: delineation of the DMN based on Yeo et al. (2011). Colors denote the probability for each vertex being a part of the DMN. The scale goes from 0.15 (red) to 1 (yellow). Lower panel: atrophy and expansion from the maps in Fig. 1 were extracted for vertices inside versus outside DMN and plotted in terms of z-scores. Atrophy and expansion are larger for DMN vertices than vertices outside DMN. All contrasts were significant (P < 0.05, corrected).

Path Modeling

Using all variables as predictors and AD atrophy as dependent in a multiple regression yielded R2 = 0.57. A path model (SPSS Amos version 21) was constructed to disentangle variance across all variables (Fig. 6). Cortical expansion between macaque and human was the single exogenous variable; AD was the endogenous variable, with DMN affiliation, development, and aging as mediating variables. Paths were drawn from cortical expansion to all variables and from all variables to AD. Additional paths were drawn from DMN to aging and development and from development to aging. All paths were significant (P < 0.05, corrected). Aging was related to AD by partial β = 0.67, compared with 0.12 for DMN and 0.10 for cortical expansion. Expansion was further related to aging by β = 0.30 and to DMN by β = 0.15. DMN was related to aging by β = 0.12. The indirect contribution from expansion to AD through aging was 0.20. DMN had an indirect contribution through aging of 0.08. Developmental expansion was related to cortical expansion (β = 0.11), DMN (β = 0.13), aging (β = 0.18) and to a smaller degree directly to AD atrophy (β = 0.04). The indirect contribution from development to AD through aging was 0.12.

Figure 6.

Path model of shared and unique contributions to distribution of AD pathology. A path diagram was constructed to show how the variance was distributed among variables. Arrows show relationships between any 2 variables in terms of independent (start of line) versus dependent (end of arrow); colors and thickness of lines correspond to standardized partial path weights.

Figure 6.

Path model of shared and unique contributions to distribution of AD pathology. A path diagram was constructed to show how the variance was distributed among variables. Arrows show relationships between any 2 variables in terms of independent (start of line) versus dependent (end of arrow); colors and thickness of lines correspond to standardized partial path weights.

As a final test, the estimations were re-run with the biomarker-defined low-risk group of elderly included instead of the full sample of elderly. The general results upheld, but with a moderate reduction in the relationship between cortical expansion and aging (from β = 0.30 to β = 0.19) and aging and AD (from β = 0.67 to β = 0.56), and an increase in the relationship between cortical expansion and AD (β = 0.10 to β = 0.19).

Discussion

Atrophy as a Function of Cortical Expansion

Different rates of cortical expansion between the macaque and human brains explained 12.2% of the distribution of atrophy in AD. One hotspot of expansion, TPJ and parts of lateral temporal cortex, showed highly elevated rates of atrophy in AD. Conversely, medial parts of the occipital cortex showed low expansion and resistance to atrophy. However, the results also revealed deeply problematic aspects with the view that high-expanding regions are specifically targeted in AD. MTL is the earliest and most heavily affected region in AD (Price and Morris 1999; McDonald et al. 2009) but shows relatively low rates of expansion between macaque and human (Bardet et al. 2007; Hill et al. 2010) and across several simian species (Chaplin et al. 2013). This “coldspot” of expansion showed accelerated atrophy, with the MTL and medial parietal memory network showing marked anti-correlation between degree of expansion versus AD. Rodent models are frequently used to study learning, memory, and spatial navigation (Colgin et al. 2008), because the MTL system is evolutionary preserved, with similarities in organization and function across many species. It has also been shown that a part of the retrosplenial cortex in primates, the prostriate cortex, has evolutionary old histological characteristics, with some very primitive visual response properties (Yu et al. 2012).

These low-expanding and AD-vulnerable regions overlapped well with the Papez’ circuit, which is important for normal episodic memory function. The fact that such cognitive functions are reduced due to degeneration of low-expanding brain regions and networks prompts the question of to what degree they really represent human-specific adaptations. Since storage and transmission of knowledge in preliterate societies should depend highly on individual brain capacity, evolution of a human-specific memory capacity, with corresponding expansion of relevant cortical regions, could be plausible (Bufill and Carbonell 2004). In accordance with such an account, there are aspects of human memory vulnerable to AD (Addis et al. 2009) that may not be shared with other primate species, such as episodic simulation and long-term foresight–mental time travels (Tulving 1983, but see Schwartz and Evans 2001). A prime candidate for supporting such functions is the DMN, likely involved in aspects of complex cognition such as episodic simulations (Addis et al. 2009), self-projection (Buckner and Carroll 2007) and mental time travels (Ostby et al. 2012). These are functions that may possess qualities that could be human-specific (Tulving 1983; Suddendorf 2013, but see Schwartz and Evans 2001 and Corballis 2013). However, DMN-like activity develops quickly during the first year of infancy (Gao et al. 2013), likely prior to establishment of more stable, extensive, and prolonged episodic memory and mental time travel abilities, and has been observed in a number of species, including non-human primates (Rilling et al. 2007; Barks et al. 2015; Belcher et al. 2013), rats (Lu et al. 2012), and mice (Sforazzini et al. 2014). A meta-analysis of fMRI data in monkeys showed consistent task-related deactivation within a network including medial prefrontal and medial and lateral parietal cortex, thus very similar to the human DMN (Mantini et al. 2011), indicating that DMN may not be exclusive to humans. There is even evidence that DMN in chimpanzees, like in humans, plays a role in social cognition (Barks et al. 2015). A recent study of humans and monkeys identified 2 human-specific resting-state networks, which did not include DMN (Mantini et al. 2013). Although it is very difficult to make inferences from brain activity to cognitive function, especially across species and particularly during uncontrolled task-free conditions, comparative data do not support the view that the DMN reflects the brain's most human and complex parts. This also means that the function of the DMN not necessarily represents human-specific cognition. Of course we do not know whether the DMN has the same function across species, and there could be microstructural differences that lead to differences in DMN function.

Regardless of whether certain aspects of episodic memory differ between humans and other primates (Corballis 2013; Suddendorf 2013), it is not evident that these aspects are the most vulnerable early in AD. Spatial navigation is an early AD symptom related to episodic memory (Bellassen et al. 2012; Lithfous et al. 2013), also dependent on the MTL system, which can hardly be said to be a human-specific cognitive adaptation. The basis of spatial navigation is rooted in grid cells in the entorhinal cortex (Fyhn et al. 2004), which is the first region to harbor increased amounts of neurofibrillary tangles and other AD pathology (Braak and Braak 1991; Van Hoesen et al. 1991), although likely not β-amyloid in the initial phases (mentioned later). Together with the occipital cortex, MTL was the major “coldspot” region in the expansion of the cerebral cortex, with a high degree of relative preservation across several primate species. Thus, even though we cannot exclude the possibility that certain human-specific cognitive functions may be impaired in AD, there seems to be no clear connection between the early cognitive symptoms and brain pathology on the one hand, and cortical expansion across primate species and human-specific higher-order cognitive functions on the other.

Specificity of Atrophy: Aging versus AD

Despite discrepancy in the critical medial DMN and MTL regions, there were also some overlaps between cortical expansion and AD atrophy. A further aim of this study was to test whether high-expanding regions with accelerated atrophy were preferentially targeted in AD compared with aging. The expansion–atrophy relationship was similar in magnitude in aging and AD. Partialling out the effect of aging caused the relationship between cortical expansion and AD to drop from 12.2% to 2.25% shared variance. This suggests that rather than being specifically targeted by AD, some high-expanding areas may represent regions of high vulnerability to different conditions, aging in particular, in line with a systems-vulnerability view (Jagust 2013). The hotspot region in the TPJ/lateral temporal cortex showed accelerated decline in AD but was heavily affected by aging as well. To the extent that high-expanding regions are specifically targeted in AD, we would not expect to see the same pattern of decline in normal aging. The same general pattern was upheld when only the healthy elderly negative for amyloid and homozygote for APOE ε3 were included, although the relationships naturally were somewhat attenuated.

A common denominator for low-expanding regions that are vulnerable to AD and normal aging could be high levels of unique types of plasticity (Cotman et al. 1993; Neill 1995, 2012; Mesulam 1999). MTL and DMN play special roles in memory, with high demands for life-long plasticity (Aimone et al. 2010; Lazarov et al. 2010). MTL also harbors the only area of adult human neurogenesis besides the olfactory bulb and the dentate gyrus (Eriksson et al. 1998; Lotsch et al. 2013). This state of affairs is markedly different in a number of other species, including other primates, where new neurons may be added throughout life in areas of the neocortex (Gould et al. 2001; Bernier et al. 2002, but see Rakic 2002). Evolution must progress with a delicate balance of plasticity and stability and appears to have favored stability over plasticity in humans (Rakic 2002, 2004; Bhardwaj et al. 2006). Hence, although the human MTL is not very different from other primate species in terms of anatomy, function (Clark and Squire 2013) or neurogenesis (Rakic 2002), what is special about the human MTL may rather be its role as an evolutionary old and plastic area within an otherwise much more developed and expanded neocortex.

Altered neurogenesis in the hippocampus has been proposed as an early critical event in AD (Mu and Gage 2011), and lesions in the limbic system could have consequences for association cortices function due to reciprocally interconnections (Mesulam 1999). Of note, many of the structures on the surface of the basal forebrain and within MTL that comprise olfactory cortex are also sites of initial pathology (Price et al. 1991; Wang et al. 2010), and along with episodic memory and navigation (Drago et al. 2011), olfactory function (Stamps et al. 2013) is among the first to decline in AD. The downside of being lone areas involved in, or surrounding or tightly interconnected to areas being involved in, this type of plasticity in the human brain may be increased vulnerability, with accumulation of negative impacts through life (mentioned later). Different mechanisms related to neuroplasticity are then manifested in other cortical regions. For instance, dendritic spines may represent a primary site of structural plasticity throughout the cortex, and spine density and plasticity is reduced in aging (Jacobs et al. 1997; Esiri 2007; Freeman et al. 2008; Bloss et al. 2011; Benavides-Piccione et al. 2013). Importantly, while tangle accumulation starts in MTL, spreading to isocortical association areas in later AD stages (Braak and Braak 1991), medial parietal and prefrontal DMN regions are characterized by increased concentrations of β-amyloid in initial phases (Sperling et al. 2009). Thus, the 2 hallmark histopathological criteria for AD originate in regions characterized by different mechanisms of plasticity, that are relatively preserved during evolution while showing high rates of atrophy.

The cost of maintained plasticity may be increased vulnerability to factors that can trigger cognitive decline (Bufill et al. 2013). Perturbation of neuroplasticity has been proposed as a fundamental principle that could potentially account for the clinical and neuropathological features of AD and that amyloid depositions and neurofibrillary tangles are manifestations of the same underlying phenomenon (Mesulam 1999). The potential for neuroplasticity is higher in the limbic system than in other parts of the cerebral cortex, which increases its vulnerability to neurofibrillary degeneration. A consequence of this view may be that late-onset AD is the manifestation of a failure to keep up with the “increasingly more burdensome work of plasticity” (Mesulam 1999).

Interestingly, this calls for a life-span perspective on aging and AD, in that the same regions are sensitive even to prenatal factors. For instance, recent research demonstrated effects of normal variation in neonatal characteristics such as birth weight on MTL (Walhovd et al. 2012) and DMN structures (Raznahan et al. 2012). Thus, rather than regarding such networks as specifically targeted in AD, a life-span systems-vulnerability view may explain dysfunction of these systems across a range of insults and conditions. Effects of APOE4, the major risk allele for sporadic AD (Corder et al. 1993), have been found on gray matter volume in MTL in neonates (Knickmeyer et al. 2014) and in the entorhinal cortex and the hippocampus in children and adolescents (Shaw et al. 2007). The likely existence of life-long gene × environment interactive negative influences indicates that antagonistic pleiotropy is not the only explanation for brain degeneration in aging-related dementias such as AD. The possibility of extreme sensitivity to initial conditions, and developmental contributions to emergence of pathology due to allostatic load (Lenroot and Giedd 2011), may fit better with a systems-vulnerability account for aging-related degeneration.

Common and Shared Effects

The final aim of the study was to test the distribution of variance between DMN, degree of differential cortical expansion between species, development, aging, and AD. Atrophy was higher within the DMN, and DMN showed on average more expansion than non-DMN regions. Interestingly, DMN and cortical expansion were mainly independent predictors of AD. Although there was a small indirect contribution to AD atrophy from cortical expansion through DMN, the proposed link between DMN and expansion in explaining AD-vulnerability was not strong. DMN and cortical expansion also contributed to explain AD atrophy indirectly through the influence of age. The indirect contribution from expansion to AD through aging was 0.20, compared with the direct contribution of 0.10. This suggests that expansion and DMN affiliation explain the anatomical distribution of atrophy in AD to a substantial extent through their influence on atrophy in aging. The same was true for developmental expansion, which as expected was related to cortical expansion between macaque and human (Hill et al. 2010; Fjell, Westlye et al. 2015), with a direct relationship to AD of 0.04 and an indirect relationship through aging of .12.

Limitations and Conclusion

The human brain is not a direct result of evolutionary adaptation of living primates, as humans and other primates have extinct common ancestors. As pointed out by Rakic (2009), comparative anatomical and DNA sequencing data indicate that macaques and humans belong to different branches on the phylogenetic tree (Rakic 2009), and as such, one cannot draw inferences about evolution directly from cortical differences between them. Thus, inter-species comparisons can only be used to test hypotheses about aspects of the brain that are more or less human-specific, or at least, more or less different between humans and other living primate species. This way, inter-species comparisons can be very useful in informing us about the anatomical correspondence between AD atrophy and degree of “human uniqueness” in regional cortical size. Further, during hominid evolution, cortical adaptations in addition to expansion have certainly occurred (Geschwind and Rakic 2013), for example, related to gene expression (Bufill et al. 2013), microstructural changes (Chen et al. 2013), and network reorganization (Buckner and Krienen 2013). Given that AD appears to be a unique human disease, we would expect overlap between disease mechanisms and unique aspects of the human brain at some levels. The present results indicate that a simple expansion-vulnerability model of cortical atrophy may not be warranted, instead favoring an approach focusing on why regions well preserved across several primates seem to be critical in the chain of pathology. As speculated, it may be worth closing in on the interplay between phylogenetically old and preserved mechanisms of plasticity, as seen for example in the dentate gyrus, and high-expanding association cortices.

Another issue of discussion is the delineation of DMN. It has been suggested to fractionate DMN into an MTL and a dorsal medial prefrontal subsystem including TPJ and lateral temporal cortices, with posterior cingulate and anterior medial prefrontal cortex as hubs (Andrews-Hanna et al. 2010). Following this scheme, the MTL system would be preserved across primates and characterized by high degree of atrophy. The prefrontal would be more expanded, with some regions less (medial prefrontal) and others more (lateral temporal) prone to atrophy. Thus, the DMN is not easily characterized as a unitary system in terms of either expansion or AD atrophy.

The data also demonstrate that aging is the major factor explaining the anatomical distribution of atrophy in AD. Differential degree of cortical expansion, affiliation to DMN, and developmental timing all contribute to explain the regional distribution of AD atrophy to a substantial extent through their influence on aging. Thus, a systems-vulnerability approach to AD, with the aim of trying to understand why certain brain regions and neural networks are vulnerable to different detrimental influences accumulating through life (Jagust 2013), may be a more promising approach than focusing on the effect of the disease in isolation. What is now needed are proper mechanistic models to explain the observed correlations between expansion and vulnerability across aging and AD on a molecular neurobiological level.

Funding

This research was supported by grants from the Norwegian Research Council and the European Research Council to AMF and KBW.

Notes

We are grateful to William Jagust and James Rilling for valuable comments to a previous version of the manuscript. Conflict of Interest: None declared.

References

Addis
DR
Pan
L
Vu
MA
Laiser
N
Schacter
DL
2009
.
Constructive episodic simulation of the future and the past: distinct subsystems of a core brain network mediate imagining and remembering
.
Neuropsychologia
 .
47
:
2222
2238
.
Aimone
JB
Deng
W
Gage
FH
2010
.
Adult neurogenesis: integrating theories and separating functions
.
Trends Cogn Sci
 .
14
:
325
337
.
Andreasen
NC
O'Leary
DS
Cizadlo
T
Arndt
S
Rezai
K
Watkins
GL
Ponto
LL
Hichwa
RD
1995
.
Remembering the past: two facets of episodic memory explored with positron emission tomography
.
Am J Psychiatry
 .
152
:
1576
1585
.
Andrews-Hanna
JR
Reidler
JS
Sepulcre
J
Poulin
R
Buckner
RL
2010
.
Functional-anatomic fractionation of the brain's default network
.
Neuron
 .
65
:
550
562
.
Bardet
J
Essen
DK
Feron
C
Gouat
P
2007
.
Evaluation of the social bond: a new method tested in Mus spicilegus
.
Comptes Rendus Biol
 .
330
:
837
843
.
Barks
SK
Parr
LA
Rilling
JK
2015
.
The default mode network in chimpanzees (Pan troglodytes) is similar to that of humans
.
Cerebral Cortex
 .
25
:
538
544
.
Belcher
AM
Yen
CC
Stepp
H
Gu
H
Lu
H
Yang
Y
Silva
AC
Stein
EA
2013
.
Large-scale brain networks in the awake, truly resting marmoset monkey
.
J Neurosci
 .
33
:
16796
16804
.
Bellassen
V
Igloi
K
de Souza
LC
Dubois
B
Rondi-Reig
L
2012
.
Temporal order memory assessed during spatiotemporal navigation as a behavioral cognitive marker for differential Alzheimer's disease diagnosis
.
J Neurosci
 .
32
:
1942
1952
.
Benavides-Piccione
R
Fernaud-Espinosa
I
Robles
V
Yuste
R
Defelipe
J
2013
.
Age-based comparison of human dendritic spine structure using complete three-dimensional reconstructions
.
Cereb Cortex
 .
23
8
:
1798
1810
.
Bernier
PJ
Bedard
A
Vinet
J
Levesque
M
Parent
A
2002
.
Newly generated neurons in the amygdala and adjoining cortex of adult primates
.
Proc Natl Acad Sci USA
 .
99
:
11464
11469
.
Bhardwaj
RD
Curtis
MA
Spalding
KL
Buchholz
BA
Fink
D
Bjork-Eriksson
T
Nordborg
C
Gage
FH
Druid
H
Eriksson
PS
et al
2006
.
Neocortical neurogenesis in humans is restricted to development
.
Proc Natl Acad Sci USA
 .
103
:
12564
12568
.
Bloss
EB
Janssen
WG
Ohm
DT
Yuk
FJ
Wadsworth
S
Saardi
KM
McEwen
BS
Morrison
JH
2011
.
Evidence for reduced experience-dependent dendritic spine plasticity in the aging prefrontal cortex
.
J Neurosci
 .
31
:
7831
7839
.
Braak
H
Braak
E
1991
.
Neuropathological stageing of Alzheimer-related changes
.
Acta Neuropathologica
 .
82
:
239
259
.
Buckner
RL
2012
.
The serendipitous discovery of the brain's default network
.
NeuroImage
 .
62
:
1137
1145
.
Buckner
RL
Carroll
DC
2007
.
Self-projection and the brain
.
Trends Cognit Sci
 .
11
:
49
57
.
Buckner
RL
Krienen
FM
2013
.
The evolution of distributed association networks in the human brain
.
Trends Cognit Sci
 .
17
12
:
648
665
.
Bufill
E
Blesa
R
Augusti
J
2013
.
Alzheimer's disease: an evolutionary approach
.
J Anthropological Sci
 .
91
:
135
157
.
Bufill
E
Carbonell
E
2004
.
Are symbolic behaviour and neuroplasticity an example of gene-culture coevolution?
Revista de Neurologia
 .
39
:
48
55
.
Chaplin
TA
Yu
HH
Soares
JG
Gattass
R
Rosa
MG
2013
.
A conserved pattern of differential expansion of cortical areas in simian primates
.
J Neurosci
 .
33
:
15120
15125
.
Chen
X
Errangi
B
Li
L
Glasser
MF
Westlye
LT
Fjell
AM
Walhovd
KB
Hu
X
Herndon
JG
Preuss
TM
et al
2013
.
Brain aging in humans, chimpanzees (Pan troglodytes), and rhesus macaques (Macaca mulatta): magnetic resonance imaging studies of macro- and microstructural changes
.
Neurobiol Aging
 .
34
:
2248
2260
.
Clark
RE
Squire
LR
2013
.
Similarity in form and function of the hippocampus in rodents, monkeys, and humans
.
Proc Nat Acad Sci USA
 .
110
(Suppl 2)
:
10365
10370
.
Colgin
LL
Moser
EI
Moser
MB
2008
.
Understanding memory through hippocampal remapping
.
Trends Neurosci
 .
31
:
469
477
.
Corballis
MC
2013
.
Mental time travel: a case for evolutionary continuity
.
Trends Cognit Sci
 .
17
:
5
6
.
Corder
EH
Saunders
AM
Strittmatter
WJ
Schmechel
DE
Gaskell
PC
Small
GW
Roses
AD
Haines
JL
Pericak-Vance
MA
1993
.
Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families
.
Science
 .
261
:
921
923
.
Cotman
CW
Pike
CJ
Cummings
BJ
1993
.
Adaptive versus pathological plasticity. Possible contributions to age-related dementia
.
Advanc Neurol
 .
59
:
35
45
.
Dale
AM
Fischl
B
Sereno
MI
1999
.
Cortical surface-based analysis. I. Segmentation and surface reconstruction
.
Neuroimage
 .
9
:
179
194
.
Drago
V
Babiloni
C
Bartres-Faz
D
Caroli
A
Bosch
B
Hensch
T
Didic
M
Klafki
HW
Pievani
M
Jovicich
J
et al
2011
.
Disease tracking markers for Alzheimer's disease at the prodromal (MCI) stage
.
J Alzheimer's Dis
 .
26
(Suppl 3)
:
159
199
.
Eriksson
PS
Perfilieva
E
Bjork-Eriksson
T
Alborn
AM
Nordborg
C
Peterson
DA
Gage
FH
1998
.
Neurogenesis in the adult human hippocampus
.
Nat Med
 .
4
:
1313
1317
.
Esiri
MM
2007
.
Ageing and the brain
.
J Pathol
 .
211
:
181
187
.
Ewers
M
Frisoni
GB
Teipel
SJ
Grinberg
LT
Amaro
E
Jr
Heinsen
H
Thompson
PM
Hampel
H
2011
.
Staging Alzheimer's disease progression with multimodality neuroimaging
.
Prog Neurobiol
 .
95
:
535
546
.
Finch
CE
Austad
SN
2012
.
Primate aging in the mammalian scheme: the puzzle of extreme variation in brain aging
.
Age
 .
34
:
1075
1091
.
Fischl
B
Sereno
MI
Dale
AM
1999
.
Cortical surface-based analysis. II: inflation, flattening, and a surface-based coordinate system
.
Neuroimage
 .
9
:
195
207
.
Fischl
B
Sereno
MI
Tootell
RB
Dale
AM
1999
.
High-resolution intersubject averaging and a coordinate system for the cortical surface
.
Hum Brain Mapp
 .
8
:
272
284
.
Fjell
AM
McEvoy
L
Holland
D
Dale
AM
Walhovd
KB
Alzheimer's Disease Neuroimaging I
.
2013
.
Brain changes in older adults at very low risk for Alzheimer's disease
.
J Neurosci
 .
33
:
8237
8242
.
Fjell
AM
Walhovd
KB
Westlye
LT
Ostby
Y
Tamnes
CK
Jernigan
TL
Gamst
A
Dale
AM
2010
.
When does brain aging accelerate? Dangers of quadratic fits in cross-sectional studies
.
NeuroImage
 .
50
:
1376
1383
.
Fjell
AM
Westlye
LT
Amlien
I
Tamnes
CK
Grydeland
H
Engvig
A
Espeseth
T
Reinvang
I
Lundervold
AJ
Lundervold
A
et al
2015
.
High-expanding cortical regions in human development and evolution are related to higher intellectual abilities
.
Cerebral Cortex
 .
25
:
26
34
.
Folstein
MF
Folstein
SE
McHugh
PR
1975
.
“Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician
.
J Psychiatr Res
 .
12
:
189
198
.
Freeman
SH
Kandel
R
Cruz
L
Rozkalne
A
Newell
K
Frosch
MP
Hedley-Whyte
ET
Locascio
JJ
Lipsitz
LA
Hyman
BT
2008
.
Preservation of neuronal number despite age-related cortical brain atrophy in elderly subjects without Alzheimer disease
.
J Neuropathol Exp Neurol
 .
67
:
1205
1212
.
Fyhn
M
Molden
S
Witter
MP
Moser
EI
Moser
MB
2004
.
Spatial representation in the entorhinal cortex
.
Science
 .
305
:
1258
1264
.
Gao
W
Gilmore
JH
Shen
D
Smith
JK
Zhu
H
Lin
W
2013
.
The synchronization within and interaction between the default and dorsal attention networks in early infancy
.
Cereb Cortex
 .
23
:
594
603
.
Geschwind
DH
Rakic
P
2013
.
Cortical evolution: judge the brain by its cover
.
Neuron
 .
80
:
633
647
.
Gould
E
Vail
N
Wagers
M
Gross
CG
2001
.
Adult-generated hippocampal and neocortical neurons in macaques have a transient existence
.
Proc Natl Acad Sci USA
 .
98
:
10910
10917
.
Greicius
MD
Srivastava
G
Reiss
AL
Menon
V
2004
.
Default-mode network activity distinguishes Alzheimer's disease from healthy aging: evidence from functional MRI
.
Proc Natl Acad Sci USA
 .
101
:
4637
4642
.
Gusnard
DA
Raichle
ME
Raichle
ME
2001
.
Searching for a baseline: functional imaging and the resting human brain
.
Nat Rev Neurosci
 .
2
:
685
694
.
Hill
J
Inder
T
Neil
J
Dierker
D
Harwell
J
Van Essen
D
2010
.
Similar patterns of cortical expansion during human development and evolution
.
Proc Natl Acad Sci USA
 .
107
:
13135
13140
.
Holland
D
Dale
AM
Alzheimer's Disease Neuroimaging I
.
2011
.
Nonlinear registration of longitudinal images and measurement of change in regions of interest
.
Med Image Anal
 .
15
:
489
497
.
Holland
D
McEvoy
LK
Dale
AM
Alzheimer's Disease Neuroimaging I
.
2012
.
Unbiased comparison of sample size estimates from longitudinal structural measures in ADNI
.
Human Brain Map
 .
33
:
2586
2602
.
Jacobs
B
Driscoll
L
Schall
M
1997
.
Life-span dendritic and spine changes in areas 10 and 18 of human cortex: a quantitative Golgi study
.
J Comp Neurol
 .
386
:
661
680
.
Jagust
W
2013
.
Vulnerable neural systems and the borderland of brain aging and neurodegeneration
.
Neuron
 .
77
:
219
234
.
Jagust
WJ
Mormino
EC
2011
.
Lifespan brain activity, beta-amyloid, and Alzheimer's disease
.
Trends Cognit Sci
 .
15
:
520
526
.
Knickmeyer
RC
Wang
J
Zhu
H
Geng
X
Woolson
S
Hamer
RM
Konneker
T
Lin
W
Styner
M
Gilmore
JH
2014
.
Common variants in psychiatric risk genes predict brain structure at birth
.
Cerebr Cortex
 .
24
:
1230
1246
.
Lazarov
O
Mattson
MP
Peterson
DA
Pimplikar
SW
van Praag
H
2010
.
When neurogenesis encounters aging and disease
.
Trends Neurosci
 .
33
:
569
579
.
Lenroot
RK
Giedd
JN
2011
.
Annual research review: developmental considerations of gene by environment interactions
.
J Child Psychol Psychiatry Allied Disciplines
 .
52
:
429
441
.
Lithfous
S
Dufour
A
Despres
O
2013
.
Spatial navigation in normal aging and the prodromal stage of Alzheimer's disease: insights from imaging and behavioral studies
.
Ageing ResRev
 .
12
:
201
213
.
Lotsch
J
Schaeffeler
E
Mittelbronn
M
Winter
S
Gudziol
V
Schwarzacher
SW
Hummel
T
Doehring
A
Schwab
M
Ultsch
A
Forthcoming
2013
.
Functional genomics suggest neurogenesis in the adult human olfactory bulb
.
Brain Struct Funct
 . .
Lu
H
Zou
Q
Gu
H
Raichle
ME
Stein
EA
Yang
Y
2012
.
Rat brains also have a default mode network
.
Proc Natl Acad Sci USA
 .
109
:
3979
3984
.
Lustig
C
Snyder
AZ
Bhakta
M
O'Brien
KC
McAvoy
M
Raichle
ME
Morris
JC
Buckner
RL
2003
.
Functional deactivations: change with age and dementia of the Alzheimer type
.
Proc Natl Acad Sci USA
 .
100
:
14504
14509
.
Mantini
D
Corbetta
M
Romani
GL
Orban
GA
Vanduffel
W
2013
.
Evolutionarily novel functional networks in the human brain?
J Neurosci
 .
33
:
3259
3275
.
Mantini
D
Gerits
A
Nelissen
K
Durand
JB
Joly
O
Simone
L
Sawamura
H
Wardak
C
Orban
GA
Buckner
RL
et al
2011
.
Default mode of brain function in monkeys
.
J Neurosci
 .
31
:
12954
12962
.
McDonald
CR
McEvoy
LK
Gharapetian
L
Fennema-Notestine
C
Hagler
DJ
Jr
Holland
D
Koyama
A
Brewer
JB
Dale
AM
Alzheimer's Disease Neuroimaging I
.
2009
.
Regional rates of neocortical atrophy from normal aging to early Alzheimer disease
.
Neurology
 .
73
:
457
465
.
Mesulam
MM
1999
.
Neuroplasticity failure in Alzheimer's disease: bridging the gap between plaques and tangles
.
Neuron
 .
24
:
521
529
.
Morris
JC
1993
.
The clinical dementia rating (CDR): current version and scoring rules
.
Neurology
 .
43
:
2412
2414
.
Mu
Y
Gage
FH
2011
.
Adult hippocampal neurogenesis and its role in Alzheimer's disease
.
Mol Neurodegener
 .
6
:
85
.
Neill
D
1995
.
Alzheimer's disease: maladaptive synaptoplasticity hypothesis
.
Neurodegeneration
 .
4
:
217
232
.
Neill
D
2012
.
Should Alzheimer's disease be equated with human brain ageing? A maladaptive interaction between brain evolution and senescence
.
Ageing Res Rev
 .
11
:
104
122
.
Orban
GA
Van Essen
D
Vanduffel
W
2004
.
Comparative mapping of higher visual areas in monkeys and humans
.
Trends Cognit Sci
 .
8
:
315
324
.
Ostby
Y
Walhovd
KB
Tamnes
CK
Grydeland
H
Westlye
LT
Fjell
AM
2012
.
Mental time travel and default-mode network functional connectivity in the developing brain
.
Proc Natl Acad Sci USA
 .
109
:
16800
16804
.
Price
JL
Davis
PB
Morris
JC
White
DL
1991
.
The distribution of tangles, plaques and related immunohistochemical markers in healthy aging and Alzheimer's disease
.
Neurobiol Aging
 .
12
:
295
312
.
Price
JL
Morris
JC
1999
.
Tangles and plaques in nondemented aging and “preclinical” Alzheimer's disease
.
Ann Neurol
 .
45
:
358
368
.
Rakic
P
2009
.
Evolution of the neocortex: a perspective from developmental biology
.
Nat Rev Neurosc
 .
10
:
724
735
.
Rakic
P
2002
.
Neurogenesis in adult primate neocortex: an evaluation of the evidence
.
Nat Rev Neurosc
 .
3
:
65
71
.
Rakic
P
2004
.
Neuroscience: immigration denied
.
Nature
 .
427
:
685
686
.
Rapoport
SI
Nelson
PT
2011
.
Biomarkers and evolution in Alzheimer disease
.
Prog Neurobiol
 .
95
:
510
513
.
Raznahan
A
Greenstein
D
Lee
NR
Clasen
LS
Giedd
JN
2012
.
Prenatal growth in humans and postnatal brain maturation into late adolescence
.
Proc Natl Acad Sci USA
 .
109
:
11366
11371
.
Reisberg
B
Franssen
EH
Souren
LE
Auer
SR
Akram
I
Kenowsky
S
2002
.
Evidence and mechanisms of retrogenesis in Alzheimer's and other dementias: management and treatment import
.
Am J Alzheimer's Dis Other Dementias
 .
17
:
202
212
.
Rilling
JK
Barks
SK
Parr
LA
Preuss
TM
Faber
TL
Pagnoni
G
Bremner
JD
Votaw
JR
2007
.
A comparison of resting-state brain activity in humans and chimpanzees
.
Proc Natl Acad Sci USA
 .
104
:
17146
17151
.
Rosa
MG
Tweedale
R
2005
.
Brain maps, great and small: lessons from comparative studies of primate visual cortical organization
.
Philos Transact RoySoc B
 .
360
:
665
691
.
Schwartz
BL
Evans
S
2001
.
Episodic memory in primates
.
Am J Primatol
 .
55
:
71
85
.
Sforazzini
F
Schwarz
AJ
Galbusera
A
Bifone
A
Gozzi
A
2014
.
Distributed BOLD and CBV-weighted resting-state networks in the mouse brain
.
NeuroImage
 .
87
:
403
415
.
Shaw
LM
Vanderstichele
H
Knapik-Czajka
M
Clark
CM
Aisen
PS
Petersen
RC
Blennow
K
Soares
H
Simon
A
Lewczuk
P
et al
2009
.
Cerebrospinal fluid biomarker signature in Alzheimer's disease neuroimaging initiative subjects
.
Ann Neurol
 .
65
:
403
413
.
Shaw
P
Kabani
NJ
Lerch
JP
Eckstrand
K
Lenroot
R
Gogtay
N
Greenstein
D
Clasen
L
Evans
A
Rapoport
JL
et al
2008
.
Neurodevelopmental trajectories of the human cerebral cortex
.
J Neurosci
 .
28
:
3586
3594
.
Shaw
P
Lerch
JP
Pruessner
JC
Taylor
KN
Rose
AB
Greenstein
D
Clasen
L
Evans
A
Rapoport
JL
Giedd
JN
2007
.
Cortical morphology in children and adolescents with different apolipoprotein E gene polymorphisms: an observational study
.
Lancet Neurol
 .
6
:
494
500
.
Sperling
RA
Laviolette
PS
O'Keefe
K
O'Brien
J
Rentz
DM
Pihlajamaki
M
Marshall
G
Hyman
BT
Selkoe
DJ
Hedden
T
et al
2009
.
Amyloid deposition is associated with impaired default network function in older persons without dementia
.
Neuron
 .
63
:
178
188
.
Stamps
JJ
Bartoshuk
LM
Heilman
KM
2013
.
A brief olfactory test for Alzheimer's disease
.
J Neurol Sci
 .
333
:
19
24
.
Suddendorf
T
2013
.
Mental time travel: continuities and discontinuities
.
Trends Cognit Sci
 .
17
:
151
152
.
Tamnes
CK
Walhovd
KB
Dale
AM
Ostby
Y
Grydeland
H
Richardson
G
Westlye
LT
Roddey
JC
Hagler
DJ
Jr
Due-Tonnessen
P
et al
Alzheimer's Disease Neuroimaging I
.
2013
.
Brain development and aging: overlapping and unique patterns of change
.
NeuroImage
 .
68
:
63
74
.
Tulving
E
1983
.
Elements of episodic memory
 .
New York
:
Oxford University Press
.
Van Essen
DC
Dierker
DL
2007
.
Surface-based and probabilistic atlases of primate cerebral cortex
.
Neuron
 .
56
:
209
225
.
Van Essen
DC
Glasser
MF
Dierker
DL
Harwell
J
2012
.
Cortical parcellations of the macaque monkey analyzed on surface-based atlases
.
Cerebr Cortex
 .
22
:
2227
2240
.
Van Hoesen
GW
Hyman
BT
Damasio
AR
1991
.
Entorhinal cortex pathology in Alzheimer's disease
.
Hippocampus
 .
1
:
1
8
.
Walhovd
KB
Fjell
AM
Brown
TT
Kuperman
JM
Chung
Y
Hagler
DJ
Jr
Roddey
JC
Erhart
M
McCabe
C
Akshoomoff
N
et al
2012
.
Long-term influence of normal variation in neonatal characteristics on human brain development
.
Proc Natl Acad Sci US A
 .
109
:
20089
20094
.
Wang
J
Eslinger
PJ
Doty
RL
Zimmerman
EK
Grunfeld
R
Sun
X
Meadowcroft
MD
Connor
JR
Price
JL
Smith
MB
et al
2010
.
Olfactory deficit detected by fMRI in early Alzheimer's disease
.
Brain Res
 .
1357
:
184
194
.
Yeo
BT
Krienen
FM
Sepulcre
J
Sabuncu
MR
Lashkari
D
Hollinshead
M
Roffman
JL
Smoller
JW
Zollei
L
Polimeni
JR
et al
2011
.
The organization of the human cerebral cortex estimated by intrinsic functional connectivity
.
J Neurophysiol
 .
106
:
1125
1165
.
Yu
HH
Chaplin
TA
Davies
AJ
Verma
R
Rosa
MG
2012
.
A specialized area in limbic cortex for fast analysis of peripheral vision
.
Curr Biol
 .
22
:
1351
1357
.

Author notes

Some of the data used in the preparation of this article were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data, but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf.